论文部分内容阅读
研究了动力系统模态参数的ITD(Ibrahim Time Domain)识别法原理,指出它的实质在于自由振动函数列阵u(t)按模式以最小二乘拟合矩阵A和B。讨论了以虚拟站扩展有效测量站和设置过量待识别模态吸收测量噪声的细节。 提出了自由振动信息综合的概念。指出:在测量站位置和数量一定的条件下,不同来源的位移、速度和加速度自由振动时间历程或它们的时移后的时间历程的线性组合,都可以作为一个供识别计算用的“自由振动”。这样,在一次识别计算中,将可以充分利用手头有的一次或多次实测数椐,使模态参数识别更完全或提高信噪比。重点研究了一次自由振动数据的综合:自由振动信息随机减量,并提出设置一个新的用户选择参数NRNDEC(Number of Random Decrement Times)——随机减量次数。 提出了两种检验识别得的模态参数可靠性的方案:1.在实测自由振动信息中混入若干个人为的“检验模态”;2.模态分量因子识别和自由振动预计。前者是把计算机模拟识别试验成果用于实际识别的桥梁,很简单,但很直观,很有效。
The principle of identification of the dynamical system modal parameter ITD (Ibrahim Time Domain) is studied, and the essence of it is that the free vibration function array u (t) is modeled by the least square fitting matrix A and B. The details of expanding an effective measurement station with a virtual station and setting excessive measurement modes to absorb the measurement noise are discussed. Put forward the concept of free vibration information synthesis. It is pointed out that the linear combination of time history of displacement, velocity and acceleration of different sources or their time-history after time-shift can be used as a “free vibration ”. In this way, in a single recognition calculation, one or more actual measurements 手 at hand can be fully utilized to make the modal parameter identification more complete or improve the signal-to-noise ratio. This paper focuses on the synthesis of a free vibration data: random decrement of free vibration information, and proposes to set a new user selection parameter NRNDEC (Number of Random Decrement Times) - the number of random decrements. Two schemes for testing the reliability of the modal parameters are proposed: 1. Several “test modalities” are mixed in the measured free vibration information; 2. The modal component factor identification and the free vibration prediction. The former is a bridge that identifies the experimental results of computer simulation for actual identification. It is simple, but very intuitive and effective.