Transformer oil chromatography anomaly identification method based on random matrix

• Main Authors: GAO YANG, ZHU TAO, ZHANG YIZHOU, HE XING, AI QIAN, WAN YILUN, TANG LEI, PAN MEIQI
• Format: Patent
• Language: Chinese; English
• Published: 28.05.2024
• Subjects: INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIRCHEMICAL OR PHYSICAL PROPERTIES; MEASURING; PHYSICS; TESTING

The invention relates to the technical field of transformers, in particular to a transformer oil chromatography anomaly identification method based on a random matrix, which comprises the following steps: S1, reducing transformer oil chromatography sample data; s2, simulating an oil chromatographic data negative sample based on the generative model; and S3, constructing an oil chromatography anomaly identification method by using a random matrix big data analysis tool. According to the method, the rough set is adopted to reduce the dimension of the oil chromatography data, and compared with the mode that all the oil chromatography data are directly used as input data, the number of the reduced data is smaller, and the dimension is lower; the WGAN-GP is adopted to generate oil chromatography negative sample data, so that the number of positive and negative samples can be balanced, and the detection effect on the transformer oil chromatography abnormal sample is improved; the high-dimensional features of the oi