论文部分内容阅读
基于零阶 T- S模糊模型 ,本文指出了一种实用的输入输出数据进行复杂系统的模糊神经元网络建模的方法 .该方法的主要目的是解决模糊神经元网络中结构辨识的困难 .一种所谓语言模型被用来寻找系统输入变量的最佳组合 .这样 ,使得在输入变量选择阶段 ,我们既不需要估计模糊模型的参数 ,也不需要确定模糊规则数 .由于预先运用基于 FCM的自适应模糊聚类方法确定模糊神经元网络合理的结构 ,并设置网络的初始权值 ,从而可提高网络的训练速度 .此外 ,为提高模糊神经元网络的辨识精度 ,本文还提出了一种改进的隶属函数 .一个实际建模问题的仿真验证了本文方法的有效性 .
Based on the zero-order T-S fuzzy model, this paper presents a practical method of input and output data for the complex system of fuzzy neural network modeling method. The main purpose of this method is to solve the problem of structural identification in fuzzy neural network. The so-called language model is used to find the best combination of input variables in the system so that we do not need to estimate the parameters of the fuzzy model or determine the number of fuzzy rules in the input variable selection stage. Adapting the fuzzy clustering method to determine the reasonable structure of the fuzzy neural network and setting the initial weights of the network can improve the training speed of the network.In addition, in order to improve the identification accuracy of the fuzzy neural network, an improved Membership function.A simulation of the actual modeling problem verifies the effectiveness of the proposed method.