China
Africa
Explore chapters and articles related to this topic
Structural risk assessment and management through the capture of dynamic movement.
Published in Nigel Powers, Dan M. Frangopol, Riadh Al-Mahaidi, Colin Caprani, Maintenance, Safety, Risk, Management and Life-Cycle Performance of Bridges, 2018
The on-site measurements of acceleration are processed using standard and proprietary digital signal processing (DSP) techniques to facilitate the extraction of relevant dynamic properties of the structural system. The main output of the DSP process is the decomposition of the time-series acceleration data into its frequency content (Power Spectral Density). The spectrum is interpreted to identify the modes of vibration (frequencies of resonance, amplitudes and modal ratios) of the structure. This forms the baseline Dynamic Signature and is essentially a fingerprint of how the structure is currently behaving with resonances appearing as isolated peaks in the plots. This baseline Dynamic Signature provides the datum with which future dynamic responses can be compared for assessment of changes to the structure’s performance.
Fundamentals of Method of Moments for Artificial Materials
Published in Filippo Capolino, Theory and Phenomena of Metamaterials, 2017
Christophe Craeye, Xavier Radu, Filippo Capolino, Alex G. Schuchinsky
Series acceleration formulations, like the Shanks and Levin-T transforms, have been used successfully [67–71] to evaluate periodic GFs. Alternative techniques of convergence acceleration in periodic GFs, particularly Ewald and Kummer transformations, have shown excellent computational efficiency. Algorithms based upon the Ewald method [72] are extensively discussed in [61–64,73–81]. Approaches based on the Kummer method are described in [51,65,82,83]. It is necessary to mention that other efficient schemes have been reported for the periodic GF acceleration. For example in [84] the Veysoglu’s transformation (see also [59,65,66]) is used to achieve exponential convergence for 2-D problems periodic in one direction.
Bayesian damage detection on full-scale pole structure with anchor bolt tension loosening
Published in Joan-Ramon Casas, Dan M. Frangopol, Jose Turmo, Bridge Safety, Maintenance, Management, Life-Cycle, Resilience and Sustainability, 2022
Let Ot be defined as time series acceleration of data for anomaly detection. To test where there are significant changes in features between reference and anomaly detection data, features of the reference data Λ˜featr the anomaly detection data Λ˜featt are set, and hypothesis test is conducted considering fol lowing assumptions. Null hypothesis H0: Λ˜featr=Λ˜feattAlternative hypothesis H1: Λ˜featr≠Λ˜feattThe hyper parameters that determine the posterior probability distribution are not altered between the reference and anomaly detection data
Step length and grade effects on energy absorption and impact attenuation in running
Published in European Journal of Sport Science, 2020
Michael Baggaley, Gianluca Vernillo, Aaron Martinez, Nicolas Horvais, Marlene Giandolini, Guillaume Y. Millet, W. Brent Edwards
Time- and frequency-domain analyses were performed on the axial acceleration signals from the tibia- and sacrum-mounted accelerometers. It is well documented that high frequency components (>10 Hz) of segmental accelerations are produced by foot-ground impact (Shorten & Winslow, 1992). For time-domain analysis, a high-pass Butterworth filter with a cutoff frequency of 5-Hz was first applied to isolate the impact component of acceleration. Peak acceleration magnitudes were then quantified. For frequency-domain analyses, the time-series acceleration signals were separated into individual segments from the instant of peak negative tibial acceleration prior to foot contact through the subsequent 0.3-s (Figure 1(A)). The data were then zero padded to a total of 1024 points. The power spectral density was then calculated, with a frequency resolution of 1.95 Hz, using a Fourier transform (Figure 1(B)) (Shorten & Winslow, 1992). Impact attenuation between the tibia and sacrum at each frequency within the 10–30 Hz frequency range (Bediz, Nevzat Özgüven, & Korkusuz, 2010; Shorten & Winslow, 1992) was quantified (Figure 1(C)) with a transfer function:where TFi is the attenuation (in dB) between the tibia (PSDtibia,i) and sacrum (PSDsacrum,i) power spectral densities for the i-th frequency bin within the 10–30 Hz frequency range. An average TF within the 10–30 Hz range was subsequently determined.
Efficient reevaluation of surface displacements in a layered elastic half-space
Published in International Journal of Pavement Engineering, 2020
Sebastian Andersen, Eyal Levenberg, Mathias B. Andersen
This paper contributes to the body of work dealing with expediting layered elastic calculations. It proposes a new method for efficient reevaluation of surface displacements useful for optimisation processes. The method does not involve spline approximation (i.e. integrand interpolation) nor does it entail any kind of extrapolation techniques, series acceleration or asymptotic representation. In turn, it is based on generalising and further improving the ‘subtraction and addition’ technique suggested in Chen (1971) and in Khazanovich and Wang (2007). The development is specifically tailored for inverse analysis situations, wherein system responses at several predefined locations must be repeatedly evaluated with new layer properties. The paper commences by restating the standard layered elastic formulation for evaluating surface displacements. It then identifies and illustrates where high calculation costs reside. The new proposed method is presented and explained next, followed by a demonstration for a typical pavement system. The paper concludes with a short discussion on the merits of the proposed approach, and lists some potential ideas for further upgrading.
Consolidation solution of soil around a permeable pipe pile
Published in Marine Georesources & Geotechnology, 2020
Sihai Wang, Pengpeng Ni, Zheng Chen, Guoxiong Mei
The variations of excess pore-water pressure at r = 2r0 with different θ calculated using the proposed semi-analytical solution are compared with the numerical results in Figure 5. It can be observed that the results obtained from these two different methods are in good agreement. In addition, the techniques of the series acceleration method proposed by Shanks (1955) and the numerical inversion of Laplace transform adopted from Stehfest (1970) are all effective in providing a satisfactory calculation accuracy.