Multiobjective Operation Optimization of Wastewater Treatment Process Based on Reinforcement Self-Learning and...

Area of Science:

• Environmental Engineering
• Artificial Intelligence
• Process Optimization

Background:

• Wastewater treatment processes (WWTP) face challenges with nonstationary, time-varying dynamics impacting operational efficiency.
• Optimizing WWTP for both effluent quality and energy consumption requires advanced control strategies.

Purpose of the Study:

• To develop a multiobjective operation optimization method for WWTP.
• To reduce energy consumption and ensure effluent quality in dynamic wastewater treatment systems.

Main Methods:

• Online sequential random vector functional-link (OS-RVFL) neural networks for online parameter learning.
• A knowledge base storing typical optimization cases for guiding subsequent optimizations.
• Reinforcement self-learning-based multiobjective particle swarm optimization (RSL-MOPSO) algorithm with selective information feedback.

Main Results:

• The proposed RSL-MOPSO algorithm effectively adjusts particle motion trends based on optimization feedback.
• Recorded wastewater state parameters improve solution quality and calculation efficiency.
• The selective information feedback mechanism ensures algorithm diversity and convergence.

Conclusions:

• The developed method successfully reduces energy consumption in WWTP.
• The approach effectively ensures high effluent quality.
• Intelligent decision-making selects optimal setpoints from the Pareto frontier for controllers.