Study On The Prediction Of Land Need Used For Construction Based On Genetic Algorithm And BP Neural Network

Along with the demand of construction land increasing, the growing contradiction between human and land has been gradually revealed. According to the social and economic development target, prediction of construction land scale is very important for reasonable land use planning and land policy. It has been predicted on the scale of construction land are the main methods used multiple linear regression, gray model and artificial neural network. This article will establish the construction land scale prediction model of genetic algorithm GA-BP neural network based on the study of BP neural network. And precision of the model is inspected according to relevant data on the CZT urban agglomeration1996-2009, the research result indicates that:There are many social and economic indicators influence CZT urban agglomeration, but according to their correlation coefficient with the scale of construction land, they are arranged in the order:total population, urban and rural residents’local and foreign savings deposit year-end, total retail sales of social consumer goods, the tertiary industry and urban population.Theoretical research indicate that BP neural network existing defects. To optimize the BP neural network’s initial weights and threshold values by taking advantage of genetic algorithms’powerful ability of global optimization, which combines genetic algorithm with BP neural network efficiently, and GA-BP neural network model is established. Utilizing construction land and social economic data in CZT urban agglomeration1996-2009, BP neural network model and GA-BP neural network model are set up respectively. Experimental results show that BP neural network model optimized by genetic algorithm is more accurate and efficient in comparison with the traditional BP neural network model.