论文部分内容阅读
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。
In order to improve the accuracy and practicability of the highway natural division, based on the data field theory, combined with the actual needs of highway construction, highway natural division index analysis method is studied. Based on the analysis of the basic principle of data field, a spatial interpolation algorithm of highway zoning index based on data field theory is proposed. The method of using the field strength variation function to determine the weight of each sample data is discussed in detail. The spatial interpolation algorithm of zoning index and geo-basic data Combined with the spatial data processing platform, taking into account the demand of highway construction, a single element index analysis method based on data field theory is proposed. The CLPDavg values were predicted by single factor analysis, inverse distance weighting and geostatistical interpolation. The results of three methods were cross-validated. The results show that using the single factor index analysis method based on data field theory, the road natural zoning index distribution map has higher accuracy and is more realistic.