论文部分内容阅读
仿真系统的可信度检验已经是仿真研究的一个重要组成部分,也是评价飞行模拟器能否反映客观实际的重要依据。论文根据飞行模拟器静态飞行性能验证的实际特点,讨论了进行飞行模拟器静态飞行性能验证对数据的要求,介绍了模拟器静态数据一致性检验的方法,并提出对于样本量较少的试飞数据采用Bootstrap法将小样本观测量处理产生大样本观测子样,再对两较大样本观测量进行非参数检验,检验结果相容后再进行参数假设检验,最终得出样本相容性较好,试飞数据与仿真数据属于同一样本分布。结论表明,对于样本观测量相差较大的数据,可以通过自助Bootstrap法扩大小样本容量,再进行假设检验,可获得很好的检验效果,为处理样本相差较大观测量相容性检验提供方法。
The reliability test of the simulation system is an important part of the simulation research and an important basis for evaluating whether the flight simulator can reflect the objective reality. According to the actual characteristics of static flight performance verification of flight simulator, the paper discusses the data requirements of flight simulator static flight performance verification, introduces the method of consistency test of static data of simulator, and proposes that for the test data with small sample size Bootstrap method is used to process small sample observations to generate large sample observation sub-samples, then non-parametric test is performed on the two larger sample observations. After the test results are compatible, then the parameter hypothesis test is performed, and finally the sample compatibility is better. Test data and simulation data belong to the same sample distribution. The conclusion shows that for the data with large difference between sample observations, the self-help Bootstrap method can be used to expand the sample size and then test the hypothesis to obtain a good test result, providing a method for testing the compatibility of large-scale observational samples.