郭辉;

• 1:安徽理工大学测绘学院
• 2:中国矿业大学(北京)地球科学与测绘工程学院

摘要(Abstract)：

针对二次曲面模型在地形起伏较大区域用于GPS高程转换中存在较大的模型误差的问题,该文构建了二次曲面-RBF神经网络组合的GPS高程转换模型,组合模型中用二次曲面拟合高程异常中的中长波项,用RBF神经网络来泛化高程异常去除中长波后的残余项,并进行了二次曲面模型、RBF神经网络模型及二次曲面-RBF组合模型的实测数据GPS高程转换、比较分析与精度评定。实例结果表明:该组合模型比二次曲面模型的转换精度提高了22%,比RBF神经网络模型的转换精度提高了40%,该组合模型的转换方法可行、精度优于单一模型。

关键词(KeyWords)： 组合模型;GPS高程;RBF神经网络;二次曲面拟合;高程异常

Abstract:

Keywords:

基金项目(Foundation): 中国煤炭工业协会2015年度科学技术研究指导性计划项目(MTKJ2015-313)

作者(Authors): 郭辉;

DOI: 10.16251/j.cnki.1009-2307.2018.02.007

参考文献(References)：

• [1]高宁,高彩云,吴良才.GPS高程迭加拟合模型的研究[J].西安科技大学学报,2009,29(3):339-343.(GAO Ning,GAO Caiyun,WU Liangcai.Research of combined model for GPS height[J].Journal of Xi’an University of Science and Technology,2009,29(3):339-343.)
• [2]吴寒,姚宜斌.附加随机参考模型的GPS高程转换方法[J].大地测量与地球动力学,2010,30(3):67-70.(WU Han,YAO Yibin.GPS height conversion method with additional stochastic model[J].Journal of Geodesy and Geodynamics,2010,30(3):67-70.)
• [3]高伟,徐绍铨.GPS高程分区拟合转换正常高的研究[J].武汉大学学报(信息科学版),2004,29(10):908-911.(GAO Wei,XU Shaoquan.Subregional fitting and transforming GPS height into normal height[J].Geomatics and Information Science of Wuhan University,2004,29(10):908-911.)
• [4]任超,吴伟,黄征凯,等.基于AIC准则的RBF神经网络在GPS高程拟合中的应用[J].测绘科学,2013,38(2):77-79.(REN Chao,WU Wei,HUANG Zhengkai,et al.RBF neural network application in GPS height fitting based on AIC criterion[J].Science of Surveying and Mapping,2013,38(2):77-79.)
• [5]徐绍铨,李振洪,吴云孙.GPS高程拟合系统的研究[J].武汉测绘科技大学学报,1999,24(4):336-340.(XU Shaoquan,LI Zhenhong,WU Yunsun.Research of GPS abnormal height interpolation system[J].Journal of Wuhan Technical University of Surveying and Mapping,1999,24(4):336-340.)
• [6]赵建虎,刘经南,张红梅.顾及非格网数据考虑地形改正的GPS水准高程拟合[J].武汉测绘科技大学学报,1999,24(4):346-350.(ZHAO Jianhu,LIU Jingnan,ZHANG Hongmei.Height determination with GPS leveling data considering non-grid data and effect of terrain[J].Journal of Wuhan Technical University of Surveying and Mapping,1999,24(4):346-350.)
• [7]史俊莉,李生平.考虑地形改正的GPS高程拟合研究[J].测绘通报,2015(7):66-67.(SHI Junli,LI Shengping.GPS height fitting research considering topographic correction[J].Bulletin of Surveying and Mapping,2015(7):66-67.
• [8]朱卫东,李全海.基于标准化动量BP神经网络的GPS高程转换[J].大地测量与地球动力学,2010,30(2):123-125.(ZHU Weidong,LI Quanhai.Conversion of GPS height based on standardization momentum BP neural network[J].Journal of Geodesy and Geodynamics,2010,30(2):123-125.)
• [9]张莉,李羽荟.RBF神经网络在GPS高程拟合中的应用[J].地理空间信息,2011,9(5):23-25.(ZHANG Li,LI Yuhui.Application of RBF neural networks to GPS leveling elevation fitting[J].Geospatial Information,2011,9(5):23-25.)
• [10]刘建,王琪洁,王小辉,等.基于遗传算法的RBF神经网络在GPS高程拟合中的应用[J].测绘科学,2013,38(2):143-145.(LIU Jian,WANG Qijie,WANG Xiaohui,et al.Application of RBF neural network to GPS elevation fitting based on genetic algorithm[J].Science of Surveying and Mapping,2013,38(2):143-145.)
• [11]王新志,陈伟,祝明坤.样本数据归一化方式对GPS高程转换的影响[J].测绘科学,2013,38(6):162-165.(WANG Xinzhi,CHEN Wei,ZHU Mingkun.Influence of sample data normalization ways on GPS elevation transformation[J].Science of Surveying and Mapping,2013,38(6):162-165.)
• [12]张昊,王琪洁,朱建军,等.样本数据预处理对基于BP神经网络的GPS高程拟合的影响[J].大地测量与地球动力学,2011,31(2):125-128.(ZHANG Hao,WANG Qijie,ZHU Jianjun,et al.Influence of sample data preprocessing on BP neural network-based GPS elevation fitting[J].Journal of Geodesy and Geodynamics,2011,31(2):125-128.)
• [13]刘长建,柴红洲,吴洪举,等.GPS水准高程等级的评定[J].测绘通报,2008(8):1-3.(LIU Changjian,CHAI Hongzhou,WU Hongju,et al.Evaluating height class of GPS-leveling[J].Bulletin of Surveying and Mapping,2008(8):1-3.)
• [14]陈绍杰,秦永洋,孙兴平,等.顾桥矿区GPS基准网高程系统转换研究[J].矿山测量,2009(6):61-65.(CHEN Shaojie,QIN Yongyang,SUN Xingping,et al.Research on height system transformation of the GPS datum network in Guqiao coal mine[J].Mine Surveying,2009(6):61-65.)
• [15]中华人民共和国国家质量监督检验检疫总局,中国国家标准化管理委员会.国家三、四等水准测量规范:GB/T 12898—2009[S].北京:中国标准出版社,2009.(General Administration of Quality Supervision,Inspection and Quarantine of the People’s Republic of China,Standardization Administration of the People’s Republic of China,Specifications for the third and fourth order leveling:GB/T 12898—2009[S].Beijing:Standards Press of China,2009.)