论文部分内容阅读
为了解决实际电力系统运行条件经常变化和系统中存在不确定性等问题，提出使用神经网络和模糊控制理论以实现具有自学习功能的电力系统神经模糊负荷频率控制。控制器主要由一个含有两个隐含层的神经网络组成，连同输入层和输出层一起顺序完成模糊控制的每一步操作。在定义了适当的目标函数之后，采用ＢＰ算法，神经网络可自动由受控系统获取学习样本，并对网络的参数（连接权）进行优化，仿真结果表明了该方法的有效性。
In order to solve the problems of frequent changes of operating conditions and uncertainties in the system, the neural network and fuzzy control theory are proposed to realize the self-learning power system fuzzy control based on the fuzzy frequency. The controller is mainly composed of a neural network with two hidden layers, together with the input layer and the output layer to complete each step of fuzzy control in order. After the appropriate objective function is defined, the BP algorithm is used. The neural network can automatically obtain the learning samples from the controlled system and optimize the network parameters (connection rights). The simulation results show the effectiveness of the method.