融合传递机理的履带车辆系统级振动状态关联模型研究

Research on the System Level Vibration State Correlation Model for Tracked Vehicles Fusing Transmission Mechanisms

针对履带车辆的振动预测,提出了一种融合传递机理的履带车辆系统级振动状态关联模型。首先对履带车辆的结构进行分析,明确振动传递路径并提出多层次关联模型架构;然后结合深度学习技术构建关联模型,并通过关键位置激励载荷参数筛选对模型进行优化;最后使用真实车辆振动数据集进行振动状态预测。结果表明,与未融合传递机理的关联模型相比,融合传递机理的履带车辆振动关联模型在6个振动指标的预测精度上均获得提升,证明了融合传递机理的振动预测方法的有效性。

For vibration prediction of tracked vehicles,a system-level vibration state correlation model fusing transmission mechanisms is proposed.Firstly,based on the structure of tracked vehicles,the vibration transmission path is clarified,and a multi-level correlation model architecture is determined.Then,the correlation model is constructed using deep learning approaches and optimized by selecting key position excitation load parameters.Finally,real vehicle vibration dataset is used for vibration state prediction.Compared with the method without fusing transmission mechanisms,the proposed correlation model fusing transmission mechanisms improves the prediction accuracy of six vibration indicators,which verifies the effectiveness of the vibration prediction method fusing transmission mechanism.