东南大学学报（自然科学版）Issue(z2)：375-379,5.DOI:10.3969/j.issn.1001-0505.2013.S2.034
基于重力场模型和神经网络融合的大范围测区 GPS 高程转换方法
GPS height conversion method based on combination of gravity field model and neural network in large scale
摘要
Abstract
In order to improve the accuracy of GPS ( global positioning system) height conversion in large scale, based on the conventional GPS height conversion methods such as plane fitting, quadric surface fitting and polyhedral function fitting and so on, a new GPS height conversion method based on the combination of the gravity field model and neural network is presented.A large scale measur-ing range in Jiangsu Province is taken as an example.The comparison of the GPS height conversion conventional methods and the proposed method based on the gravity field model ( EGM2008 , EIGEN-6C2,EIGEN6C) and neural network are carried out.The results show that the conversion accuracies of plane fitting, quadric surface fitting and polyhedral function fitting are 1.367 2,0.122 4 and 0.130 6 m, respectively.However, the conversion accuracies of the methods combining EGM2008, EIGEN-6C2, EIGEN-6C with neural network are 0.041 1,0.038 1 and 0.039 2 m, re-spectively.Therefore, the combination of the gravity field model and neural network can greatly im-prove the precision of GPS height transformation in large scale.关键词
重力场模型/神经网络/GPS高程转换Key words
gravity field model/neural network/GPS ( global positioning system) height conversion分类
天文与地球科学引用本文复制引用
孙腾科,赵杏杏,张永磊..基于重力场模型和神经网络融合的大范围测区 GPS 高程转换方法[J].东南大学学报（自然科学版）,2013,(z2):375-379,5.基金项目
国家自然科学基金资助项目（41204016，41274017）、江苏省科技支撑计划资助项目（BE2010316）． （）
