大电机定子绝缘损伤成像检测方法研究

为了准确评估定子绝缘结构健康状况,基于导波理论进行定子绝缘结构损伤成像检测研究。首先根据定子绝缘材料特性参数,建立定子线棒结构有限元模型。对定子绝缘损伤检测中导波的激励、传播和接收进行数值模拟,并建立横向裂纹、纵向裂纹、气隙和分层四种典型定子绝缘损伤模型。利用希尔伯特变换提取损伤散射信号到达时间作为损伤特征,采用概率成像方法对定子绝缘中的损伤状况进行成像检测。数值仿真结果表明:损伤散射导波飞行时间是表征损伤位置的有效特征,A0模态损伤散射波对定子绝缘损伤更敏感,概率成像能够直观地检测出定子绝缘中的损伤并能进行准确定位,为进一步应用导波进行大电机定子绝缘状态评估提供有效的参考信息。

Damage Detection of Large Generator Stator Insulation Using Guided Wave-based Imaging Method

In order to accurately assess the structure condition of stator insulation, the basic theory of guided wave and probability-imaging method are employed in this study for stator insulation damage detection of large generator. The excitation, propagation and acquisition of guided wave for stator insulation damage detection are simulated by numerical simulation. Finite Element models of four typical insulation damage, i.e. void, delamination, longitudinal and transverse crack are established. Hilbert Transform (HT) was used to extract the time of flight (ToF) of damage scattered wave for damage location. Numerical simulation results reveal that, ToF is an effective feature that can represent the damage location; the A0 mode Lamb wave is more sensitive to the insulation damage. the probability-imaging method can used to not only detect but also accurately locate the specific insulation damage location，which can provide effective information for further study of the stator insulation condition assessment based on the guided wave.