Abstract：In power electronic devices，the supporting capacitor is a crucial component. To ensure the safety of

the system，monitoring the status of the capacitor is particularly important. A neural network approach was

employed，and analyzing the correlation between control strategies and circuit topology parameters to determine

two input variables and one output variable for the neural network. Subsequently，the training was conducted using

a BP neural network，and the trained model was imported into Simulink simulation software to validate the

correctness of the theory and the accuracy of the monitoring results. Finally，experiments were carried out using a

experimental platform to verify the feasibility of the proposed solution. The research results indicate that compared

to previous approaches，the proposed solution can achieve online monitoring of capacitor capacitance without

increasing system complexity.

      Key words：condition monitoring；neural networks；support capacitor；rectifier converter