On-line, simultaneous identification of multiple modes for flexible structures using adaptive filtering techniques

Abstract

Using adaptive filtering techniques, on-line system identification algorithms have been investigated to identify simultaneously modal parameters of multiple modes for flexible structures. The existing adaptive filter based, on-line modal parameter identification algorithm is designed to identify a single mode at a time. Thus, the algorithm must be supplemented by a band-pass filter to extract from the measurement signal the frequency contents in the vicinity of a desired mode. This filtering approach is not effective for flexible structures with high modal density. The algorithms presented herein are explored mainly to overcome this drawback of the existing algorithm. In this paper, adaptive filters are designed to allow simultaneous identification of multiple modes without using the band-pass filters. The design process and the performance of the adaptive filters are presented for the simultaneous identification of two or more modes. The algorithms have been implemented on a DSP (digital signal processor) system to evaluate the performance of on-line identification. A cantilevered beam was used as a test article along with a shaker and an accelerometer. The algorithm for the identification of two modes runs successfully but a difficulty has been encountered for the simultaneous identification of three or more number of modes due to the complexity of the algorithm. This capability of identifying multiple modes simultaneously will help improve the performance of the on-line modal parameter identification algorithm for flexible structures with high modal density.