A Powell PSO optimized random forest fault diagnosis method for high-voltage circuit breakers is proposed, in order to improve the accuracy of the state identification of mechanical vibration signals of high-voltage circuit breakers. Firstly, the fault simulation model of high-voltage circuit breaker is built on MATLAB, and the closing process is simulated to collect vibration signals; Then the mechanical vibration signal of the circuit breaker is denoised by wavelet transform, and the eigenvalues are extracted; Secondly, Powell PSO fusion algorithm is constructed to optimize the random forest algorithm; Finally, the eigenvectors are input into the optimized random forest, and the classifier of Powell PSO optimized random forest algorithm is constructed, which finally realizes the high-precision identification of the mechanical fault state of high-voltage circuit breakers. The experimental results show that the new fault diagnosis method has high overall recognition rate and high application value.