Full download ebooks at ebookmeta.

com

Proceedings of the 27th International Conference

on Systems Engineering, ICSEng 2020: 182 (Lecture
Notes in Networks and Systems, 182) Henry Selvaraj
(Editor)
https://ebookmeta.com/product/proceedings-of-the-27th-
international-conference-on-systems-engineering-
icseng-2020-182-lecture-notes-in-networks-and-
systems-182-henry-selvaraj-editor/

OR CLICK BUTTON

DOWLOAD NOW

Download more ebook from https://ebookmeta.com

More products digital (pdf, epub, mobi) instant
download maybe you interests ...

Microelectronics Circuits and Systems Select

Proceedings of 7th International Conference on
Micro2020 Lecture Notes in Electrical Engineering 755

https://ebookmeta.com/product/microelectronics-circuits-and-
systems-select-proceedings-of-7th-international-conference-on-
micro2020-lecture-notes-in-electrical-engineering-755/

Web Information Systems Engineering – WISE 2021: 22nd

International Conference on Web Information Systems
Engineering, WISE 2021, Melbourne, VIC, Australia,
October 26-29, 2021 Proceedings, Part I Wenjie Zhang
https://ebookmeta.com/product/web-information-systems-
engineering-wise-2021-22nd-international-conference-on-web-
information-systems-engineering-wise-2021-melbourne-vic-
australia-october-26-29-2021-proceedings-pa/

Advances in Deep Learning, Artificial Intelligence and

Robotics: Proceedings of the 2nd International
Conference on Deep Learning, Artificial ... 2020
(Lecture Notes in Networks and Systems) Luigi Troiano
(Editor)
https://ebookmeta.com/product/advances-in-deep-learning-
artificial-intelligence-and-robotics-proceedings-of-the-2nd-
international-conference-on-deep-learning-
artificial-2020-lecture-notes-in-networks-and-systems-luigi-tro/

Advances in Networked Based Information Systems The

24th International Conference on Network Based
Information Systems NBiS 2021 Lecture Notes in Networks
and Systems Springer
https://ebookmeta.com/product/advances-in-networked-based-
information-systems-the-24th-international-conference-on-network-
based-information-systems-nbis-2021-lecture-notes-in-networks-
Intelligent Tutoring Systems 18th International
Conference ITS 2022 Bucharest Romania June 29 July 1
2022 Proceedings Scott Crossley

https://ebookmeta.com/product/intelligent-tutoring-systems-18th-
international-conference-its-2022-bucharest-romania-
june-29-july-1-2022-proceedings-scott-crossley/

Intelligent Systems and Applications Proceedings of the

2021 Intelligent Systems Conference IntelliSys Volume 1
294 Lecture Notes in Networks and Systems Kohei Arai
(Editor)
https://ebookmeta.com/product/intelligent-systems-and-
applications-proceedings-of-the-2021-intelligent-systems-
conference-intellisys-volume-1-294-lecture-notes-in-networks-and-
systems-kohei-arai-editor/

Proceedings of the International Conference of

Mechatronics and Cyber MixMechatronics 2018 Lecture
Notes in Networks and Systems Book 48 Gheorghe I
Gheorghe
https://ebookmeta.com/product/proceedings-of-the-international-
conference-of-mechatronics-and-cyber-
mixmechatronics-2018-lecture-notes-in-networks-and-systems-
book-48-gheorghe-i-gheorghe/

Semantic Systems In the Era of Knowledge Graphs 16th

International Conference on Semantic Systems SEMANTiCS
2020 Amsterdam The Netherlands September 7 10 2020
Proceedings Eva Blomqvist
https://ebookmeta.com/product/semantic-systems-in-the-era-of-
knowledge-graphs-16th-international-conference-on-semantic-
systems-semantics-2020-amsterdam-the-netherlands-
september-7-10-2020-proceedings-eva-blomqvist/

Digital Transformation Technology Proceedings of ITAF

2020 Lecture Notes in Networks and Systems 224 Dalia
A. Magdi

https://ebookmeta.com/product/digital-transformation-technology-
proceedings-of-itaf-2020-lecture-notes-in-networks-and-
systems-224-dalia-a-magdi/
Lecture Notes in Networks and Systems 182

Henry Selvaraj
Grzegorz Chmaj
Dawid Zydek Editors

Proceedings of the
27th International
Conference
on Systems
Engineering,
ICSEng 2020
Lecture Notes in Networks and Systems

Volume 182

Series Editor
Janusz Kacprzyk, Systems Research Institute, Polish Academy of Sciences,
Warsaw, Poland

Advisory Editors
Fernando Gomide, Department of Computer Engineering and Automation—DCA,
School of Electrical and Computer Engineering—FEEC, University of Campinas—
UNICAMP, São Paulo, Brazil
Okyay Kaynak, Department of Electrical and Electronic Engineering,
Bogazici University, Istanbul, Turkey
Derong Liu, Department of Electrical and Computer Engineering, University
of Illinois at Chicago, Chicago, USA; Institute of Automation, Chinese Academy
of Sciences, Beijing, China
Witold Pedrycz, Department of Electrical and Computer Engineering,
University of Alberta, Alberta, Canada; Systems Research Institute,
Polish Academy of Sciences, Warsaw, Poland
Marios M. Polycarpou, Department of Electrical and Computer Engineering,
KIOS Research Center for Intelligent Systems and Networks, University of Cyprus,
Nicosia, Cyprus
Imre J. Rudas, Óbuda University, Budapest, Hungary
Jun Wang, Department of Computer Science, City University of Hong Kong,
Kowloon, Hong Kong
The series “Lecture Notes in Networks and Systems” publishes the latest
developments in Networks and Systems—quickly, informally and with high quality.
Original research reported in proceedings and post-proceedings represents the core
of LNNS.
Volumes published in LNNS embrace all aspects and subfields of, as well as new
challenges in, Networks and Systems.
The series contains proceedings and edited volumes in systems and networks,
spanning the areas of Cyber-Physical Systems, Autonomous Systems, Sensor
Networks, Control Systems, Energy Systems, Automotive Systems, Biological
Systems, Vehicular Networking and Connected Vehicles, Aerospace Systems,
Automation, Manufacturing, Smart Grids, Nonlinear Systems, Power Systems,
Robotics, Social Systems, Economic Systems and other. Of particular value to both
the contributors and the readership are the short publication timeframe and the
world-wide distribution and exposure which enable both a wide and rapid
dissemination of research output.
The series covers the theory, applications, and perspectives on the state of the art
and future developments relevant to systems and networks, decision making, control,
complex processes and related areas, as embedded in the fields of interdisciplinary
and applied sciences, engineering, computer science, physics, economics, social, and
life sciences, as well as the paradigms and methodologies behind them.
Indexed by SCOPUS, INSPEC, WTI Frankfurt eG, zbMATH, SCImago.
All books published in the series are submitted for consideration in Web of
Science.

More information about this series at http://www.springer.com/series/15179

Henry Selvaraj Grzegorz Chmaj
• •

Dawid Zydek
Editors

Proceedings of the 27th

International Conference
on Systems Engineering,
ICSEng 2020

123
Editors
Henry Selvaraj Grzegorz Chmaj
Department of Electrical and Computer Department of Electrical and Computer
Engineering Engineering
University of Nevada Las Vegas University of Nevada Las Vegas
Las Vegas, NV, USA Las Vegas, NV, USA

Dawid Zydek
Department of Electrical and Computer
Engineering
University of Nevada Las Vegas
Las Vegas, NV, USA

ISSN 2367-3370 ISSN 2367-3389 (electronic)

Lecture Notes in Networks and Systems
ISBN 978-3-030-65795-6 ISBN 978-3-030-65796-3 (eBook)
https://doi.org/10.1007/978-3-030-65796-3
© Springer Nature Switzerland AG 2021
This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part
of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission
or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar
methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are exempt from
the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this
book are believed to be true and accurate at the date of publication. Neither the publisher nor the
authors or the editors give a warranty, expressed or implied, with respect to the material contained
herein or for any errors or omissions that may have been made. The publisher remains neutral with regard
to jurisdictional claims in published maps and institutional affiliations.

This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Preface

ICSEng 2020

This important series of conferences arose from technical cooperation among

Wrocław University of Science and Technology (Poland), Coventry Polytechnic
(UK), Wright State University (USA) and the University of Nevada, Las Vegas
(USA). Prior to 1980, a series of International Conferences on Systems Science had
been held in Wroclaw, Poland. In 1980, it was decided that the conference would
change venue on alternate years, be held in Coventry, UK, and, at the same time,
change emphasis from science to engineering when hosted in UK. Consequently,
the First and Second International Conferences on Systems Engineering were held
in Coventry, UK, in 1980 and 1982. In 1982, it was decided that the engineering
series of conferences would be held in the USA each third year. The Third and Fifth
International Conferences on Systems Engineering were held in Dayton, Ohio,
USA. The Seventh, Ninth, Eleventh, Thirteenth, Fifteenth, Seventeenth,
Nineteenth, Twenty-First, Twenty-Third and Twenty-Fifth International
Conferences on Systems Engineering were held in Las Vegas, Nevada, USA. In the
year 2018, the twenty-sixth conference was held at the University of Technology
Sydney, Australia. This year 2020, the Twenty-Seventh International Conference
on Systems Engineering is held in Las Vegas, Nevada, USA. Due to unforeseen
circumstances arising out of the COVID-19 outbreak, the conference is held
remotely.
Research in the discipline of systems engineering is an important concept in the
advancement of engineering and information sciences. Systems engineering
attempts to integrate many of the traditional engineering disciplines to solve large
complex functioning engineering systems, dependent on components from all the
disciplines. The research papers contained in these proceedings reflect the state
of the art in systems engineering from all over the world and should serve as vital
references to researchers to follow.
We received 92 submissions. Each paper was reviewed by at least two inde-
pendent reviewers. A total of 44 papers were selected as full papers that gives an

v
vi Preface

acceptance rate of 48%. The University of Nevada, Las Vegas, is very pleased to
host this Twenty-Seventh International Conference on Systems Engineering.
We would like to thank all the authors, reviewers, participants and student
volunteers for making the conference a success. We wish all the participants a
fruitful conference.

Henry Selvaraj
ICSEng 2020 General Chair
Grzegorz Chmaj
Dawid Zydek
ICSEng 2020 Organizing Committee Co-chairs
Conference Organization

ICSEng 2020
General Chair
H. Selvaraj University of Nevada, Las Vegas, USA

Organizing Committee Co-chairs

G. Chmaj University of Nevada, Las Vegas, USA
D. Zydek University of Nevada, Las Vegas, USA

Steering Committee Chair

H. Selvaraj University of Nevada, Las Vegas, USA

Steering Committee
H. Selvaraj University of Nevada, Las Vegas, USA
G. Chmaj University of Nevada, Las Vegas, USA
D. Zydek University of Nevada, Las Vegas, USA
J. Swiatek Wrocław University of Science and Technology,
Poland
P. Swiatek Wrocław University of Science and Technology,
Poland
K. Burnham University of Wolverhampton, UK

Organizing Committee
G. Chmaj University of Nevada, Las Vegas, USA
D. Zydek University of Nevada, Las Vegas, USA
B. Tiwari University of Nevada, Las Vegas, USA
S. Sharma University of Nevada, Las Vegas, USA

vii
viii Conference Organization

Program Committee
ICSEng 2020
R. J. Baker University of Nevada, Las Vegas, USA
G. Borowik Deep Dive Lab, Poland
A. P. F. Carvalho University of Sao Paulo at Sao Carlos, Brazil
Z. Chaczko University of Technology Sydney, Australia
G. Chmaj University of Nevada, Las Vegas, USA
M. Ezra Universiti Tunku Abdul Rahman, Malaysia
T. Fujimoto Toyo University, Japan
L. Gewali University of Nevada, Las Vegas, USA
T. F. Gonzalez University of California at Santa Barbara, USA
T. Gorski Polish Naval Academy, Poland
A. Iyer Virginia Commonwealth University, USA
L. Jozwiak Eindhoven University of Technology,
Netherlands
J. Kacprzyk Polish Academy of Sciences, Poland
A. Kasprzak Wrocław University of Science and Technology,
Poland
A. Khamis Military Technical College, Cairo, Egypt
M. H. Kolekar Indian Institute of Technology Patna, India
L. Koszalka Wrocław University of Science and Technology,
Poland
T. Luba Warsaw University of Technology, Poland
H. Migliore Portland State University, USA
S. Muthaly Asia Pacific University of Technology &
Innovation, Malaysia
I. Pozniak-Koszalka Wrocław University of Science and Technology,
Poland
J. Rozenblit University of Arizona, USA
H. Selvaraj University of Nevada, Las Vegas, USA
D. Selvathi Mepco Schlenk Engineering College, India
S. Singh University of Nevada, Las Vegas, USA
P. Stubberud University of Nevada, Las Vegas, USA
S. Stubberud Boeing Company, USA
M. Thoma University of Hannover, Germany
A. Vadivel SRM AP, Amaravati, India
R. Venkat University of Nevada, Las Vegas, USA
L. Wang Harbin Institute of Technology, China
M. Wozniak Wrocław University of Science and Technology,
Poland
D. Zydek University of Nevada, Las Vegas, USA
Conference Organization ix

Sponsors
Contents

AeroSpace Systems
A Track Registration Association Technique – Feature Analysis . . . . . . 3
Stephen C. Stubberud and Kathleen A. Kramer
Investigation of Wind Energy Impact on Power Systems Stability
Using Lyapunov Exponents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Mojtaba Amiri, Maryam Dehghani, Alireza Khayatian,
Mohsen Mohammadi, Navid Vafamand, and Jalil Boudjadar
Motor Current Signature Analysis Using Shapelet . . . . . . . . . . . . . . . . . 23
Maryam Iraji, Maryam Dehghani, Mohsen Mohammadi, Navid Vafamand,
and Jalil Boudjadar
A Prototype Telepresence Robot for Use in the Investigation of Ebola
and Lassa Virus Threatened Villages in Nigeria . . . . . . . . . . . . . . . . . . 34
S. D. Monk, E. Jorgensen, A. McCulloch, R. Peacock,
I. Sangprachatanaruk, J. Azeta, C. A. Bolu, and C. J. Taylor
Hamiltonian-Based Libration Point Orbit Control on Manifold
of Constant Energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
Keum W. Lee and Sahjendra N. Singh
State and Control Input Limitations in Sliding Mode Control
with System Uncertainties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Marek Jaskuła and Piotr Leśniewski
Pointing Orientations for Manipulator End-Effectors . . . . . . . . . . . . . . . 71
Douglas R. Isenberg
Non-switching Reaching Law Based DSMC Strategies in the Context
of Robustness Comparison . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
Paweł Latosiński and Monika Herkt

xi
xii Contents

Intelligent Systems
Different Approaches to Threshold Estimation for Variation Mode
Decomposition-Based Signal Denoising . . . . . . . . . . . . . . . . . . . . . . . . . . 95
Krzysztof Brzostowski and Jerzy Świątek
Multi-criteria Optimization of Freight Parking Planning Along
a Single Corridor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
Grzegorz Filcek and Maciej Hojda
Synthesis of Index Generation Function Using Linear
and Functional Decomposition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
Grzegorz Borowik, Tadeusz Łuba, Dariusz Wąsicki, and Grzegorz Chmaj
Differential Evolution with Taguchi Crossover Using
Polar Coordinates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
Peter Stubberud
Using Machine Learning Techniques to Estimate the Remaining
Useful Life of a System with Different Types of Datasets . . . . . . . . . . . . 139
Carlos Lemus and Shahram Latifi

