基于非线性超声导波的大电机定子绝缘声学检测与损伤识别研究

Insulation structure aging is the main root cause of stator insulation invalidation and failure outage of large generator, and insulation damage formation and development are the concrete manifestation. However, due to lacking of accurate insulation aging model, the existing structure health monitoring (SHM) methods of stator insulation are still stay on experiences and macro statistics level, which cannot quantitatively describe the aging process and failure mechanism of stator insulation. Based on the nonlinear ultrasonic theory and by utilizing the nonlinear effect ultrasonic guided wave coupling micro-damage of insulation structure, this project study tiny insulation micro-damage signal acquisition, damage feature extraction and state performance representation under complex conditions, illustrating the action law that insulation damage generating nonlinear ultrasonic features and the quantitative relationships that damage variable vs. material nonlinear coefficients vs. ultrasonic features. Based on the ultrasonic excitation and Fiber Bragg Grating (FBR) sensing, researches are focused on studying the detection methods of insulation micro-damage, pinpointing the mechanism of interaction between ultrasonic excitation and fiber Bragg grating, acquiring the acoustic field distribution characteristics of ultrasonic guided wave propagation in stator insulation, meanwhile, the ultrasonic response properties of FBR and the matching character of FBR with ultrasonic excitation. Identification methods of insulation micro-damage based on multi-source information are being studied so as to monitor initiate damage and structure health condition of stator insulation. Research on this project has great theory significance and utility value for the development of guided wave based structure health monitoring technique of stator insulation.

绝缘结构老化是导致大型发电机定子绝缘失效和引起故障停机的主要原因，具体表现为绝缘损伤的形成与发展。因缺乏准确的绝缘老化模型，现有的绝缘健康性监测方法仍停留在经验和宏观统计水平上，而对其老化过程及失效机理还不能定量描述。项目基于非线性超声理论，利用超声导波与绝缘结构微小损伤耦合的非线性效应，研究复杂条件下绝缘微小损伤信号获取、损伤特征提取及状态性能表征，揭示绝缘损伤产生非线性超声特征的作用规律，获取损伤变量－材料非线性系数－超声特征之间的定量关系。研究基于超声激励和光纤光栅传感的绝缘微小损伤检测方法，明确超声激励与光纤光栅传感的作用机理，获取超声导波在定子绝缘中的声场分布特征、光纤光栅的超声响应特性及超声激励与光纤光栅传感的匹配特性。研究基于多源信息的绝缘微小损伤识别方法，以期在定子绝缘损伤出现早期监测其健康状况。项目研究对发展基于导波的定子绝缘结构健康监测技术具有重要的理论意义和实用价值。

大型发电机定子绝缘老化状态在很大程度上决定了发电机的寿命及运行可靠性。定子绝缘老化与绝缘内部（或表面）微观损伤的形成和发展有着密切的关系。目前对定子绝缘老化机理还不能定量描述，对于反映绝缘老化的特征参量仍停留在经验和宏观统计的水平上，而对其微观本质即结构性损伤的形成与发展仍缺乏系统的分析和研究。因此，及时、准确地识别出绝缘损伤的形状、大小和位置等信息对于定子绝缘老化状态评估具有十分重要的意义。.针对大电机定子绝缘微小损伤（亚毫米级损伤）检测问题，基于非线性超声导波理论，利用超声导波与绝缘结构微小损伤耦合的非线性效应，研究了复杂条件下绝缘微小损伤信号获取、损伤特征提取及绝缘微小损伤特性表征方法。研究了基于非线性超声导波的定子绝缘微小损伤检测与定位方法。研究了基于多特征融合的定子绝缘微小损伤识别方法，以实现对定子绝缘早期损伤及健康状况进行有效监测。.研究结果表明：基于定子绝缘单损伤模型，利用Lamb波与绝缘损伤耦合的非线性效应，能够获取损伤变量－材料非线性系数－超声特征之间的定量关系。利用超声导波产生的非线性二次谐波不仅可以有效检测出定子绝缘微小损伤（亚毫米级微小损伤），而且还能够对损伤进行定位。提出了一种基于Lamb波复合特征的损伤成像方法，能够有效识别绝缘损伤的形状和大小。在此基础上，提出了考虑定子绝缘损伤几何参数的多特征融合方法，能够对定子绝缘内部气隙、分层、贯穿性裂纹损伤及表面损伤等典型损伤类型进行有效识别。.通过开展项目研究，主要解决了定子绝缘结构中微小损伤模型的建立与优化机制，以及超声导波与绝缘结构微小损伤的非线性耦合特征提取及特性表征、多源激励与波包信号解调方法等关键基础性问题，形成了基于非线性超声导波的大电机定子绝缘结构健康监测和微小损伤识别的基础理论和方法，能够和已有的绝缘状态监测方法形成优势互补，为准确评估大电机定子绝缘状态奠定理论和技术基础。