Oil immersed transformer fault detection method based on gas on oil surface

The invention provides an oil-immersed transformer fault detection method based on gas on an oil surface, and belongs to the technical field of oil-immersed transformer fault detection.The method comprises the steps that historical operation parameters and internal gas data of an oil-immersed transformer are collected in real time, and a mathematical model of the operation state and gas concentration change of the transformer is established; and the incidence relation between the two is analyzed, and a gas-operation parameter change matrix is constructed. A training set is established based on data of a fault-free transformer and a known fault transformer, and a neural network model capable of identifying the fault type of the to-be-detected transformer is trained. And finally, carrying out real-time monitoring and fault diagnosis on the to-be-detected transformer. According to the method, internal data of the transformer are fully utilized, a complete fault diagnosis system is established, and the technical