摘要
建设用地需求量预测对合理规划建设用地规模以及有效控制建设用地数量具有重要意义。本文应用多元线性回归模型(MLR)、GM(1,1)灰色系统模型和基于主成分分析的BP神经网络模型(PCA-BP)建立3种预测模型对2020年连云港的建设用地需求量进行预测。结果表明:3种模型的预测精度均较高,其中PCA-BP神经网络模型预测的平均绝对误差(MAE)、误差均方根(RMSE)以及平均相对误差(MRE)在3种模型中均最小,预测精确度最高,多元线性回归模型次之,灰色GM(1,1)模型略低;根据精度评价结果最终确定以3种模型预测结果的平均值作为2020年连云港建设用地需求量的预测值,得出预测结果为189506.02 hm^2。
Construction land demand prediction plays a significant role in planning the scale of construction land legitimately and controlling the number of construction land effectively. In this paper, three kinds of prediction models that multivariate linear regression(MLR), GM(1,1) gray model and BP neural network model(PCA-BP) based on principal component analysis are established to predict the construction land demand of Lianyungang in 2020. The results show: The prediction accuracy of the three models are all relatively high, for the three models, the average absolute error(MAE), root mean square error(RMSE) and the mean relative error(MRE) of the PCA-BP neural network model prediction are all smaller than the other two models, the accuracy of this prediction model is the highest, followed by multiple linear regression model, gray GM(1,1) model is slightly lower;Finally choose the average of the prediction results of the three models as the predicted value of Lianyungang construction land demand in 2020, the prediction result is 189,506.02 ha.
作者
王丹
杨小艳
郑剑
陈龙高
胡洋洋
高卫东
Wang Dan;Yang Xiaoyan;Zheng Jian;Chen Longgao;Hu Yangyang;Gao Weidong(School of Geography,Geomatics and Planning,Jiangsu Normal University,Xuzhou Jiangsu 221116,China;School of Environment and Spatial Informatics,China University of Mining & Technology,Xuzhou Jiangsu221116,China;Land and Resources Bureau of Lianyungang City,Lianyungang Jiangsu 222001,China)
出处
《科技通报》
2019年第1期60-65,共6页
Bulletin of Science and Technology
基金
国家自然科学基金项目(项目编号:41601087)
国家自然科学基金项目(项目编号:41271121)
江苏省普通高校研究生科研创新计划项目(KYZZ_0388)
江苏高校优势学科建设工程项目资助(PAPD)
