ETRI-Knowledge Sharing Plaform

KOREAN
논문 검색
Type SCI
Year ~ Keyword

Detail

Conference Paper Comparison of Leak Detection Techniques Using Water Pipeline Vibration Sensor Data
Cited 0 time in scopus Share share facebook twitter linkedin kakaostory
Authors
Mi-seon Kang, Kwang-Ju Kim, Jung-won Yu, Hyan-su Bae, In-Su Jang
Issue Date
2023-10
Citation
International Conference on Consumer Electronics (ICCE) 2023 : Asia, pp.33-36
Publisher
IEEE
Language
English
Type
Conference Paper
DOI
https://dx.doi.org/10.1109/ICCE-Asia59966.2023.10326371
Abstract
This paper conducts a comparative study on leak detection techniques using vibration sensor data installed in water pipeline networks. By applying machine learning techniques, we have implemented a technology to detect and classify leaks in water pipelines, thereby enabling the prediction of leaks in the water mains. The vibration sensor dataset used in this study consists of leak detection sensor data installed in water supply meters and valve chambers within the water supply network. This dataset includes data provided by the public data portal (AI hub) as well as investigation data provided by a company. The investigation data were collected in the field by leakage detection experts. We designed leak classification models using various machine learning techniques and measured and analyzed the performance of each dataset using feature importance.
KSP Keywords
Classification models, Feature Importance, Machine Learning technique(MLT), Pipeline networks, Public Data, Vibration sensor, Water mains, Water pipelines, Water supply network, comparative study, data portal