Research on A Fuzzy Clustering Partition Analysis Method for Circuit Fault Diagnosis
Xiaomin Xie, Yueqin Wang, Renwei Dou, Kun Hu, Ying Hong · 2021 3rd International Conference on Applied Machine Learning (ICAML) · 2021
To cluster circuits, in order to improve the fault resolution, reduce the scale of neural network, improve the generalization ability of neural network and reduce the computational complexity, the idea of hierarchical fault diagnosis is adopted. That is to say, the fault feature information is classified, and all fault modes are divided into several categories, and each type of fault is identified by a small-scale neural network. The specific mean fuzzy clustering with four clustering centers is carried out on the fault feature samples input to the neural network, and the voltage value of the test point is selected as the sample data in practical application.