With the widespread use of electronic transformers in the power grid, transformer faults have become an urgent problem to be solved.In order to solve this problem, this paper designs the online monitoring system of electronic transformer, designs the composition and core unit of the system, and applies BP neural network and particle swarm optimization to the fault diagnosis of electronic transformer. The advantages and disadvantages of the improved neural network algorithm and BP neural network algorithm are compared and analyzed by numerical examples, the results show that the fault diagnosis method adopted in this paper is effective and feasible.The research work provides a reference for the development of transformer line monitoring system in China.