Robustness of neuro-evolution in urban drainage system control under communication failures: comparing centralized and decentralized control schemes

Water Research, Journal Year: 2025, Volume and Issue: 282, P. 123733 - 123733

Published: April 27, 2025

Real-time control (RTC) in urban drainage systems can effectively mitigate flooding and Combined Sewer Overflow (CSO) spills. Recently neuro-evolution has shown promise RTC, which relies on communication to receive real-time state information send signals. However, the impact of system failures this approach is not fully understood. This study aims evaluate robustness for operation under various failure scenarios, focusing both centralized decentralized schemes. The considered include transient disruptions observation or action process prolonged sensor disconnections. simulation results from Astlingen benchmarking network indicate that performance total CSO volume reduction ranks as follows: > baseline strategy: Equal Filling Degree (EFD). In terms robustness, outperforms disconnections, while excels handling maintaining local stability during Nevertheless, surpass EFD strategy smaller effectiveness degradations reduced variability. provides insights into failures, especially respective advantages schemes, contributing development more resilient systems.

Language: Английский

Leveraging LSTM-based neuro-evolution for enhanced real-time control in urban drainage systems

Water Research X, Journal Year: 2025, Volume and Issue: unknown, P. 100353 - 100353

Published: May 1, 2025

Language: Английский