【摘要】 基于机械设备运行过程中的监测数据及状态信息，提出一种基于经验模态分解（Empirical Mode Decomposition, EMD）和支持向量机（Support Vector Machine, SVM）的机械设备可靠性评估方法。该方法采用EMD方法完成数据预处理，剔除状态信号中的噪声；利用相关系数准则，选取合适的模态分量（Intrinsic Mode Function, IMF）重构原始信号；采取时域分析方法提取多维特征向量，从单调能力和鲁棒能力两个维度完成特征向量评价与筛选；采用SVM方法计算各特征向量到分类面的距离，建立设备可靠度评估模型。最后，基于某滚动轴承数据集，完成可靠性评估案例分析。更多还原

【Abstract】 Based on the monitoring data and state information of mechanical equipment during the operation, an empirical mode decomposition(EMD) and support vector machine(SVM) based reliability assessment method for mechanical equipment was proposed in this paper. EMD method is used to complete the data preprocessing and eliminate the noise in the state signals. Using the correlation coefficient criterion,the intrinsic mode function(IMF) was selected to reconstruct the original signals. The multi-dimensional feature vector is extracted by time domain analysis, and the feature vector is evaluated and screened from two dimensions of monotonic capability and robust capability. SVM method is used to calculate the distance between each feature vector and the classification surface, and the equipment reliability evaluation model is established. Based on a rolling bearing data set, the reliability assessment case analysis is completed.更多还原