Formulation of shear strength of slender RC beams using gene expression programming, part II: With shear reinforcement
Introduction
The flexural behavior of concrete elements is relatively well-recognized. This has resulted in establishing consistent expressions for the design of these elements in several codes of practice. Some studies have also been conducted to investigate the shear mechanism of concrete members over the last couple of decades. However, the shear transfer mechanisms in concrete elements are not as straightforward as the flexural behavior of these elements and still needs more studies. Most of the expressions provided in the current code provisions are mainly based on experimental studies. They indeed follow both the sectional and strut and tie model to predict the shear capacity of concrete beams and significantly vary from code to code which provides a wide range of factor of safety against failure. Understanding the mechanism of the shear transfer in concrete elements has always been challenging over the past decades. Extensive research has been carried out in this field.
Ritter [1] and Morsch [2] proposed the first analytical model, i.e., truss analogy, to calculate the shear capacity of a reinforced concrete beam. Many other researchers have also published other expressions to predict the shear capacity of concrete beams. However, most of the currents standard codes of practice are using semi-empirical formulations. Recent analytical methods of determining the shear capacity of reinforced concrete beams, including the disturbed stress field model (DSFM) [3], [4], modified compression field theory (MCFT) [5], [6], [7], and the softened truss model [9], [10] are mainly developed based on the truss model assumption. Recently, Arslan [11] proposed two different formulas for normal and high strength reinforced concrete beams with stirrups.
Artificial neural networks (ANNs) have been recently utilized in many fields including engineering (e.g. [12]). ANNs has the ability of self-organization and self-learning. They can efficiently employ the prior acquired knowledge to respond to the new set of information quickly and automatically. There have been some researches with the specific objective of applying ANNs to the evaluation of the shear strength of RC beams with shear reinforcement (e.g. [13]). The ANNs were shown to be a powerful tool for predicting the shear strength of concrete beams. Despite the acceptable performance of ANNs, there is no efficient procedure for the selection of structures of such networks. Hence, many experiments have to be performed to obtain an appropriate configuration. Also, ANNs do not usually give a definite function to calculate the outcome using the input values.
Genetic programming (GP) [14] is a new alternative approach to ANNs for the aim of nonlinear modeling. The major advantage of GP lies in its powerful ability to generate prediction equation. GP follows the principle of Darwinian natural selection for evolving the models. It is known as an extension of genetic algorithms (GAs). The main difference between these two approaches is that in GP the evolving programs (individuals) are parse trees rather than fixed-length binary strings. GP and its variants have successfully been applied to various kinds of structural engineering problems (e.g. [15], [16], [17]).
Different variants of GP have been proposed to improve the traditional GP [18]. Gene expression programming (GEP) [19] is a robust variant of GP. The traditional GP representation is based on the evaluation of a single tree (model) expression. GEP evolves computer programs of different sizes and shapes encoded in linear chromosomes of fixed length. Some of the GEP applications to concrete structures modeling include prediction of the strength of concrete under triaxial compression [20] and prediction of the compressive strength of high-performance concrete [21].
This study investigates the feasibility of using GEP for simulating the complex behavior of the shear capacity of RC beams with stirrups. A new relationship is developed between the shear strength and several influencing parameters using a database gathered from 55 experimental studies. A comparative study is further conducted between the results obtained by the proposed model and those of the building codes. The GEP model is developed based on reliable experimental results collected from the literature.
Section snippets
Shear strength mechanisms
To determine the maximum shear capacity of reinforced concrete beams, the shear strength of the stirrups should be added to those provided by the concrete shear capacity.
Gene expression programming
GP may be defined as a supervised machine learning technique that searches a program space instead of a data space. Symbolic regression is typically carried out through the standard GP to evolve a population of computer programs. The GP-based methods are substantially useful in cases where traditional methods are fairly complicated to be carried out, or the specific models of mathematical physics do not exist in detail. GEP is a natural development of GP first invented by Ferreira [19]. A basic
Model development
The primary objective of this paper is to develop a new formulation for the shear strength of slender RC beams with stirrups using the GEP approach. The GEP model was developed using six influencing input parameters as follows:where V (kN) is the shear strength of RC beam without stirrups, bw (mm) is the web width, d (mm) is the effective depth, a/d represents the shear span to depth ratio, fc (MPa) is the concrete compressive strength, ρl (%) shows the amount of longitudinal
Performance analysis and comparative study
According to Smith [80], if a model gives R > 0.8, and the error value is at the minimum, there is a strong correlation between the predicted and measured values. The model can, therefore, be judged as good. It can be observed from Fig. 4 and Table 3 that the GEP model with high R and low MAE and RMSE values precisely predicts the shear strength. The proposed model has both predictive ability (low MAE and RMSE values) and generalization performance (similar MAE and RMSEvalues) [81].
