Abstract:Due to existing defects of traditional additive hazard model (AHM) in describing the influence of equipment condition to failure rate, an AHM with equipment age and health index(HI) as variables was established using HI to describe the equipment condition. The process of model solving ,which is based on the essence of the model, is also presented through three steps: Firstly, to solve the baseline hazard function, the failure rate development curve is acquired using improved EMD algorithm which is based on LS-SVM, then the curve is optimized with Genetic Algorithm; secondly, the NGDM model in Grey Theory is used to solve the covariate function; thirdly, joint optimization of the baseline hazard function and the covariate function is carried out. Simulation results have verified that the fitting error is less than 2% while the prediction error is less than 3%. The author conducts preliminary test of the applicability of AHM model with the failure rate sequence of a certain relay protection device using the model solving process and expounds the usage method of AHM model in the end.