基于振动模态特征分析的车辆悬架系统状态监测

发布时间：2018-04-21 20:48

本文选题：模态参数识别 + 悬架系统　；参考：《太原理工大学》2017年硕士论文

【摘要】：近年来,人们对汽车平顺性和安全性的要求越来越高。悬架系统作为直接影响车辆安全性、平顺性和操稳性的重要部件,如果其发生故障可能会造成严重的经济与人身安全损失。统计发现,不同故障原因类型中悬架部件损坏造成的故障比率非常高。因此,有必要对悬架的故障诊断及在线监测进行深入的研究,从而降低悬架故障导致的损失。本研究致力于研究对悬架系统的在线监测方法,从而及时发现悬架早期故障并加以排除来有效地降低事故发生的概率,最大限度地保证行车安全,并提高汽车的行驶平顺性。但是由于路面情况复杂,路面不平度对轮胎的激励一般为特定频段的随机激励,如何对悬架系统的运行状态进行在线监测一直是故障诊断领域的研究难点。随机子空间法作为一种将响应信号作为输入的模态识别算法,具有信号采集方便、适用于随机振动状态和高噪声工况下鲁棒性强等优点受到了越来越多学者的关注。但其识别准确性受阻尼比的影响较大,而悬架系统的阻尼比较高(在20%-30%之间),这限制了随机子空间法在车辆悬架系统监测中的应用。因此本文提出将改进后的平均相关子空间方法(ACS-SSI)应用于悬架在线监测中,采用多次平均后的相关函数信号取代原算法采用的响应信号作为算法输入,从而大大提升了算法在复杂工况下的识别精度。然后建立七自由度线性车辆振动模型,仿真识别了在不同阻尼比、路面激励及噪声条件下的模态参数,以此判断其对识别结果的影响,并验证在悬架状态监测上应用平均相关随机子空间算法的合理性。其次,判断各个模态参数对故障的灵敏度,并基于此判断建立了基于振型和模态能量差异法作为依据的在线监测方法。最后,设计传感器布置方案对车身姿态的相关参数进行采集,最终确定了利用9轴MEMS惯性传感器采集车身垂向振动,车身俯仰角速度以及车身侧倾角速度,并进行悬架系统不同的故障形式的监测试验,从而验证了在线监测算法的可信度。
[Abstract]:In recent years, the requirements of vehicle ride comfort and safety are becoming higher and higher. Suspension system is an important component which directly affects vehicle safety, ride comfort and operating stability. If it breaks down, it may cause serious economic and personal safety losses. According to statistics, the failure rate of suspension parts is very high in different fault cause types. Therefore, it is necessary to study deeply the fault diagnosis and on-line monitoring of suspension, so as to reduce the loss caused by suspension failure. This study is devoted to study the on-line monitoring method of suspension system, so as to find the early fault of suspension in time and remove it to reduce the probability of accident, to ensure the safety of driving to the maximum extent, and to improve the ride comfort of the vehicle. However, due to the complexity of pavement conditions, the road roughness to the tire excitation is generally random excitation in a specific frequency band, how to monitor the running state of suspension system online has been a difficulty in the field of fault diagnosis. As a modal recognition algorithm which takes response signal as input, stochastic subspace method has been paid more and more attention by more and more scholars because of its convenience in signal acquisition, good robustness under random vibration and high noise condition, and so on. However, the accuracy of identification is greatly affected by the damping ratio, and the damping ratio of suspension system is high (between 20% and 30%), which limits the application of stochastic subspace method in vehicle suspension system monitoring. In this paper, the improved average correlation subspace method (ACS-SSI) is applied to the on-line monitoring of suspension, and the response signal of the original algorithm is replaced by the multi-average correlation function signal as the input of the algorithm. Thus, the recognition accuracy of the algorithm under complex working conditions is greatly improved. Then the vibration model of a seven-degree-of-freedom linear vehicle is established, and the modal parameters under different damping ratio, road excitation and noise are identified, and the influence of the modal parameters on the identification results is judged. The rationality of applying the average correlation random subspace algorithm to suspension condition monitoring is verified. Secondly, the sensitivity of each modal parameter to the fault is judged, and an on-line monitoring method based on mode shape and modal energy difference method is established. Finally, the sensor layout scheme is designed to collect the relative parameters of the body attitude. Finally, the 9 axis MEMS inertial sensor is used to collect the vertical vibration of the body, the pitch angle velocity of the body and the roll angular velocity of the body. The reliability of the online monitoring algorithm is verified by testing the different fault forms of suspension system.
【学位授予单位】：太原理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：U463.33

【参考文献】