基于自然激励技术和HHT变换的重力坝模态分析

Modal parameter identification of gravity dam based on natural excitation technique and Hilbert-Huang transform

在自然激励技术和 Hilbert-huang变换相结合的基础上，利用水口水电站的强震监测资料，对水口混凝土重力坝进行模态参数识别。发展了将带通滤波技术与该时频域模态识别技术相结合的方法，有效地避免了信号分解时所产生的频率混杂现象，从而提高了参数识别精度和可靠性。首先利用结构动力学的简例验证了该方法的正确性，然后将该方法应用于混凝土重力坝的模态识别问题。计算结果表明，采用本文的时频域 HHT＋NExT方法识别的重力坝模态频率稍大于传统的频域峰值点拾取法和时域 ARX法，与有限元计算结果更为接近，识别的阻尼比也稍大于峰值法和 ARX法，略小于水工抗震设计规范规定取值范围5%～10%的下限5%。

Modal identification of the Shuikou concrete gravity dam has been conducted based on natural ex-citation technique and Hilbert-Huang transform under four different seismic excitations on it. A new ap-proach is proposed by bandpass filtering technology combining with the modal identification method, which avoids the phenomenon of frequency mixed effectively during the signal decomposition, thus improving the accuracy and reliability of the modal parameter identification. Firstly,a simple example of structural dynam-ics was used to verify the correctness of the method, and then this method was used for concrete gravity dam modal parameter identification for the first time. The results show that the frequencies identified by HHT＋NExT are slightly larger than the result of the peak-picking method and ARX model, and close to the results of FEM. Damping ratios identified by HHT＋NExT are also somewhat larger than the results of the peak-picking method and ARX model, and meanwhile, they are slightly less than the design specifica-tion ranges namely between 5% ~ 10% of the lower limit of 5% in specifications for seismic design of hy-draulic structures.