Environmental Process Modeling

Environmental processes involve complex interactions among physical, chemical, biological, and operational mechanisms, often exhibiting nonlinear and dynamic behavior.

Our research group develops mathematical and computational models for environmental systems by integrating mechanistic knowledge with data-driven approaches. A major focus is the development of digital twins and physics-informed models for prediction, interpretation, optimization, and control.

Specifically, we focus on:

• Mathematical modeling of environmental, chemical, and biological processes.

• Digital twins for environmental process systems.

• Physics-informed and hybrid modeling.

• Dynamic modeling and prediction of wastewater treatment processes.

• System identification, calibration, and parameter estimation.

• Explainable modeling of physical and biochemical mechanisms.

Environmental Systems Engineering

Modern environmental systems must achieve treatment performance, operational stability, energy efficiency, and sustainability simultaneously.

Our research group develops intelligent optimization and control frameworks by integrating process modeling, artificial intelligence, mathematical optimization, and control. We aim to advance environmental systems toward predictive, adaptive, and autonomous operation while considering physical, biological, and operational constraints.

Specifically, we focus on:

• Process optimization for environmental systems.

• AI-based supervisory and autonomous control.

• Reinforcement learning and multi-agent reinforcement learning.

• Physics-informed reinforcement learning.

• Predictive control and real-time decision support.

• Fault detection, sensor reconstruction, and resilient operation.

• Reliability- and risk-aware environmental system operation.

• Sustainable operation considering energy, carbon emissions, and operating costs.

• ESG-oriented environmental system assessment.

Renewable Energy System Optimization

Achieving carbon neutrality requires intelligent coordination of renewable energy generation, storage, energy demand, and system operation.

Our research group develops forecasting, optimization, and decision-support methods for renewable energy systems, with particular emphasis on RE100, net-zero, and carbon-neutral strategies for environmental and industrial systems.

Specifically, we focus on:

• Renewable energy generation and demand forecasting.

• Optimal integration of solar, wind, and energy storage.

• RE100 and net-zero transition strategies.

• Integrated optimization of energy generation, storage, and consumption.

• Intelligent energy management for environmental infrastructure.

• Techno-economic and environmental assessment.

• Multi-objective optimization of energy, cost, and carbon emissions.

Environmental Big Data & AI

Environmental systems generate large volumes of heterogeneous data from sensors, monitoring networks, laboratory measurements, and meteorological observations. These data are often nonlinear, nonstationary, high-dimensional, and incomplete.

Our research group develops statistical and machine-learning approaches to extract meaningful information from environmental data and support prediction, monitoring, and decision-making.

Specifically, we focus on:

• Environmental time-series forecasting.

• Water-quality, air-quality, and hydrological prediction.

• Multivariate and multimodal environmental data analysis.

• Spatiotemporal modeling.

• Advanced deep-learning models for environmental forecasting.

• Anomaly detection and early-warning systems.

• Data-driven soft sensors and sensor reconstruction.

• Explainable AI for identifying environmental drivers and mechanisms.

Environmental Computer Vision & Bio-Vision-AI

Images and videos provide valuable information about biological organisms, environmental conditions, behavioral responses, and abnormal events that cannot always be captured using conventional sensors.

Our research group develops Environmental Computer Vision and Bio-Vision-AI methods for automated monitoring, interpretation, and decision support in environmental and biological systems.

Specifically, we focus on:

• Detection, classification, and segmentation of biological organisms.

• Vision-based environmental and biological monitoring.

• Object tracking in complex biological environments.

• Analysis of behavioral and morphological changes.

• Detection of disease, stress, and abnormal conditions.

• Computer vision for smart aquaculture systems.

• Multimodal learning integrating visual and environmental sensor data.

• Explainable computer vision and Bio-Vision-AI.