论文部分内容阅读
针对脉冲燃烧风洞中的测力系统,提出了一种动态标定方法。利用力锤在模型表面上不同位置,沿不同方向施加一系列集中载荷,由输入载荷和天平输出辨识出该表面对应的单位脉冲响应函数(UIRF),再将各表面对应的UIRF加权得到系统的UIRF,加权系数由试验状态下各表面的压力分布确定。辨识某表面对应的UIRF时,通过将其参数化使反卷积问题转化为参数优化问题以回避问题的病态特性。求解参数优化问题时,先用遗传算法搜索到参数全局最优解的近似值,再以此作为单纯形方法的初值继续优化得到参数最优值。在ANSYS中模拟了动态标定过程,考虑了实际试验中输出应变含有较大噪声的情况,验证了这种动态标定方法的准确性。
Aiming at the force measurement system in pulsed combustion wind tunnel, a dynamic calibration method is proposed. A series of concentrated loads are applied in different directions by using a hammer at different positions on the surface of the model. The unit impulse response function (UIRF) corresponding to the surface is identified by the input load and the balance output, and then the corresponding UIRFs of each surface are weighted UIRF, the weighting coefficient is determined by the pressure distribution of each surface under test conditions. When identifying a UIRF corresponding to a surface, the deconvolution problem is transformed into a parametric optimization problem by parameterizing it to avoid the morbid characteristic of the problem. When solving the parameter optimization problem, the approximate value of global optimal solution of the parameter is firstly searched by using genetic algorithm, and then the optimal value of the parameter is continuously optimized by using the initial value of the simplex method. The dynamic calibration process is simulated in ANSYS, considering the fact that the output strain in the actual test contains more noise, which verifies the accuracy of this dynamic calibration method.