Transformer fault diagnosis method based on QIA optimization BP neural network

For application of the traditional BP neural network has many disadvantages, such as slow convergence rate, low accuracy and poor adaptive ability. In this paper, an algorithm based on quantum immune optimization BP neural network (quantum immune algorithm BP neural network, QIA-BP) for transformer fault diagnosis has been proposed. An example of fault diagnosis based on dissolved gas analysis in oil is shown that the QIA-BP algorithm can improve the accuracy of fault diagnosis and reach the effective identification of transformer faults. It provides a new way for fault diagnosis of power transformer.