论文部分内容阅读
基于局部特征量的神经网络方法对变压器的故障诊断进行了新的探索，建立了相应的故障诊断的神经网络压缩模型；实例诊断结果证明了这种方法的有效性。本方法能够诊断传统的三比值法无法诊断的故障，对于其他系统的故障诊断，也具有一定的指导意义。
The neural network method based on local feature vectors is used to explore the fault diagnosis of the transformer. The corresponding neural network compression model of fault diagnosis is established. The result of the example diagnosis proves the effectiveness of this method. The method can diagnose the failure which can not be diagnosed by the traditional three-ratio method and has certain guiding significance for the fault diagnosis of other systems.