An adaptive fault diagnosis method of power transformers based on combining oversampling and cost‐sensitive learning. Issue 6 (9th June 2021)

• Record Type: Journal Article
• Authors: Zhang, Lijing; Sheng, Gehao; Hou, Huijuan; Zhou, Nan; Jiang, Xiuchen
• Is Part Of: IET smart grid. Volume 4:Issue 6(2021)
• Page Start: 623
• Page End: 635
• Publication Date: 2021-06-09
• Subjects: learning (artificial intelligence) -- belief networks -- pattern classification -- fault diagnosis -- power transformers -- power engineering computing -- data analysis -- power system faults; Smart power grids -- Periodicals; Computer science -- Periodicals; Energy industries -- Periodicals; Broadcasting -- Periodicals; 333.79110285
• DOI: 10.1049/stg2.12044
• Languages: English
• ISSNs: 2515-2947

Abstract

Abstract: Dissolved gas analysis is an important technique for the insulation condition assessment and incipient fault diagnosis of power transformers. However, the performance of the traditional ratio methods can be hardly improved due to the overreliance on absolute ratio threshold. In this paper, a novel method combining oversampling and cost‐sensitive learning is proposed to improve the diagnosis accuracy of all fault types of transformers. The radial‐based oversampling (RBO) method is adopted to synthesise samples for the complex fault classes. With the newly generated samples, the deep belief network (DBN) can effectively learn the features of complex fault classes and distinguish them from the other fault classes. Moreover, by integrating a cost matrix into the loss function, the parameters of DBN are adaptively updated so as to ensure the correct classification of the fault class with less samples. Based on the oversampling and cost‐sensitive learning, the proposed method can form suitable classification boundaries amongst thermal, discharge and complex fault classes. The effectiveness and generalisation capability of the proposed method are verified by case studies in a real‐world fault dataset of power transformers with multi‐source samples. The results demonstrate that the proposed method improves the classification accuracies in all fault classes, especially in the complex fault classes. The overall accuracy can be reach over 90% by applying both RBO and cost‐sensitive learning.