基于试验数据的魔术轮胎模型参数估计

为了准确估计魔术轮胎模型参数并构建可信的轮胎动力学模型,基于某型轮胎的纵向力及侧向力试验数据,以测试数据和模型拟合数据残差加权平方和最小为目标函数,采用麦夸尔特法（Levenberg-Marquardt）与全局优化法相结合的混合优化算法,分别对8个轮胎纵向力及10个轮胎侧向力模型参数进行估计.拟合优度统计检验结果表明魔术轮胎模型回归效果非常显著,拟合精度较高.纵向力及侧向力模型拟合曲线与试验测点拟合较好,纵向刚度及侧偏刚度测试结果与计算结果的最大相对误差分别只有8.48％和9.96％,具有很好的一致性,验证了参数估计的准确性.

Identification of Magic Tyre Mode Parameters Based on Experimental Data

In order to estimate magic tyre model parameters with more confidence, the weighted sum of squares residual （WSSR） between experimental data and fitted data is taken as objection function, a kind of hybrid optimizing algorithm has been adopted to estimate eight longitudinal force magic model parameters and ten lateral force magic model parameters based on experimental data of tire longitudinal force and lateral force. The applied optimizing approach is the combination of Levenberg-Marquardt local method and global optimization method. The regression effect of magic tire model is very significant and fitting accuracy is high in testing of goodness of fit. The accuracy of parameter estimation is verified by the small relative error （8.48% and 9.96%） between experimental values and fitted values of longitudinal stiffness and cornering stiffness.