基于磁声多物理场耦合效应的汽车转向节缺陷快速成像检测方法研究

The research on defect detection of vehicle is an important part to ensure the driving safety. At present, it is impossible to detect the internal defect of steering knuckle fast and accurately, In view of this, the coupling effect of multiphysics field is proposed to detect the vehicle defect in this project. The new detection method combines the high contrast of electromagnetic imaging with the high spatial of sonography, and it is hoped to detect internal defects of steering knuckle fast and highly precisely. In this project, theory of magnetoacoustic tomography coupled multiphysics field is studied, which combines electromagnetic field with acoustic field. Influence of magnetic field parameters on the signal-to-noise of acoustic signal measured and image resolution are investigated, and the optimal measurement conditions are explored. Fast newton’s one-step error reconstructor is developed based on newton’s one-step error reconstructor. The application of magnetoacoustic tomography coupled multiphysics field is expanded in the detection of defects of vehicle structural parts, and the new techniques and methods are provided for the defect detection of vehicle steering especially.

汽车转向节的缺陷检测研究是保障车辆行驶安全的一项重要内容，目前汽车转向节的缺陷检测存在的一个难题是无法准确快速地实现转向节内部缺陷的检测，鉴于此，本项目提出基于磁声多物理场耦合效应的汽车转向节缺陷检测成像新方法。新方法结合了电磁成像高对比度和超声成像高分辨率的优点，能够实现转向节内部缺陷的高精度和快速检测。本项目研究电磁场、声场多物理场耦合成像理论，研究缺陷检测成像系统涉及的电磁场激励参数、高灵敏度信号检测方法和快速图像重建算法，考察激励磁场参数对测量声信号的信噪比和图像分辨率的影响规律，探索最佳的测量条件，基于一步牛顿法开发与之相适应的快速牛顿一步误差图像重建新方法。此方法拓展了磁声多物理场耦合成像方法在汽车结构件缺陷检测中的应用，特别为汽车转向节缺陷检测提供新的技术和方法。

汽车转向节的缺陷检测存在的一个难题就是无法实现其内部缺陷的早期检测。本项目提出了基于磁声多物理场耦合的转向节快速成像检测方法。利用电场磁场的激励在转向节内产生超声波，由此获得转向节的高分辨率图像。本项目研究了成像检测方法的理论基础，对耦合电磁学、声学和固体物理学进行了理论分析，揭示了电磁、超声等能量形式在高电导率物体内传输规律、互相作用原理、能量转换特点与信号产生机理等。本项目研究多物理场耦合成像检测的正问题，提出并实现了电磁超声耦合检测方法；设计电磁超声换能器，改善信号接收的效果；开发了基于维纳滤波反卷积的时间反演图像重建算法，并搭建电磁超声多物理场耦合成像检测试验平台，实现快速图像重建。实验证明本项目提出的检测方法可以获得高分辨率的电导率图像,能够突破传统汽车转向节缺陷检测的限制——仅能检测转向节表面或者接近表面的缺陷，实现转向节内部缺陷的检测。