基于故障录波器信息的故障诊断系统

对人工神经网络（ANN）方法应用于变电站故障诊断进行了研究，提出了一种基故障录波器信息的故障诊断系统。该系统中，核心部分是三层前向BP网络分别对变电站内的变压器、线路和母线构造相应的故障诊断ANN模型，对每种保护、断路器和自动装置构造相应的动作性能诊断模块。仿真计算结果表明：基于波器信息的将ANN应用于变电站故障诊断的方法是一种有效的方法。

Implementation of Intelligent Fault Dianosis System Based on Fault Recorder Information

The application of Artificial Neural Network (ANN) to fault diagnosis of substation is studied. An implementation manner of fault diagnosis system based on fault recorder information is put forward. In the structure of the presented system three layer feed forward BP networks are used as the kernel part of the fault diagnosis. For transformers, lines and busbars in substation the corresponding ANN models for fault diagnosis are established, at the same time, the corresponding action performance diagnosis modules for all kinds of protections, circuit breakers and automation devices are created. Simulation results show that it is effective to apply ANN to substation fault diagnosis on the basis of fault recorder information.