Machine Learning and Analytics

Random Generation of Visibility Aware Polygons . . . . . . . . . . . . . . . . . 151
Laxmi Gewali and Bibek Subedi
Assessment of Groundwater Quality and Quantity Using Supervised
Machine Learning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
Ashika Mettu and Shahram Latifi
An Efficient Approach for Selecting QoS-Based Web Service Machine
Learning Models Using Topsis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172
Miguel Angel Quiroz Martinez, Josue Leonardo Moncayo Redin,
Erick David Alvarado Castillo, and Luis Andy Briones Peñafiel
Automated Agronomy: Evaluation of Fruits Ripeness Using Machine
Learning Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
Grzegorz Chmaj, Saugat Sharma, and Henry Selvaraj

Internet of Things
Adaptation-Enabled Architecture for Internet of Things Systems . . . . . 195
Patryk Schauer and Łukasz Falas
Internet of Things Systems as Complex Services . . . . . . . . . . . . . . . . . . 205
Patryk Schauer, Łukasz Falas, and Arkadiusz Warzyński
Communication Services in the Internet of Things . . . . . . . . . . . . . . . . . 213
Patryk Schauer, Łukasz Falas, and Arkadiusz Warzyński
Contents xiii

A Smart Environment Monitoring Application for Mobile Internet

of Things . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223
Samia Tasnim, Ahmon Ferguson, Brooke Gordon, Chad Gordon,
Kishwar Ahmed, and Idongesit Mkpong-Ruffin

Special Session: Applied Media Informatics and Technology

Interaction Design for Email Client: User Movements Reflected in
Gesture-based Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
Ziran Fan and Takayuki Fujimoto
A Support System for Habitual Journaling . . . . . . . . . . . . . . . . . . . . . . 247
Chie Shiraishi and Takayuki Fujimoto
“Ready-Made Computing“ for Cost-Effective Service Performance . . . . 258
Kazuya Murata and Takayuki Fujimoto
Design of Smartphone Software Keyboard for Sensory Experience . . . . 267
Yui Tanaka and Takayuki Fujimoto
Problems with the Corporate Information System for Work Efficiency
Improvement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
Wangjie Xu, Takayuki Fujimoto, and Ziran Fan
Re-experiencing the ‘Process of Perception’: Consideration
of ‘Memory’ Reproduction with Data Mapping Structure
of Brain Cortex . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287
Ning Zhou and Takayuki Fujimoto
Prototyping and Evaluation of Drawing Software to Provide
Analog-Like Feeling by Sound . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297
Yulana Watanabe, Yui Tanaka, and Takayuki Fujimoto
Facial Features in Stylized Images Typically Used for Face
Recognition and Comparison of Impression Between Stylized Images
and AI Images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307
Miki Sunakawa and Takayuki Fujimoto
Examination of Tourism Promotion Using Online Videos . . . . . . . . . . . 314
Shunsuke Aoki and Takayuki Fujimoto
SNS Application with “Peeping Function” to Prevent Information
Leakage on Social Media . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 324
Taishi Nemoto and Takayuki Fujimoto
Proposal and Prototype of Video Playback Method Based
on Walking Speed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 332
Tianyu Zhu, Takayuki Fujimoto, and Ziran Fan
xiv Contents

A Research of the Utilization of an Educational Space

as an Advertising Medium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 342
Kanata Itoh and Takayuki Fujimoto
Examination of Problems and Improvements of Virtual Library
Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 348
Xuezhen Li, Takayuki Fujimoto, and Ziran Fan
Enhanced Sign Changes for SEMG Onset Detection . . . . . . . . . . . . . . . 358
Mateusz Magda, Jerzy Świątek, and Piotr Wyrostek

Special Session: Optimal, Robust, and Adaptive Control Systems

Optimal Energy Resources Scheduling of Hybrid Diesel/Battery Ships
in Shallow Waters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 369
Mohsen Banaei, Jalil Boudjadar, and Mohammad-Hassan Khooban
Linear Quadratic Regulator for Helicopter Model with a Prescribed
Degree of Stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 379
Kaushik Raj, Santosh Kumar Choudhary, and Venkatesan Muthukumar
A Safety-Driven Cost Optimization for the Real-Time Operation
of a Hybrid Energy System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 390
Jalil Boudjadar and Mohammad Hassan Khooban

Special Session: Software Engineering

Context Driven Software Development . . . . . . . . . . . . . . . . . . . . . . . . . . 403
Arunkumar Khannur
A Framework for the Semiotic Quality of User Stories . . . . . . . . . . . . . 413
Pankaj Kamthan and Nazlie Shahmir
On Marketable User Stories for Customer Satisfaction . . . . . . . . . . . . . 423
Pankaj Kamthan and Nazlie Shahmir

Special Session: Engineering of Cyber-Physical Systems

Hierarchical Versioning to Increase Compatibility in Signal-Oriented
Vehicle Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 435
Andreas Vetter and Eric Sax
Towards Data-Driven Digital Twins for Smart Manufacturing . . . . . . . 445
Deena P. Francis, Sanja Lazarova-Molnar, and Nader Mohamed
Contents xv

A KNN Approach for Blockchain Based Electronic Health

Record Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 455
Sazia Parvin, Sonia Farhana Nimmy, Sitalakshmi Venkatraman,
Sa’ed Abed, and Amjad Gawanmeh

Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 465

AeroSpace Systems
 A Track Registration Association Technique –
Feature Analysis

Stephen C. Stubberud1 and Kathleen A. Kramer2(&)

1
General Atomics – EMS, San Diego, CA, USA
scstubberud@ieee.org
2
Department of Electrical Engineering, University of San Diego,
San Diego, CA, USA
kramer@sandiego.edu

Abstract. A recent track-to-track association technique was developed to

associate tracks from multiple sensor platforms where position and heading
errors were incorporated into the reports. When this issue, known as target
registration, is present, standard association routines often result in misassoci-
ations or non-associations between the tracks of the same target. This new
technique develops a track representation that is invariant in translation errors
and insensitive to rotation errors. It has been demonstrated to be effective for
registration problems with reasonable heading errors and any translation errors.
In this work, the features used in the track representation are analyzed for their
effectiveness and sensitivity for various registration error sizes, providing the
foundation for the weighting of the features to enhance registration association.

Keywords: Target registration
Registration association
Target tracking

Translation error
Rotation error

1 Introduction

Target registration occurs when multiple target tracking platforms are reporting their
tracks to other platforms, but bias errors in one or more readings causing the potential
for misassociation. Standard association routines are based on nearest neighbor or the
chi-squared-metric [1, 2].
Figures 1 and 2 illustrate the registration problem. In Fig. 1, two platforms are
tracking the same target. If both platforms knew their local coordinates and headings
exactly, their reported tracks would be as shown. Figure 2 illustrates the difference in
reported tracks if one of them had a bias in its reported position, and an offset in its
heading. Standard association routines in such as case can undergo gate out. A gate out
occurs when the association score exceeds a threshold value that defines a score that
indicates that the no association should be allowed.
In [3], an association approach for registration was presented. The approach was an
extension of the association routine first developed in [4, 5]. This image-correlation
technique for general track association gave the foundation for a track representation
that is invariant when registration occurs. The problem with the technique is that when
uncertainty in the track is present, the correlation falls apart.

© Springer Nature Switzerland AG 2021

H. Selvaraj et al. (Eds.): ICSEng 2020, LNNS 182, pp. 3–11, 2021.
https://doi.org/10.1007/978-3-030-65796-3_1
4 S. C. Stubberud and K. A. Kramer

A recently developed registration-association technique [6] used the idea of from

[3] of track representation that would be invariant under registration. This association
scheme used the chi-squared metric as the foundation. The key difference between this
technique and the standard application is the track vector representation. The core idea
was to develop a registration-invariant representation of tracks. The results indicated
that the technique is invariant for translation but not for rotation. It demonstrated
insensitivity to rotation of the tracks but not a true invariance.

Fig. 1. Two sensor platforms provide two track reports of the same target.

In this paper, the individual features of the vector representing the tracks are
analyzed. The impact on the association score for each feature is analyzed. The impact
is then used to estimate weighting on the feature to determine is the insensitivity over
different regions of rotation can be increased. Section 2 provides a summary overview
the registration-association technique. In Sect. 3, the feature’s effects on the association
score under different rotation conditions are analyzed. This followed in Sect. 4 with
proposed weighting schemes in the association routine.

2 Registration Association Technique

The standard tracking algorithm is the extended Kalman filter (EKF). The track state,
usually referred to as x, is a Gaussian random vector with an associated covariance,
often denoted as P. The standard association is referred to as the chi-squared metric.
For standard track-to-track association, the metric is defined as

v2 ¼ ðxPlatA  xPlatB ÞT ðPPlatA þ PPlatB Þ1
ðxPlatA  xPlatB Þ ð1Þ

Since this is the multiplication of two Gaussian random variables, the metric has a
chi-squared distribution. The vector of errors between the two tracks is weighted by the
 A Track Registration Association Technique – Feature Analysis 5

Fig. 2. Sensor Platform 2 has a registration error in reporting, both translation and rotation.

vector uncertainties. Thus, differences that are associated with large uncertainties are
weighted less while small differences are amplified when associated with small
uncertainties. An analogy would be close stars in astronomy as compared to missed
putts in golf.
The standard vector used in track association is a vector comprised of position,
velocity, and, sometimes, acceleration. The position-velocity vector is defined as

xT ¼ ½ x x_ y y_  ð2Þ

Each track-pair score is then entered into an assignment matrix after being pro-
cessed as

ev
2
ð3Þ

An assignment matrix that provides the maximum total score is produced by

applying a version of the Munkres algorithm [7] upon the full set of track-pair scores.
A new track-vector is defined in Eq. (4) as the distance traveled, the direction travelled,
the change in speed, and the target heading for each time step. This vector represents
the track differently than Eq. (2) which would be impacted significantly by translation
and rotation errors.
2  2 1 3
2~ 3 ~xn þ ~y2n 2
dn 6      7
6 7
6 h~ 7 6 tan1 ~~yxnn  tan1 ~~xynn 1 7
6 n7 6 1 7
6 7¼6  2 1 7 ð4Þ
4 ~sn 5 6 ~ ~
x_ n þ y_ n
2 2 7
6 7
u~n 4     5
tan1 yx__ nn
6 S. C. Stubberud and K. A. Kramer

where
2 3 2 3
xen xn  xn1
6 ye 7 6 y  y 7
6 n7 6 n n1 7
6e 7¼6 7: ð5Þ
4 x_ n 5 4 x_ n  x_ n1 5
y_en y_ n  y_ n1

It is important to note that the fourth element in Eq. (4) is only the current heading
information provided by the track state. Unlike the first (distance) element and third
(speed change) element, which were both based upon current and previous state ele-
ments, the heading angle change proved too noisy to be effective, even with smoothing
over time, as used in the second element for the position angle.
The chi-squared metric to determine the association score is from Sensor Plat-
form A to Sensor Platform B therefore generated as.
 T  1  
v2Reg ¼ V Track pltA  V Track pltB HPpltA H T þ HPpltB H T V Track pltA  V Track pltB ;
ð6Þ

where
3 02 2~ 3 1
d~n dn
B ~ 7 C
  B6 6 hn 7
6~
6 hn 7
7 C
V Trackplt A B
 V Trackplt B ¼ B6 7 6 7 C ð7Þ
4 5 4 5 C
@ ~sn ~sn A
u~ n Trackplt A ~ n Trackplt B
u

and the matrix P is the associated standard track four-state position and velocity
covariance matrix. The Jacobian is defined in Eq. (8) as.
2 @hd @hh @hs @hu 3 2 @hd @hh
3
@x @x @x @x @x @x
6 @hd @hu 7 6 @hd @hh
0 7
6 @hh @hs 7 6 7
6 @y @y @y @y 7 6 @y @y 7
H ¼6 7¼6 @hu 7
6 @hd @hh @hs @hu 7 6 @hs 7
4 @ x_ @ x_ @ x_ @ x_ 5 4 0
@ x_ @ x_ 5
@hd @hh @hs @hu @hs @hu
@ y_ @ y_ @ y_ @ y_ @ y_ @ y_

with
@hd @hs 2x_
@x ¼ 2x
1 @x ¼ 1
ðx2 þ y2 Þ2 ðx_ 2 þ y_ 2 Þ2
@hd @hs
@y ¼ 2y
1 @y ¼ 2_y
1
ðx2 þ y2 Þ2 ðx_ 2 þ y_ 2 Þ2 ð8Þ
@hh y @hu _y
¼
@x ðx2 þ y2 Þ @x ðx_ þ y_ 2 Þ
2

@hh @hu x_
¼ ðx2 þx y2 Þ ¼
@y @y ðx_ 2 þ y_ 2 Þ
 A Track Registration Association Technique – Feature Analysis 7

Since a single timestep may not be reliable, a small window of track points is used
to create an association score
Xwinsize 1
regassocscore ¼ k¼0
v2Reg ðn  kÞ: ð9Þ

The implementation of the metric in the assignment is scaled so that the distances
do not become zero in the exponent and cause computational zeros to be formed. This
causes the order of the matrix to have an effect on the selection. The scale factor is in
the exponent is shown in Eq. (10) where the subscript indicates a two-norm.

kHPpltA H T þ HPpltB H T k2 ð10Þ

3 Feature Analysis

Each of the elements of the feature vector in Eq. (4) provides a portion of the score.
Each has an impact on the score which varies based on the rotation. To analyze this
impact, each feature will be applied alone to the association problem for a variety of
rotation angles. The measure to be used is the correct association.
A three-track scenario, as defined in [5], is used. As seen in Fig. 3, there are two
sensor platforms, each with range-bearing sensors. Platform 1 is located at 300 m north
of the origin point and 100 m east of the origin point. Its sensor has a range accuracy of
0.1 m and a bearing accuracy of 0.0975°. Platform 2 is located at 0m north of the origin
and -100 m east of the origin. Its sensor’s range accuracy is 0.05 m, and its bearing
accuracy is 0.3°. Target 1 performs an S-curve maneuver over the scenario. The second
target flies a straight-line trajectory with constant speed at the beginning of the scenario
followed by a constant acceleration. Target 3 decelerates over a straight-line trajectory.
The window size was selected to be 5.

Fig. 3. A three-target and two-sensor scenario

8 S. C. Stubberud and K. A. Kramer

The performance of each feature is measured for a variety of rotations. These

rotations are 0°, 2°, 5°, 10°, 15°, 20°, 45°, 60°, and 90°.
In Tables 1, 2, 3 and 4, the performance was measured for each feature for the
aforementioned rotation angles. The results were broken down into four different
elements: Correct Association, Missed Associations – Straight Line Target to Straight
Line Target, Missed Associations – Straight Line Target to S-curve Target, All
Associations Wrong.

