基于灰色—BP神经网络耦合模型的杨凌示范区建设用地规模预测

建设用地是一个地区发展的重要物质载体,合理确定建设用地规模是地区理性发展和土地利用管理的重要依据。以杨凌示范区1997-2008年社会经济数据和建设用地数据为基础,运用Pearson相关分析确定了建设用地规模变化驱动因子,通过GM（1,1）模型预测出各驱动因子2009-2015年的数值。以1997-2008年建设用地规模及驱动因子数据为基础运用误差反向传递学习算法对神经网络进行训练,预测出了杨凌示范区2009-2015年的建设用地规模。

Prediction of Construction Land in Yangling Demonstration Based on Grey-ANN Model

Construction land is the carrier of a region＇s economic and social development,reasonable forecast of the construction land size is the basis of rational development and land management.The driving forces of the change in construction land are analyzed by the Pearson correlation analysis based on the social and economic and the construction land size data of Yangling Demonstration from 1997 to 2008.The data of driving forces from 2009 to 2015 is forecasted by GM（1,1）.Back Propagation network was trained and tested by the data from 1997 to 2008,and then it forecasts the construction land size in Yangling Demonstration from 2009 to 2015.