Prediction Of Highway Freight Volume In Henan Province Based On PCA-AFSA-SVM

With the rapid development of the world economy,the degree of the globalized economy is deepening day by day,and the influence among countries,between economies and between economies is more far-reaching,and the globalization of logistics network is followed.In this situation,China is also actively preparing for the establishment of a more comprehensive and mufti-dimensional logistics and transportation network.In this transport network,railway transport,water transport,air transportation,road transport,pipeline transportation and so on play their respective roles.However,in many modes of transportation,road transportation occupies a dominant position because of its strong agility,flexibility and rapidity.Therefore,the country needs to carry on the mufti-directional construction to the road transportation.And in the course of construction,the highway Freight volume is an important factor which needs to be considered and planned,and its accurate prediction results play an important role in guiding the construction of road transportation network and the development of transport economy.Based on the above background and purpose,this paper mainly carries on the forecast research to the Henan Province highway freight volume.First of all,this paper summarized the main research results at home and abroad,fully analyzed the characteristics of highway freight volume forecasting methods,combined with these characteristics to learn the relevant theoretical knowledge,for the selection of prediction model laid the foundation for this paper;Secondly,the influence factors of highway freight volume are explained from many aspects,and a relatively perfect index system is set up,and each index is pointed out.The correlation analysis was carried out to prove the rationality of the index selection,and the principal component analysis(PCA)was used to reduce the dimension of the index data.Then,the cubic exponential smoothing method,multiple regression prediction,SVM and PCA-SVM model are used to forecast the highway freight volume in Henan Province,and theforecast results are compared and analyzed.Finally,on the basis of reducing the principal component dimension of the index,the artificial fish swarm algorithm is used to optimize the offspring of the main parameter penalty factor C and the radial basis kernel function parameters of SVM into the SVM.Establishment of PCA-AFSA-SVM Model to per-estimate Highway Freight Volume in Henan Province The feasibility and applicability of PCA-AFSA-SVM model in highway freight volume forecasting are explained by comparing and analyzing the results with other forecasting methods.On this basis,the highway freight volume of Henan Province in 2018 is predicted and the results are analyzed.