当前位置: X-MOL 学术Int. J. Civ. Eng. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Identification of Joint Discrepancy in Steel Truss Bridge Using Hilbert Transform with root-MUSIC and ESPRIT Techniques
International Journal of Civil Engineering ( IF 1.8 ) Pub Date : 2021-01-13 , DOI: 10.1007/s40999-020-00597-2
Anshul Sharma , Pardeep Kumar , Hemant Kumar Vinayak , Suresh Kumar Walia

In the present paper, the first three modal frequencies obtained through recorded vibration signal of a steel truss bridge is investigated at three different vehicular speeds. The denoising and extraction of modal frequencies from vibration response signals of steel truss bridge is performed using Hilbert transform (HT) in combination with high-frequency resolution enhancing techniques. The modal frequency extracting techniques applied includes Fast Fourier transform (FFT), multiple signal classification (MUSIC), and estimation of signal parameters by rotational invariance technique (ESPRIT). The comparisons of the outcomes of HT-FFT, HT-root-MUSIC, and HT- ESPRIT are computed. It is observed that HT-FFT computed results with the low resolution which hindered in obtaining distinct modal frequencies while HT-MUSIC and HT-ESPRIT achieved the most denoised, robust and reliable outcomes. The HT-ESPRIT outperforms HT-MUSIC in the elimination of unwanted noise. Further, the application of the sliding window HT-ESPRIT method clearly shows the variation of modal frequencies with the change in vehicular speeds. The first three analytical modal frequencies of 4.56 Hz, 10.44 Hz and 16.66 Hz frequencies are compared with the experimental frequencies obtained through vibrations at 10 km/h, 20 km/h and 30 km/h vehicular speeds, respectively. The joints (also denoted as nodes) 6, 7, and 8 exhibited irregular behaviour with no or minimum frequency peaks due to increased flexibility of member at these locations.



中文翻译:

根-MUSIC和ESPRIT技术的希尔伯特变换识别钢桁架桥的共同差异

在本文中,研究了通过记录的钢桁架桥的振动信号在三种不同的车速下获得的前三个模态频率。使用希尔伯特变换(HT)结合高频分辨率增强技术,对钢桁架桥的振动响应信号进行模态频率的去噪和提取。应用的模态频率提取技术包括快速傅立叶变换(FFT),多信号分类(MUSIC)和通过旋转不变技术(ESPRIT)估计信号参数。计算了HT-FFT,HT-root-MUSIC和HT-ESPRIT的结果比较。可以看出,HT-FFT以低分辨率计算结果,这阻碍了获得截然不同的模态频率,而HT-MUSIC和HT-ESPRIT则获得了最大的降噪,稳健和可靠的结果。HT-ESPRIT在消除不需要的噪声方面胜过HT-MUSIC。此外,滑动窗口HT-ESPRIT方法的应用清楚地表明了模态频率随车速的变化而变化。将前三个分析模态频率分别为4.56 Hz,10.44 Hz和16.66 Hz的频率与通过以10 km / h,20 km / h和30 km / h的车速振动所获得的实验频率进行比较。关节(也称为节点)6、7和8表现出不规则的行为,没有或只有最小的频率峰值,这是由于这些位置的构件的柔韧性增加了。强大而可靠的结果。HT-ESPRIT在消除不需要的噪声方面胜过HT-MUSIC。此外,滑动窗口HT-ESPRIT方法的应用清楚地表明了模态频率随车速的变化而变化。将前三个分析模态频率分别为4.56 Hz,10.44 Hz和16.66 Hz的频率与通过以10 km / h,20 km / h和30 km / h的车速振动所获得的实验频率进行比较。关节(也称为节点)6、7和8表现出不规则的行为,没有或只有最小的频率峰值,这是由于这些位置的构件的柔韧性增加了。强大而可靠的结果。HT-ESPRIT在消除不需要的噪声方面胜过HT-MUSIC。此外,滑动窗口HT-ESPRIT方法的应用清楚地表明了模态频率随车速的变化而变化。将前三个分析模态频率分别为4.56 Hz,10.44 Hz和16.66 Hz的频率与通过以10 km / h,20 km / h和30 km / h的车速振动所获得的实验频率进行比较。关节(也称为节点)6、7和8表现出不规则的行为,没有或只有最小的频率峰值,这是由于这些位置的构件的柔韧性增加了。滑动窗口HT-ESPRIT方法的应用清楚地表明了模态频率随车速的变化。将前三个分析模态频率分别为4.56 Hz,10.44 Hz和16.66 Hz的频率与通过以10 km / h,20 km / h和30 km / h的车速振动所获得的实验频率进行比较。关节(也称为节点)6、7和8表现出不规则的行为,没有或只有最小的频率峰值,这是由于这些位置的构件的柔韧性增加了。滑动窗口HT-ESPRIT方法的应用清楚地表明了模态频率随车速的变化。将前三个分析模态频率分别为4.56 Hz,10.44 Hz和16.66 Hz的频率与通过以10 km / h,20 km / h和30 km / h的车速振动所获得的实验频率进行比较。关节(也称为节点)6、7和8表现出不规则的行为,没有或只有最小的频率峰值,这是由于这些位置的构件的柔韧性增加了。

更新日期:2021-01-13
down
wechat
bug