Machine Learning and Simulation-Optimization Coupling for Water Distribution Network Contamination Source Detection

Abstract

This paper presents and explores a novel methodology for solving the problem of a water distribution network contamination event, which includes determining the exact source of contamination, the contamination start and end times and the injected contaminant concentration. The methodology is based on coupling a machine learning algorithm for predicting the most probable contamination sources in a water distribution network with an optimization algorithm for determining the values of contamination start time, end time and injected contaminant concentration for each predicted node separately. Two slightly different algorithmic frameworks were constructed which are based on the mentioned methodology. Both algorithmic frameworks utilize the Random Forest algorithm for classification of top source contamination node candidates, with one of the frameworks directly using the stochastic fireworks optimization algorithm to determine the contamination start time, end time and injected contaminant concentration for each predicted node separately. The second framework uses the Random Forest algorithm for an additional regression prediction of each top node’s start time, end time and contaminant concentration and is then coupled with the deterministic global search optimization algorithm MADS. Both a small sized (92 potential sources) network with perfect sensor measurements and a medium sized (865 potential sources) benchmark network with fuzzy sensor measurements were used to explore the proposed frameworks. Both algorithmic frameworks perform well and show robustness in determining the true source node, start and end times and contaminant concentration, with the second framework being extremely efficient on the fuzzy sensor measurement benchmark network.

Quantifying Mixing in Sewer Networks for Source Localization

“…The dye tracing and numerical modeling undertaken here have been done with consideration for source localization, i.e., finding the location of an event later monitored downstream. One approach to source localization based on a downstream measurement is that of matching a downstream record against a downstream prediction (Sokáč 2018;Grbčić et al 2021). Zehnder (2021) identifies matching the spread of the recorded profile as a key factor.…”

“…(1) in their work. Both Grbčić et al (2021) and Zehnder (2021) used hydraulic and solute transport model outputs to train machine learning models to identify pollutant sources. These approaches tackle the challenges of unknown injection volume, type, and duration, but require output from accurate modeling of the system/flows.…”

Application of modified seagull optimization algorithm with archives in urban water distribution networks: Dealing with the consequences of sudden pollution load

“…L Grbčić et al [ 19 ] developed Machine Learning and Simulation-Optimization Coupling for Detecting Pollution Sources in Water Distribution Networks. Two frameworks were constructed, using the Random Forest algorithm for categorization and either the stochastic fireworks optimization algorithm or MADS for optimization.…”

Novel Eulerian Approach with Cellular Automata Modelling to Estimate Water Quality in a Drinking Water Network

“…Ortega et al (2020) use a Bayesian approach for the source of contamination determination, using an algorithm that reports the probability that each node is the source to explain the correlations between the sampling positions, defining a classification. Grbčić et al (2021) present a complete framework for identifying the source of contamination (with a machine learning algorithm), the contamination times (with a stochastic fireworks optimization algorithm), and the injected contaminant concentration (through optimization and Random Forest algorithm).…”