【作者】 柳强；

【导师】 赵国材；

【作者基本信息】 辽宁工程技术大学 ， 控制理论与控制工程， 2006， 硕士

【摘要】 在模糊CCC算法的基础上提出了一种改进措施,并通过理论分析和仿真分析说明了改进的模糊CCC算法的有效性;构造了神经网络实现了改进的模糊CCC算法——模糊CCC神经网络,给出了模糊CCC神经网络的学习算法;同时针对常规遗传算法(GA)的不足,提出了一种嵌入式协同进化算法(ECA),给出了其数学模型,同时给出了自适应交叉、变异算子以提高算法效率;结合ECA算法全局搜索能力强和BP算法局部搜索能力强的特点,提出了一种基于ECA的模糊CCC神经网络控制器的优化策略;针对汽包水位的特性以及传统方法的弊端,最后将该控制策略应用于锅炉汽包水位控制系统,并引入前馈模糊控制器来克服“虚假水位”现象,最后给出了仿真结果,验证了该控制策略的有效性与优越性。更多还原

【Abstract】 This paper proposes a improved algorithm based on improving a CCC algorithm of fuzzy controller and testifies the superiority of the improved algorithm by theory analysis and simulation results ; The paper designs a Neural Net to actualize the improved CCC algorithm——Fuzzy CCC Neural Net, and then gives a learning algorithm; And Aimed at the limitation of genetic algorithm, the paper proposes a embedded co-evolutionary algorithm (ECA) and gives its mathematical model while giving the adaptive crossover and mutation in order to improve the efficiency of the algorithm. And combining the superiority of the global search ability of ECA and the better local search ability of BP algorithm, this paper proposes a control strategy of Fuzzy CCC Neural Net based on ECA. Because of characteristics of boiler water system and the disadvantage of traditional method, the control strategy was applied to a boiler water control system, and adopts a feedforward fuzzy control to resolve the“falsehood water level”.And simulation results verify the availability and superiority of the proposed method.更多还原