Development of a Wastewater Treatment Digital Twin for Carbon Neutrality and Sustainable Aquatic Ecosystem Services
With the declining population, the shortage of skilled engineers in wastewater treatment is becoming a serious issue. At the same time, achieving carbon neutrality by 2050 and ensuring the sustainable provision of aquatic ecosystem services in receiving water bodies require more advanced wastewater treatment management, including energy-saving and energy-generating operations, as well as process control that maintains appropriate nutrient concentrations in treated effluent. In this study, we are developing a Hybrid ODE (ordinary differential equation) model that integrates artificial intelligence (AI) and machine learning with the internationally standardized wastewater treatment simulator, the IWA Activated Sludge Models. Hybrid ODE is an emerging concept that combines mechanistic models with AI technologies and is increasingly being applied across various fields. While balancing predictive accuracy, reproducibility, and interpretability remains a major challenge, we propose a novel model structure to address these issues. Ultimately, we aim to develop a real-time wastewater treatment digital twin capable of flexibly optimizing energy efficiency and nutrient control in response to continuously varying influent water quality and quantity.
Recent Publications:
- Hososhima, Y., Guang-yao Zhao, Fujita, M. (2025) Real-time estimation of nitrifying bacteria concentration by using ammonia sensors installed in circulating nitrification and denitrification process, Japanese Journal of JSCE, 81 (17), 25-25004 (in Japanese). doi: 10.2208/jscejj.25-25004.
- Zhao G, Furumai H, Fujita M (2025) Parallel hybrid ordinary differential equation for modeling biological phosphorus removal modified for enhanced predictive performance and physical interpretability, Journal of Environmental Management 380, 124932. doi: 10.1016/j.jenvman.2025.124932.
- Zhao G, Furumai H, Fujita M (2024) Supporting data–enhanced hybrid ordinary differential equation model for phosphate dynamics in municipal wastewater treatment, Bioresource Technology 409, 131217. doi: 10.1016/j.biortech.2024.131217.
- Zhao G, Suzuki S, Deng J, Fujita M (2022) Machine learning estimation of biodegradable organic matter concentrations in municipal wastewater, Journal of Environmental Management 323, 116191. doi: 10.1016/j.jenvman.2022.116191.
