中文摘要：

介绍了GOCE卫星重力梯度数据系统误差的常用求定方法，提出了一种联合卫星轨迹交叉点不符值和现有重力场模型的系统误差综合标定方法．给出了分步解算和整体平差两种解算方法及相应的计算步骤．分步解算是先利用卫星轨迹交叉点不符值确定含尺度影响的偏差漂移项，然后对观测值进行偏差漂移改正，并利用现有重力场模型计算尺度和偏差，最后对偏差漂移进行尺度改正．整体平差是以现有重力场模型标校方程作为观测方程，以卫星交叉点不符值作为条件约束，由此采用附加条件的参数平差模型，整体解算尺度、偏差和偏差漂移．论文最后利用算例验证了本文综合标校方法的有效性，结果表明：无论分步解算或整体平差，标校出的偏差参数相对误差最大值均小于1．2％，平均值小于0．5%；尺度因子相对误差最大值小于2％，平均值小于0．8％；偏差漂移参数相对误差最大值小于0．4％，平均值小于0．2％．整体平差结果的精度高于分步解算结果精度．

英文摘要：

After a brief review of the methods of systemic error calibration for GOCE gravity gradients, a synthetic calibration method is developed based on satellite crossover differences and the existing gravity field model. Two kinds of algorithms, namely step by step and combined adjustment are presented. The step by step algorithm solves the bias drift influenced by scale factor firstly. Then it is moved from the gravity gradient observations, and the existing gravity field model is used to calibrate the scale factor and bias. Lastly, the originally solved bias drift including scale influence is corrected by the solved scaled factor. The combined adjustment uses the exiting gravity field information to calibrate observed gravity gradients as observation equation, and the crossover differences as conditional equation. So the parameter adjustment model including conditional constrain is adopted to solve the problem. A numerical example from the simulated gravity gradients of radial component of GOCE satellite is performed. The results show that the new synthetic calibration method can efficiently solve the systematic error parameter using both of above algorithms. The maximum relative errors （RE） of bias is less than 1. 2%, the average is less than 0. 5 %. The maximum RE of scale factor is less than 2 G, the average is less than 0.8 G. The maximum RE of bias drift is less than 0.4%, the average is less than 0.2%. The calibration precision of the combined adjustment is higher than that of the step by step algorithm.