Search

Highlights • Shear strength predictions from existing models differ from experimental results. • Predictions by MCFT(R2k) and CCC bear the closest comparison to experimental results. • Design values of current shear design provisions compare well, except for Eurocode 2. • Eurocode 2 shear design formulation may be uneconomical at low levels of stirrups. Abstract An accurate method for predicting shear strength for reinforced concrete beams is paramount since shear failure is catastrophic and can lead to grave consequences. Different standards and guidelines use different models for shear resistance predictions. Their stipulations on shear design differ considerably from one another, resulting in different design procedures and safety performance. This contribution assessed the mean and design value predictions of shear models for beams with shear reinforcements in currents codes and published technical literature. The models investigated includes the EC2 VSIM shear model, the ACI 318 (2011) shear model, the Fib Model Code 2010 (MC-10 (III)) shear model, the best-estimate prediction by Modified Compression Field Theory (MCFT) based analysis program Response 2000 (R2k) and the Compression Chord Capacity model (CCC). The mean and design value predictions from the various methods are compared to one another and to experimental results, over the parametric range of shear reinforcement, concrete strength and beam depth. The assessment revealed that the mean value predictions of the different models differ considerably from one another and from experimental observations. The mean value predictions from MCFT (R2k) and CCC predictions bear the closest comparison to that of the experimental observations for the range of shear reinforcement ρ w f ywm investigated. The mean value predictions of the EC2 VSIM was shown to significantly underpredict capacity for slightly shear-reinforced concrete beams. The shear method produced the most conservative mean value predictions out of all the methods investigated at shear reinforcement ρ w f ywm ≤ 1 M P a. The design value analysis revealed that the design values of the various shear design methods compare well, except for EC2 VSIM at low levels of design shear reinforcement (ρ w f ywd ≤ 1 M P a). [ABSTRACT FROM AUTHOR]

• Machine learning -based shear strength models are proposed for SFRC beams. • The models are in close alignment with the experimentally observed shear strength and the existing models. • The models provide more accurate and unbiased predictions. • Partial resistance safety factors are proposed for the uncertainties related to shear design of SFRC beams. • Sensitivity of SFRC beam shear models to shear span to depth ratio is a cause for concern. Shear failure in reinforced concrete beams poses a critical safety issue since it may occur without any prior signs of damage in some cases. Many of the existing shear design equations for steel fiber reinforced concrete (SFRC) beams include significant uncertainty due to failure in reflecting the phenomenology of shear resistance accurately. Given these, adequate reliability evaluation of shear design provisions for SFRC beam is of high significance, and increased accuracy and minimisation of variability in the predictive model is essential. This contribution proposes machine learning (ML) based methods - Gaussian Process regression (GPR) and the Random Forest (RF) techniques - to predict the ultimate shear resistance of SFRC slender beams without stirrups. The models were developed using a database of 326 experimental SFRC slender beams obtained from previous studies, utilising 75% for model training and the remainder for testing. The performance of the proposed models was assessed by statistical comparison to experimental results and to that of the state-of-practice existing shear design models (f ib Model Code 2010, German guideline, Bernat et al. model). The proposed ML-based models are in close alignment with the experimentally observed shear strength and the existing predictive models, but provide more accurate and unbiased predictions. Furthermore, the model uncertainty of the various resistance models was characterised and investigated. The ML-based models displayed the lowest bias and variability, with no significant trend with input parameters. The inconsistencies observed in the predictions by the existing shear design formulations at the variation of shear span to effective depth ratio is a major cause for concern; reliability analysis is required. Finally, partial resistance safety factors were proposed for the model uncertainty associated with the existing shear design equations. [ABSTRACT FROM AUTHOR]