Dynamic Scheduling Algorithm Based on Evolutionary Reinforcement Learning for Sudden Contaminant Events Under Uncertain Environment

For sudden drinking water pollution event, reasonable opening or closing valves and hydrants in a distribution network (WDN), which ensures the isolation discharge of contaminant as soon possible, is considered an effective emergency measure. In this paper, we propose scheduling algorithm based on evolutionary reinforcement learning (ERL), can train good policy by combination computation (EC) (RL). Then, optimal guide operation real time sensor information, protect people from risk contaminated water. Experiments verify our achieve results effectively reduce impact events.