论文部分内容阅读
为了克服基于模式识别的智能控制的不足 ,提出了一种把神经网络和遗传算法与基于模式识别的智能控制相结合的控制方法。新方法中的神经网络采用遗传算法进行学习和训练 ,用来预测控制效果 ,从而修正特征模式集中的控制参数与控制规则。仿真研究表明 ,新方法由于增加了自适应机构 ,降低了对特征模式集中的经验数据和控制参数的依赖程度 ,提高了系统的性能 ,且便于特征模式集的设计。
In order to overcome the deficiencies of intelligent control based on pattern recognition, a control method combining neural network and genetic algorithm with intelligent control based on pattern recognition is proposed. The neural network in the new method uses genetic algorithm to learn and train, which can be used to predict the control effect so as to correct the control parameters and control rules in the feature mode. The simulation results show that the new method can reduce the dependence on empirical data and control parameters in feature mode, improve the system performance and facilitate the design of feature mode set due to the addition of adaptive mechanism.