动车组齿轮箱传感器优化布置

Optimum Layout of Sensors in EMU Gearbox

为了解决动车组齿轮箱结构健康监测中的传感器优化布置问题,针对有效独立法存在的不足,引入比例系数的概念,用模态振型比例系数评价两测点间的信息独立程度,提出了一种能够同时满足所选测点模态可测性和避免信息冗余性的测点优化布置算法,并提出了基于振动模态的置信准则评价测试点选取方法的优劣。首先建立齿轮箱箱体模型,采用Block lanczos算法提取模态参数;其次构建模态的Fisher信息矩阵;再次利用比例系数-有效独立法对齿轮箱节点迭代筛选出传感器布置位置的6个最优节点;最后运用模态保证准则验证了齿轮箱传感器布置的合理性。结果表明:比例系数-有效独立法是一种更优的测点布置方法,用其研究分析齿轮箱结构的振动模态信息,能更加精确地诊断出齿轮箱的故障特征。该方法有助于准确分析齿轮箱箱体的振动特性,为高速列车齿轮箱的故障诊断和健康预测提供理论依据。

In order to solve the problem existing in optimal arrangement of sensors in the structural health monitoring of EMU gearboxes,and considering the defects of the effective independent method,the concept of proportional coefficient is introducedhere,and the modal mode proportional coefficient is used to evaluate the degree of information independence between two measuring points.An optimal arrangement algorithm of measuring points that can satisfy the modal measurability of the selected measuring points and avoid information redundancy at the same time is proposed.And a confidence criterion based on vibration mode is also put forward to evaluate the advantages and disadvantages of the test point selection method.Firstly,the gearbox box model is established,and the Block lanczos algorithm is used to extract the modal parameters.Secondly,the Fisher information matrix of the modal is constructed.The proportional coefficient-effective independent method is reused to iteratively screen out the 6 optimal sensor placement positions for the gearbox nodes.Finally,the rationality of the sensor arrangement of the gearbox is verified by the modal guarantee criterion.As revealed by the results,the proportional coefficient-effective independent method is used to study and analyze the vibration modal information of the gearbox structure,which is a better measuring point arrangement method and can more accurately diagnose the fault characteristics of the gearbox.The method helps to accurately analyze the vibration characteristics of the gearbox casing,and provides a theoretical basis for the fault diagnosis and health prediction of high-speed train gearboxes.