Table 1. Distance travel feature association performance

Sensor bias Track associations: 30
Angle (deg.) Mis-assoc: Mis-assoc: Mis-assoc: Mis-assoc: All
None (All correct) Straight-to-Straight Straight-to- S-Curve
0 18 0 12 0
2 18 0 12 0
5 22 0 8 0
10 30 0 0 0
15 30 0 0 0
20 29 0 1 0
45 30 0 0 0
60 26 0 4 0
90 18 0 12 0

Table 2. Direction change feature association performance

Sensor bias Track associations: 30
Angle (deg.) Mis-assoc: Mis-assoc: Mis-assoc: Mis-assoc: All
None (All correct) Straight-to-Straight Straight-to- S-Curve
0 2 9 8 11
2 1 9 9 11
5 2 8 15 5
10 12 2 13 3
15 14 2 11 3
20 19 0 8 3
45 26 0 2 2
60 29 0 0 1
90 2 10 7 11
 A Track Registration Association Technique – Feature Analysis 9

Table 3. Speed change feature association performance

Sensor bias Track associations: 30
Angle (deg.) Mis-assoc: Mis-assoc: Mis-assoc: Mis-assoc: All
None (All correct) Straight-to-Straight Straight-to- S-Curve
0 25 0 5 0
2 30 0 0 0
5 30 0 0 0
10 30 0 0 0
15 30 0 0 0
20 30 0 0 0
45 28 0 2 0
60 28 0 2 0
90 25 0 5 0

Table 4. Heading feature association performance

Sensor bias Track associations: 30
Angle (deg.) Mis-assoc: Mis-assoc: Mis-assoc: Mis-assoc: All
None (All correct) Straight-to-Straight Straight-to- S-Curve
0 20 9 1 0
2 6 23 1 0
5 7 22 1 0
10 19 10 1 0
15 26 2 2 0
20 26 1 2 1
45 22 1 2 5
60 20 1 8 1
90 4 1 2 23

As can be seen, the most effective single features are the distance-traveled feature,
d, and the speed-change feature, s. These features would provide 81.9% and 94.8%
correct associations on average. The least effective as a solo feature is direction change,
h, at 39.6% average correct associations.

4 Feature Weighting Results

A variety of weighting values for the chi-squared vector features were compared. Based
on the analysis of Sect. 3, Features 1 and 3 were considered exceedingly useful. They
were each given equal weights for all of the comparisons. The direction and heading
features, Features 2 and 4, were varied from values between 0 and 1. Table 5 lists the
correct associations for each angle and an average percentage of correct associations.
10 S. C. Stubberud and K. A. Kramer

As the results show, the weighting can have a small improvement in performance.
If feature weighting were to be used in implementation, a weighting with Features 1
and 3 equally weighted and Feature 2 weighted less than Features 1 and 3 but more
than Feature 4 would be best. All features should be used. The analysis shows that the
technique is effective up to 20 degrees of rotation in most cases and the outrageous
heading error of 90 degrees has deleterious effects. If the 90-degree heading error is not
considered, the top four performing weighting schemes increase to between 95% to
98.75% accuracy.

Table 5. Correct associations based on feature weighting

Angle (deg.) Feature weights (4)
1 1 1 1 1 1 1
1 0 0 0.5 0.5 0.25 0.5
1 1 1 1 1 1 1
1 0 0.5 0.5 0 0.5 0.25
0 29 21 30 30 20 30 30
2 29 19 30 30 21 30 30
5 28 24 29 30 26 30 30
10 29 30 29 30 27 30 30
15 29 30 29 30 29 29 30
20 30 30 26 28 28 28 29
45 26 30 25 29 30 27 30
60 28 25 22 25 28 24 28
90 7 21 7 6 20 7 8
Avg. 87.0 85.2 84.1 88.2 84.8 87.0 90.7

5 Conclusion

A new target registration-association scheme was analyzed to determine the effec-

tiveness of the features selected. The analysis showed that individual features of
position change and speed acceleration were the most impactful. When combined in
different weightings, the least impactful individual feature was quite useful. Weighting
of the features used in the association metric can have an impact. This impact appears
to be noticeable but relatively small, approximately 3% absolute improvement.
All features, the distance travelled, the direction change, the speed change, and the
heading angle, should be used in the association score. Further research to improve the
representation where the features both translation and rotation are invariant and
improved weighting techniques are being considered.
 A Track Registration Association Technique – Feature Analysis 11

References
1. Blackman, S., Popoli, R.: Design and Analysis of Modern Tracking System. Artech House,
Boston (1999)
2. Bar-Shalom, Y., Blair, W.D. (eds.): Multitarget - Multisensor Tracking: Applications and
Advances, vol. III. Artech House, Boston (2000)
3. Stubberud, S.C.: Track registration using image correlation. In: Proceedings of the 2015
Aerospace Conference, Big Sky, MT, pp. 1–7 (2015)
4. Stanek, C.J., Javidi, B., Yanni, P.: Performance assessment of frequency plane filters applied
to track association and sensor registration. Proc. SPIE 5094, 323–328 (2003)
5. Stanek, C.J.: Gridlocking and correlation methods and arrangements. U.S. Patent
US680399B2 (2004)
6. Stubberud, S.C., Kramer, K.A.: A target registration association technique. In: Submitted to
the IEEE Transactions of Aerospace Electronic Systems (2020)
7. Giernacki, W.: Robust control with optimization of robustness index. In: 2012 IEEE
International Conference on Systems, Man, and Cybernetics (SMC), Seoul, South Korea,
pp. 2389–2394 (2012)
 Investigation of Wind Energy Impact on Power
Systems Stability Using Lyapunov Exponents

Mojtaba Amiri1, Maryam Dehghani1, Alireza Khayatian1,

Mohsen Mohammadi2, Navid Vafamand1(&), and Jalil Boudjadar3
1
School of Electrical and Computer Engineering, Shiraz University, Shiraz, Iran
n.vafamand@shirazu.ac.ir, navidvafamand@gmail.com
2
School of Mechanical Engineering, Shiraz University, Shiraz, Iran
3
Department of Engineering, Aarhus University, Aarhus, Denmark

Abstract. The world tendency is to replace all generators that use fossil fuels
by renewable energy. Although this is a very important step toward keeping the
environment clean, this paper shows that by replacing wind energy by the
conventional generators, the relative stability becomes worse. This does not
mean that the paper is against such replacement, but makes an alert to be aware
of such important issue and to implement appropriate controllers for increasing
the relative stability of the power systems equipped with renewable energy
sources. In this work, a data-driven approach is used to assess the stability of
power grids by just using voltage time-series data. Lyapunov exponent
(LE) index is presented as an effective tool to analyze complex systems stability.
If the maximum LE (MLE) is positive (negative), then the system is unstable
(stable). In the present paper, the MLE is computed for a finite-time interval for
the sample system of the New England 39-Bus system. In the end, the effect of
renewable wind energy on the stability of the system has been investigated and
by substituting the wind farm in different areas of the system, the impact of
adding a wind farm to the system in different locations is found.

Keywords: Lyapunov exponent
Power system
Stability assessment
Wind

energy

1 Introduction

Evaluating the complex power system stability is a critical issue which can be done
through various approaches [1]. The model equations of power systems are highly
nonlinear and these systems are high order systems. There are many approaches to
assess the stability of systems. Many of these methods use the system model. On the
other hand, the nonlinearity of the power systems, as well as the high order of the
systems, is itself a strong reason for the high computational burden which requires
expensive hardware and increases cost. Therefore, the methods that do not use model
equations to analyze stability can be very effective. Because of the importance and
complexity of these systems, the online and model-free voltage stability assessment is
very important.

© Springer Nature Switzerland AG 2021

H. Selvaraj et al. (Eds.): ICSEng 2020, LNNS 182, pp. 12–22, 2021.
https://doi.org/10.1007/978-3-030-65796-3_2
 Investigation of Wind Energy Impact on Power Systems Stability 13

In [2], Liu et al. used LE to evaluate the behavior of power systems for the first
time. By using measurement data, they have measured the transient chaos of the power
system with LE. That method is applied on a three-machine system. In [3], an appli-
cable LE-based approach is suggested for the online stability evaluation of rotor angle
in the presence of perturbations. The method presented in [3] is an extension of the
results achieved by Lui et al. [2]. In [3], the relation of the MLE with the rotor angle
stability in power systems is studied. The effectiveness of that method for the 200-Bus
system has been investigated. In [4], a data-driven approach is developed to assess the
voltage stability in power systems. The algorithm of [4] is an extension of the method
proposed by Rosenstein et al. [5], in which, the LE is used as an indicator for deter-
mining stability. The authors of [6] proposed an algorithm to evaluate the rotor angle
stability of a power system. That approach is not a model-based approach and can use
phasor measurement unit (PMU) fuzzy measurement data. In that algorithm, the LE
measures the degree of convergence or divergence of the rotor angle trajectories. In [7],
a robust stability analysis is performed for a power system by the means of the LE to
analyze the stability of the voltage and rotor angle. The author of [7] calls that method
as “THRUST”. Contrary to references [8, 9] that have been used only for small sys-
tems, the THRUST has been also utilized to evaluate the stability of a small system
(39-Bus) and a very large system (13029-Bus). In [10], Wei et al. proposed a data-
based approach using LE theory to analyze rotor angle stability. The MLE is obtained
by a method based on the recursive least squares from the measured data of the rotor
angles. The results of that study were applied to the Northeast Power Coordinating
Council 140-Bus and the New England 39-Bus systems and have yielded good results.
In [11], Ge et al. presented an improved method of voltage stability assessment. The
effect of voltage fluctuations on the computation procedure of the LE is studied and a
phase correction method is presented to compensate for the negative effects of voltage
fluctuations. Simulations have been carried out on the Guangdong Province Power
Network of China.
In this research, the LE-based method is used for stability assessment. LE is
introduced by Lyapunov [12] and is a powerful and efficient method to analyze the
stability of a power system [13]. LE is calculated by two methods: I) model-based, and
II) model-free (data-driven) [14, 15]. This paper uses a model-free (data-driven) and
online approach of computation of the MLE to analyze the stability of the power
system using only voltage time-series data. If the MLE value is positive, the system is
unstable and if the LE gets a negative value, the system is stable. In other words, this
paper is an extension of [16–19] in assessing system stability using online data from
system, such as PMU data. Moreover, the use of renewable energies power plants
instead of fossil fuel power plants to reduce pollution and costs is increasing. Replacing
a synchronous generator with a renewable energy system reduces the relative stability
of the power system [20–23]. In the present paper, the effect of replacing a wind farm
with a synchronous generator on the stability of a New England 39-Bus system is
investigated. The effect of replacing a wind farm with all of the synchronous generators
of the system is investigated by using an online model-free approach of MLE com-
putation for several different cases. Power system simulations are done by DIgSILENT.
The paper is organized as follows. The low-inertia systems are represented in
Sect. 2. Section 3 represents the Date-Driven-based stability assessment approach.
14 M. Amiri et al.

In Sect. 4, simulation results on the New England 39-Bus system are presented to
assess the effect of substituting a wind farm with a specific synchronous generator of
the system. Finally, the conclusions are in Sect. 5.

2 Low-Inertia Systems

The main purpose of synchronous generators is to generate active power in order to

adjust the frequency and voltage. The inertia of the synchronous generators affects the
dynamic behavior of the system shortly after fault or disturbance in the power balance.
Research has shown that low-inertia generators such as wind turbines that can be used
in a power system instead of synchronous generators, cause the system to be unable to
maintain voltage and frequency of the system in the desired value [20]. Here, a system
with low inertia is examined shortly after a fault occurs. The equation of a conventional
system with both synchronous and non-synchronous generators is as follows [20]:

M x_ ðtÞ ¼ Ps ðtÞ þ Pns ðtÞ  Pl ðtÞ  Pj ðtÞ; ð1Þ

where M shows the total inertia of the system, xðtÞ is average frequency, Pns and Ps
show the active power of non-synchronous and synchronous generators, respectively.
Pl and Pj show the loads and losses, respectively. Bigger values of M increase the
system’s control ability to and robustness against load fluctuations and faults. By
increasing the number of non-synchronous generators, the value of M is reduced,
causing an increase in the frequency fluctuations and even system failure to maintain
the voltage at a certain value. In order investigate this effect, a data-driven tool is used.

3 Data-Driven-Based Stability Assessment

Lyapunov introduced two methods to analyze the systems stability: I) LE-based

approach, II) Lyapunov function-based approach [12]. The second method is very
common, but it is a model-based approach and for stability analysis should have an
accurate model of the system. Additionally, in the second method, finding a proper
Lyapunov function is very challenging. On the other hand, the LE-based method can be
used in both model-based and model-free approaches. In this paper, a model-free
approach is used to assess the stability of the power system in an online procedure. LEs
can convey information about the system relative stability. The largest LE is known as
MLE [24]. The sign of MLE is used for stability investigation. Negative MLE refers to
a stable system, while positive MLE shows instability.
The LE generally shows the rate of separation and closeness of two trajectories
which are initially so close to each other. Two trajectories which are starting from very
close initial conditions may exponentially diverge or may converge to a finite set, such
as a fixed point, a limit cycle, a periodic or quasi-periodic motion, or a chaotic attractor.
Briefly, if the value of k is positive, it indicates that the two trajectories are separated
and if the value of k is negative, two trajectories will become close to each other. This
is called chaotic if the initial conditions are perturbed and the system becomes
 Investigation of Wind Energy Impact on Power Systems Stability 15

unpredictable. Attractors of a system with one or more positive LEs are called strange
or chaotic. The LE is highly dependent on the initial conditions of the system.
The pioneer numerical algorithm to compute the MLE of the system from time
series was introduced in [5], which was unable to compute the MLE of the big data and
also this algorithm is not robust to delay time changes. System information can be also
obtained from system state-space equations. In the words, it is not enough to only use
set of time-series data or simply a measurement variable. To solve this problem, the
measurement variables should be increased [25] by embedding data in the m-
dimensional delay coordinates. Suppose voltage time series data is available in the form
fv1 ; v2 ; . . .; vk g, the delay coordinates are as follows:


V i ¼ vi ; vi þ s ; . . .; vi þ ðm1Þs ! V ¼ ðV 1 ; V 2 ; . . .; V M ÞT ; ð2Þ

in which i ¼ 1; . . .; M, M ¼ k  ðm  1Þs, m is the embedding dimension and s is the

reconstruction delay time. If the number of data is infinite and the data are not noisy,
there is no obstacle in choosing any value of s, but the data are often noisy and finite
[26]. The method of calculating m is well described in [27], and here the value of m is
chosen equal to one.
The MLE computation algorithm of this paper is adapted from [4]. MLE is cal-
culated from the time-series data of the grid voltages. In this paper the MLE is shown
by k. Data-driven MLE computation algorithm is
provided in Fig. 1, where kðkDtÞ
computes the MLE at time kDt. Vt ¼ v1t ; v2t ; . . .; vnt in (2) displays the time-series data
of the n bus voltages of the system. If the value of the MLE in (2) becomes positive
(negative), the system is unstable (stable).

