Despite the increasing popularity of photonic Doppler velocimetry (PDV) in shock wave experiments, its capability of capturing low particle velocities while changing rapidly is still questionable. The paper discusses the performance of short time Fourier transform (STFT) and continuous wavelet transform (CWT) in processing fringe signals of fast-changing low velocities measured by PDV. Two typical experiments are carried out to evaluate the performance. In the laser shock peening test, the CWT gives a better interpretation to the free surface velocity history, where the elastic precursor, main plastic wave, and elastic release wave can be clearly identified. The velocities of stress waves, Hugoniot elastic limit, and the amplitude of shock pressure induced by laser can be obtained from the measurement. In the Kolsky-bar based tests, both methods show validity of processing the longitudinal velocity signal of incident bar, whereas CWT improperly interprets the radial velocity of the shocked sample at the beginning period, indicating the sensitiveness of the CWT to the background noise. STFT is relatively robust in extracting waveforms of low signal-to-noise ratio. Data processing method greatly affects the temporal resolution and velocity resolution of a given fringe signal, usually CWT demonstrates a better local temporal resolution and velocity resolution, due to its adaptability to the local frequency, also due to the finer time-frequency product according to the uncertainty principle.
A baseline-free damage identification method is proposed to identify damages in metallic sandwich panels with truss core in the article. The method is based on flexibility matrix and gapped smoothing method, with damage index defined
DI
m . The weight coefficient m is introduced to consider the effect of damages on both low-order modes and high-order modes. Numerical simulations and experiments are conducted to evaluate the present method. Besides, damage index is also defined by processing
DI
m with Teager energy operator, and comparisons between
DI
m and are also carried out. Results show that the proposed method is effective in detecting single damage and multiple damages of the same or different extent. The weight coefficient m plays a very important role in identification of multiple damages of different styles. When comparing with , it is found that the present index
DI
m is better at suppressing the singularity caused by contact nodes and detecting of multiple damages which contain small or slight damages.