On-Line Fault Diagnosis Method for Power Transformer Based on Missing Data Repair

Abstract

Data quality is an important factor affecting the accuracy of transformer fault diagnosis. In order to reduce the impact of missing data, an on-line fault diagnosis method using a loop iterations of improved k-Nearest Neighbour (kNN) and multi-class SVMs based on the missing data repair is proposed in this paper. In the kNN method, the improved Manhattan distance weighted by the negative exponent of the correlation coefficient is designed to measure the distance between samples. On one hand, the influence of the strong correlation indicators on the missing data can be highlighted to improve the accuracy of data repair. On the other hand, the improved Manhattan distance is suitable for an efficient search strategy based on the k-d tree which can achieve the fast search for massive historical data and meet the real-time demand of on-line diagnosis. Diagnosis test results show that the proposed method can keep the high diagnostic accuracy on the incomplete data and realize the efficient on-line fault diagnosis for transformers.