Genetic Algorithm Based Fault Diagnosis and Repair Strategy in Computer Hardware System

With the rapid development of the information age, computers have become the dominant core force for the forward development of society. The good or bad operation status of a computer directly affects people’s work and even their lives. Therefore, how to obtain a more accurate diagnosis method has been the focus of many scholars’ research. Based on this research background, the study firstly introduces the computer hardware system fault feature extraction method based on wavelet packet decomposition, and then further optimizes the BP neural network by using the improved genetic algorithm to construct a GA-BP neural network fault diagnosis model after the basic theoretical study of genetic algorithm. At the same time, the effectiveness of the model proposed in this paper is proven through simulation experiments and empirical evidence. After analyzing, it was found that the improved and optimized BP neural network using genetic algorithm has good training performance. In the optimized computer hardware system fault diagnosis model training error experiment, the convergence speed of the proposed model has been improved, the number of iterations has been reduced, and the error has been decreased. The error remains at about 0.003, which can be concluded that the GA-BP network fault diagnosis model designed in this paper has a higher accuracy rate for the identification of computer hardware system faults. It can synthesize the characteristics of regular maintenance, depending on the situation and the aftermath of maintenance, and target fault repair for different computer system hardware failures.