The Demand Prediction of Power Emergency Material Based on Grey-Markov Chain Model

Abstract

In this research paper, the Grey-Markov chain model is proposed to solve the problem in the demand prediction of power emergency material. We first conducted preliminary prediction with the Grey model, and then proceeded to modify the prediction result with the Markov chain model. We also collected the data from 20 earthquakes, both in China and abroad and made predictions of three indexes including the households of power outage, crew in emergency maintenance and the number of damaged oil-immersed transformer. The testing results indicated that the Grey model can be separately used to predict the demand for the power emergency material, but the precision is not quite satisfying; the Markov chain model can modify the relative error of the Grey model, and the average absolute error of the three indexes decreased from 6.20%, 6.16% and 5.46% to 5.77%, 5.38% and 4.56%, respectively.