Offline RL control for semiconductor wastewater treatment (JCR: 6.5%, IF: 6.7)

Prof. SungKu Heo (Corresponding author) published a research article on offline reinforcement learning (RL)-based autonomous operation for semiconductor wastewater treatment in the Journal of Water Process Engineering. This study addresses the challenge of treating wastewater containing tetramethylammonium hydroxide (TMAH), a toxic compound that inhibits nitrification and disrupts conventional biological processes. The proposed offline RL-based feedforward control system enables adaptive and data-driven operation under diverse influent conditions, improving cost efficiency, effluent quality, and TMAH removal performance compared to fixed-schedule strategies. The findings highlight the potential of RL as a robust decision-support framework for intelligent and autonomous wastewater treatment systems. The paper is accessible via the following DOI link: https://doi.org/10.1016/j.jwpe.2026.109875.

Greenest A.I. Lab
Greenest A.I. Lab
Led by Prof. SungKu Heo

We work for digital and autonomous solutions for a climate-resilient future, and push boundaries across autonomous systems, decarbonization, and circular innovation.