The proposed
Conclusions
In this study, a robust variant of GP, namely GEP is employed to assess the shear resistance of RC beam with stirrups. The shear strength was formulated in terms of several affecting factors (b, d, fc, a/d, ρl, ρw) representing the behavior of the shear strength. A widely dispersed database including both the HSC and NSC reinforced beams was used for the model development. It was observed that the GEP-based model is capable of predicting the shear strength of RC beam with stirrups to a high
References (84)
- et al.
Strut-and-tie modelling of reinforced concrete beams using genetic algorithms optimization
Constr. Build. Mater.
(2009) - et al.
Prediction of principal ground-motion parameters using a hybrid method coupling artificial neural networks and simulated annealing
Comput. Struct.
(2011) - et al.
Modeling mechanical performance of lightweight concrete containing silica fume exposed to high temperature using genetic programming
Constr. Build. Mater.
(2010) - et al.
Multi-stage genetic programming: a new strategy to nonlinear system modeling
Inf. Sci.
(2011) - et al.
A new predictive model for compressive strength of HPC using gene expression programming
Adv. Eng. Softw.
(2012) - et al.
Formulation of shear strength of slender RC beams using gene expression programming, part I: without shear reinforcement
Automat. Constr.
(2014) - et al.
Prediction of compressive and tensile strength of limestone via genetic programming
Expert Syst. Appl.
(2008) - et al.
A hybrid computational approach to derive new ground-motion attenuation models
Eng. Appl. Artif. Intell.
(2011) - et al.
Experimental study on high-strength concrete beams failing in shear
Eng. Struct.
(2005) - et al.
A novel QSPR model for prediction of lower flammability limits of organic compounds based on support vector machine
J. Hazard. Mater.
(2009)
Die Bauweise Hennebique
Schweiz. Bauzeitung
Disturbed stress field model for reinforced concrete: formulation
ASCE J. Struct. Eng.
Disturbed stress field model for reinforced concrete: implementation
ASCE J. Struct. Eng.
Development of the 2004 CSA A23.3 shear provisions for reinforced concrete
Can. J. Civ. Eng.
The modified compression field theory for reinforced concrete elements subjected to shear
J. Am. Concr. Inst.
Predicting the response of reinforced concrete beams subjected to shear using the modified compression field theory
ACI Struct. J.
Shear strength prediction for deep beams with web reinforcement
Struct. J. Am. Concr. Inst.
Behavior of reinforced concrete membrane elements in shear
Struct. J. Am. Concr. Inst.
Shear strength of reinforced concrete beams with stirrups
Mater. Struct.
Modeling and simulation of shear resistance of R/C beams using artificial neural network
J. Franklin Inst.
Genetic Programming, on the Programming of Computers by Means of Natural Selection
A linear genetic programming application to CFRP confined concrete cylinders
Mater. Struct.
Expression programming techniques for formulation of structural engineering systems
Gene expression programming: a new adaptive algorithm for solving problems
Complex Syst.
A novel approach to strength modeling of concrete under triaxial compression
J. Mater. Civil Eng. – ASCE
Gene Expression Programming: Mathematical Modeling by an Artificial Intelligence
Shear strength of members without transverse reinforcement
Can. J. Civ. Eng.
Shear strength of reinforced lightweight concrete beams of normal and high strength concrete
Mag. Concr. Res.
Detailing of stirrup reinforcement
ACI Struct. J.
Effect of concrete strength and minimum stirrups on shear strength of large members
ACI Struct. J.
Shear strength of reinforced concrete beams
ACI J. Proc.
Staggered shear design of concrete beams: large scale tests
Can. J. Civ. Eng.
Diagonal tension in reinforcedconcrete beams
ACI J. Proc.
How safe are our large, lightly reinforced concrete beams, slabs and footings?
ACI Struct. J.
Behavior and strength of reinforced haunched beams using high strength concrete
ACI J. Proc.
Cited by (43)
Prediction of compressive strength of brick columns confined with FRP, FRCM, and SRG system using GEP and ANN methods
2024, Journal of Engineering Research (Kuwait)Prediction of moment redistribution capacity in reinforced concrete beams using gene expression programming
2023, StructuresCitation Excerpt :Gene expression programming (GEP) is one of the soft computing modelling tool that has been frequently used in many civil engineering studies to predict mathematical relations based on experimental studies as an efficient alternative to traditional regression. There are many studies in the literature performed on RC beams using GEP such as, prediction of flexural strength for beams reinforced with FRP bars [8], proposing a formulation for shear strength of slender RC beams [9], and proposing a model for shear strength of steel fibers RC beams [10]. However, in the literature review, it is not available any study regarding the determination of the moment redistribution with the GEP model.