中文摘要：

为评估不同模型的粗差探测性能,使用多余观测数、内部可靠性和外部可靠性等指标对飞行器航迹测量数据处理中的3种典型模型的粗差探测能力进行了定量评估和比较分析。建立了不同模型间性能比较的等效规则,按照典型的工程应用场景进行了仿真实验。结果表明：大数据集模型的系统结构较强,具有更优的单粗差探测性能,但大模型对于多粗差探测是不利因素;在实际应用中,应根据粗差发生的统计特征,选取数据量大小合适、结构强度高的模型作为优选模型。

英文摘要：

To evaluate the gross-error detection performance of different models, the redundancy number, internal and external reliability index etc. were introduced to quantitatively compare and analyze the gross- error detection performance of three typical models used in the aircraft tracking-data process. The equivalent rules for index comparison between different models were established; the simulation experiment in conformity with the actual engineering application was carried out. The results show that the model using a larger data set has stronger structural strength and better single gross-error detection per- formance,but it is drawback to gross-error detection. It is suggested that in practice, the model with stronger structural strength and the proper size is chosen as the optimal model according to the statistical characteristics of gross-error occurrence.