Forecasting Method for Lockage Freight Volume of Three Gorges Dam Area Based on Optimized Grey Buffer Operator

Abstract

In order to effectively predict the lockage freight volume of the Three Gorges Dam area, an improved grey prediction model for the lockage freight volume is proposed under the action of the optimized variable weight weakening buffer operator. By exploring the trend of the original data of freight volume, it is found that the original data sequence has the oscillation properties of "small sample" and "shock wave interference". The classical GM(1,1) model is difficult to simulate well for non-homogeneous exponential sequences and trend mutation sequences. So the method of constructing optimized variable weight weakening buffer operator is proposed by introducing accumulative transformation and translation transformation. On this basis, a new grey prediction model adapted to the lockage freight volume of the Three Gorges Dam area is established, and a case forecast is performed. By comparing and analyzing the prediction accuracy of the improved gray model and the traditional prediction model, it can be clearly seen that the improved gray model prediction accuracy of the optimized variable weight weakening buffer operator is significantly improved.