Research on modeling method of thermal system based on big data

In this paper, we managed to identify a typical thermal process of an ultra supercritical unit on the basis of DSC data obtained from a power plant rather than doing experiments. With screened reliable data and data processing, this system identification was performed by applying Particle Swarm Optimization(PSO) to establish mathematical model. As a result, the model-predicted data showed an excellent agreement with these data from actual operations. The model can be used to quantitively study the properties of thermal systems with various operating conditions, and it provides guidance for system optimization.