Aiming at the fluctuation of maintenance and repair costs in substation life cycle cost, a traditional grey model and an improved grey model are adopted respectively, to forecast the maintenance and repair costs of a substation in the next 3 years in order to optimize cost allocation strategy. Simulation results show that the prediction accuracy of both models is one grade; while both the average relative error and the posterior error ratio of the improved model are lower than those of the traditional one. The prediction accuracy of the improved model is hence higher than that of the traditional one, and can be suitable to predict the maintenance and repair costs of a substation. Finally, the improved grey model is used to predict the maintenance and repair costs during 2019 to 2021 of a specified substation in a city.