Artificial neural network based prediction for frp-confined concrete under cyclic loading
By: Gopinath, Smitha.
Contributor(s): Ramesh Gopal.
Publisher: USA Springer 2022Edition: Vol.103(4), Dec.Description: 1015-1028p.Subject(s): Humanities and Applied SciencesOnline resources: Click here In: Journal of the institution of engineers (India): Series ASummary: Artificial neural network (ANN)-based model is developed for predicting the stress and strain enhancement of FRP wrapped concrete. An experimental database is considered in the investigations, which includes confinement using carbon, glass and aramid FRP on concrete. The aspect ratio, form of FRP wrap, number of confining layers, unconfined power, confining pressure and FRP characteristics are used as input parameters. Performance of proposed ANN models was evaluated by considering two indices, coefficient of determination and measure of root mean square error. The hoop strain, which is a main influencing parameter in confining pressure and dilation of FRP-confined concrete, is also predicted using ANN, for which no model predictions are available as of today. The predictive accuracy of some of the currently available models from the literature has been assessed by estimating the stress and strain enhancement due to FRP confinement by evaluating the root mean square error. The findings from the investigations show that ANN-based models can accurately predict the response close to experimental response and competitive enough compared to already existing mathematical models. The results pave way toward opening up the scope of data-driven models for the design of FRP confinement for structures.Item type | Current location | Call number | Status | Date due | Barcode | Item holds |
---|---|---|---|---|---|---|
Articles Abstract Database | School of Engineering & Technology Archieval Section | Not for loan | 2023-0341 |
Artificial neural network (ANN)-based model is developed for predicting the stress and strain enhancement of FRP wrapped concrete. An experimental database is considered in the investigations, which includes confinement using carbon, glass and aramid FRP on concrete. The aspect ratio, form of FRP wrap, number of confining layers, unconfined power, confining pressure and FRP characteristics are used as input parameters. Performance of proposed ANN models was evaluated by considering two indices, coefficient of determination and measure of root mean square error. The hoop strain, which is a main influencing parameter in confining pressure and dilation of FRP-confined concrete, is also predicted using ANN, for which no model predictions are available as of today. The predictive accuracy of some of the currently available models from the literature has been assessed by estimating the stress and strain enhancement due to FRP confinement by evaluating the root mean square error. The findings from the investigations show that ANN-based models can accurately predict the response close to experimental response and competitive enough compared to already existing mathematical models. The results pave way toward opening up the scope of data-driven models for the design of FRP confinement for structures.
There are no comments for this item.