中文摘要：

目前，土地利用变化中的一个显著特点是城市化扩张，而城市化扩张是一个动态的、非线性的、多反馈回路的复合系统。神经网络在非线性系统的建模和预测中具有明显的优越性。以泰安市城区及周边为研究区域，依据近年来的土地利用变化数据，用BP算法构造预测模型，模拟近年来市区面积的变化情况，并据此预测以后多年的城市建成区面积。结果表明，方法可行，预测精度较高。

英文摘要：

At present, a remarkable characteristic of changes in land use is urbanized expansion, which is a dynamic, nonlinear, multi-feedback loop＇s complex system. The neural network applied in the modeling and forecasting of nonlinear system is of obvious superiority. This article takes urban area and perimeter of Tai＇an City as the study region to structure forecasting model using BP algorithm and according to the change data of land use in recent years, and to simulate the changes of urban district area in recent years, then forecast the urban district area in the future. The result indicates that this method is feasible, and of high forecasting precision.