Fig. 1. Data-driven MLE computation algorithm

16 M. Amiri et al.

4 Simulation Results

The New England 39-Bus system is considered as the case study and simulated in
DIgSILENT for different cases. First of all, stability analysis of this system is inves-
tigated for six cases. Then, by using MLE, the impact of wind energy on the system
stability is investigated for three cases. The New England 39-Bus System has 10
generators, 12 transformers, 39 buses, and 34 lines. This system is a reduced order of
the transmission system in the USA [28].
By using the data-driven approach presented in Fig. 1 to compute the MLE, the
stability is analyzed in six cases which are mentioned below. The MLE is computed
only using bus voltage time-series data.
Case 1)
A short-circuit at 16th bus is simulated and it is cleared after 180 ms.
Case 2)
A short-circuit at 16th bus is simulated and it is cleared after 200 ms.
Case 3)
A short-circuit at 31th bus is simulated and it is cleared after 210 ms.
Case 4)
A short-circuit at 16th bus is simulated and it is cleared after 230 ms.
Case 5)
A short-circuit at the beginning of Line 2–3 is simulated and it is cleared
after 230 ms.
Case 6) A short-circuit at the beginning of Line 2–3 is simulated and it is cleared
after 250 ms.
The 39 bus voltage trajectories in case 1 are shown in Fig. 2 for 10 s. The MLE
evolution of the system in case 1 is shown in Fig. 3. The value of the MLE in this case
became negative and we can realize that the system is stable. It should be mentioned
that the voltage magnitudes are shown in p. u.
The 39 bus voltage trajectories in case 2 are shown in Fig. 4 for 5 s. Figure 5
shows the MLE evolution of the system in case 2. The value of the MLE, in this case,
became positive and one can realize that the system is unstable.

Fig. 2. 39 bus voltage magnitudes in p.u. in case 1

 Investigation of Wind Energy Impact on Power Systems Stability 17

Fig. 3. The evolution of the MLE of the system in case 1

Fig. 4. 39 bus voltage magnitudes in p. u. in case 2

Fig. 5. The evolution of the MLE of the system in case 2

18 M. Amiri et al.

The 39 bus voltage trajectories in case 3 are given in Fig. 6 for 5 s. Figure 7
illustrates the MLE evolution of the system for the case 3. As can be seen in Fig. 7, the
value of the MLE, in this case, became negative and one can realize that the system is
stable. The simulation results of the other cases for 5 s are shown in Table 1.

Fig. 6. 39 bus voltage magnitudes in p. u. in case 3

Fig. 7. The evolution of the MLE of the system in case 3

Table 1. Simulation results of six cases

Cases The MLE values after 5 s Stability
Case 1 −0.3144 Stable
Case 2 +0.6408 Unstable
Case 3 −0.3316 Stable
Case 4 +0.4984 Unstable
Case 5 −0.6119 Stable
Case 6 +0.5918 Unstable
 Investigation of Wind Energy Impact on Power Systems Stability 19

For an investigation of wind energy impact on the power system stability, each of
the synchronous generators of the 39-Bus system is replaced by a wind farm,
respectively, and the impact of this substitution is investigated in three cases. It should
be mentioned that the New England 39-Bus system is stable in these three cases when
the substitution is not done. By calculating the MLE of the system using voltage time-
series data, we can analyze the stability of the system in each case. The three cases are
as follows:
Case 1) A short-circuit at 16th Bus is simulated and is cleared after 180 ms.
Case 2) A short-circuit at 31th Bus is simulated and is cleared after 210 ms.
Case 3) A short-circuit at the beginning of “Line 02–03” is simulated and by
switching off the faulted line, this fault is cleared after 230 ms.
Wind farms are connected to the New England 39-Bus system in parallel with
synchronous generators as given in Fig. 8. Each of the synchronous generators is
replaced with a wind farm specified by the active power generation capacity same as
the original generator and the impact of this substitution is investigated by computation
of the MLE of the system using voltage time-series data in 5 s for 3 mentioned cases.
The results of this substitution of the wind farm with synchronous generators are shown
in Table 2. The table shows values of the MLE after 5 s.

Fig. 8. Parallel connection of wind farm with a synchronous generator to the system
20 M. Amiri et al.

Table 2. The MLE values of the system in different cases after 5 s

MLE
Case 1 Case 2 Case 3
Without DFIG −0.3144 −0.3316 −0.6119
Replacing with G1 +0.6591 +0.6971 +0.7543
Replacing with G2 −0.1594 −0.1654 −0.1691
Replacing with G3 +0.5441 +0.3658 +0.3702
Replacing with G4 +0.6927 +0.4920 +0.4721
Replacing with G5 +0.6556 +0.4469 +0.3719
Replacing with G6 +0.7842 +0.4183 +0.3702
Replacing with G7 +0.1241 +0.5167 +0.5016
Replacing with G8 +0.6477 +0.5318 +0.5549
Replacing with G9 −0.2114 −0.2773 −0.3160
Replacing with G10 +0.4785 +0.6119 +1.0570

The MLE values in the table show that with the displacement of the wind farm in
areas of the system, in most cases, the system becomes unstable and, in some cases, the
relative stability is decreased.
As shown in Table 2, by substitution of 39 buses system synchronous generators in
most cases, the MLE became positive and in two cases reduced relative system sta-
bility. According to this table, the Lyapunov exponent theory can be utilized as a tool
for locating the wind farm.

5 Conclusion

In this paper, first, the low-inertia systems in introduced. Then, the Lyapunov exponent
theorem is introduced and by using the model-free method of computation of the MLE,
the stability of the system is analyzed. It is observed that this method can assess the
stability of power systems in an online manner with low computational burden. Then,
by replacing the wind farm with each of the synchronous generators of a power system,
system stability is evaluated. It is found that these substitutions may reduce relative
stability or make the system unstable. This study proves the importance of controller
design strategies when the renewable energy sources are added to the power system.

References
1. Liu, M., Bizzarri, F., Brambilla, A.M., Milano, F.: On the impact of the dead-band of power
system stabilizers and frequency regulation on power system stability. IEEE Trans. Power
Syst. 34, 3977–3979 (2019)
2. Liu, C.-W., Thorp, J.S., Lu, J., Thomas, R.J., Chiang, H.-D.: Detection of transiently chaotic
swings in power systems using real-time phasor measurements. IEEE Trans. Power Syst. 9
(3), 1285–1292 (1994)
 Investigation of Wind Energy Impact on Power Systems Stability 21

3. Yan, J., Liu, C.-C., Vaidya, U.: PMU-based monitoring of rotor angle dynamics. IEEE
Trans. Power Syst. 26(4), 2125–2133 (2011)
4. Dasgupta, S., Paramasivam, M., Vaidya, U., Ajjarapu, V.: Real-time monitoring of short-
term voltage stability using PMU data. IEEE Trans. Power Syst. 28(4), 3702–3711 (2013)
5. Rosenstein, M.T., Collins, J.J., De Luca, C.J.: A practical method for calculating largest
Lyapunov exponents from small data sets. Physica D 65(1–2), 117–134 (1993)
6. Dasgupta, S., Paramasivam, M., Vaidya, U., Ajjarapu, V.: PMU-based model-free approach
for real-time rotor angle monitoring. IEEE Trans. Power Syst. 30(5), 2818–2819 (2014)
7. Khaitan, S.K.: THRUST: a Lyapunov exponents based robust stability analysis method for
power systems. In: 2017 North American Power Symposium (NAPS), pp. 1–6. IEEE (2017)
8. Verdejo, H., Vargas, L., Kliemann, W.: Stability of linear stochastic systems via Lyapunov
exponents and applications to power systems. Appl. Math. Comput. 218(22), 11021–11032
(2012)
9. Dasgupta, S., Paramasivam, M., Vaidya, U., Ajjarapu, V.: PMU-based model-free approach
for short term voltage stability monitoring. In: 2012 IEEE Power and Energy Society
General Meeting, pp. 1–8. IEEE (2012)
10. Wei, S., Yang, M., Qi, J., Wang, J., Ma, S., Han, X.: Model-free MLE estimation for online
rotor angle stability assessment with PMU data. IEEE Trans. Power Syst. 33(3), 2463–2476
(2017)
11. Ge, H., et al.: An improved real-time short-term voltage stability monitoring method based
on phase rectification. IEEE Trans. Power Syst. 33(1), 1068–1070 (2017)
12. Lyapunov, A.M.: The general problem of the stability of motion. Int. J. Control 55(3), 531–
534 (1992)
13. Amiri, M., Dehghani, M., Khayatian, A., Mohammadi, M.: Lyapunov exponent based
stability assessment of power systems. In: 2019 6th International Conference on Control,
Instrumentation and Automation (ICCIA), pp. 1–5. IEEE (2019)
14. Khodadadi, H., Khaki-Sedigh, A., Ataei, M., Jahed-Motlagh, M.R.: Applying a modified
version of Lyapunov exponent for cancer diagnosis in biomedical images: the case of breast
mammograms. Multidimensional Syst. Signal Process. 29(1), 19–33 (2016)
15. Pikovsky, A., Politi, A.: Lyapunov exponents: a tool to explore complex dynamics.
Cambridge University Press, New York (2016)
16. Dehghani, M., Shayanfard, B., Khayatian, A.R.: PMU ranking based on singular value
decomposition of dynamic stability matrix. IEEE Trans. Power Syst. 28(3), 2263–2270
(2013)
17. Shayanfard, B., Dehghani, M., Khayatian, A.: Optimal PMU placement for full observability
and dynamic stability assessment. In: 2011 19th Iranian Conference on Electrical
Engineering, pp. 1–6. IEEE (2011)
18. Mohammadi, H., Khademi, G., Dehghani, M., Simon, D.: Voltage stability assessment using
multi-objective biogeography-based subset selection. Int. J. Electric. Power Energy Syst.
103, 525–536 (2018)
19. Mohammadi, H., Khademi, G., Simon, D., Dehghani, M.: Multi-objective optimization of
decision trees for power system voltage security assessment. In: 2016 Annual IEEE Systems
Conference (SysCon), pp. 1–6. IEEE (2016)
20. Milano, F., Dörfler, F., Hug, G., Hill, D.J., Verbič, G.: Foundations and challenges of low-
inertia systems. In: 2018 Power Systems Computation Conference (PSCC), pp. 1–25. IEEE
(2018)
21. Toor, A., et al.: Energy and performance aware fog computing: a case of DVFS and green
renewable energy. Future Gener. Comput. Syst. 101, 1112–1121 (2019)
22. Khooban, M.H., Vafamand, N., Boudjadar, J.: Tracking control for hydrogen fuel cell
systems in zero-emission ferry ships. Complexity 2019 (2019)
22 M. Amiri et al.

23. Vafamand, N., Khooban, M.H., Dragičević, T., Boudjadar, J., Asemani, M.H.: Time-delayed
stabilizing secondary load frequency control of shipboard microgrids. IEEE Syst. J. 13(3),
3233–3241 (2019)
24. Best, E.A.: Stability Assessment of Nonlinear Systems Using the Lyapunov Exponent. Ohio
University, Athens (2003)
25. Takens, F.: Detecting strange attractors in turbulence. In: Dynamical Systems and
Turbulence, Warwick 1980, pp. 366–381. Springer (1981)
26. Fraser, A.M., Swinney, H.L.: Independent coordinates for strange attractors from mutual
information. Phy. Rev. A 33(2), 1134 (1986)
27. Kennel, M.B., Brown, R., Abarbanel, H.D.: Determining embedding dimension for phase-
space reconstruction using a geometrical construction. Phys. Rev. A 45(6), 3403 (1992)
28. Pai, M.: Energy Function Analysis for Power System Stability. Springer, New York (2012)
 Motor Current Signature Analysis Using
Shapelet

Maryam Iraji1, Maryam Dehghani1, Mohsen Mohammadi2,

Navid Vafamand1(&), and Jalil Boudjadar3
1
Electrical and Computer Engineering, Shiraz University, Shiraz, Iran
n.vafamand@shirazu.ac.ir
2
School of Mechanical Engineering, Shiraz University, Shiraz, Iran
3
Department of Engineering, Aarhus University, Aarhus, Denmark

Abstract. In the contemporary industrial world, induction motors have special

place in the technological advancements. Therefore, detecting faults in these
systems is gaining more and more attentions. This paper proposes a new
approach for the fault detection and classification in induction motors. In the
developed approach, shapelet concept is employed. The main distinguished
advantage of shapelet is that they can extract features in dataset based on a local
shape-similarity. To make the suggested shapelet algorithm more accurate and
efficient, a new criterion for the minimum and maximum number of all classes
members is introduced. Compared to the previous shapelet feature extraction
techniques, the proposed technique has high accuracy and speed. For case study
of a sample induction motor, which is performed in healthy and broken rotor bar
situations, the shapelet approach is exploited to identify any possible faults.
Simulation results show the satisfactory performance of the suggested shapelet
technique for detecting broken rotor bar fault.

Keywords: Shapelet algorithm
Fault detection
Induction motor faults

1 Introduction

Fault detection methods become one of the most critical issues as in the most parts of
industry, faults are likely to occur and it can cause devastating damage in equipments.
In this situation using methods that can detect fault accurately and fast is very
important. Shapelet is a machine learning method that can find local shape-based
differences in signals accurately and fast. So Shapelet method can be employed as one
of the most accurate way to detect faults. In 2009, Eamonn Keogh et al. introduced
shapelet concept as a new classification method for the first time [1]. After that pioneer
work, the shapelet attracts a lot of attention because of its advantages. For instance, the
shapelet can classify datasets high accurately; it can gain some insight into the data; and
it works in both time and frequency domains [1]. Some extensions of the approach of
[1] are logical-shapelet concept [2], shapelet transform [3], unsupervised-shapelet (U-
shapelet), which is a novel method for clustering data [4], and multivariate shapelet
detection (MSD) for multivariate time series [5]. As Shapelet algorithm can recognize
local shape differences, it has been successfully applied on various systems to find

© Springer Nature Switzerland AG 2021

H. Selvaraj et al. (Eds.): ICSEng 2020, LNNS 182, pp. 23–33, 2021.
https://doi.org/10.1007/978-3-030-65796-3_3
24 M. Iraji et al.

faulty behaviors [6–8]. However, the process of selecting a proper shapelet is very
time-consuming and needs a high computational burden.
Until now, a wide variety of fault detection methods have been presented for the
induction motors (IMs). The IMs are widely exploited in industry, as they are highly
robust and have low cost maintenance, power efficiency and reliability. Though,
continuous care should be taken for their maintenance as the environmental situation
could cause serious faults, like broken rotor bar or shorted stator winding in the IM
systems [9–11]. The fault detection methods can be classified in two ways; invasive and
non-invasive methods [12]. Motor current signature analysis (MCSA) is one of the
popular non-invasive methods. Cusido et al. developed a method that use the convo-
lution of wavelet-based functions [13]. However, to the best knowledge of the authors,
the application of shapelet method in the IMs is not investigated yet, which is the main
motivation of this work.
In this paper, shapelet algorithm with new criteria for the maximum and minimum
number of all class members, is employed for online fault detection and classification in
the IMs. These criteria are defined based on the distance of timeseries from each other
and the number of timeseries that are closer to shapelet candidate than the other ones.
To show the merits of the suggested approach, a train dataset is made by simulating a
normal IM. Then, a type of fault is exerted on the train dataset. The proposed
shapelet algorithm with new criteria is used to discover proper shapelet in order to
detect and classify occurring faults of the IM, accurately. Finally, the chosen shapelet is
utilized to detect faults on a numerical IM system.
The rest of the paper is organized as follow; in Sects. 2 and 3, the study tools for
this paper such as the theory of shapelet and the model of IMs are elaborated. In
Sect. 4, new criteria for shapelet algorithm and the novel method for fault detection and
classification are proposed. In Sect. 5, the case study is presented that consists an IM
with healthy condition and broken rotor fault. Then, the shapelet are utilized in each of
these situations and fault is detected. Section 6 consists of the conclusion about using
shapelet for fault detection.

