摘要
由于模糊聚类将故障样本等同进行模糊划分,且受初始值影响,故提出将PSO-WFCM算法用于变压器油中溶解气体的故障诊断。该算法选取油中气体作为故障特征量,利用粒子群算法得到最佳初始聚类中心,用以指导模糊聚类求取最终的聚类中心。实验结果表明,其弥补了模糊聚类的不足,还提高了变压器的诊断性能。
Because the samples are divided equally and the division is sensitive to the initial value in fuzzy clustering method, this paper applies PSO-WFCM algorithm to detection of dissolved gas for fault diagnosis of power transformer. In PSO-WFCM algorithm, the dissolved gas in oil is taken as the characteristic quantity of fault, and PSO is adopted to obtain the optimal initial cluster centers for achieving the final cluster centers via fuzzy clustering. Experiments show that the proposed method makes up for the deficiency of fuzzy clustering method, and improves fault diagnosis for transformer.
出处
《高压电器》
CAS
CSCD
北大核心
2014年第1期72-76,共5页
High Voltage Apparatus
