A MODIFIED ALGORITHM FOR RANDOM DECREMENT TECHNIQUE
As a typical pretreatment of time-domain signals, Random Decrement Technique (RDT) can identify the system free response under white-noise excitation. But, for a long time, this method doesn’t possess its accurate mathematical description. Consequently, RDT cannot be widely used in engineering application. Fatally, when RDT is applied to multi-degree of freedom (MDOF) system, the results show some critical defections like low precision, poor stability and so on. In response to all these shortcomings, this article proposes a Modified Random Decrement Technique (MRDT) which includes two aspects. On the one hand, this article supplements the mathematical description of RDT, through introducing relaxed “vector elemental triggering conditions, breaks the restriction of classical triggering condition demands, the system free responses under any specified initial condition can be reconstructed. On the other hand, the original signal-to-noise ratio (SNR) of reconstructed responses is highest in the initial period of time. This article proposes a simple time-domain extension method which utilizes the high SNR extended to the entire period of time. In addition, this study carries out some simulations. The numerical results show that MRDT algorithm, applied to MDOF system, greatly improves the numerical stability and accuracy of the results, injects new vitality to Random Decrement Technique.
random decrement technique MRDT vector elemental triggering conditions time-domain eztension
Bin Chen XiaoXu Zhang Hanwen Song
Department of Mechanics and Engineering Science, Fudan University, Shanghai College of Aeronautics and Astronautics and Mechanical, Tongji University, Shanghai
国际会议
The Fifth International Conference on VETOMAC-V(第五届振动工程及机械技术国际会议)
武汉
英文
321-328
2009-08-27（万方平台首次上网日期，不代表论文的发表时间）