2 Theory of Shapelet

A time series is a temporal ordered list of real-valued data that is recorded in a fixed
interval. Suppose there is a set of n time series T ¼ fT1 T2 . . .Tn g, where each Ti has
m samples Ti ¼ hti1 ti2 . . .tim i. For the sake of simplicity, it is assumed that all series in
dataset have the same length. Though, inspired by [14], the results of this paper can be
extended to the case that the series do not have the same length. Shapelet is a subset of
one of the time series in the dataset and it is employed to extract a specific feature of
each class, in order to find the best touchstone for achieving fast and accurate classi-
fication. Shapelet is used for finding local shape-based similarity, so shapelet must be
normalized and non-variable to scale and offset [14]. Generally, the shapelet discovery
has three main steps
 Motor Current Signature Analysis Using Shapelet 25

2.1 Generating Candidates

There are time series with length m and the goal is to find a candidate with length l and
l can change in the range of ½min; max, while min  3 and max  m. All normalized
subseries of Ti that have l samples are put in Wi;l and Wi;l for all time series of T are
collected in Wl . Finally, collection of Wl for all possible amount of l for dataset T is W.




Wi;l ¼ ti;1 ; . . .; ti;l ; ti;2 ; . . .; ti;l þ 1 ; . . .; ti;ml þ 1 ; . . .; ti;m ð1Þ
 
Wl ¼ W1;l [ W2;l [ . . . [ Wn;l ð2Þ

W ¼ fWmin [ . . . [ Wmax g ð3Þ

2.2 Shapelet Distance Calculation

The distance between two subseries with the same length by square Euclidean distance
is calculated as

X
m
distðS; RÞ ¼ ðsi  ri Þ2 ð4Þ
i¼1

where S and R are time series, si and ri are their components and m is their length. For
computing the distance between a time series by length m and a subseries by the length
l, while l  m, minimum amount of squared Euclidean distance between normalized
subseries S and all normalized subseries inside Ti with the same length as the S (each of
then is named as R) must be found.

dSi ¼ min distðS; RÞ ð5Þ

R2Wi;l

The list of distance between the shapelet candidate S and the time series T is:


DS ¼ dS;1 ; dS;2 ; . . .; dS;n ð6Þ

2.3 Shapelet Assessment

The purpose of shapelet assessment is to determine how accurate this shapelet classifies
the dataset based on DS . For specifying the quality of shapelet, information gain (IG) is
used as a basic method for the shapelet assessment [7]. In this manner, DS is put in
order and IG must be calculated for each split point sp, where a split point is the
average between each two consecutive distances in Ds . For calculating the IG of each
26 M. Iraji et al.

sp, DS must be divided into AS and BS . The AS consists of all distances of DS \sp and
the BS consists of the rest of distances. So, the IG of sp is
 
jAS j jBS j
IGðDS ; spÞ ¼ H ðDS Þ  H ðAS Þ þ H ðBS Þ ð7Þ
jBS j jDS j

where jDS j is the number of all members in DS and HðDS Þ is defined as

H ðDS Þ ¼ pðAS Þ logðpðAS ÞÞ  pðBS Þ logðpðBS ÞÞ ð8Þ

where pðAs Þ and pðBs Þ are the proportion of objects in classes As and Bs with respect to
the overall dataset, respectively as

pðAs Þ ¼ jAs j=jDs j; pðBs Þ ¼ jBs j=jDs j ð9Þ

The IG for the candidate S, (i.e. IGS ) is

IGS ¼ max IGðDS s; pÞ ð10Þ

sp2Ds

Optimal Split Point (OPS) is the sp for which the IGS has the maximum value.

3 Modeling Faults in Induction Motors

In this section, dynamical representation of the IM in the healthy and broken rotor bars
situations are presented. An IM is an extremely symmetrical electromagnetic system.
Fault leads to a specified percent of asymmetry. So, broken rotor bars by inducting a
specific degree of unbalancing are modelled in the rotor parameters [15].

3.1 Model of Healthy Induction Machine

In the qd0 reference frame, the IM system is modelled by the following differential
equations of the voltages:
2 3
0 1 0
4 1 dkqd0
Vsqd0 ¼ rsqd0 iqd0
s þ x 0 0 5kqd0
s þ s
ð11Þ
dt
0 0 0
2 3
0 1 0
4 dkqd0
Vrqd0 ¼ rrqd0 iqd0
r þ ðx  xr Þ 1 0 0 5kqd0
r þ
r
ð12Þ
dt
0 0 0
 Motor Current Signature Analysis Using Shapelet 27

where x ¼ dh=dt, xr ¼ dhr =dt; rs;r

qd0
¼ rs;r I3 and
2 3
L11 0 0 L14 0 0
 6 0 L22 0 0 L25 0 7
6 7 qd0
kqd0 6 0 0 L33 0 0 0 7 is
s ¼6 7 qd0 ð13Þ
kqd0 6 L41 0 0 L44 0 0 7 ir
r 4 0 L52 0 0 L55 0 5
0 0 0 0 0 L66

The elements of Lij are elaborated in [15].

3.2 Model of Induction Machine with Broken Rotor Bars

Broken rotor bar causes asymmetry in the IM rotor and it leads to non-symmetrical
stator current distribution, so unbalanced current is the reason of breaking other rotor
bars during long time [11]. In broken rotor bar fault in induction motor, rotor becomes
asymmetric by changing in rr
2 3
rr11 rr12 rr13
rrnew ¼ rr þ 4 rr21 rr22 rr23 5 ð14Þ
rr31 rr31 rr33

The elements of Drrqd0 are found in [15].

4 Proposed Approach

The proposed method has two main parts. The first part is the offline process that
consists of modeling the IM and its faults in order to make a comprehensive train
dataset. In addition, new criteria for specifying the number of members in all classes
are proposed for the shapelet algorithm described in Sect. 2. In the second part, the
obtained shapelet is employed to detect faults online.

4.1 Offline Step

This step consists of two main processes: generating train dataset and shapelet dis-
covery. The first requirement for finding shapelet is to generate a comprehensive train
dataset that includes all kinds of time series from all operating situations. For the
application of the fault detection of IMs, the train dataset consists of two healthy IM
and IM with broken rotor bar (BRB) fault working situations. A sample filtered and
normalized Fourier transform of stator current of phase a for healthy IM and an IM with
broken rotor bars is shown in Fig. 1. It is obvious from Fig. 1 that whenever each type
of fault is applied to the IM, several peaks appear in frequencies around the supply
frequency. The difference in their shapes is a good criterion for detection and classi-
fication of occurring faults. Therefore, shapelet can be utilized to classify the differ-
ences in the shapes of the time series.
28 M. Iraji et al.

Fig. 1. Filtered and normalized Fourier transform of stator current: (a) healthy condition (b) IM
with broken rotor bars.

In the conventional shapelet algorithm, which is given in Sect. 2, the number of

members in the classes are not predefined. For a dataset that consists of several classes
two of which have with negligible differences, if no limit is exerted on the maximum
number of members in each class, the shapelet algorithm may not distinguish these two
classes. Otherwise, if no limit is determined for the minimum number of members in
each class, a sample class with various members and small difference among its own
members may be considered as a different class. Thereby, it is necessary to choose the
number of classes members systematically. To achieve this goal, the minimum number
of each class member is considered as Minn .it can be written as [10]:

Minn ¼ D=ðk þ 1Þ ð16Þ

where k is the number of classes, n is the number of members for each class and D is
the number of all members in dataset.
The maximum number of each class is considers as Maxn and it can be calculated
by

M ¼ maxððk  1ÞD=2k  1; D=kÞ ð17Þ

Maxn ¼ maxðM; Minn þ floor ðM  Minn =2ÞÞ ð18Þ

The algorithm of calculating Minn and Maxn is shown in Fig. 2. As can be seen in
Fig. 2, the Minn are calculated from (16) and at the first step. After that, M must be
calculated from (17) and the value of Maxn is achieved by it. The algorithm of IG
calculation based on the Minn and Maxn is provided in Fig. 3. In comparison with the
conventional IG algorithm, this algorithm considers a range for numbers of objects in
each class. In the algorithm that is shown in Fig. 3, if the number of remaining objects
is less than Maxn , all of them must be put in DA and DB is provided with artificial
objects that do not exist in the dataset. Also, the split point (sp) is considered as a
summation of maximum distance and 2=3 of the last shapelet’s split point.
 Motor Current Signature Analysis Using Shapelet 29

Fig. 2. The algorithm of calculating Minn and Maxn

4.2 Online Step

In this step, first it must be determined whether fault is happened in IM or not. If yes, is
it BRB or none of them? In order to do it, a detection and classification method is
proposed. In this method, firstly, the distance vector between Fourier transform of
stator current and shapelet are calculated. Secondly, each array of this vector is com-
pared to the threshold of the corresponding shapelet. If the array is not less than
threshold, it must be eliminated. Third, minimum of the remaining arrays is chosen and
its corresponding shapelet is considered as this stator current time series feature. If this
shapelet is a representative of healthy IM class, it means fault has not happened. If the
chosen shapelet belongs to IM with BRB fault class, it means this fault has happened in
IM. If all of the arrays are eliminated, it means IM is faulty but the type of fault is not
broken rotor bar. The algorithm of the proposed method is shown in Fig. 4.
30 M. Iraji et al.

Fig. 3. The algorithm of calculating IG based on Minn and Maxn

5 Case Study

At the first step, a 50 Hz, 6 poles, three phase induction motor is modelled in healthy
situation based on [13]. A dataset of Fourier transform of stator current for phase a, for
healthy condition and 6 broken rotor bar condition are made for train dataset. It consists
of 50 series in frequency domain. This dataset is shown in Fig. 5.
Shapelet Algorithm with new criteria is employed in this step, in order to find best
shapelet for accurate classification. For each condition of the induction motor, a
specific shapelet is found. The shapelets of healthy motor and motor with BRB fault are
 Motor Current Signature Analysis Using Shapelet 31

presented in Fig. 6. It is obvious that the shapelet of IM with broken rotor bar fault
consists peaks of signal that caused by fault. on the other hand, the shapelet of
Healthy IM consists smooth signal without peaks. After finding Shapelets, a test dataset
with 800 time series is made for performance evolution. This dataset consist of 400
timeseries in healthy situation and 400 timeseries for IM with 6 broken rotor bars. The
test dataset is shown in Fig. 7. Finally, discovered Shapelets are used to classify test
dataset, the algorithm of the proposed method for classification is shown in Fig. 4.

Fig. 4. Algorithm of the proposed method

Fig. 5. The train dataset

Fig. 6. (a) Shapelet of healthy IM (b) Shapelet of IM with 6 broken rotor bars
Other documents randomly have
different content
 CHAPTER XIV.
DEMOBILISATION.

(a) Auby and Douai.

