Transformer DGA fault diagnosis method based on PNN network

The invention belongs to the technical field of power equipment monitoring, and relates to a PNN network-based transformer DGA fault diagnosis method, which comprises the following steps: selecting monitoring data of components and content of dissolved gas in transformer oil to form a DGA data set, carrying out minimum-maximum normalization processing on the DGA data set, initializing smoothing parameters of the PNN network, and carrying out fault diagnosis on the smoothing parameters of the PNN network; building a PNN network according to the initialization parameters, and inputting the training set into the PNN network; the smoothing parameter of the PNN network is used as an optimization parameter, and the fault diagnosis accuracy of the PNN network is used as fitness; an improved snake optimization algorithm is adopted to obtain optimal fitness and corresponding smoothing parameters; and constructing a transformer DGA fault diagnosis model based on the PNN network based on the optimal smoothing parameter,