摘要
针对网络化控制系统中模糊控制器的量化因子和比例因子采用传统经验方法难以整定的问题,提出了一种改进量子粒子群(IQPSO)算法对模糊控制器量化因子和比例因子进行优化。该方法将ABC算法中的搜索算子作为变异算子引入到QPSO算法中,使得IQPSO算法较好地克服了QPSO算法保持种群多样性差容易早熟收敛的缺陷,并以ITAE指标作为IQPSO算法的适应度函数对模糊控制器进行优化。典型工业过程仿真结果表明,IQPSO优化的模糊控制器具有比PID控制器和标准QPSO优化的模糊控制器更好的控制性能和适用性。
The traditiaonal experience method is difficult to design measure-parameter and scal-parameter of fuzzy controller in networked control systems(NCS) , this paper advanced a new method to select these parameters depended on improved quantum particle swam optimization ( IQPSO ) algorithm. To improve the performance of quantum particle swam optimization (QPSO) algorithm, this paper proposed an adaptive mutation QPSO algotithm based on search operator of artificial bee colony(ABC) al- gorithm. The method used the ITAE indx as the fitness function of the IQPSO algorithm to optimize the fuzzy controller parame- ters. Simulation results of the typical indusrieal process show that the optimal fuzzy controller using IQPSO has better control performance and adaptability than the optimal fuzzy controller using QPSO and the optimal PID controller using IQPSO.
出处
《计算机应用研究》
CSCD
北大核心
2013年第8期2301-2303,2314,共4页
Application Research of Computers
基金
国家自然科学基金资助项目(60964003)