基于决策树与脉冲神经膜系统的输电网故障诊断方法

为有效处理电网故障诊断过程的不确定和不完备信息，提出一种基于决策树与模糊推理脉冲神经膜系统的输电网故障诊断方法:首先采用权重网络分割法将电网分割为若干小型子网，再利用决策树算法对原始故障决策表进行训练，并约减故障信息，提取输电网故障产生式规则；然后利用模糊推理脉冲神经膜系统的强大知识并行推理和模糊信息处理能力，建立基于 FRSNPS 的故障诊断模型，实现输电网故障诊断；最后，以 IEEE14 节点标准系统为对象进行仿真实验和分析。实验结果表明，该方法在单类型和多类型故障信息丢失时，依然能够诊断出正确故障元件。

A Fault Diagnosis Method of Power Transmission Networks Based on Decision Trees and Spiking Neuron P Systems

To deal with the uncertain and incomplete fault information, a fault diagnosis method of power transmission networks based on decision trees and fuzzy reasoning spiking neural P system is proposed. Firstly, the target network is divided into several small subnets by the weight network segmentation method, and then the original fault decision table is trained by a decision tree algorithm to reduce the fault information and extract the fault production rules for the target transmission network. Then, fault diagnosis models based on fuzzy reasoning spiking neural P systems are built to find faulty sections. Finally, case studies are carried out with the IEEE 14 bus test system. Experimental results show that the proposed method can diagnose faulty sections with high fault tolerance for the fault information.