Application of neural networks and an adapted wavelet packet for generating artificial ground motion

Asadi, Azita; Fadavi, Manouchehr; Bagheri, Abdollah; Amiri, Gholamreza Ghodrati

doi:10.12989/sem.2011.37.6.575

Cited by 10 publications

(3 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In an attempt to produce more artificial earthquake accelerograms from, available data compatible with the specified response spectra or the design spectra, Amiri et al [22] introduced a method based on wavelet packet transform and stochastic neural networks. Using artificial neural network and wavelet packet transform Asdi et al [23] presented a numerical method for the decomposition of artificial earthquake records consistent with any arbitrarily specified target response spectra requirements. In order to generate spectrum-compatible near-field artificial earthquake accelerograms, Amiri et al [24] introduced a new methodology based on particle swarm optimization, wavelet packet transform techniques, and multilayer feed-forward neural networks.…”

Section: Introductionmentioning

confidence: 99%

Simulation of spectrum-correspondent accelerogram by using artificial neural networks

Izadi¹,

Mohammadi²

2016

J. vibroeng.

View full text Add to dashboard Cite

Regarding the scarcity of appropriate recorded earthquakes, and the ever-increasing use of dynamic time history analyses for more accurate calculation of structures response, the simulation of artificially produced records necessary. In this study, accelerograms are simulated from the response or design spectrum by using generalized regression neural networks. In the training phase the response spectrum is used as the input for the simulating network, and the corresponding accelerogram as the output. Accelerograms achieved from some recorded earthquakes of Iran are used for training the neural network. The appropriate accuracy, and high speed of training are the properties of the network. After training the network, accelerogram corresponding to the design spectrum of Iranian code of practice for seismic resistance design of buildings is generated. Similar procedures can be carried out for design spectrum of other cods to achieve the corresponding records.

show abstract

Section: Introductionmentioning

confidence: 99%

Simulation of spectrum-correspondent accelerogram by using artificial neural networks

Izadi¹,

Mohammadi²

2016

J. vibroeng.

View full text Add to dashboard Cite

show abstract

“…Where   x  is the relative acceleration response of an SDOF with a period of T and damping ratio of 5% under the ETEFs, and   is the median acceleration spectra of normalized GMs. In this study, the GMs suite recommended by [28] is used as target motions. These ground motions are recorded from large magnitude events ( 6.5 M  ) at sites located greater than or equal to 10 km from fault rupture.…”

Section: Generation Of Endurance Time Excitationsmentioning

confidence: 99%

Evaluation of the Efficiency of Mother Wavelet Functions for Simulating Endurance Time Excitations

Mashayekhi,

Harati,

Mirfarhadi

et al. 2024

Scientia Iranica

View full text Add to dashboard Cite

The Endurance Time (ET) method is employed as a dynamic time history technique to analyze structures under artificially intensified acceleration time histories, known as Endurance Time Excitation Functions (ETEFs). Prior studies have shown that discrete wavelet transform (DWT) is an effective approach for generating ETEFs by representing signals as transform coefficients determined through optimization procedures. However, the impact of the chosen mother wavelet function on simulated ETEF accuracy remains unexplored. This study introduces a methodology to investigate the influence of mother wavelet functions on simulated ETEFs. Specifically, 31 mother wavelet function candidates from four families (Daubechies, Coiflet, Symlet, and Bio-Orthogonal) are examined. Results reveal that the choice of the mother wavelet function can lead to approximately 15% variation in simulated ETEFs' accuracy. The Daubechies wavelet family stands out as the preferred choice, exhibiting a diminished impact compared to alternative families. Remarkably, this wavelet family is associated with an importance factor of 5.5%, significantly lower than the 13% observed for the other families. Within the Daubechies family, db12 demonstrates optimal efficiency in generating linear response-based ETEFs. The research highlights the superiority of the Daubechies wavelet family, offering valuable insights to enhance ETEF simulation accuracy and reliability for effective ET method implementation.

show abstract

“…Owing to the ability to learn and generalise interactions among many variables, the artificial neural network technology has considerable potential in the modelling problems. Ghaboussi and Lin (1997); Lin and Ghaboussi (2001); Ghodrati Amiri and Bagheri (2008); Ghodrati Amiri et al (2009) and Asadi et al (2011) developed innovative methodologies for the generation of artificial earthquake accelerograms using neural networks. In the field of attenuation relationship prediction, owing to the uncertainties inherent in the variables describing the earthquake source, the difficulty in defining broad categories to classify site, a lack of understanding of wave propagation processes and ray path characteristics from source to site, the estimations from attenuation regression analyses are often inaccurate (García et al, 2007).…”

Section: Introductionmentioning

confidence: 99%

Estimation of spectral acceleration based on neural networks

Bakhshi¹,

Bagheri²,

Amiri³

et al. 2014

Proceedings of the Institution of Civil Engineers - Structures

View full text Add to dashboard Cite

This study presents an effective method based on artificial intelligence to predict spectral acceleration from the data of the 'Next generation attenuation' project. The proposed method uses the learning abilities of artificial neural networks to expand the knowledge of the mapping from earthquake parameters to spectral accelerations, which results in spectral accelerations for special frequencies. In this paper, the Levenberg-Marquardt algorithm is applied for training neural networks. For each type of faults, considered earthquake parameters included in the input of the neural network are moment magnitude, distance from the recording site to epicentre, hypocentre depth and average shear-wave velocity in the top 30 m. To test the adequacy of the trained neural networks, they were examined through their training set and new data. The obtained results demonstrate that neural networks can be effectively and reliably employed in the estimation of spectral acceleration.

show abstract

Application of neural networks and an adapted wavelet packet for generating artificial ground motion

Cited by 10 publications

References 17 publications

Simulation of spectrum-correspondent accelerogram by using artificial neural networks

Simulation of spectrum-correspondent accelerogram by using artificial neural networks

Evaluation of the Efficiency of Mother Wavelet Functions for Simulating Endurance Time Excitations

Estimation of spectral acceleration based on neural networks

Contact Info

Product

Resources

About