Leveraging Evolutionary Surrogate-Assisted Prescription in Multi-Objective Chlorination Control Systems

2025-08-26

AI Summary

Addressing the conflict between multi-objective optimization—ensuring residual chlorine compliance, minimizing disinfection by-products (DBPs), and reducing energy consumption—and prohibitively costly trial-and-error in real-world drinking water chlorination systems, this paper proposes the Evolutionary Surrogate-assisted Prescription (ESP) framework. ESP uniquely integrates evolutionary algorithms, interpretable surrogate modeling, and multi-objective Bayesian optimization to establish a simulation-real-world co-training paradigm for operational strategy development. This integration significantly enhances policy generalizability and transfer efficiency under dynamic water quality disturbances. Evaluated on real-world data from an operational water treatment plant in the inaugural IJCAI-2025 Drinking Water Chlorination Challenge, ESP achieved a 32% reduction in residual chlorine control error and a 27% decrease in trihalomethane (THM) formation compared to baseline methods. These results demonstrate both the engineering feasibility and methodological novelty of ESP, advancing the state of the art in intelligent, adaptive water disinfection control.

Technology Category

• Search and Optimization: Mixed Discrete/Continuous Search
• Intelligent Robots: Learning & Optimization for ROB
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

This short, written report introduces the idea of Evolutionary Surrogate-Assisted Prescription (ESP) and presents preliminary results on its potential use in training real-world agents as a part of the 1st AI for Drinking Water Chlorination Challenge at IJCAI-2025. This work was done by a team from Project Resilience, an organization interested in bridging AI to real-world problems.

Problem

• Optimizing multi-objective chlorination control using evolutionary methods
• Training real-world agents for water treatment systems
• Bridging AI solutions to practical drinking water challenges

Innovation

• Evolutionary Surrogate-Assisted Prescription for multi-objective control
• AI agents training for drinking water chlorination systems
• Bridging artificial intelligence to real-world water challenges