System Technology & Eco-friendly Processes (STEP) Lab at Pusan National University develops system technologies such as modeling, optimization, model predictive optimal control, data analysis, and machine learning, then applies them to process systems to achieve various objectives such as production enhancement, energy saving, reduction of carbon emission, and real time control. In particular, we are focusing on developing eco-friendly process-related technologies to prevent ecosystem change and realize sustainable development.
Efficiency enhancement in eco-friendly processes through systems engineering approach
The need for environmental technologies and eco-friendly processes to prevent climate and ecosystem changes caused by pollution and specific hazardous substances and realize sustainable industrial development is growing globally. As carbon emission policy becomes an issue, various processes have been developed to capture and utilize carbon dioxide. In particular, Korea is the 6th largest emitter of greenhouse gases in the world, so it is urgent to establish a national strategy with the launch of the Paris Climate Agreement. STEP laboratory develops fundamental models to accurately predict the state change of various eco-friendly process systems, such as carbon capture and utilization processes and electrochemical hydrogen production processes, and derives optimal operating conditions and optimal control policy to improve the yield and energy efficiency of processes by applying system technologies based on the constructed model and measurement data.
Smart plant technologies based on fundamental model and artificial intelligence
Based on the accumulation of big data via recent development of information and communication technology, the development of computational resources, and the development of artificial intelligence (AI) through the development of deep learning algorithms, the 4th industrial revolution is affecting the global industry. In the chemical process systems engineering field, smart plant technology is emerging where the upstream and downstream processes are connected in real time beyond the factory automation in the simple O&M manner, process information is stored as big data, and the artificial intelligence-based systems actively operate the plant via data-driven learning and adaptation. STEP laboratory develops smart plant techniques combining model-based and data-driven frameworks where the basic model-based optimization methods built on the prior knowledge of the system is integrated with the data-driven methods to more efficiently derive optimal operating conditions and control policy by overcoming the limitations of the existing smart plant technologies that rely almost on process data. By this, we plan to secure more advanced original technologies and lay the groundwork for the domestic smart plant industry to take a further leap forward.
Database establishment of digital twins for eco-friendly processes
Despite the obvious environmental benefits of recently developed eco-friendly processes, most are still in the basic research stage at lab-scale. STEP laboratory develops fundamental models that can accurately represent the dynamic characteristics of these processes, constructs digital twins by coding, and then establishes the database for commercialization.