After a night at Haulchin and a second at Douchy, the Battalion

embussed for Auby. The road lay entirely through the country which
had recently been evacuated by the enemy. Everywhere were signs
of his occupation and departure. The route was necessarily a
circuitous one, for there were many canals in the district and every
bridge had been destroyed by the enemy during his withdrawal. Only
a few temporary military bridges were in existence, the bare
minimum necessary to supply the advancing British armies. Auby
was almost deserted when the Battalion arrived, and, as it was a
large village and the Battalion was small, there were ample billets.
Like all villages recently vacated by the Germans, it was in a filthy
condition; but the actual billets were fairly clean, having been
occupied for a few days by troops of the 63rd Division. The church
was in ruins, as was also the chemical works which was the
mainstay of the place. But, otherwise, not much damage had been
done, for there had been no protracted fighting in the district.
Reorganisation and training were begun at once. Though the
Battalion was very low in strength, it might soon be called on to fight
again. For fighting was still in progress, and, though the enemy
retreat was quickly becoming a rout, few realised before the
armistice how thoroughly beaten the Germans were. Only two
platoons per company could be organised, but large drafts were
hoped for soon. On November 8th the Battalion was inspected by
the G.O.C., 147th Infantry Brigade, who commended it on its good
turn-out, and also on its recent fighting record.
On the morning of November 11th came news of the armistice,
which had been concluded with the enemy. Hostilities were to cease
at 11-0 a.m., and all troops were to remain in the positions occupied
at that hour. The news came rather as a surprise to most. Sweep-
stakes on the date of the cessation of hostilities had been got up in
the last few days by both the 4th and 7th Battalions; officers of the
former netted the proceeds of both. There was little time to make
preparations for celebrating the occasion. At 11-0 a.m. the Battalion
Band turned out and paraded the streets. A Battalion concert was
hurriedly arranged for the evening and went off very well, in spite of
the total absence of beer, which could not be procured in time.
Brigadier-General H. H. S. Morant, D.S.O., was present, and said a
few words to the men; his pious wish that the armistice had been
postponed a little, in order that he might have seen the Battalion
again in action, was greeted with many cries of dissent. Though, of
course, everyone was glad that the war was over, there were
undoubtedly some who viewed the event with rather mixed feelings.
To those who had lived for more than four years with the one great
purpose of defeating the enemy, it seemed almost that the object of
their lives had been taken away. And there is also a spirit of
camaraderie and good-fellowship on the battlefield and in the
trenches which no peacetime conditions can wholly reproduce.
The Battalion had come to Auby expecting to remain for, at most,
a week or two. As events turned out it spent four months and a half
there—by far the longest period it ever spent at one place abroad.
The time was marked by few events of importance. Towards the end
of November a draft, about 200 strong, arrived. This was rather a
surprise, particularly as many of the men were farmers who, it was
expected, would be in one of the earliest classes for demobilisation.
Early in December it was decided to bring the Battalion Colours out
to France, and a colour party[21] was sent to England to fetch them.
The Colours had been deposited in the Parish Church of Halifax since
the summer of 1915. On December 8th they were handed over to
the colour party by the Rev. Canon A. E. Burn, D.D., after the
morning service, and were safely conveyed to France.
 During the earlier part of December, ceremonial drill was the main
occupation. On December 16th the 49th Division was inspected by
the G.O.C., XXII. Corps, under whom it had served so long, though,
at the time of the review, it was in the VIII. Corps. This was the first
and only time that the whole division paraded together in France.
Christmas was celebrated with due honours. As was the general
custom, the officers’ dinner was held on Christmas Eve and the
sergeants’ dinner on Boxing Day, in order that Christmas Day itself
might be wholly devoted to the men. A Rugby match between the
officers and sergeants of the Battalion was played during the
afternoon. The “form” of the officers was not very convincing;
possibly this was the result of the festivities of the previous night.
Altogether, the festive season passed very happily.
A few days after Christmas the 147th Infantry Brigade gave a
“treat” to the children of Auby, and never until then was it realised
how many youngsters there were in the village. Indeed, it was
suspected that many had come in specially from the neighbourhood.
A cinematograph show in the afternoon was followed by a tea, after
which a real Father Christmas—the Medical Officer’s store of cotton
wool had been heavily drawn upon, and few would have recognised
Sergt. E. Jones in the benevolent-looking old man—helped to hand
out a present to each child. But the most striking event of the
entertainment was the wonderful enthusiasm of the children as they
joined in the singing when the band struck up the “Marseillaise.”
Meanwhile, the Battalion had not been idle. Some work was
carried out, trying to restore the damage done by the enemy in the
district. Much of the country had been flooded and attempts were
made to reduce the water, though without much success. Trenches
were filled in, barbed wire entanglements were removed, and roads
were improved. An area was allotted to the Battalion for salvage
work, and a certain amount of useful material was collected. In
connection with this work, a most unfortunate accident occurred
early in December. During their removal to the salvage dump some
trench mortar shells exploded, killing two men of A Company and
wounding several more. About once a week one whole company
moved into Douai, which was about four miles away, to provide
guards at the station there for forty-eight hours.
Some training was carried out every morning, and the afternoons
were entirely devoted to games. The Battalion had a fairly good
football field at its disposal, and this was in use every fine day. Much
of both the training and recreation was in connection with the
Divisional Competition, for the shield presented by Major-General N.
J. G. Cameron. In these competitions the Battalion finished second,
and was very unlucky not to do even better. Its greatest triumph was
the winning of the competition for the best platoon in the Division.
The Battalion was represented by No. 5 Platoon of B Company,
under Sec.-Lieut. W. G. Bradley, with Sergt. F. B. Birtwhistle as
platoon sergeant. After winning the Brigade eliminating competition
with ease, this platoon came up against the platoons of the 1/5th
Battalion West Yorkshire Regt. and 1/4th Battalion York and
Lancaster Regt., winners in their respective brigades. The
competition was most comprehensive and occupied three whole
days. The platoons were inspected in drill order and fighting order;
they had competitions in marching, firing with rifle, Lewis gun and
revolver, close and extended order drill, ceremonial, and guard
mounting. The competition ended with a small tactical scheme.
Right from the start good shooting put No. 5 Platoon ahead, and it
never lost this position, winning comfortably.
A Company won the Brigade inter-company Association football
competition, but came down rather badly when it played a company
of the D.A.C. in the divisional competition. Both Battalion teams
reached the finals in the tug-of-war, but neither won. In the Brigade
boxing championships the Battalion had two firsts and a second, but
its representatives did little afterwards. About the middle of January
several successes[22] were gained in the 147th Infantry Brigade
Gymkhana. An officers’ Rugby XV. was started and had great
success, though most of the players had had no experience of the
game before. The Battalion Rugby XV. was the best in the Brigade,
but, unfortunately, the divisional Rugby competition was never
played. Altogether, the Battalion was well to the fore both in military
training and in sports.
With the armistice, the army education scheme was taken up
strongly. The lack of trained teachers and the deficiency in books
and stationery were serious obstacles, but they did not prevent a
great deal of useful work being done. At least one hour a day was
set apart for education, and classes were attended by everyone.
General subjects were taught by platoon commanders, with the help
of any competent N.C.O’s or privates they could discover. Special
classes were started in French and Spanish, mathematics and
commercial subjects. Many lectures were given, mainly by outside
lecturers; in particular, three lectures on the recent history of Egypt,
which were given by the Brigadier and were largely based on his
personal experiences, were much enjoyed. A Battalion debating
society was started and had several successful evenings, the chief
being a mock court-martial of “William Hohenzollern,” carried out by
the officers. Major W. C. Fenton’s rendering of the title part, and
Sec.-Lieut. R. M. Leddra’s impersonation of “Little Willie,” fairly
brought down the house.
Reading and recreation rooms were opened in the village. The
Battalion canteen continued its activities, but found the competition
of the numerous estaminets which were being opened very strong.
Whist drives were introduced and proved very popular; the Battalion
was particularly fortunate in the receipt of large numbers of splendid
prizes from the Halifax Courier Fund. For a short time an officers’
club was in being, but it was not a great success.
The Battalion had not been long at Auby before the civilian
inhabitants began to return. This had its advantages, but it had
disadvantages too. On the whole the men got on very well with the
natives, and the opening of shops and estaminets was very much
appreciated. But, as more and more people returned, the billeting
question for a time became rather serious.
 Early in January demobilisation on a large scale began. Though
much desired by most, this had also its melancholy aspect. It was
sad to see the men, who had been so much to the Battalion,
gradually melting away. Especially was this realised when a farewell
whist drive was held on January 20th, in honour of Major W. C.
Fenton, M.C., Sergt. F. Smith, and many men, who were to leave the
next morning. Day after day, as one saw the lorry-loads of friends
leaving for the Corps Concentration Camp, one realised how quickly
a chapter of one’s life was drawing to its close. By the end of
February over 270 had left, besides a large number who had been
demobilised on leave. Towards the end of February a draft of 8
officers and 169 other ranks left to join the 13th Battalion Duke of
Wellington’s Regt. at Dunkerque. By the middle of March the
Battalion was but a skeleton, its effective strength being below 100.
On March 20th, what remained of the Battalion was moved to
Douai, where the cadres of the 49th Division were being
concentrated. Here the Battalion was soon reduced to cadre strength
—4 officers and 46 other ranks. Before the cadre started for England
it had been reduced still lower. Time hung very heavily on the hands
of the few people who were still left. There were not enough to do
any training, the necessary duties did not occupy much time, and
there was practically nothing to do in Douai. Apart from one or two
cinemas, and the estaminets, there were few forms of amusement.
Even the strongest adherents of military life began to long for
demobilisation. Almost daily there were rumours that the Battalion
was to leave at some early date. But the weeks dragged on until
May was past before the first move was made.

(b) The Return of the Cadre.[23]

On June 7th, after many false rumours and one lot of cancelled
orders, the cadre at last left Douai. For the past month the weather
had been very hot and dry, and it was a broiling morning when the
Battalion vehicles and stores were entrained. The officers, with the
loose baggage, occupied one of the well-known cattle trucks, which
was made quite comfortable with a table, some forms and chairs;
the men occupied third-class compartments. Early in the afternoon
the train started. On the way to Arras a glimpse was obtained of the
old camping ground at Feuchy. Then, circling round the ruins of Lens
and passing Bethune and Hazebrouck, the cadre arrived at a siding
near Dunkerque, just before dusk. The night was spent at what was
known as the “dirty” camp—the blankets provided certainly deserved
the name. The next morning the Battalion moved to No. 3 Camp
where it remained for nine days.
No. 3 Camp was the centre through which passed all men and
cadres on their way for demobilisation by the Dunkerque route. It
was managed entirely by the 13th Battalion Duke of Wellington’s
Regt., so the cadre found itself in the midst of friends. Officers’ and
Sergeants’ Messes were thrown open; everything that was wanted
was immediately forthcoming, if available. Two very pleasant
officers’ reunion dinners were held in Dunkerque, and the Officers’
Mess of the 13th Batt. Duke of Wellington’s Regt. held a special
guest night for the officers of the 147th Infantry Brigade. Altogether,
the tedium of the stay on the coast was very much relieved. Strange
to say, No. 3 Camp was pitched on the exact site which the Battalion
had occupied in July, 1917. But how changed was the country! Then
the Battalion had seen nothing but a flat grass patch and a waste of
sand hills, on which to pitch its tents. Now the whole area was
covered by immense camps, mostly of huts, which would
accommodate many thousands of men. The time near Dunkerque
passed uneventfully. The cadre remained there rather longer than
was usual, partly as a result of a violent wind storm which delayed
sailings for two or three days.
On June 16th the cadre sailed. The morning was spent in loading
the vehicles on to the cargo vessel—S.S. “Clutha”—which was to
carry them to Southampton. Among the loading party were several
transport men who had assisted in a similar operation at
