D-最优试验设计在动力调谐陀螺测试中的应用  

Application of D-optimal Test Design Method in Testing of Dynamically Tuned Gyroscope

作　　者：富立[1] 刘文丽[1] 

机构地区：[1]北京航空航天大学自动化科学与电气工程学院,北京100083

出　　处：《航空学报》2008年第2期467-471,共5页Acta Aeronautica et Astronautica Sinica

基　　金：国家"973"基础研究项目

摘　　要：为了提高动力调谐陀螺的静态漂移误差测试精度,针对简化静态漂移误差模型,并且考虑陀螺安装角误差的影响,依据D-最优试验设计理论以及正交设计的均匀分散性试验设计思路,设计了一种满足D-最优性质的正交八位置试验方案,并在整个正交空间内采用全局搜索的方法验证了其D-最优性。在相同试验环境下,采用力反馈位置试验法,以全正交空间预测误差平方和最小为评价准则,对该八位置试验方案和国军标GJB1183—91中的八位置试验方案进行比较与评价,试验结果表明该八位置试验方案优于国军标GJB1183—91中八位置试验方案。In order to obtain a better estimate of the acceleration-insensitive drift rate, acceleration-sensitive drift rate coefficients and misalignment angles of the assumed performance model of the two-degree of freedom dynamically tuned gyroscope, a new D-optimal eight-position test method is proposed, according to the D-optimal test design method and the symmetric separate design idea of orthogonal design. A search method in the whole orthogonal space is used to verify the D-optimal property of the new test method. Under the same test conditions and test equipment required, gyro torque rebalance test method is employed to compare the proposed eight-position test method with standard test method in the General specification for dynamic tuned gyro in the whole orthogonal space, using the square of predictive error minimum approach. The experimental results show that this eight-position test method is at least 4 times more accurate than that in the General specification for dynamic tuned gyro used in inertial navigation system.

关 键 词：最优试验设计 动力调谐陀螺 误差模型 漂移测试 八位置 

分 类 号：V241.559[航空宇航科学与技术—飞行器设计]