LAUSR

This paper firstly introduced the theory and implementation of the least squares complex frequency domain (LSCF) method based on the truncated singular value decomposition (SVD) for noise reduction. The simulation results of signals with different signal noise ratios (SNR) show that this algorithm is stable and has excellent capacity of resisting noise disturbance. On this basis, an online detection system for natural frequency of brake discs is developed. This system takes the LAN-XI date acquisition module and LABSHOP software of B&K as platform and uses Visual Basic (VB) and MATLAB mixed programming technique for its secondary development. The LSCF method based on the truncated SVD is applied for curve fitting and modal parameter extracting. The qualified criterion is that the stably appearing natural frequencies are in the tolerance ranges. The practical application shows that this system intuitively and accurately displays the detection results and performs stably and efficiently.