基于数据场理论的公路自然区划指标分析方法研究

发布时间：2018-02-20 03:10

本文关键词： 道路工程自然区划空间分析交叉验证数据场区划指标　出处：《公路交通科技》2017年04期 　论文类型：期刊论文

【摘要】：为了提高公路自然区划的准确性与实用性,基于数据场理论,结合公路工程建设的实际需求,进行公路自然区划指标分析方法研究。在分析数据场基本原理的基础上,提出基于数据场理论的公路区划指标空间插值算法,详细论述了利用场强变差函数确定各样本数据权重的方法;将区划指标空间插值算法与地理基础数据和空间数据处理平台相结合,同时考虑公路工程建设需求,提出了基于数据场理论的公路区划单要素指标分析方法。利用单要素指标分析方法、反距离权重法和地统计的插值方法对施工不利日数(CLPDavg)值进行预测,对3种方法预测结果进行交叉验证。结果表明,利用基于数据场理论的单要素指标分析方法得到的公路自然区划指标分布图具有更高的精度,更符合实际。
[Abstract]:In order to improve the accuracy and practicability of highway natural regionalization, based on the theory of data field and the actual demand of highway engineering construction, the analysis method of highway natural regionalization index is studied. A spatial interpolation algorithm of highway zoning index based on data field theory is proposed, and the method of determining the weight of each sample data by using field intensity variation function is discussed in detail. The spatial interpolation algorithm of regionalization index is combined with the geographic basic data and spatial data processing platform, and the needs of highway construction are taken into account at the same time. Based on data field theory, a method of single factor index analysis for highway regionalization is proposed, which is used to predict the number of construction unfavorable days and CLPDavg value by using single factor index analysis method, inverse distance weight method and interpolation method of geostatistics. The results show that the distribution map of highway natural regionalization index obtained by single factor index analysis method based on data field theory has higher accuracy and is more practical.
【作者单位】：山东交通学院;山东省交通规划设计院;济南城市建设投资集团有限公司;
【基金】：山东省自然科学基金项目(ZR2012EEL29;ZR2016EEM42) 山东省高等学校科技计划项目(J16LA17)
【分类号】：U412

【参考文献】