摘要：This paper presents a hybrid damage detection method based on continuous wavelet transform (CWT) and modal parameter identification techniques for beam-like structures. First, two kinds of mode shape estimation methods, herein referred to as the quadrature peaks picking (QPP) and rational fraction polynomial (RFP) methods, are used to identify the first four mode shapes of an intact beam-like structure based on the hammer/accelerometer modal experiment. The results are compared and validated using a numerical simulation with ABAQUS software. In order to determine the damage detection effectiveness between the QPP-based method and the RFP-based method when applying the CWT technique, the first two mode shapes calculated by the QPP and RFP methods are analyzed using CWT. The experiment, performed on different damage scenarios involving beam-like structures, shows that, due to the outstanding advantage of the denoising characteristic of the RFP-based (RFP-CWT) technique, the RFP-CWT method gives a clearer indication of the damage location than the conventionally used QPP-based (QPP-CWT) method. Finally, an overall evaluation of the damage detection is outlined, as the identification results suggest that the newly proposed RFP-CWT method is accurate and reliable in terms of detection of damage locations on beam-like structures.
