Oil Pollution Risk Identification Method of Oilimmersed Transformer Based on Attention Mechanism and Maximum Entropy

To accurately grasp the change of oil pollution risk of the oil-immersed transformer, a risk identification method of oil pollution of the oil-immersed transformer based on attention mechanism and maximum entropy is proposed. The data of each gas in the oil chromatography of an oil-immersed transformer is processed to determine the data acquisition results of the oil pollution risk of the oil-immersed transformer. An attention mechanism and maximum entropy pretreat the oil pollution risk of an oil-immersed transformer, and the identification steps for the oil pollution risk of the oilimmersed transformer are constructed. The experimental results show that the oil-immersed transformer is relatively stable during identification, with no obvious signs of failure. The oil pollution risk identification results of the oil-immersed transformer obtained by this method are basically consistent with the actual values, which further verifies the superiority of this method.