Pre-diagnosis technology for short-circuit withstand capability of distribution transformer based on big data

Abstract

Abstract The short-circuit withstand capability test of distribution transformer has the characteristics of long test period, high cost, strong destructiveness, low qualification rate, and can not be verified by simulation or calculation. It has become the bottleneck restricting the engineering and industrialization development of transformer industry. Based on experimental big data, this paper uses the expert diagnosis method of fuzzy mathematics to establish a superior evaluation model. The typical design process of distribution transformers is scored by experts, and the short-circuit passing rate is calculated according to the real short-circuit test data. Linear regression analysis is used to establish the regression equation, and a pre-diagnosis technique for short-circuit withstand capability of distribution transformers is proposed. At the same time, the validity of the model is verified by the actual case, and the accurate pre-diagnosis of the resistance to short-circuit capability is realized. It breaks through the common key technologies of the transformer industry and provides an early warning for the quality control of power grid pro-ducts and the elimination of hidden danger products. It has important theoretical guiding significance and engineering application value.