Power Generation Prediction Of Three Major Renewable Energy Resources Based On Grey Model With Parameters Combination Optimization Method

The grey prediction model is a core component of the grey system theory.Because of the outstanding performance in the field of studying small data and poor information system prediction,it attracts attention and research from a wide range of scholars.It has been widely used in many fields of social and economic life.However,the existing grey prediction models can hardly be fully adapted to the widening practical application scenarios.They need to be continuously optimized and improved to meet practical needs.Based on the modeling process,the combination optimization method of parameters in the grey prediction model is proposed.On the basis of GM(1,1),TDGM(1,1)and TWGM(2,1),three new grey models with parameters combination optimization are constructed.They are applied to predict the power generation of three major renewable energy resources in ChinaFirstly,the parameters combination optimization method of grey prediction model is proposed.The accumulating order,the background value coefficient and the initial condition are three important parameters that affect the structure and performance of the grey prediction model.They are transformed from fixed values into variables respectively.According to the grey prediction model modeling process,the parameters combination optimization method is proposed.The method combines the advantages of accumulating order,background value coefficients and initial conditions,which can improve the performance of the grey prediction model as a whole.Secondly,grey prediction model is constructed based on the parameters combination optimization method.Applying the parameters combination optimization method,the corresponding parameters combination optimization models are constructed based on GM(1,1),TDGM(1,1)and TWGM(2,1).Specifically,taking the three-parameter combination optimization grey prediction model as an example,the modeling mechanism and parameters estimation method of the GM(1,1)model,TDGM(1,1)model and TWGM(2,1)model with three-parameter combination optimization are described respectively.And the time response function and inverse accumulating restored function of the three-parameter combination optimization models are derived respectively.Thirdly,power generation of three major renewable energy resources are predicted.Hydro,wind and photovoltaic power generation,which have the highest renewable energy generation capacity,are selected as the research subjects.Firstly,the buffer operator is applied to preprocess the photovoltaic power generation data.Then the GM(1,1)model with parameters combination optimization is applied to simulate and analyze the photovoltaic power generation.The results show that the GM(1,1)model with three-parameter combination optimization is superior to other parameters combination optimization.And the model is applied to predict the photovoltaic power generation in China from 2021 to 2025.Secondly,the TDGM(1,1)model of parameters combination optimization is constructed to simulate and analyze the hydropower generation.ARIMA model and SVM model are selected for comparative analysis.The results show that the TDGM(1,1)model with three-parameter combination optimization has better performance.The model is applied to realize the reasonable prediction of China’s hydropower generation from 2021-2025.Finally,the TWGM(2,1)model with parameters combination optimization is constructed to simulate and analyze the wind power generation.The results show that the TWGM(2,1)model optimized by three-parameter combination performs better than other parameter optimization models.And the model is applied to predict China’s wind power generation from 2021 to 2025.