Determination of the PID controller parameters by Modified Binary Particle Swarm Optimization algorithm

Abstract

The control of the bed temperature which is the key variable in Circulating Fluidized Bed Boiler (CFBB) system is a challenging problem due to its characteristics of the strong nonlinearity, large time delay and time-varying. It is very difficult to tune its PID controller based on the traditional PID tuning methods such as the classic Z-N method. Therefore, a Modified Binary Particle Swarm Optimization (MBPSO) algorithm is introduced to tune the parameters of bed temperature controller of CFBB in this paper, which is easy to implement and works with excellent optimization ability. To test and verify the effectiveness of the proposed method, the control performance on CFBB's bed temperature control is compared with Z-N method, and the tuning methods based on the standard PSO and discrete binary PSO. The simulation results show that PID tuning based MBPSO is valid and outperforms the ones based on the original discrete binary PSO algorithm and Z-N method, while the standard PSO fails to tackle this problem.