BP神经网络和GM（1，n）模型在吉林省四平市建设用地面积预测中的应用比较

[目的]科学准确地预测城市建设用地,有利于把握城市发展的速度,了解城市化发展进程,为相关政府部门掌握土地利用情况,制定土地总体规划提供科学依据。[方法]把四平市作为研究对象,从"城市—农村"合力视角构建影响因子,利用因子分析探讨影响建设用地扩张的相关性,对指标进行筛选,在此基础上利用BP神经网络和灰色模型对四平市2012,2013和2014年建成区面积进行预测,最后对预测结果进行比较分析。[结果]通过预测与比较分析可知,BP神经网络结果的相对误差分别为0.8%,1.1%和2%,而灰色GM(1.1)模型预测结果相对误差分别为0.04%,0%和3.2%。可以看出,BP神经网络预测出的结果与实际相比较误差均在2%以内。[结论]BP神经网络预测的结果较精确,运用该方法可以有效提高预测的精度。

Comparison of GM(1,n) and BP Neural Network Model in Predicting Construction Lands in Siping City, Jilin Province

[Objective] The paper aims to compare the accurracy of BP neutral network and GM(1,n) in predicting construction land changes,which is beneficial to understand the urban development and provide refernces for general land planning.[Methods] With Siping City as the research object,we selected impact factors with the perspective of "city-rural integration" and used factor analysis to estimate the influence of construction land expansion and choose indicators.We then simulated and compared the predictions of construction land in 2012,2013 and 2014 in Siping City using the BP neural network and grey model.[Results]The relative error with BP neural network was 0.8 ％,1.1％ and 2 ％,and the gray GM (1.1) model was 0.04％,0％ and 3.2％ respectively.The BP neural network are better than GM(1.1) model.[Conclusion]BP neural network can provide a higher accuracy.