Leak Detection and Classification of Water Pipeline Data Using LSTM Auto-Encoder with Xavier Initialization

Daehee Lee; Hyunseung Choo; Jongpil Jeong

doi:10.1109/BCD57833.2023.10466341

Leak Detection and Classification of Water Pipeline Data Using LSTM Auto-Encoder with Xavier Initialization

Sungkyunkwan University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

4 Scopus citations

Abstract

Modern cities operate extensive water supply networks. In Korea, the water supply network coverage reaches 98.9%, but more than 10% of water is still leaking every year. Previously, leaks in water pipelines were detected using flow meters and pressure meters, but these methods could only detect leaks after they occurred, necessitating on-site investigations to accurately pinpoint the location of the leak. In the United States, aging water and sewage facilities incur economic losses of hundreds of billions of dollars every year. To address this issue, this paper proposes an LSTM (Long Short-Term Memory) Auto-Encoder model, enhanced with Xavier Initialization, that can detect leaks using water pipeline data and automatically identify the location and extent of the leakage. This model achieved 93.7% performance by applying the Xavier Initialization technique to the LSTM Auto-Encoder model and achieved relatively high accuracy compared to the model using only LSTM Auto-Encoder. The results of this paper can serve as an effective predictive tool for monitoring water pipeline leaks.

Original language	English
Title of host publication	2023 IEEE/ACIS 8th International Conference on Big Data, Cloud Computing, and Data Science, BCD 2023
Editors	Jongwoo Park, Ngo Thi Phuong Lan, Sungtaek Lee, Tran Anh Tien, Jongbae Kim
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	69-74
Number of pages	6
ISBN (Electronic)	9798350373615
DOIs	https://doi.org/10.1109/BCD57833.2023.10466341
State	Published - 2023
Event	8th IEEE/ACIS International Conference on Big Data, Cloud Computing, and Data Science, BCD 2023 - Ho Chi Minh City, Viet Nam Duration: 14 Dec 2023 → 16 Dec 2023

Publication series

Name	2023 IEEE/ACIS 8th International Conference on Big Data, Cloud Computing, and Data Science, BCD 2023

Conference

Conference	8th IEEE/ACIS International Conference on Big Data, Cloud Computing, and Data Science, BCD 2023
Country/Territory	Viet Nam
City	Ho Chi Minh City
Period	14/12/23 → 16/12/23

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 11 Sustainable Cities and Communities

Keywords

Auto-Encoder
Leak Detection
LSTM
Water Pipeline Data
Xavier Initialization

Access to Document

10.1109/BCD57833.2023.10466341

Cite this

Lee, D., Choo, H., & Jeong, J. (2023). Leak Detection and Classification of Water Pipeline Data Using LSTM Auto-Encoder with Xavier Initialization. In J. Park, N. T. P. Lan, S. Lee, T. A. Tien, & J. Kim (Eds.), 2023 IEEE/ACIS 8th International Conference on Big Data, Cloud Computing, and Data Science, BCD 2023 (pp. 69-74). (2023 IEEE/ACIS 8th International Conference on Big Data, Cloud Computing, and Data Science, BCD 2023). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/BCD57833.2023.10466341

Lee, Daehee ; Choo, Hyunseung ; Jeong, Jongpil. / Leak Detection and Classification of Water Pipeline Data Using LSTM Auto-Encoder with Xavier Initialization. 2023 IEEE/ACIS 8th International Conference on Big Data, Cloud Computing, and Data Science, BCD 2023. editor / Jongwoo Park ; Ngo Thi Phuong Lan ; Sungtaek Lee ; Tran Anh Tien ; Jongbae Kim. Institute of Electrical and Electronics Engineers Inc., 2023. pp. 69-74 (2023 IEEE/ACIS 8th International Conference on Big Data, Cloud Computing, and Data Science, BCD 2023).

@inproceedings{246597021753418abae49ca01200f067,

title = "Leak Detection and Classification of Water Pipeline Data Using LSTM Auto-Encoder with Xavier Initialization",

abstract = "Modern cities operate extensive water supply networks. In Korea, the water supply network coverage reaches 98.9\%, but more than 10\% of water is still leaking every year. Previously, leaks in water pipelines were detected using flow meters and pressure meters, but these methods could only detect leaks after they occurred, necessitating on-site investigations to accurately pinpoint the location of the leak. In the United States, aging water and sewage facilities incur economic losses of hundreds of billions of dollars every year. To address this issue, this paper proposes an LSTM (Long Short-Term Memory) Auto-Encoder model, enhanced with Xavier Initialization, that can detect leaks using water pipeline data and automatically identify the location and extent of the leakage. This model achieved 93.7\% performance by applying the Xavier Initialization technique to the LSTM Auto-Encoder model and achieved relatively high accuracy compared to the model using only LSTM Auto-Encoder. The results of this paper can serve as an effective predictive tool for monitoring water pipeline leaks.",

keywords = "Auto-Encoder, Leak Detection, LSTM, Water Pipeline Data, Xavier Initialization",

author = "Daehee Lee and Hyunseung Choo and Jongpil Jeong",

note = "Publisher Copyright: {\textcopyright} 2023 IEEE.; 8th IEEE/ACIS International Conference on Big Data, Cloud Computing, and Data Science, BCD 2023 ; Conference date: 14-12-2023 Through 16-12-2023",

year = "2023",

doi = "10.1109/BCD57833.2023.10466341",

language = "English",

series = "2023 IEEE/ACIS 8th International Conference on Big Data, Cloud Computing, and Data Science, BCD 2023",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "69--74",

editor = "Jongwoo Park and Lan, \{Ngo Thi Phuong\} and Sungtaek Lee and Tien, \{Tran Anh\} and Jongbae Kim",

booktitle = "2023 IEEE/ACIS 8th International Conference on Big Data, Cloud Computing, and Data Science, BCD 2023",

