Study on optimization control method of bed temperature for circulating fluidized bed

The combustion system of Circulating Fluidized Bed (CFB) boiler is a strong-coupled non-linear system with large delay. Bed temperature is one of the most important parameters, which influences the safety of unit and couples with the main steam pressure strongly. A reliable control method is beneficial to suppress the temperature fluctuation. In this paper, a fuzzy PI controller based on Genetic Algorithm (GA) is designed for the model of bed temperature, of which the parameters are optimal in the region to ensure the control effect. On the basis of mechanism analysis and gray correlation analysis, a multivariable model is built. The coal feed and the primary air are selected as inputs, the bed temperature and main steam pressure are selected as outputs. Then the model is decoupled by partial feed-forward method, and the GA is used to optimize the parameters of the fuzzy PI controller. Simulations are carried out in different conditions and the results indicate that the performance of optimized fuzzy PI controller based on proposed model is better than fuzzy PI controller.