Southampton, more than four years before, when the Battalion
sailed for France. In the afternoon the cadre embarked on S.S. “St.
George,” and, about 3-0 p.m., the vessel moved away from the quay.
Among her passengers was Capt. W. B. B. Yates, who had that
morning arrived at Dunkerque on leave from the Murman Coast.
Slowly the ship passed out of Dunkerque harbour and moved along
the French coast, until opposite Calais. The day was perfect. A bright
sun blazed down upon the deck, and there was scarcely a ripple on
the water. Every now and then a fountain of water would rise at
some miles’ distance, followed by the sound of a deep ‘boom,’ as
some mine was destroyed. Opposite Calais the ship turned north
towards the English coast, and, in the early evening, the coast of
France dipped from view, and all said good-bye to the land which
held for them so many memories. A run along the English coast
brought the “St. George,” in the early morning, into Southampton
Water, where it anchored off Netley until about 8-0 a.m.
Soon after breakfast the cadre landed at Southampton Docks and
the work of unloading began. As the vehicles were slung by the
great cranes out of the ship, they were run by the men into a large
covered shed to await entrainment. A change had been made in the
programme. When it left France, the cadre had expected to go to a
camp at Fovant and remain there until stores had been checked
over. But on arrival in England, it was found that only the stores
were to go to Fovant, while the cadre was to proceed direct to
Halifax. As soon as the vehicles were loaded the cadre might leave.
Everyone worked with a will, and, by tea time, everything was on
the train. Then all marched to Southampton West Station to entrain
for London. While the cadre was waiting there, the train carrying the
battalion’s vehicles passed through the station on its way to Fovant,
and a shout of unholy glee was the farewell of the transport men to
their beloved (?) limbers. An uneventful journey brought the party to
London where all spent the night at the Buckingham Palace Hotel.
Shortly after 3-0 p.m. on June 18th, the cadre reached Halifax.
On the platform Brigadier-General R. E. Sugden, C.M.G., D.S.O.,
Lieut.-Colonel H. S. Atkinson, T.D., Lieut.-Colonel J. Walker, D.S.O.,
and many other old officers of the Battalion were waiting to meet
the party and give it a rousing welcome. Outside the station was
Sergt.-Drummer H. Deane, with a band which he had collected. A
colour escort had been sent down from the depot of the 3rd
Battalion Duke of Wellington’s Regt. But, best of all, hundreds of
demobilised men of the Battalion—the men who had made its
history—had paraded there to welcome their old comrades.
“Demobilised men of the 1/4th Duke of Wellington’s Regt. rally round
their old Battalion” was the inscription on the banner which they
carried.
The cadre was photographed at the Station, and then it fell in and
moved off. Notice of its coming had been very short, but the streets
were gay with flags and packed with cheering crowds. Preceded by
the band, and followed by the Colours with their escort and the
demobilised “Dukes,” who once again “moved to the right in fours,”
the cadre marched up Horton Street and round to the Town Hall,
where it was welcomed home by the Mayor. The Commanding
Officer thanked the Mayor for all the town and district had done for
the Battalion while overseas, and for the day’s welcome home. Then,
turning to the men, he said farewell to them in words which all will
remember:—
“The Battalion will now pass to another command, and I take this
opportunity of thanking the men for their loyalty to me, to the
regiment, and to the town and district they have represented. Those
of us who have been spared to come through this great conflict hold a
very sacred trust. We must ever remember the comrades we have left
lying on the battle-fields of France and Flanders. Let us try to prove
worthy of their sacrifice. They have died that we might live; and on
our return to civilian life we must continue to serve our country, our
town, and our homes in the same spirit of loyalty, cheerfulness and
trust that pulled us through four years of war. If you do that you can
look to the future with happiness, certain that your great efforts of
the last few years have not been made in vain.”
 THE CADRE AT HALIFAX, JUNE 18th, 1919.
The cadre was then entertained by the Mayor at the Drill Hall, and
afterwards the men dispersed to their homes for the night.
The next day the men paraded for the last time and went to
Ripon for dispersal. And when the time came to say the last good-
bye there were many who regretted that, henceforward, their paths
would lie apart. With the warmest of handshakes, and the deepest
feelings of friendship and respect, we parted one from another, and
the history of the 1/4th Battalion Duke of Wellington’s (West Riding)
Regiment in the Great War came to an end.
 APPENDIX I.
ITINERARY OF THE BATTALION.
Date of Move to.
Move.
4.8.14. Halifax Mobilisation of the Battalion.
5.8.14. Hull By train
11.8.14. Immingham By boat.
13.8.14. Great Coates By march route.
15.9.14. Riby Park By march route.
17.10.14. Marsden By train.
5.11.14. Doncaster By train.
St. Martin’s Camp, By train to Folkestone; by S.S.
14.4.15.
Boulogne “Invicta” to Boulogne.
By march route to Hesdigneul; by
15.4.15. Estaires train to Merville; by march route
to Estaires.
22.4.15. Doulieu By march route.
24.4.15. La Croix Lescornez By march route.
No. 3 Section, Relieved 3rd Bn. Worcestershire
26.4.15.
Fleurbaix Sector Regt.
29.4.15. La Croix Lescornez
Inter-relieving with 1/6th Bn. D. of W.
2.6.15. No. 3 Section
Regt.
5.5.15. La Croix Lescornez
8.5.15. Croix Blanche By march route.
9.5.15. No. 3 Section
13.5.15. La Croix Lescornez
17.5.15. No. 3 Section
21.5.15. La Croix Lescornez
25.5.15. No. 3 Section Inter-relieving with 1/6th Bn. D. of W.
29.5.15. La Croix Lescornez Regt.
2.6.15. No. 3 Section
6.6.15. La Croix Lescornez
12.6.15. No. 3 Section
18.6.15. La Croix Lescornez
25.6.15. Sailly-sur-la-Lys
26.6.15. Doulieu
27.6.15. Farms near Bailleul
29.6.15. Flêtre By march route.
Wood near St. Jans ter
30.6.15.
Biezen
Canada Wood, near
7.7.15.
Elverdinghe
Relieved 2nd Bn. Royal Dublin
8.7.15. Lancashire Farm Sector
Fusiliers.
13.7.15. Canal Bank (Bde. Res.)
18.7.15. Lancashire Farm Sector
Wood near Oosthoek Inter-relieving with 1/5th Bn. D. of W.
24.7.15.
(Divl. Res.) Regt.
30.7.15. Lancashire Farm Sector
5.8.15. Canal Bank (Bde. Res.)
Glimpse Cottage
11.8.15.
Sector Inter-relieving with 1/7th Bn. D. of W.
Saragossa Farm (Bde. Regt.
16.8.15.
Res.)
18.8.15. Ypres Left Sector
Malakoff Farm (Bde. Inter-relieving with 1/5th Bn. D. of W.
21.8.15.
Res.) Regt.
24.8.15. Ypres Left Sector
Coppernollehoek (Divl.
26.8.15. Relieved by 1/4th Bn. Y. and L. Regt.
Res.)
 Relieved 1/6th Bn. W. Yorks. Regt.
8.9.15. Turco Farm Sector
Regt.
15.9.15. Canal Bank (Bde. Res.)
21.9.15. Turco Farm Sector Inter-relieving with 1/5th Bn. D. of W.
Elverdinghe (Bde. Regt.
27.9.15.
Res.)
Camp near Woesten-
2.10.15.
Poperinghe Road.
Glimpse Cottage
14.10.15. Relieved 1/5th Bn. W. Yorks. Regt.
Sector
21.10.15. Canal Bank (Bde. Res.) Relieved by 1/4th Bn. K.O.Y.L.I.
30.10.15. Ypres Left Sector Relieved 1/7th Bn. D. of W. Regt.
Coppernollehoek (Divl. By motor bus. Relieved by 1/6th Bn.
3.11.15.
Res.) D. of W. Regt.
11.11.15. Ypres Left Sector
Malakoff Farm (Bde. Inter-relieving with 1/5th Bn. D. of W.
15.11.15.
Res.) Regt.
19.11.15. Ypres Left Sector
Malakoff Farm (Bde.
23.11.15. Relieved by 1/6th Bn. D. of W. Regt.
Res.)
Coppernollehoek (Divl.
27.11.15.
Res.)
Malakoff Farm (Bde.
5.12.15.
Res.)
9.12.15. Ypres Left Sector Relieved 1/5th Bn. D of W. Regt.
Malakoff Farm (Bde.
13.12.15.
Res.) Inter-relieving with 1/6th Bn. D. of W.
17.12.15. Ypres Left Sector Regt.
20.12.15. Elverdinghe
27.12.15. Coppernollehoek
30.12.15. Poperinghe
Camp near St. Jans ter
31.12.15. By march route.
Biezen
1.1.16. Houtkerque
15.1.16. Wormhoudt
2/3.2.16. Camps En Amienois By march route to Esquelbecq; by
train to Longueau; by march
 route to Ailly; by motor bus to
Camps En Amienois.
11.2.16. Picquigny
12.2.16. Molliens-au-Bois By march route.
13.2.16. Warloy Baillon
Right Section, Authuille
28.2.16. Relieved 1/4th Bn. K.O.Y.L.I.
Trenches
4.3.16. Bouzincourt
5.3.16. Authuille Defences
6.3.16. Mailly-Maillet By march route.
29.3.16. Harponville
30.3.16. Naours
23.4.16. Hedauville By motor bus.
12.5.16. Aveluy Wood
1.6.16. Martinsart Wood
15.6.16. Aveluy Wood
By march route.
20.6.16. Vadencourt Wood
27.6.16. Senlis
28.6.16. Vadencourt Wood
B Assembly Trenches,
30.6.16. By march route, via Senlis.
Aveluy Wood
1.7.16. Crucifix Corner, Aveluy By march route.
2.7.16. Johnstone’s Post Relieved 1/5th Bn. W. Yorks. Regt.
Right Sub-sector,
5.7.16. Relieved 1/5th Bn. D. of W. Regt.
Thiepval Wood
B Assembly Trenches, Relieved by 1/6th Bn. D. of W. Regt.
7.7.16.
Aveluy Wood and 1/5th Bn. K.O.Y.L.I.
Thiepval Wood, Right Relieved 1/6th Bn. D. of W. Regt.
8.7.16.
Sub-sector and 1/5th Bn. K.O.Y.L.I.
Relieved by 1/5th and 1/7th Bns. D.
11.7.16. „ Bde. Res.
of W. Regt.
 „ Left Sub-
14.7.16.
sector
17.7.16. „ Bde. Res.
„ Left Sub-
21.7.16.
sector
Inter-relieving with 1/5th Bn. D. of W.
25.7.16. „ Bde. Res.
Regt.
„ Left Sub-
31.7.16.
sector
6.8.16. „ Bde. Res.
„ Left Sub-
12.8.16.
sector
Relieved by 9th Bn. Loyal North
19.8.16. Raincheval
Lancs. Regt.
27.8.16. Forceville By march route.
2.9.16. Martinsart Wood By march route.
Thiepval Wood,
2/3.9.16. For attack on German line.
Assembly Parallels
3.9.16. Martinsart Wood By march route.
7.9.16. Hedauville
15.9.16. Martinsart Wood By march route.
16.9.16. Crucifix Corner, Aveluy
Leipsig Redoubt
18.9.16.
(Support).
Leipsig Redoubt (Front
21.9.16. Relieved 1/5th Bn. D. of W. Regt.
Line)
By march route. Relieved by 12th Bn.
24.9.16. Lealvillers
Middlesex Regt.
25.9.16. Halloy
27.9.16. Humbercamps By march route.
28.9.16. Bienvillers-au-Bois
Hannescamps, Left Relieved 2nd Bn. Royal Welsh
29.9.16.
Sub-sector Fusiliers.
3.10.16. Souastre
Inter-relieving with 1/5th Bn.
Hannescamps, Left K.O.Y.L.I.
9.10.16.
Sub-sector
16.10.16. Bienvillers-au-Bois Relieved by 1/6th Bn. D. of W. Regt.
18.10.16. Humbercamps
19.10.16. St. Amand By march route.
21.10.16. Souastre (Bde. Res.)
24.10.16. Y Sector, Fonquevillers Relieved 1/6th Bn. W. Yorks. Regt.
30.10.16. Souastre
5.11.16. Y Sector
11.11.16. Souastre Inter-relieving with 1/5th Bn. D. of W.
17.11.16. Y Sector Regt.
23.11.16. Souastre
29.11.16. Y Sector
By march route. Relieved by 1/5th
5.12.16. Warlincourt
Bn. Sherwood Foresters.
6.12.16. Halloy By march route.
Berles-au-Bois and By march route. Relieved 2nd Bn.
7.1.17.
Humbercamps Bedfordshire Regt.
B1 Sub-sector, Berles-
10.1.17.
au-Bois
14.1.17. Humbercamps
18.1.17. B1 Sub-sector Inter-relieving with 1/5th Bn. D. of W.
Berles-au-Bois and Regt.
22.1.17.
Humbercamps
26.1.17. B1 Sub-sector
30.1.17. Humbercamps
By march route. Relieved 8th Bn.
1.2.17. Rivière (Bde. Res.)
K.R.R.C.
2.2.17. F1 Sub-sector, Rivière Relieved 7th Bn. K.R.R.C.
7.2.17. Rivière
11.2.17. F1 Sub-sector
16.2.17. Rivière Inter-relieving with 1/5th Bn. D. of W.
20.2.17. F1 Sub-sector Regt.
22.2.17. Bailleulval
26.2.17. F1 Sub-sector
28.2.17. Bailleulval Relieved by 2/9th Bn. London Regt.
1.3.17. Souastre
2.3.17. Halloy By march route.
6.3.17. Neuvillette
By march route to Doullens; by train
7/8.3.17. La Fosse to Merville; by march route to La
Fosse.
Senechal Farm (Bde. By march route. Relieved 12th Bn.
9.3.17.
Res.) London Regt.
13.3.17. Ferme du Bois Sector
17.3.17. Senechal Farm
23.3.17. Ferme du Bois Sector
29.3.17. Senechal Farm
4.4.17. Ferme du Bois Sector
10.4.17. Senechal Farm Inter-relieving with 1/5th Bn. D. of W.
16.4.17. Ferme du Bois Sector Regt.
22.4.17. Senechal Farm
28.4.17. Ferme du Bois Sector
4.5.17. Senechal Farm
10.5.17. Ferme du Bois Sector
16.5.17. Senechal Farm
25.5.17. Estaires By march route.
26.5.17. Sailly-sur-la-Lys By march route.
Relieved 2/5th Bn. Loyal North
27.5.17. Cordonnerie Sector
Lancs. Regt.
Rouge de Bout (Bde.
3.6.17. Inter-relieving with 1/5th Bn. D. of W.
Res.)
Regt.
11.6.17. Cordonnerie Sector
Relieved by 2/4th Bn. Loyal North
15.6.17. Estaires
Lancs. Regt.
18.6.17. Sailly Labourse By motor bus.
19.6.17. Philosophe (Bde. Res.) Relieved 1st Bn. Leicestershire Regt.
25.6.17. St. Elie Sector Relieved 1/6th Bn. D. of W. Regt.
1.7.17. Tower Keep, Vermelles Relieved by 9th Bn. Suffolk Regt.
By march route to Philosophe; by
3.7.17. L’Epinette
motor bus to L’Epinette.
 By march route to Merville; by train
13.7.17. Mardyck Camp, St. Pol to Dunkerque; by march route
to St. Pol.
18.7.17. Bray Dunes
20.7.17. Ghyvelde By march route.
31.7.17. La Panne Bains
By motor bus to beyond Oost
Lombartzyde Right Dunkerque; by march route to
3.8.17.
Sub-sector trenches. Relieved 1/5th Bn.
K.O.Y.L.I.
Presque L’Isle
9.8.17.
Defences Inter-relieving with 1/6th Bn. D. of W.
Lombartzyde Right Regt.
13.8.17.
Sub-sector
By march route. Relieved by 20th Bn.
16.8.17. Oost Dunkerque
Royal Fusiliers.
By march route. Relieved 2nd Bn.
17.8.17. Oost Dunkerque Bains Argyle and Sutherland
Highlanders on coast defence.
27.8.17. Surrey Camp
29.8.17. La Panne
13.9.17. Bray Dunes
23.9.17. Coudekerque
24.9.17. Wormhoudt
25.9.17. Buysscheure
28.9.17. Audenthun Area By march route.
Longue Croix (Staple
30.9.17.
Area)
Clyde Camp, near
3.10.17.
Watou
Red Rose Camp,
4.10.17.
Vlamertinghe
Pommern Castle (Bde.
4.10.17. Relieved 1st Bn. Canterbury Regt.
Res.)
Front Line near Relieved 3rd Bn. Canterbury and 3rd
5.10.17.
Abraham Heights Bn. Auckland Regts.
Relieved by 2/5th Bn. Manchester
6.10.17. Pommern Castle
Regt.
 Attacked in support of 146th Infantry
9.10.17. Peter Pan
Brigade.
10.10.17. X Camp, St. Jean Relieved by New Zealand Div.
No. 3 Area,
16.10.17.
Vlamertinghe
24.10.17. A Camp, Winnezeele By march route.
Farms near
27.10.17.
Steenvoorde
9.11.17. Canal Area, near Ypres By motor bus.
12.11.17. Swan Area, near Ypres By march route.
Anzac Ridge (Bde.
19.11.17. Relieved 1/7th Bn. W. Yorks. Regt.
Res.)
Molenaarelsthoek
23.11.17. Relieved 1/6th Bn. D. of W. Regt.
Sector
27.11.17. Gordon House Area Relieved by 1/5th Bn. K.O.Y.L.I.
28.11.17. Vancouver Camp By march route.
6.12.17. Dragoon Camp By march route.
7.12.17. Keerselaarhoek Sector Relieved 4th Bn. Suffolk Regt.
10.12.17. Dragoon Camp
Inter-relieving with 1/5th Bn. D. of W.
15.12.17. Keerselaarhoek Sector Regt.

