基于层次分析法和神经网络电网安全态势分析

摘要

The power grid safety situation analysis is realized based on the evaluation index. Since the nonlinear relation⁃ship among the safety assessment indexes causes the bad effect of security situation analysis,an evaluation method of power grid security situation analysis based on analytic hierarchy process(AHP)and neural network is proposed. The AHP is used to layer the multiple logical structures of the power grid,and the factors effecting on the security situation is set with weight. And then the weight of each influence factor is acquired,and according to which the influence factor with high power grid security interfe⁃rence is selected. The neural network model was trained with historical evaluation data to accomplish the power grid safety as⁃sessment. Experimental results show that the improved method can evaluate the power grid security accurately,and has high evaluation precision.%对于电网的安全态势分析是依据评估指标完成的。由于安全评估指标间存在非线性关系，导致对安全态势分析的效果不佳。提出一种基于层次分析法和神经网络的电网安全态势分析评估方法，运用层次分析法对电网的多个逻辑结构进行分层，对影响安全态势的因素进行权值设定，获取各影响因素的权重，按照权重大小选择对电网安全干扰较高的影响因素，通过历史评估数据对神经网络模型进行训练后，完成电网的安全评估。实验结果说明改进方法可对电网安全进行准确评估，并且评估精度较高。