Multivariate Optimization Model Based On New Kernel And Degree Of Greyness And Its Application In Smog

Smog is one of the important judgment indexes of air quality assessment,so it has practical significance to accurately predict it.In view of the large degree of greyness of smog data,the MGM(1,m)model based on interval grey number sequence with known distribution information will effectively extend the application range of the model.On this basis,MGM(1,m)optimization models are established considering the time-delay of sample data,the nonlinearity between variables,the time-varying of model parameters and the influence of relevant factors on the system characteristic variables.Finally,the optimization models are applied to the simulation and prediction of smog.The specific research contents are as follows:In the second chapter,the interval grey number sequence with known probability function is taken as the original data,and the time-delay and nonlinear characteristics of the model are considered to construct the time-delay nonlinear MGM(1,m|?,?)model based on the new kernel and degree of greyness.The time-delay parameter is determined by the maximum of grey time-delay absolute correlation degree,and the nonlinear parameter is determined by the minimum of average relative error.Finally,in order to verify the practical significance of the model,the smog related data of Nanjing city are used for simulation and prediction,and compared with other models.Considering all the results,the simulation accuracy and prediction accuracy of the new model are higher.In the third chapter,on the premise that the distribution information of interval grey number sequence is known,the quadratic time-varying term is introduced to optimize the parameters of discrete MGM(1,m)model based on kernel and degree of greyness sequences,so that the characteristics of parameters changing with time in the process of model construction are fully considered.And then the characteristics of model multiplication parameters are studied and analyzed.Finally,the steps of the new model are described,and the practical application is carried out by using the smog related data in Nanjing.Through the comparison and analysis with other grey models,it is proved that the quadratic time-varying MGM(1,m)model based on the new kernel and degree of greyness sequences has better simulation and prediction effect on the whole.In the fourth chapter,aiming at the interval grey number sequence with known distribution information,the grey MGM(1,m,N)model with m related factors influencing variables acting on N-1 system characteristic behavior variables is constructed based on the new kernel and degree of greyness sequences calculated from interval grey number sequence.Then,the upper and lower bounds of interval grey number sequence are obtained according to the simulated and predicted values,and the error test is carried out.Finally,the new model is applied to the smog data of Nanjing and compared with the comparative model.The results show that the new model is feasible and effective in theory and practice.