Design of Clustering Adaptive Fuzzy Control Based on Genetic Algorithm

In view of the boiler having a non-linear, non-minimum phase characteristics, instability, time-delay and load disturbance, the steam separator dynamic instability, sensor noise as well as precise mathematical model being difficult to establish, the paper presented a new type of clustering adaptive fuzzy controllers applied to non-linear, multi-interference and strong coupling of the boiler feed water systems. Improved genetic algorithm is used to optimize the parameters of the membership of fuzzy controller to enhance the self-adaptive of fuzzy controller. And adaptive clustering algorithm is used again to make up the non-adaptive and non-integrity of fuzzy control, So the boiler feed water system achieves self - adaptive control. Simulation results show that the program´s adaptability, real-time, robustness is very strong, Even if the peaking units changes operating conditions, the system still maintained a good control performance and operating results, So it has a strong practical engineering value.