A new wavelet‐Hilbert transform based sparse component analysis (WHT‐SCA) method is presented for online system identification in indeterminate conditions. The instantaneous phase ratios of output signals are obtained by using a wavelet‐Hilbert transform based filter; and the out‐of‐phase data, that causes errors in identification accuracy, is detected and eliminated. Then, modal parameters of the structure are identified through existing relationships between the dispersion of filtered data in the frequency domain. Subsequently, to demonstrate the capability of the online identification, a new controller is introduced by combining the WHT‐SCA and a semi‐active tuned mass damper (STMD), resulting in creation of smart structures. The performance of the proposed method and controller is investigated through examples. The results demonstrate that, modal parameters of structures are identified accurately even with noise contamination and limited number of sensors. Also, the STMD is effectively robust against any variations in modal parameters of the structure.

中文翻译：

---

使用小波-希尔伯特变换和稀疏分量分析的结构在线系统识别

提出了一种基于小波-希尔伯特变换的稀疏分量分析（WHT-SCA）方法，用于不确定条件下的在线系统识别。通过使用基于小波-希尔伯特变换的滤波器获得输出信号的瞬时相位比。并检测并消除了导致识别精度错误的异相数据。然后，通过频域中滤波数据的色散之间的现有关系来识别结构的模态参数。随后，为了演示在线识别的能力，通过结合WHT-SCA和半主动调谐质量阻尼器（STMD）引入了新的控制器，从而创建了智能结构。通过实例研究了所提出的方法和控制器的性能。结果表明，即使存在噪声污染和有限数量的传感器，也可以准确识别结构的模态参数。而且，STMD可以有效抵抗结构的模态参数的任何变化。