Urban Construction Land Demand Forecast Method

Urban constructive land use is the carrier on which cities carry out their economic and social activities. With the speeding-up of the urbanization in the countryside, the demand of the constructive land has increased significantly, which has intensified the contradiction between the supply and demand of the urban constructive land use and made the rational space structure and distribution in specific regions more necessary. The existing urban construction land scale has not been able to satisfy the new time, the new duty need. The new turn revision of overall land use planning is being carried on; in order to provide the foundation to make the overall land use planning scientifically, there search on predicting of construction land demand should be in urgent in the new era background.The paper utilizes the general land use planning as research object. We introduced the basic content and process methods of the forecast for urban construction land systematically, focusing on the selection and improvement of forecast models for urban construction land.In the first place, the paper studied the commonsense forecasting methods and evaluated them from suitable condition, for example: the linear regression method, the grey model method, the generating function method and so on and have carried on the appraisal to these methods; And then the multi-dimensional modeling method and the combining forecasting method of BP neural network has been introduced.Combines the studied area's alter data of land utilizing and socioeconomic statistical data from 1996 to 2005,Using the correlate analysis, the grey relation analysis and the principal components analysis to filtrate leading factor of influencing urban constructive land use ,and then predicting model of regression , predicting model of grey system , multi-dimensional forecasting model and combined forecasting model of BP neural network has been constructed .Through real diagnosis, we analyzed and compared BP network and combined forecasting with grey relation analysis and the principal components . The results showed that the forecasting model which based on BP network through compared with each other which illuminates that the model of construction land demand prediction has some value both in theory and in practice and it provides profitable reference for construction land demand forecast father study and government's decision-making.Aiming trivial calculate process of BP, this paper builds BP ,imports the sample vector ,and trains the bridge construction network with the help of the neural network toolbox of MATLAB program, determining the training function the number of hidden layer and the active function for BP, thus makes the forecast function in the bridge construction into reality obtaining a good export result in the forecast function in the construction land into reality by obtaining a good export result in the construction demand land of f 2010 and 2020.Finally, using GIS to express the growth direction of construction land in the future with land construction' present situation and development dynamic of the future.