Feature extraction of Wiener kernel fault diagnosis based on improved particle swarm annealing hybrid optimization algorithm

This paper, which the particle swarm algorithm and simulated annealing algorithm is improved and fused, has formed the particle swarm annealing hybrid optimization algorithm. The performance of IPSSAO is evaluated by MATLAB and three typical functions are optimized with IPSSAO algorithm, PSO and ASPSO algorithm. The simulation results show that IPSSAO algorithm optimization effect is better than the other two algorithms. On this basis, IPSSAO algorithm feature extraction method is proposed. The feature extraction method is applied to nonlinear analog circuit fault feature extraction. Then IPSSAO intelligent extraction method of the nonlinear analog circuit Wiener kernel fault features is put forward and is verified by examples. Experimental results show that this method achieves the circuit diagnosis effective feature extraction and improves the diagnosis accuracy in fault diagnosis system based on BP neural network.