Tracking Algorithm of Natural Vibration Modes Using Image Operation

Abstract

A new algorithm using image operation was developed for identifying and tracking a natural vibration mode that is one of the most important factors in structural design. It enables the automatic survey of parameters in the vibration design of structures even when mode switching (crossing) occurs. Mode switching is the change of order in which natural vibration modes appear. The developed algorithm performs the following tasks. First, the absolute value of each component of a vector corresponding to a natural vibration mode is represented as a gray-scale image by a post processor. Each of the gray-scale images is then binarized. The numbers of groups consisting of pixels with the same color are counted in each binary image, and the numbers are used for distinguishing the natural vibration mode to be tracked from the other modes. A structural optimization system was developed. It consists of the software routine using the developed algorithm, a commercial optimization software, pre and post processors, and a calculation solver. The effectiveness of the algorithm using a cantilever plate whose natural vibration mode appearance order varies in changing the design parameters was tested and verified. A new reinforcing rib was then added to the cantilever plate that requires re-meshing. The effectiveness of the algorithm was also tested and verified on the modified plate.