Comparison on variations of natural frequencies for wooden structural models with and without damages part 2, shaking table based results

Following part 1, this part 2 reports the shaking table results, and the comparisons between the simulations and the experiments. Because the small size of our shaking table (only 1 in by 1 in), the model structures have a special scale reduction, which have been precisely calculated without loss of real structural characteristics. Input waves are completely same with those in part 1. The shaking table result has also clearly shown that for one construction type, the natural frequency decreased following the increase of damages. But for the other type structure, the natural frequency remains unchanged even the damage increases. Those results completely identify with the simulation results from ANSYS in part 1. One strange result appears that even if the degree of damages increases, the natural frequency remains unchanged. Such results will have minus effect for the research itself of structural health monitoring. It is a fact that many monitoring methods depend on the decrease of natural frequencies while we have damages in the building, but the result of this report clearly have shown that sometimes the frequency unchanged, which will in turn show the impossibility of the frequency based health monitoring system. It is a big confusion but we believed the problems will be solved this year, because we planned to do same experiment on the real building this year, and the results on the real building will be reported.