基于粗集规则编码的神经网络控制器设计

Design of Neural Network Controller Based on Rule Encoded by Rough Sets

摘要: 利用通过粗糙集产生的控制规则对神经子网络进行编码,用遗传算法独立进化每一个子网络,把进化后的子网络用改进的遗传算法通过适当连接形成最后的神经网络。利用该神经网络进行在线控制,并和PID控制效果相比较,证明了其有效性。

Abstract: The control rules for encoding neural sub-networks are generated by rough sets. The neural sub-networks are evolved and combined into an integrated neural network by genetic algorithm. Such neural network can be used as a controller for control systems online. The simulation results prove the effectiveness of the control mechanism.