基于证据支持矩阵的特征权重融合的风电机组故障诊断

Fault Diagnosis of Wind Turbine Based on Feature Weight Fusion Method with Evidence Support Matrix

针对风电机组齿轮箱故障特征的不确定性、复杂性和多元性的特点,提出基于证据支持矩阵特征权重的融合新方法,建立故障诊断模型。分析了影响证据冲突的冲突因子和证据距离,利用这两个因子构造证据支持度矩阵;求解了该证据支持度矩阵最大特征值对应的特征向量,并将此作为证据的权重,利用证据组合公式进行融合;最后将其用于风电机组齿轮箱故障诊断。实验结果表明,该方法可较好提高风电机组齿轮箱故障诊断的效率和准确率。

In view of uncertainty, complexity and diversity in the fault characteristics of the wind power unit＇s gearbox, a method of feature weight fusion based on evidence support matrix is proposed to construct the model. Two factors, the conflict coefficient and evidence distance, are analyzed, which affect the evidence conflict. An evidence support matrix is constructed. The char- acteristic vector related to the maximum eigenvalue of the matrix is calculated as an evidence weight. The evidence weights are fused using the evidence combination formula. The method is applied to fault diagnosis of a wind turbine gearbox. The results show that the proposed method can enhance efficiency and accuracy of the fault diagnosis of wind turbine gearbox.