17.12.17. Argyle Camp Relieved by 1/7th Bn. W. Yorks. Regt.

18.12.17. Halifax Camp By march route.
By train to Hellfire Corner; by march
Molenaarelsthoek
23.12.17. route to trenches. Relieved
Sector
1/5th Bn. K.O.Y.L.I.
Garter Point (Bde.
29.12.17. Relieved by 1/5th Bn. D. of W. Regt.
Res.)
Infantry Barracks,
4.1.18. Relieved by 1/8th Bn. W. Yorks. Regt.
Ypres
12.1.18. Devonshire Camp By march route.
By train from Brandhoek to Caestre;
26.1.18. St. Silvestre Cappel
by march route to billets.
By march route to Ebblinghem; by
5.2.18. Moulle train to Watten; by march route
to billets.
10.2.18. Buysscheure By march route.
11.2.18. St. Silvestre Cappel By march route.
Infantry Barracks, By march route to Caestre; by train
21.2.18.
Ypres to Ypres.
22.2.18. Reutel Sector Relieved 1st Bn. Otago Regt.
2.3.18. Maida Camp
10.3.18. Reutel Sector
Inter-relieving with 1/5th Bn. Y. and L.
Westhoek and Railway Regt.
18.3.18.
Wood Dugouts
27.3.18. Reutel Sector
Relieved by 1st Bn. Leicestershire
3.4.18. Maida Camp
Regt.
9.4.18. Camp near Reninghelst By march route.
By motor bus to La Crèche; by
10.4.18. Erquinghem march route, via Le Veau, to
Erquinghem.
10.4.18. Nieppe
11.4.18. S.E. of Bailleul
12.4.18. Bailleul
By march route. (Battle of the Lys).
15.4.18. S. of St. Jans Cappel
18.4.18. St. Jans Cappel
20.4.18. Mont Noir
21.4.18. Mont des Cats By march route.
21.4.18. Poperinghe By march route.
25.4.18. Ouderdom
25.4.18. Millekruisse (Bde. Res.)
Millekruisse (Front By march route. (Battle of Kemmel.)
1.5.18.
Line)
3.5.18. Camp S. of Poperinghe
Road Camp, St. Jans
4.5.18. By march route.
ter Biezen
Camp near St. Martin-
14.5.18. By motor bus.
au-Laert
Road Camp, St. Jans
19.5.18. By motor bus.
ter Biezen
22.5.18. Penton Camp, Proven By march route.
Road Camp, St. Jans
26.5.18. By march route.
ter Biezen
 By light railway to Vlamertinghe; by
3.6.18. Zillebeke Sector march route to trenches.
Relieved 15th Bn. K.R.R.C.
Bde. Res. near
11.6.18. Relieved by 1/7th Bn. D. of W. Regt.
Goldfish Chateau
Siege Camp (Divl.
20.6.18. Relieved by 1/5th Bn. Y. and L. Regt.
Res.)
Sector N. of Menin
29.6.18. Relieved 1/7th Bn. W. Yorks. Regt.
Road
15.7.18. Siege Camp Relieved by 1/5th Bn. Y. and L. Regt.
Bde. Res. near
23.7.18. Relieved 1/6th Bn. W. Yorks. Regt.
Goldfish Chateau
Relieved 2nd Bn. 118th American Inf.
1.8.18. Zillebeke Sector
Regt.
Relieved by 3rd Bn. 117th American
7.8.18. Brielen Line
Inf. Regt.
8.8.18. Siege Camp Relieved by 1/4th Bn. K.O.Y.L.I.
16.8.18. Menin Road Sector Relieved 1/5th Bn. Y. and L. Regt.
Relieved by 5th Bn. Argyle and
20.8.18. Wood near Oosthoek
Sutherland Highlanders.
By light railway to Proven; by train to
23.8.18. Nielles-lez-Ardres Audruicq; by march route to
billets.
By march route to Nortkerque; by
28.8.18. Siracourt and Beauvois train to Wavrans; by march
route to billets.
1.9.18. Camblain L’Abbé By motor bus.
Wakefield Camp, near
13.9.18. By march route.
Roclincourt
23.9.18. Feuchy By march route.
6.10.18. Bivouacs near Buissy By motor bus.
Sunken road S. of
9.10.18.
Haynecourt
Railway embankment
10.10.18. near By march route.
Escaudœuvres
Assembly positions on
11.10.18.
Iwuy—Rieux Road
11.10.18. Ridge S.E. of Iwuy Captured in battle.
 Vordon Wood, W. of La
12.10.18. Following up retreating enemy.
Selle River
Area S.W. of Avesnes-
14.10.18. Relieved by 1/7th Bn. W. Yorks. Regt.
le-Sec
16.10.18. Vordon Wood Relieved 1/7th Bn. W. Yorks. Regt.
Relieved by 1st Bn. Warwickshire
18.10.18. Naves
Regt.
21.10.18. Le Bassin Rond
27.10.18. Douchy
By march route.
Thiant—Moncheaux
28.10.18.
Road
Sunken Road S. of
31.10.18.
Maing
Bde. Support near La
1.11.18.
Rhonelle River
2.11.18. Haulchin Relieved by units of 169th Inf. Bde.
3.11.18. Douchy By march route.
4.11.18. Auby By motor bus.
20.3.19. Douai By march route.
A Camp, near
7.6.19. By train.
Dunkerque
No. 3 Camp, near
8.6.19. By march route.
Dunkerque
By S.S. “St. George”; transport on
16/17.6.19. Southampton
S.S. “Clutha.”
17.6.19. London
18.6.19. Halifax By train.
19.6.19. Ripon Dispersal Camp
 APPENDIX II.
NOMINAL ROLL OF OFFICERS WHO SERVED WITH THE BATTALION ABROAD.
In compiling the list of officers, the following rules have been
adhered to:—
1. In the case of officers who went out with the original Battalion,
the rank stated was that held on April 14th, 1915.
2. In the case of officers who joined the Battalion after April 14th,
1915, the rank stated was that held at the date the officer reported
for duty.
3. Decorations, a list of which will be found in Appendix V., have
been omitted, except the following:—
(a) Territorial Decoration. (b) Decorations won by officers before
they joined the Battalion.
4. This record is a Battalion record. Hence, only service with the
Battalion is shown.
5. The names of the officers of the original Battalion are marked *.
Akroyd, F., Sec.-Lieut. Joined, 25.11.17. Wounded near Bailleul, 12.4.18.
*Anderton, W. L., Sec.-Lieut. Bn. Bombing Officer (May–August, 1915). Killed
in action near Ypres, 21.8.15.
Anderton, R., Sec.-Lieut. Joined, 14.8.16. Transferred to 2nd Bn. D. of W.
Regt., 23.10.16.
*Andrews, M. P., Capt. O.C. A Coy. (May–August, 1915). Killed in action near
Ypres, 14.8.15.
Applewhaite, C. T., Sec.-Lieut. (6th Bn. Norfolk Regt.). Joined, 20.8.17. To
hospital sick, 16.6.18.
*Atkinson, H. S., Lieut.-Col. (T.D.). O.C. Bn. (August, 1914–May, 1915). To
hospital sick, 24.5.15.
Atkinson, R. B., Sec.-Lieut. Joined, 29.1.18. Wounded near Reutel, 16.3.18.
*Aykroyd, H. H., Sec.-Lieut. Bn. I.O. (January–September, 1916). 147th Inf.
Bde. I.O. (September, 1916–July, 1917). To hospital sick, 12.7.17.
Rejoined, 24.6.18. Adjt. (June, 1918–January, 1919). Second in
Command (January–February, 1919). Left Bn. for demobilisation, 6.2.19.
Lieut., 30.5.16. Capt., 1.6.16. A/Major, 22.1.19.
Baldwin, S., Sec.-Lieut. Joined, 21.9.16. Transferred to 8th Bn. D. of W.
Regt., 26.10.16.
 Bales, P. G., Sec.-Lieut. Joined, 21.9.16. Bn. I.O. (March–September, 1917
and May–November, 1918). A/Adjt. (September, 1917–May, 1918). Adjt.
(January–June, 1919). Demobilised with Cadre of Bn., 19.6.19.
Lieut., 1.6.16. A/Capt., 22.1.19.
Balme, S., Lieut. Bn. Signalling Officer (1915). O.C. D Coy. (February–
November, 1918). Left Bn. for demobilisation, 29.1.19.
Capt., 10.4.17.
Bamforth, H., Sec.-Lieut. Joined, 29.4.18. Demobilised on leave, 3.12.18.
*Bell, B. A., Lieut. Wounded near Fleurbaix, 26.5.15.
Benson, G. W., Sec.-Lieut. Joined, 4.1.16. Transferred to R.F.C., 10.7.16.
Bentley, J. E., Sec.-Lieut. Joined, 29.4.18. Killed in action near Roeux,
11.10.18.
Beswick, N.S., Sec.-Lieut. Joined, 22.6.15. Bn. M.G.O. To hospital sick,
15.8.15.
Biddle, V., Sec.-Lieut. Joined, 16.10.18. Transferred to 13th Bn. D. of W.
Regt., 26.2.19.
Binns, C. E., Sec.-Lieut. Promoted from the ranks; joined as an officer,
16.2.17. Wounded (gas) near Hulluch, 28.6.17. Rejoined, 9.10.18. Left
Bn. for demobilisation, 7.2.19.
Lieut., 16.8.18.
Blackwell, F. V., Sec.-Lieut. Joined, 7.1.16. Transferred to 147th T.M.B.,
12.8.16.
Blakey, E. V., Sec.-Lieut. Joined, 28.5.17. Wounded at Johnstone’s Post,
4.7.16. Rejoined, 21.9.16. O.C. C Company (October–December, 1917).
Wounded near Molenaarelsthoek, 27.12.17.
Lieut., 1.6.16. A/Capt., 20.7.17.
Booth, W. S., Sec.-Lieut. Joined, 25.5.15. Bn. Bombing Officer (1916). Killed
in action near Thiepval, 8.7.16.
Brabham, J. R. S., Sec.-Lieut. (6th Bn. Norfolk Regt.). Joined, 20.8.17. Struck
off strength of Bn. (sick in England), 27.1.18.
Bradley, W. G., Sec.-Lieut. Joined, 29.4.18. Left Bn. for demobilisation,
21.3.19.
Brice, A. E., Sec.-Lieut. Joined, 16.10.18. Transferred to 13th Bn. D. of W.
Regt., 26.2.19.
Broomhead, W. N., Hon. Lieut. and Q.M. (T.D.). Joined Bn. from 6th Bn.
Sherwood Foresters, 15.2.17. Demobilised with Cadre of Bn., 19.6.19.
Capt., 23.12.17.
Broster, R. B., Sec.-Lieut. Joined, 1.3.17. O.C. C Coy. (May–October, 1918).
Killed in action near Roeux, 11.10.18.
A/Capt., 30.12.17. Lieut., 19.6.18.
Burgoyne, H. E., Sec.-Lieut. Joined, 29.1.18. Wounded near Zillebeke,
20.6.18.
 Butler, A., Sec.-Lieut. Joined, 8.8.16. Bn. and 147th Inf. Bde. Bombing
Officer. Wounded (gas) near Nieuport, 10.8.17.
Campbell, G., Sec.-Lieut. (4th Bn. East Yorks, Regt.). Joined, 18.9.17. Struck
off strength of Bn. (sick in England), 30.8.18.
*Chambers, E. P., Major. Second in Command (August, 1914–May, 1915 and
September, 1915–April, 1916). O.C. Bn. (May–September, 1915).
Appointed Claims Officer, 49th Div., 3.4.16.
T/Lieut.-Col., 24.6.15.–15.9.15.
Charlesworth, A., Sec.-Lieut. Joined, 29.4.18. Wounded near Vlamertinghe,
26.6.18.
Chippindale, F. D., Sec.-Lieut. Joined, 19.5.17. Killed in action near
Erquinghem, 10.4.18.
Chisnall, F., Sec.-Lieut. Joined, 13.1.16. Transferred to 147th M.G. Coy.,
4.2.16.
Clarke, E., Sec.-Lieut. Joined, 29.1.18. Left Bn. for demobilisation, 18.3.19.
Copeland, J., Sec.-Lieut. Joined, 21.9.16. To hospital sick, 19.11.16.
Court, G. B., Sec.-Lieut. Joined, 28.12.15. Shell shock, 5.7.16.
Crickmer, B., Sec.-Lieut. Joined, 29.4.18. Transferred to 13th Bn. D. of W.
Regt., 22.4.19.
Crowther, G., Sec.-Lieut. Promoted from the ranks; joined as an officer,
14.3.16. Bn. I.O. and A/Adjt. (September, 1916–Mch. 1917). Killed in
action near Hulluch, 28.6.17.
Denby, I. C., Sec.-Lieut. Joined, 25.9.16. Killed in action near Hulluch,
27.6.17.
*Denning, W. F., Capt. O.C. D Coy. (June–December, 1915). To hospital sick,
9.12.15.
Donkersley, P., Sec.-Lieut. Joined, 19.5.17. Wounded near Vlamertinghe,
29.11.17.
Drew, E. B., Hon. Lieut, and Q.M. Joined, 25.11.16. To hospital sick, 31.1.17.
Eade, W. M., Hon. Capt. and Q.M. (6th Bn. Suffolk Regt.). Joined, 18.5.18.
Transferred to 1/4th Bn. Seaforth Highlanders, 20.7.18.
Edwards, A. C., Sec.-Lieut. Joined, 28.10.17. Attached to 147th T.M.B.,
6.5.18.
Entwhistle, J. W., Sec.-Lieut. Joined, 29.4.18. To hospital sick, 25.9.18.
Everitt, W. N., Sec.-Lieut. Joined, 13.9.15. O.C. A Coy. (December, 1915–
September, 1916). Killed in action N. of Thiepval, 3.9.16.
T/Lieut., 31.12.15. T/Capt., 24.6.16.
Farrar, N. T., Sec.-Lieut. Joined, 17.2.16. O.C. B Coy. (June, 1917–February,
1919). Second in Command (February–April, 1919). Left Bn. for
demobilisation, 3.4.19.
Lieut., 1.6.16. A/Capt., 12.7.17. A/Major, 7.2.19.
*Fenton, W. C., Sec.-Lieut. Wounded near Fleurbaix, 30.4.15. Rejoined,
8.1.16. Adjutant (February, 1916–June, 1918). Wounded on Belle Vue
 Spur, 9.10.17. Rejoined, 23.10.17. Second in Command (June, 1918–
January, 1919). Left Bn. for demobilisation, 21.1.19.
T/Lieut., 10.2.16. Capt., 1.6.16. A/Major, 4.6.18.
Fenton, D. H., Sec.-Lieut. Promoted from the ranks; accidentally killed before
commission announced, 8.9.15.
*Fielding, T., Hon. Lieut, and Q.M. To hospital sick, 21.8.16.
Flatow, E. W., Sec.-Lieut. Joined, 15.11.15. Wounded near Nieuport, 9.8.17.
Lieut., 1.6.16.
Fleming, F. W. O., Sec.-Lieut. Joined, 25.9.15. Gassed near Ypres and died in
hospital, 19.12.15.
Fletcher, J., Lieut. Joined, 25.11.18. Transferred to 13th Bn. D. of W. Regt.,
26.2.19.
Geldard, N., Sec.-Lieut. (6th Bn. D. of W. Regt.). Joined, 25.9.16. O.C. D
Coy. (October, 1916–October, 1917). Wounded at Nieuport, 4.8.17.
Rejoined, 11.9.17. Wounded on Belle Vue Spur, 9.10.17. Capt., 30.11.16.
Gilroy, T. T., Sec.-Lieut. Joined, 15.10.17. Wounded (at duty) near Kemmel,
25.4.18. Transferred to 13th Bn. D. of W. Regt., 22.4.19. Lieut., 1.2.19.
A/Capt., 13.2.19.
Grantham, W., Lieut. Joined, 9.12.17. Wounded and missing near Roeux,
11.10.18. Died of wounds, a prisoner in enemy hands.
A/Capt., 29.4.18.
Gumby, L., Sec.-Lieut. Joined, 9.9.17. A/Adjt. (May–September, 1918).
Attached to 147th Inf. Bde. H.Q., 13.9.18.
Hanson, H., Capt. (5th Bn. D. of W. Regt.). Joined, 23.5.17. Wounded near
Vlamertinghe and died of wounds, 1.12.17.
Hartley, J. A., Sec.-Lieut. Joined, 22.6.15. Gassed near Ypres and died in
hospital, 19.12.15.
Hatch, H. S., Sec.-Lieut. Joined, 8.8.16. To hospital sick, 3.12.17. Rejoined,
November, 1918. Left Bn. for demobilisation, 3.4.19. Lieut., July, 1918.
Hill, G.M., Sec.-Lieut. Joined, 21.9.16. Transferred to 2nd Bn. D. of W.
Regt., 23.10.16.
Hinton, W. E., Sec.-Lieut. Joined, 20.11.15. Wounded near Ypres, 11.12.15.
Rejoined, 8.6.16. To hospital sick, 24.7.16.
*Hirst, C., Lieut. Bn. T.O. (April–August, 1915). O.C. B Coy. (December,
1915–September, 1916). Killed in action N. of Thiepval, 3.9.16.
T/Capt., 22.1.16.
Hirst, A. E., Sec.-Lieut. Promoted from the ranks; joined as an officer,
26.3.16. Killed in action N. of Thiepval, 3.9.16.
Hirst, W. L., Sec.-Lieut. Joined, 25.11.16. Left Bn. for demobilisation, 1.2.19.
Lieut., 1.7.17.
Holme, R. E., Sec.-Lieut. Joined, 29.4.18. Transferred to 147th T.M.B.,
18.7.18.
 Holt, J. W., Sec.-Lieut. Joined, 25.9.16. Transferred to 8th Bn. D. of W.
Regt., 26.10.16.
Horsfall, V. A., Sec.-Lieut. Joined, 15.7.16. Killed in action N. of Thiepval,
3.9.16.
Hothersall, T., Sec.-Lieut. Joined, 16.10.18. Left Bn. for demobilisation,
28.1.19.
Huggard, B. H., Sec.-Lieut. Joined, 2.11.17. Wounded near Nieppe, 11.4.18.
Rejoined, 4.5.18. Left Bn. for demobilisation, 2.3.19. Lieut., February,
1919. A/Capt., February, 1919.
Hutton, T., Lieut. (M.C.). Joined, 19.8.18. O.C. D Coy. (November, 1918–
February, 1919). Re-posted to 2nd Bn. D. of W. Regt., 12.2.19. A/Capt.,
26.10.18.
Hyland, J. L., Sec.-Lieut. Joined, 29.4.18. Transferred to R.O.D., 31.12.18.
Illingworth, A. C., Sec.-Lieut. Joined, 14.4.16. To hospital sick, 26.10.16.
Innes, F. A., Sec.-Lieut. Joined, 25.9.15. Wounded near Ypres, 16.10.15.
Rejoined, 23.10.15. Attached to 147th Inf. Bde. H.Q., 9.8.16. Killed in
action in Thiepval Wood, 3.9.16.
Irish, F., Sec.-Lieut. Joined, 28.3.17. Bn. T.O. (October, 1917–May, 1919).
Left Bn. for demobilisation, 5.5.19.
Lieut., 25.7.18.
Jessop, T. E., Sec.-Lieut. Joined, 25.5.18. Wounded in action near Roeux,
11.10.18.
Johnson, L. L., Sec.-Lieut. (6th Bn. Norfolk Regt.). Joined, 20.8.17. Wounded
near St. Jans Cappel, 17.4.18.
Jones, R. E., Sec.-Lieut. (6th Bn. Manchester Regt.). Joined, 15.5.18.
Transferred to 13th Bn. D. of W. Regt., 26.2.19.
Jury, R., Sec.-Lieut. Joined, 19.5.17. Wounded by enemy bomb at
Dunkerque and died of wounds, 6.10.17.
Kelsall, F. H., Sec.-Lieut. Joined, 7.12.15. Wounded near Authuille, 29.2.16.
Rejoined, 28.5.16. To hospital sick, 22.7.17.
Lieut., 25.6.16.
*King, M. H., Lieut. Went to France as 147th Inf. Bde. I.O. and was extra-
regimentally employed continuously from that time.
Kirk, A., Sec.-Lieut. Joined, 21.9.16. A/Adjt. (September–December, 1916).
O.C. A Coy. (October, 1917–January, 1919). Demobilised on leave,
March, 1919.
Lieut., 1.6.16. A/Capt., 28.10.17.
Kitson, J. H., Sec.-Lieut. Joined, 29.1.18. Killed in action near Bailleul,
14.4.18.
*Learoyd, G. W. I., Lieut. Bn. M.G.O. (July, 1915–February, 1916). Transferred
to 147th M.G. Coy., 4.2.16.
Leddra, R. M., Sec.-Lieut. Joined, 29.4.18. Transferred to 13th Bn. D. of W.
Regt., 26.2.19.