Estimation of leakage rate in railway station and depot water supply networks using an improved constrained independent component analysis

Railway stations and depots in China operate small, single-source, essentially constant-pressure water networks whose leakage rate is rarely assessed because district metered areas and complete records are usually unavailable. Conventional constrained independent component analysis separates mixed flow signals, but its reference vector depends on a manually chosen threshold and is prone to false convergence. The leakage reference is here fixed to a constant vector, exploiting the near-constant background leakage under stable pressure, and the consumption reference is solved automatically from the whitened observations by augmented Lagrange multipliers and Newton iteration with negentropy maximisation, removing manual tuning. Separation requires only the inlet flow and pressure; least-squares amplitude calibration uses a single known night value. The method was validated on an EPANET (US Environmental Protection Agency water distribution modelling software) digital twin, a laboratory platform (24 conditions) and measured data from an electric multiple unit depot. Mean consumption restoration errors were below 0.07, 1.49, and 5.58%; field leakage-rate deviations were about 4 and 7 percentage points for 1- and 3-day records, with consumption correlation always above 0.91. A bootstrap 99% interval for the known-value-selection bias was within ±2.15 points. The method provides low-cost leakage-rate estimates to support renewal decisions.

Authors

Institutions

Publication Details

Journal
Journal of Hydroinformatics
Published
2026-11-20
DOI
https://doi.org/10.2166/hydro.2026.336
Primary Topic
Water Systems and Optimization
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Estimation of leakage rate in railway station and depot water supply networks using an improved constrained independent component analysis

Wenhao Li, Fan Qi, Jianyu Li, Yilong Zhang
1 links
Journal of Hydroinformatics
Water Systems and Optimization
article

Estimation of leakage rate in railway station and depot water supply networks using an improved constrained independent component analysis

Wenhao Li, Fan Qi, Jianyu Li, Yilong Zhang
article en

Abstract

Railway stations and depots in China operate small, single-source, essentially constant-pressure water networks whose leakage rate is rarely assessed because district metered areas and complete records are usually unavailable. Conventional constrained independent component analysis separates mixed flow signals, but its reference vector depends on a manually chosen threshold and is prone to false convergence. The leakage reference is here fixed to a constant vector, exploiting the near-constant background leakage under stable pressure, and the consumption reference is solved automatically from the whitened observations by augmented Lagrange multipliers and Newton iteration with negentropy maximisation, removing manual tuning. Separation requires only the inlet flow and pressure; least-squares amplitude calibration uses a single known night value. The method was validated on an EPANET (US Environmental Protection Agency water distribution modelling software) digital twin, a laboratory platform (24 conditions) and measured data from an electric multiple unit depot. Mean consumption restoration errors were below 0.07, 1.49, and 5.58%; field leakage-rate deviations were about 4 and 7 percentage points for 1- and 3-day records, with consumption correlation always above 0.91. A bootstrap 99% interval for the known-value-selection bias was within ±2.15 points. The method provides low-cost leakage-rate estimates to support renewal decisions.

Journal of Hydroinformatics
China Railway Design Corporation (China) (CN)
China Railway
Clean water and sanitation, Industry, innovation and infrastructure
Water Systems and Optimization
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon
Estimation of leakage rate in railway station and depot water supply networks using an improved constrained independent component analysis — Wenhao Li, Fan Qi, et al. · Journal of Hydroinformatics (2026) | TGRS Research Map | TGRS