Modal Analysis of Vibration Signals of Full-size Oriented Strand Board Based on ICEEMDAN−Laplace

In order to non-destructively and rapidly detect the dynamic shear modulus and modulus of elasticity of full-size oriented strand board, an intrinsic modal identification method under the conditions of the vibration method is proposed, which was achieved by a combination of Improved Complete Ensemble Empirical Modal Decomposition of Adaptive Noise(ICEEMDAN) and Laplace wavelet correlation filtering method. First, the foot-size oriented strand board vibration signals were acquired and decomposed by ICEEMDAN to obtain several intrinsic modal functions(IMF) components with different eigen-scales; secondly, the orthogonality index and correlation coefficient were used as the criterion index, and the quadratic modal decomposition was integrated to screen out the IMF components containing modal information of the full-size oriented strand board; finally, the intrinsic frequency of the full-size oriented strand board was extracted by Laplace wavelet correlation filtering, and the detection result was verified by comparing it with the static experiments and simulation analyses. The results showed that the error between the frequency experiment and the simulation results was small, within 0.8%. The correlation coefficients between the dynamic and static results of the elastic modulus in the length and width directions were 0.81282 and 0.74848, respectively, which proved that there was a significant linear relationship between the results. The in-plane shear modulus results from the dynamic and static tests also showed a significant linear relationship, with a correlation of 0.73385.