Field Data–Based Methodology for Estimating the Expected Pipe Break Rates of Water Distribution Systems

Abstract

Presented in this paper is a field data-based probabilistic approach to quantifying the expected pipe break rates of water distribution systems. Uncertain demands and variations in the roughness of pipes during their service lives are described as random variables. Sample values of these random variables are generated and input to a distribution system model to determine the resulting minimum and maximum pressures in Monte Carlo simulations. Based on an estimated break rate-maximum pressure relationship, the sample maximum pressures obtained from a Monte Carlo simulation are transformed into a sample of break rates, and the expected pipe break rate can subsequently be determined. The sample minimum pressures are used to gain a better understanding of the distribution network. This probabilistic approach is used for a part of the City of Hamilton's network in Ontario, Canada. The results show that the frequency of low-pressure events is very small, but a higher minimum pressure criterion would inevitably increase expected pipe break rates. Local field data collection is necessary to use the proposed methodology. Savings resulting from reduced pipe break rates justify costs associated with data collection.

Pipe Break Rate Assessment While Considering Physical and Operational Factors: A Methodology based on Global Positioning System and Data-Driven Techniques

“…Ultimately, they found that results of LS-SVM was more accurate than those obtained by Feed Forward Neural Network (FFNN) and Generalized Regression Neural Network (GRNN) techniques. Ghorbanian et al (2016) proposed a probabilistic approach to assess pipes break rates occurred in a part of the City of Hamilton WDN in Ontario, Canada. The performance of the proposed technique results demonstrated that the frequency of low-pressure occurrences is significantly marginal whereas a higher minimum pressure criterion would unavoidably augment expected pipe break rates.…”

Optimization of Water Pressure of a Distribution Network within the Water–Energy Nexus

“…Burst frequency generally depends on the pipe age, pipe material, external load, and climate conditions. Different models are developed to investigate this dependency [13]. Lambert et al developed the following equation [55]:…”

“…Using local field data, Ghorbanian et al [13] proposed a probabilistic approach that considers uncertain demands and pipe roughness to determine the effects of pressure on burst frequency. The Monte Carlo method was used to implement this probabilistic approach in the water network of Hamilton, Ontario, Canada.…”