基于近似模型和模式搜索法的履带车辆多体动力学模型参数修正研究

Research on Parameters Updating for Tracked Vehicle Multibody Dynamic Model Based on Approximation Model and Pattern Search Method

履带车辆多体动力学模型仿真结果的准确度直接影响其动力学性能分析结果的可信性。为提高仿真结果的准确度,研究了多体动力学模型参数的修正方法。首先建立了履带车辆多体动力学模型,通过对比仿真结果与实车测试结果,初步验证了模型的可信性;并给出了参数修正目标函数的表达式。通过参数筛选确定了待修正模型参数。采用克里格插值法,构造了待修正参数与目标函数之间的近似模型;解决了修正效率低、计算量大的问题。针对修正问题的无约束性和非线性,采用模式搜索法进行参数修正的迭代计算。修正结果表明,动力学模型仿真结果的准确度得到了提高,证明了修正方法的有效性。

The credibility of tracked vehicle dynamic performance analysis results is directly affected by the accuracy of simulation results. In order to improve the accuracy of dynamic simulation results,a method of multibody dynamic model parameters updating is researched. Firstly,a multibody dynamic model of tracked vehicle is established,and then the model credibility is validated initially according to real vehicle test results,in addition,the expression of model parameters updating objective function is given. The updating parameters are determined by means of parameter screening. To solve the problem of large calculation quantity and inefficiency of parameters updating,the approximation models of parameters to objective function are built using Kriging Interpolation Method.For the unconstraint and nonlinearity of updating problem,the iteration of parameters updating is carried out by using Pattern Search Method. The results show that the accuracy of dynamic model simulation results is improved and the availability of method of parameters updating is validated.