【作者】 魏伟；

【导师】 王中原；

【作者基本信息】 南京理工大学 ， 兵器发射理论与技术， 2009， 硕士

【摘要】 弹箭自由飞行射击试验是兵器研究中必不可少的试验手段,在弹道研究中发挥了十分重要的作用。本文主要针对弹箭自由飞行射击试验所得到的数据,对目前所采用的辨识气动力参数的方法的优缺点进行比较分析,并在此基础上提出了由弹道参数提取气动力系数的新方法。目前辨识气动参数主要采用的两种方法:Chapman-Kirk方法和最大似然方法,Chapman-Kirk法可能在某次迭代后反而使目标函数值增大,在观测数据较少或对数值较小的参数进行估值时,正规方程的系数矩阵有可能变成奇异阵,这时迭代将无法继续进行;而且当所处理的问题规模较大时,每次迭代都需要求解一个逆阵,导致计算量较大并容易产生较大的计算误差,针对这些缺点,本文对Chapman-Kirk法进行了改进,使其更能满足现代弹丸飞行数据处理的要求。并结合具体算例对改进的Chapman-Kirk方法和最大似然方法的优缺点进行比较分析。在此基础上,本文提出了一种基于气动参数变化趋势的气动参数辨识方法,并针对这种方法给出实际算例,结果表明,此种方法具有通用性,快速性和准确性,为辨识气动力参数提供了一种新的思路。更多还原

【Abstract】 Free-flight range data of projectiles is necessary for the research of weapon and also plays an important role in the study of ballistics. Using the free-fight range data, various methods of aerodynamic parameters identification are analyzed and compared with each other. Therefore, a new kind of method for aerodynamic parameters identification is proposed in this dissertation.At present, Chapman-Kirk Method (OKM) and Maximum Likelihood Method (MLM) are both in popular for aerodynamic parameters identification. However, there are some problems, when using CKM, such as the fact that the value of objective function will probably increase after some certain iterative process, or the coefficient matrices of normal equation will probably become singular as the measurement is scarce or the estimating parameters are small, or it is extremely difficult to calculate the inverse matrix during the iterative process due to large dimensions. Aiming at all the problems concluded above, CKM is improved so as to meet the requirement of data processing for modern projectiles. The improved CKM is verified and compared with MLM by virtue of an example.At last, a new kind of method for aerodynamic parameters identification, based on the trend of aerodynamic parameters, is proposed and applied to a practical calculation example. The results indicate that this new method has generality, rapidity and accuracy for identification of aerodynamic parameters. Some conclusions above are useful for further research.更多还原