Center Approach Grey BP Neural Network Prediction Model for Years of Drought Occurrence in Xinzhou District of Wuhan City

Abstract

The traditional GM(1,1) model can be employed to forecast drought in Xinzhou, but this model is not ideal due to the smaller amount of data and the larger scope of sequence changes. The center approach grey BP neural network model was adopted to predict the years of drought occurrence in Xinzhou. The scope of sequence was weakened by adjusting m value of center approach grey GM(1,1) model, then the BP neural network was applied for fitting the residuals and modifying the predictive values. The results showed that the precision of center approach grey BP neural network model was much higher than that of the single center approach GM(1,1) model and the traditional single GM(1,1) model, so this model was valid for prediction of the next drought occurrence year in Xinzhou.