基于模糊神经网络决策树的电压稳定性评估

Evaluation of Power System Voltage Stability Based on Fuzzy Neural Network Decision Tree

提出了适用于电力系统电压稳定性评估的模糊神经网络决策树模型。在生成的决策树中引入模糊神经网络技术,构建出模糊神经网络决策树模型。采用模糊神经网络中的前向神经网络BP算法对小分裂样本进行进一步处理。Matlab仿真结果表明,模糊神经网络决策树应用于电压稳定性评估中,与单独应用决策树相比,分类成功率提高了3.3%。同时,由于减小了样本维度,缩短了模糊神经网络的训练时间,更有利于实现在线电压稳定预测。

A fuzzy neural network decision tree model suitable for evaluation of power system voltage stability is proposed. By means of leading fuzzy neural network technique into the generated decision tree, a fuzzy neural network decision tree model is constructed. Using BP algorithm of feed-forward neural network in fuzzy neural network, the small disjunction samples are further processed. The results of simulation by Matlab show that applying fuzzy neural network decision tree to voltage stability evaluation the success rate of classification is 3.3% higher than that by applying decision tree method singly. Meanwhile, because the dimensionality of samples is reduced, the training time of fuzzy neural network is saved, so it is favorable to implement online voltage stability prediction.