Nonlinear Acoustic Technique for Fatigue Damage DetectionPublic Deposited
A major challenge in the field of Structural Health Monitoring (SHM) of safety-critical structures is how to effectively and efficiently detect fatigue damage, especially before there is macrocrack initiation. In this dissertation, a local ultrasonic technique is developed to assess pre-crack fatigue damage in structures. The novel feature of this technique is the employment of a nonlinear acoustic method, which has been shown to be promising to quantify the microstructural changes in materials. To validate this technique, both a laser-interferometer technique and a contact piezoelectric transducer technique are employed to calibrate the experimental setup for nonlinearity measurements. To increase the robustness of the nonlinearity measurements by using contact transducers, a method to minimize couplant-induced variations is developed, which reduces the scatter in nonlinearity measurements by around 50%. Nonlinearity tests using both bulk waves and surface waves have been conducted on three types of Inconel 718 (IN718) specimens---a monotonically loaded tensile specimen, fatigued specimens, and post-mortem fracture specimens. The change of nonlinearity with plastic deformation accumulated in the monotonically loaded specimen indicates that there is a quantitative correlation between the acoustic nonlinearity and the material nonlinearity, especially with the primary cause of plastic deformation---dislocations. This is further proven by hardness tests on the fractured IN 718 specimens. An interesting observation in the nonlinearity study of the fatigued IN 718 specimens is the two peaks in the nonlinearity parameter change during the fatigue process as the number of cycle increases. Consistent results are obtained from six fatigued specimens. A possible explanation for the two peaks may be attributed to recovery of the material and macrocrack initiation, respectively. In order to efficiently use the damage information obtained from the ultrasonic system, a damage model and a probability calculation procedure are developed to estimate the probability of macrocrack initiation. A sample problem is presented to demonstrate that this procedure can correctly forecast the macrocrack initiation based on the updated status of a structure. This potential SHM system employing the nonlinear acoustic technique can be applied with a probabilistic prognosis calculation procedure to establish a condition-based inspection of pre-macrocrack damage accumulation in structures.