基于组合赋权的IDP-CRUTIC的齿轮箱故障评价

Application of Combination Weighting Based IDP-CRITIC Model in Gearbox Failure Evaluation

摘要: 针对现有齿轮箱故障评价需要研究设备机理造成的效率底下、功能性不强的问题，本文提出了基于组合赋权的IDP-CRUTIC的齿轮箱故障评价方法。首先对传感器收集到的振动信号数据进行分析，提取相关特征。然后，绘制不同传感器在不同状态下的振动信号时间序列函数，并对这些函数的特征进行了简要分析。其次，对数据提取了平均值，方差这两个用以描述振动数据的总体趋势的特征变量，以及峰度、偏度这两个对判断齿轮箱齿轮故障有着重要作用的特征变量，并利用SPSSPRO对每一组数据进行了特征数据计算。最后，通过特征数据计算对这些变量进行了基于组合赋权的IDP-CRUTIC的评价模型进行有效性分析，选取的特征均能有效地描述齿轮箱故障。仿真和实验结果表明，所提出的方法更能描述出齿轮箱的故障评价。

Abstract: To address the existing gearbox failure evaluation needs to study the mechanism of the equipment resulting in low efficiency and low functionality, this paper proposes a combination of empower-ment-based IDP-CRITIC of gearbox fault evaluation method. Firstly, the vibration signal data col-lected by sensors are analyzed to extract relevant features. Then, the vibration signal time series functions of different sensors in different states are plotted, and the features of these functions are briefly analyzed. Next, the average and variance, which are the two characteristic variables used to describe the overall trend of the vibration data, and the kurtosis and skewness, which are the two characteristic variables that play an important role in determining the gear failure of gearboxes, were extracted from the data, and the characteristic data were calculated for each set of data by us-ing SPSSPRO. Finally, through the calculation of characteristic data, the effectiveness of these varia-bles is analyzed based on the combined weighting based IDP-CRITIC evaluation model, and the se-lected features are all effective in describing gearbox faults. The simulation and experimental re-sults show that the proposed method is more descriptive of the gearbox failure evaluation.