}

Lee, D, Choo, H & Jeong, J 2023, Leak Detection and Classification of Water Pipeline Data Using LSTM Auto-Encoder with Xavier Initialization. in J Park, NTP Lan, S Lee, TA Tien & J Kim (eds), 2023 IEEE/ACIS 8th International Conference on Big Data, Cloud Computing, and Data Science, BCD 2023. 2023 IEEE/ACIS 8th International Conference on Big Data, Cloud Computing, and Data Science, BCD 2023, Institute of Electrical and Electronics Engineers Inc., pp. 69-74, 8th IEEE/ACIS International Conference on Big Data, Cloud Computing, and Data Science, BCD 2023, Ho Chi Minh City, Viet Nam, 14/12/23. https://doi.org/10.1109/BCD57833.2023.10466341

Leak Detection and Classification of Water Pipeline Data Using LSTM Auto-Encoder with Xavier Initialization. / Lee, Daehee; Choo, Hyunseung ; Jeong, Jongpil.
2023 IEEE/ACIS 8th International Conference on Big Data, Cloud Computing, and Data Science, BCD 2023. ed. / Jongwoo Park; Ngo Thi Phuong Lan; Sungtaek Lee; Tran Anh Tien; Jongbae Kim. Institute of Electrical and Electronics Engineers Inc., 2023. p. 69-74 (2023 IEEE/ACIS 8th International Conference on Big Data, Cloud Computing, and Data Science, BCD 2023).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Leak Detection and Classification of Water Pipeline Data Using LSTM Auto-Encoder with Xavier Initialization

AU - Lee, Daehee

AU - Choo, Hyunseung

AU - Jeong, Jongpil

PY - 2023

Y1 - 2023

N2 - Modern cities operate extensive water supply networks. In Korea, the water supply network coverage reaches 98.9%, but more than 10% of water is still leaking every year. Previously, leaks in water pipelines were detected using flow meters and pressure meters, but these methods could only detect leaks after they occurred, necessitating on-site investigations to accurately pinpoint the location of the leak. In the United States, aging water and sewage facilities incur economic losses of hundreds of billions of dollars every year. To address this issue, this paper proposes an LSTM (Long Short-Term Memory) Auto-Encoder model, enhanced with Xavier Initialization, that can detect leaks using water pipeline data and automatically identify the location and extent of the leakage. This model achieved 93.7% performance by applying the Xavier Initialization technique to the LSTM Auto-Encoder model and achieved relatively high accuracy compared to the model using only LSTM Auto-Encoder. The results of this paper can serve as an effective predictive tool for monitoring water pipeline leaks.

AB - Modern cities operate extensive water supply networks. In Korea, the water supply network coverage reaches 98.9%, but more than 10% of water is still leaking every year. Previously, leaks in water pipelines were detected using flow meters and pressure meters, but these methods could only detect leaks after they occurred, necessitating on-site investigations to accurately pinpoint the location of the leak. In the United States, aging water and sewage facilities incur economic losses of hundreds of billions of dollars every year. To address this issue, this paper proposes an LSTM (Long Short-Term Memory) Auto-Encoder model, enhanced with Xavier Initialization, that can detect leaks using water pipeline data and automatically identify the location and extent of the leakage. This model achieved 93.7% performance by applying the Xavier Initialization technique to the LSTM Auto-Encoder model and achieved relatively high accuracy compared to the model using only LSTM Auto-Encoder. The results of this paper can serve as an effective predictive tool for monitoring water pipeline leaks.

KW - Auto-Encoder

KW - Leak Detection

KW - LSTM

KW - Water Pipeline Data

KW - Xavier Initialization

UR - https://www.scopus.com/pages/publications/85189935375

U2 - 10.1109/BCD57833.2023.10466341

DO - 10.1109/BCD57833.2023.10466341

M3 - Conference contribution

AN - SCOPUS:85189935375

T3 - 2023 IEEE/ACIS 8th International Conference on Big Data, Cloud Computing, and Data Science, BCD 2023

SP - 69

EP - 74

BT - 2023 IEEE/ACIS 8th International Conference on Big Data, Cloud Computing, and Data Science, BCD 2023

A2 - Park, Jongwoo

A2 - Lan, Ngo Thi Phuong

A2 - Lee, Sungtaek

A2 - Tien, Tran Anh

A2 - Kim, Jongbae

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 8th IEEE/ACIS International Conference on Big Data, Cloud Computing, and Data Science, BCD 2023

Y2 - 14 December 2023 through 16 December 2023

ER -

Lee D, Choo H , Jeong J. Leak Detection and Classification of Water Pipeline Data Using LSTM Auto-Encoder with Xavier Initialization. In Park J, Lan NTP, Lee S, Tien TA, Kim J, editors, 2023 IEEE/ACIS 8th International Conference on Big Data, Cloud Computing, and Data Science, BCD 2023. Institute of Electrical and Electronics Engineers Inc. 2023. p. 69-74. (2023 IEEE/ACIS 8th International Conference on Big Data, Cloud Computing, and Data Science, BCD 2023). doi: 10.1109/BCD57833.2023.10466341

Leak Detection and Classification of Water Pipeline Data Using LSTM Auto-Encoder with Xavier Initialization

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this