基于广义简约梯度算法的履带车辆模型参数修正

利用多体系统动力学软件RecurDyn建立了某型高速履带车辆的多刚体动力学模型。为提高模型的准确度,首先提出履带车辆动力学模型准确度静态和动态评价指标,使模型准确度量化;然后以模型主要参数为设计变量,以评价指标为响应进行归一化参数灵敏度分析,并根据分析结果确定模型修正参数;最后利用广义简约梯度优化算法分静态和动态两种工况对模型参数进行了修正。与实测值的对比结果表明,修正后履带车辆动力学模型的准确度明显提高,证明论文提出参数修正方法的可行性和有效性。该修正方法能够有效地提高履带车辆动力学模型的准确度和仿真结果的可信度,研究成果对履带车辆动力学建模与仿真工作具有重要的参考价值。

Research on Model Updating for Tracked Vehicle Dynamic Model Based on Generalized Reduced Gradient Method

A dynamic model of high mobility tracked vehicle was built up in the environment of multi-body system dynamics software RecurDyn.To improve the model accuracy,model updating method was studied.First,static and dynamic estimation indexes were built up separately for evaluating accuracy of the dynamic model,so that it could be described with numerical value.Second,parameter sensitivity analysis was carried out for selecting updating parameters,in which static estimation index and dynamic estimation index were selected as response value and the main model parameters are selected as design parameter.The updating parameters were selected according to the results of parameter sensitivity analysis.Finally,model parameter updating was carried out based on the generalized reduced gradient optimization method.The comparison results between simulation value and field test value show that model accuracy of the tracked vehicle dynamic model is improved obviously,which proves that the model updating method is feasible and effective.The model updating method can improve the accuracy of dynamic model for tracked vehicle and reliability of simulation result validity.The research results are of important reference value for the researcher of tracked vehicle dynamic modeling and simulating.