Regression based weight generation algorithm in neural network for estimation of frequencies of vibrating plates

Chakraverty, Snehashish; Singh, Varinder; Sharma, Richa

doi:10.1016/j.cma.2005.08.008

Cited by 21 publications

(7 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is also worth mentioning here that there exist various ANN training methods, such as that discussed in [5,6,12] and so the author does not claim about the optimality for the problem under analysis. As the present problem is concerned, where earthquake do depend upon various factors of the soil (ground), the parameters of building and also on the soil structure interaction.…”

Section: Numerical Results and Discussionmentioning

confidence: 98%

Neural network-based simulation for response identification of two-storey shear building subject to earthquake motion

Chakraverty

Gupta

Sharma

2009

Neural Comput & Applic

View full text Add to dashboard Cite

This article uses powerful technique of artificial neural network (ANN) models to simulate and estimate structural response of two-storey shear building by training the model for a particular earthquake. The neural network is first trained for a real earthquake data and the numerically generated responses of different floors of two-storey buildings as the training patterns. Trained ANN architecture is then used to simulate and test the structural response of different floors for various intensity earthquake data and it is found that the predicted responses given by ANN model are good for practical purposes. It is worth mentioning that although the simulation is done with numerically generated response data for particular earthquake, the idea may also be used for actual experimental (response) data.

show abstract

Section: Numerical Results and Discussionmentioning

confidence: 98%

Neural network-based simulation for response identification of two-storey shear building subject to earthquake motion

Chakraverty

Gupta

Sharma

2009

Neural Comput & Applic

View full text Add to dashboard Cite

show abstract

“…The BP neural network used in this study is a special machine learning method which focuses on intelligent predictions rather than improvements of computer algorithms, therefore the required amount of training data can be less than the traditional machine learning techniques, and a few studies have shown that training data with a size around 20 can yield acceptable training results [ 70 , 71 , 72 , 73 , 74 ]. The present study performed 28 cyclic triaxial tests on the UGM.…”

Section: Evaluation Of Critical Dynamic Stress and Final Accumulative Plastic Strainmentioning

confidence: 99%

Evaluation of Critical Dynamic Stress and Accumulative Plastic Strain of an Unbound Granular Material Based on Cyclic Triaxial Tests

et al. 2021

View full text Add to dashboard Cite

Critical dynamic stress (σcri) and accumulative plastic strain (εp) are primary indicators regarding the dynamic stability of unbound granular materials (UGMs). This study aims to seek an effective method to evaluate the dynamic stability of UGMs used in railway subgrades. First, the dynamic characteristics of an UGM used in railway subgrade bed construction were investigated by performing a series of large-scale cyclic triaxial tests, with the results showing that εp versus cycle number (N) curves can be categorized into stable, failure, and critical patterns. Grey relational analyses were then established, where the analyzed results demonstrated that the εp–N curve pattern and final accumulative plastic strain (εs) of the stable curves are strongly correlated with the moisture content (w), confining pressure (σ3), and dynamic deviator stress (σd). The analyzed grey relational grades distributed in a narrow range of 0.72 to 0.81, indicating that w, σ3, and σd have similar degrees of importance on determining the εp–N curve patterns and the values of εs of the UGM. Finally, a data processing method using a back-propagation (BP) neural network is introduced to analyze the test data, and an empirical approach is developed to evaluate the σcri (considering the effects of σ3 and w) and εs (considering the effects of σ3, w, and σd) of the UGM. The analyzed results illustrated that the developed method can effectively reflect the linear/non-linear relationships of σcri and εs with respect to σ3 and/or σd. The σcri approximately increases linearly with increasing σ3, and a simple empirical formula is proposed for the σcri. In addition, εs and its variation rate increase non-linearly with increasing σd but decrease non-linearly as σ3 increases.

show abstract

“…Another way of determining the optimal number of hidden neurons that can result in good generalization and avoid overfitting is to relate the number of hidden neurons to the number of training samples (i.e., Masters 1993;Rogers and Dowla 1994;Amari et al 1997). A number of systematic approaches have also been proposed to obtain the optimal ANN architecture (i.e., Ghaboussi and Sidarta 1998;Chakraverty et al 2006;Chakraverty 2007;Kingston et al 2008). However, none of these suggestions has been universally accepted or used.…”

Section: Ann Design and Optimisationmentioning

confidence: 99%

A comparative study for the estimation of geodetic point velocity by artificial neural networks

Yılmaz

Güllü

2014

J Earth Syst Sci

View full text Add to dashboard Cite

Space geodesy era provides velocity information which results in the positioning of geodetic points by considering the time evolution. The geodetic point positions on the Earth's surface change over time due to plate tectonics, and these changes have to be accounted for geodetic purposes. The velocity field of geodetic network is determined from GPS sessions. Velocities of the new structured geodetic points within the geodetic network are estimated from this velocity field by the interpolation methods. In this study, the utility of Artificial Neural Networks (ANN) widely applied in diverse fields of science is investigated in order to estimate the geodetic point velocities. Back Propagation Artificial Neural Network (BPANN) and Radial Basis Function Neural Network (RBFNN) are used to estimate the geodetic point velocities. In order to evaluate the performance of ANNs, the velocities are also interpolated by Kriging (KRIG) method. The results are compared in terms of the root mean square error (RMSE) over five different geodetic networks. It was concluded that the estimation of geodetic point velocity by BPANN is more effective and accurate than by KRIG when the points to be estimated are more than the points known.

show abstract

Regression based weight generation algorithm in neural network for estimation of frequencies of vibrating plates

Cited by 21 publications

References 10 publications

Neural network-based simulation for response identification of two-storey shear building subject to earthquake motion

Neural network-based simulation for response identification of two-storey shear building subject to earthquake motion

Evaluation of Critical Dynamic Stress and Accumulative Plastic Strain of an Unbound Granular Material Based on Cyclic Triaxial Tests

A comparative study for the estimation of geodetic point velocity by artificial neural networks

Contact Info

Product

Resources

About