Font Size: a A A

Prediction Of Bearing Capacity Of Central City Based On Particle Swarm Optimization And Support Vector Machine

Posted on:2018-03-11Degree:MasterType:Thesis
Country:ChinaCandidate:Z L DingFull Text:PDF
GTID:2347330512473811Subject:Applied Statistics
Abstract/Summary:PDF Full Text Request
With the acceleration of urbanization and rapid social and economic development,many cities are blindly developing and constructing,and the utilization efficiency of urban resources is low,leading to waste of land,serious shortage of fresh water and energy resources,serious pollution of the environment,serious degradation of the ecological system and increasing traffic congestion.Serious and other urban crisis risk frequency.The size of urban carrying capacity matters about whether the city can continue the healthy development and people can enjoy a higher quality of life.Therefore,people are eager to seek an effective model to predict the urban carrying capacity,in order to guide people's lives and government planning.In this paper,urban water resources,land resources,traffic and atmospheric environment as the focus of the study,combined with the Statistical Yearbook and water resources bulletin to explore existing city data.And respectively construct the urban water resources carrying capacity evaluation indicators,land resources carrying capacity evaluation index,evaluation index of traffic carrying capacity,evaluation index of atmospheric environmental carrying capacity and comprehensive carrying capacity evaluation index,with a view to solve the city in rapid development at the same time derive many problems to provide direction.In addition,in order to fully show the current situation of the development of urban carrying capacity of major cities in China,this paper selects 64 important cities,such as Beijing-Tianj in-Hebei,Yangtze River Delta,Pearl River Delta and other municipalities,provincial capitals and municipalities separately,on the basis of selected evaluation indicators Cities to study the status of urban carrying capacity.The results show that in the year of 2014,the urban comprehensive carrying capacity of our country is generally poor.More than 50%of the city's water resources carrying capacity in the state are of early warning and even crisis;more than 56%of the city's land resources carrying capacity are in a state of crisis;about 80%of urban traffic carrying capacity is extremely fragile;there were only one city with good air quality,accounting for 1.56%.All these show that:the urban carrying capacity of our city has been a serious challenge,people's normal life and the healthy development of society has been seriously affected.In this paper,we use the support vector machine(SVM)based on limited data to predict the future urban carrying capacity from the historical data of four aspects of urban carrying capacity.At first,the time series data of these four bearing capacity single column are reconstructed to generate time series matrix,and the information capacity is enlarged.The optimal carrying capacity of water resources,land resource carrying capacity,traffic carrying capacity and atmospheric environmental carrying capacity are determined.Embedding dimensions are 4,5,2,4,respectively.Then the support vector regression model is used to model the data in the sequential matrix.In view of the model results will be different because of the choice of SVR parameters,the difference is obvious.Based on the theory that the kernel function parameter sensitivity is stronger than the kernel function sensitivity,we choose the following two ways to improve the support vector machine regression prediction Model performance:First,select the default parameters in support vector machine;the other is to use PSO to select the optimal penalty factor and kernel parameter.The comparison of the two models shows that the PSO-SVM is more accurate than the general SVM,and its applicability is better.Then the PSO-SVM model was used to forecast the water resources,land resources,traffic carrying capacity and atmospheric environmental carrying capacity of Hangzhou in the next five years.The results show that the comprehensive carrying capacity of Hangzhou City is declining in the next five years,and the water resources and atmospheric environmental carrying capacity are in a good state.Generally,Hangzhou will not have a bad impact on the development of Hangzhou.While the carrying capacity of land resources and traffic carrying capacity is relatively low,most likely to Hangzhou,the future development may cause obstacles.
Keywords/Search Tags:Urban bearing capacity, Phase space reconstruction, Supp-ort vector machine, Particle swarm optimization
PDF Full Text Request
Related items