Punching shear capacity estimation of FRP-reinforced concrete slabs using a hybrid machine learning approach

Abstract

faculty of civil engineering, Institute of research and development, duy tan university, danang, Vietnam ABSTRACT Fibre-reinforced polymer (FRP) provides an alternative reinforcement for concrete flat slabs. This research proposes a hybrid machine learning model for predicting the ultimate punching shear capacity of FRPreinforced slabs. The model employs the least squares support vector machine (LS-SVM) to discover the mapping between the influencing factors and the slab punching capacity. Furthermore, the firefly algorithm (FA), a population-based metaheuristic, is utilised to facilitate the LS-SVM training. A data-set which contains actual tests of FRP-reinforced concrete slabs is utilised to construct and verify the proposed approach. The contribution of this research is to establish a hybrid machine method, based on the LS-SVM and FA algorithms, for meliorating the prediction accuracy of FRP-reinforced slabs' ultimate punching shear capacity. Experimental results demonstrate that the new model has achieved roughly 55 and 15% reductions in terms of Root Mean Squared Error compared with the formula-based and Artificial Neural Network methods, respectively.

Estimating Compressive Strength of High Performance Concrete with Gaussian Process Regression Model

“…In this section of the article, to better evaluate the performance of the GPR model, the ANN [26] and LSSVM [9,27] are employed as benchmark methods. The reasons for selecting these two benchmark models are that ANN is widely accepted as an effective tool for nonlinear function approximation and this algorithm has been successfully employed for predicting concrete strength [10,13]; LSSVM is also an advanced machine learning method featured by high modeling accuracy [28][29][30][31] and it has been recently used for modeling concrete compressive strength [5].…”

A locally weighted machine learning model for generalized prediction of drift capacity in seismic vulnerability assessments

“…In addition to the SVMR-based models mentioned previously, several other global ML models have been successfully employed in the CE domain. Artificial neural networks (ANNs) have been employed to model the ultimate deformation capacity of RC rectangular flexure-critical columns (Inel, 2007) and to predict the shear strength of RC deep beams (Pal & Deswal, 2011) and the ultimate punching shear capacity of fiber-reinforced polymer (FRP)-reinforced slabs (Vu & Hoang, 2016). Random forests (RFs) have been utilized to predict building energy consumption (Ahmad, Mourshed, & Rezgui, 2017).…”

“…Comparative studies among some of these global ML techniques have been performed by many researchers (e.g., Ahmad et al, 2017;Pal & Deswal, 2011;Vu & Hoang, 2016). Ahmad et al (2017) carried out a performance comparison between ANNs and RFs for the prediction of building energy consumption; they found that both ANNs and RFs were comparable in terms of their predictive capabilities, with the ANN-based approach slightly outperforming the RF approach in terms of root mean squared error (RMSE).…”

“…Comparative studies among some of these global ML techniques have been performed by many researchers (e.g., Ahmad et al., ; Chou et al., ; Pal & Deswal, ; Vu & Hoang, ). Ahmad et al.…”

“…() carried out a performance comparison between ANNs and RFs for the prediction of building energy consumption; they found that both ANNs and RFs were comparable in terms of their predictive capabilities, with the ANN‐based approach slightly outperforming the RF approach in terms of root mean squared error (RMSE). Several studies demonstrated the superior performance of SVMR‐based models over empirical approaches and ANNs in terms of generalization performance and overall prediction accuracy (Chou et al., ; Pal & Deswal, ; Vu & Hoang, ). In addition, previous empirical studies have also shown that the prediction performance of the LS‐SVMR is comparable to that of the SVMR (Van Gestel et al., ; Wang & Hu, ).…”