Research on fault prediction method of electronic equipment based on improved SVR algorithm

With the continuous development of electronic equipment, state-based maintenance has attracted more and more attention in the industry. As an important part of state-based maintenance, fault prediction is especially concerned. Because of its good generalization ability and anti-interference ability, SVR algorithm is widely used in the field of fault prediction. However, SVR algorithm still has the disadvantage that the smooth parameter C and kernel parameter σ
²
of the model are difficult to determine. This paper will build SVR prediction model based on Python platform, and use genetic algorithm to optimize the parameters of SVR prediction model. In this paper, experiments are carried out in voltage data set and temperature data set commonly used in electronic equipment. Experimental comparison found that compared with the original SVR model and the improved SVR model based on particle swarm optimization, the improved SVR model based on genetic algorithm has the best prediction effect.