Grey Oscillation Prediction Modeling Technology And Its Application

The real world is becoming increasingly complex.Random disturbances in systems tend to make their characteristic sequences exhibit various characteristics of oscillatory development.Therefore,the contradiction between the limitations of grey oscillatory prediction model and the complex and variable morphological feature sequences gradually deepens.For this reason,a new type of grey oscillation prediction model with parameter combination optimization is constructed based on grey system theory and dynamic optimization principles.Furthermore,this model is applied to the prediction research of hydropower generation in the European Union.The research content and results of this paper are as follows:(1)Construction of grey prediction model for oscillating sequences.After analyzing the defects of TDGM(1,1)with a relatively complete structure,a smoothing operator is firstly used to improve the smoothness of the original oscillation sequence,and the first order smoothing sequence of the original sequence is obtained;Then,TDGM(1,1)for the smooth sequence is established,and the inverse process of the grey smoothing operator is used to inverse the functional relationship between the original data and the time point,resulting in a new grey oscillation prediction model,ODGM(1,1).Finally,three types of cases(whose original sequences have different oscillation and totality)are selected to test the performance of ODGM(1,1).The results show that the new model has good reliability and adaptability for different oscillation sequences.(2)Combination optimization of parameters for the new type of grey oscillation prediction model.After analyzing the structure of ODGM(1,1),GA is used to optimize the performance parameters of the model based on the principle of dynamic optimization,thereby obtaining ODGM(1,1,inif,w,r).The modeling steps of ODGM(1,1,inif,w,r)are summarized.In addition,three sets of cases with different oscillation sequences are also applied to test the performance of ODGM(1,1,inif,w,r).The results show that the performance of ODGM(1,1,inif,w,r)is further improved on the basis of ODGM(1,1),confirming the effectiveness of parameter combination optimization of ODGM(1,1)and the adaptability and superiority of ODGM(1,1,inif,w,r)in the face of different oscillation sequences.(3)The prediction of hydropower generation in the European Union.Firstly,this paper constructs ODGM(1,1,inif,w,r)based on the historical hydroelectric power generation data of the European Union from 1994 to 2021.The modeling results show that the accuracy and stability of ODGM(1,1,inif,w,r)model are superior to other comparative models;Secondly,ODGM(1,1,inif,w,r)is applied to predict and analyze the future hydropower generation in the EU,and countermeasures and suggestions are proposed to ensure the safe and stable supply of hydropower in the EU and reasonably plan the future development based on the experimental results.