摘要
城市建设用地的准确预测是城市土地总体规划的重要决策基础。通过对影响城市建设用地主要因素的研究,提出一种基于小波神经网络的城市建设用地预测模型,并给出相应的网络学习算法。以湖南省长沙市为例,建立了基于小波神经网络的长沙市建设用地预测模型,比较分析了小波神经网络模型与灰色BP神经网络模型和传统BP神经网络模型的预测结果。分析结果表明:小波神经网络模型比灰色BP神经网络模型和传统BP神经网络模型的收敛速度快、预测精度高,在城市建设用地预测中更具应用价值。该成果为城市建设用地预测研究提供了有益参考。
Accurate forecast of urban construction land (UCL) is an important decision-making basis for general urban land plan. According to the research on principal factors influencing the UCL, we present a WNN-based UCL forecast model, and give its corresponding networks learning algorithm. Changsha city in Hunan province is used as a case, we build a WNN-based UCL forecast model for Changsha, and make the comparison and analyses on the forecast results using WNN model, grey BP neural network and traditional BP neural network model respectively. Analysis result shows that the WNN model has higher convergence speed and forecast precision than the grey BP neural network and the traditional BP neural network model, and has higher applied value in UCL forecasts. Our research work provides a useful reference for the research on UCL forecast.
出处
《计算机应用与软件》
CSCD
北大核心
2014年第10期180-182,共3页
Computer Applications and Software
基金
浙江省供销社2011年度科学研究项目(11SS88)
关键词
城市建设用地
小波神经网络
模型
预测
Urban construction land
Wavelet neural network(WNN)
Model
Forecast