固体火箭发动机参数辨识软件关键技术研究

固体火箭发动机参数辨识技术日益成熟,但相应参数辨识软件在实现过程中仍面临两个关键技术问题,即试验数据的直观交互前处理和数据匹配以及精确计算复杂几何构型装药的燃面-肉厚数据以供辨识模型模块调用。为解决上述两个问题,本文通过调用R语言内核,对试验数据进行必要的前处理和直观数据采样。将试验数据曲线与理论计算结果曲线之间的时间偏差量作为待辨识参数纳入辨识过程;并利用实体造型法,以访问几何造型平台内核的方式计算燃面-肉厚关系。通过对一组典型的翼柱形装药发动机试车数据进行辨识,验证了该方法的可行性、可信性和高效性。

Research on enabling techniques for identification software for solid rocket motor model parameters

The identification technique for solid rocket motor (SRM) model parameters is growing increasingly practical, but the realization of associated identification software is confronted with two major challenges, which are intuitive preprocessing of and adaptation to test data, as well as computing burning surface area versus web distance of grain with complex configurations for the identification procedure. To meet such challenges, the R programming language core is called for data preprocessing and intuitive data sampling, time deviation between predicted results and test data is incorporated into the identification process, and the solid modeling method is adopted, with the core of solid modeling platform accessed for computing burning surface versus web distance. Identification is carried out on test data of a SRM with fin-cylindrical grain configuration, which verifies the feasibility, trustworthiness and effectiveness of the method.