Real-time multi-objective optimization of pump scheduling in water distribution networks using neuro-evolution
Shengwei Pei,Lan Hoang,Guangtao Fu,David Butler
DOI: https://doi.org/10.1016/j.jwpe.2024.106315
IF: 7
2024-10-14
Journal of Water Process Engineering
Abstract:Pump scheduling in water distribution networks (WDNs) influences energy efficiency and supply reliability. Recently, machine-learning technologies showed promise in real-time control. However, methods sorely focused on minimizing operational costs often limit decision-makers choices. This study introduces a real-time multi-objective optimization method for pump scheduling in WDNs employing neuro-evolution, with neural networks trained by NSGA-II. The approach explores neural network-based control policies to balance system resilience and operational cost, comparing their performance against two baseline methods using NSGA-II, i.e., scenario-specific optimization (SSO) and robust optimization (RO). Simulation results of the Anytown network show that neuro-evolution performs between SSO and RO in Pareto front hypervolume, with improved outcomes using a smaller-scale neural network and a larger population. Although neuro-evolution is inferior to RO in Pareto front sparsity, it performs better than SSO and RO in pipe failure scenarios. Under default water demand scenarios, over 99 % neuro-evolution policies effectively prevent water pressure deficiencies, surpassing SSO's 83 %. Across all testing scenarios, the least-cost neuro-evolution control policy shows a 2.7 % reduction in mean operational cost and achieves an approximately an 8 % higher minimum water supply ratio compared to RO. Neuro-evolution shows promise for multi-objective real-time scheduling but needs improved performance in Pareto front sparsity.
engineering, chemical, environmental,water resources