Transformer fault diagnosis method based on multi-classification logistic regression

Transformer is the core equipment in power system, its running state directly affects the safety and economy of power grid. Traditional transformer fault diagnosis methods often rely on experience and regular detection, it is difficult to realize real-time monitoring and early warning. With the development of big data technology, combined with machine learning and artificial intelligence algorithms, the operation data of transformers can be deeply analyzed, so as to achieve rapid identification and positioning of potential faults. Taking transformer equipment as the specific research object, a fault diagnosis method based on big data fusion is proposed. After the data preprocessing of the equipment, the fault diagnosis model of the equipment is established from the bottom up through data level fusion (data association fusion), feature level fusion (preliminary diagnosis of multi-classification logistic regression) and decision level fusion (synthesis diagnosis of D-S evidence theory), and specific case analysis and method comparison are carried out. Take a typical transformer equipment as an example for experimental analysis, the results show that this method is superior to some traditional fault diagnosis algorithms and has better diagnostic capability.