Structural behaviour of high strength self compacting concrete deep beams / Mohammad Mohammadhassani

Mohammad, Mohammadhassani (2012) Structural behaviour of high strength self compacting concrete deep beams / Mohammad Mohammadhassani. PhD thesis, University of Malaya.

PDF (Thesis PhD)
Download (6Mb) | Preview


    This study is motivated by the lack of clear design provisions for the design and behaviour of deep beams. The behaviour of deep beam is significantly different from normal beams. In deep beams, the plane section does not remain planar after deformation. Presented in this study are the results of serviceability and design criteria of eight simply supported high strength self compacting concrete (HSSCC) deep beams tested to failure with variation in web reinforcement and tensile reinforcement ratios. The deflection at two points along the beam length, the web strains, tensile bars strains and the strain at concrete surface at the height of mid span section were recorded in the experimental phase. Effective input data and the corresponding deflection and strain in tie section as output data were recorded in analytical phase at all loading stages up to failure load for all the deep beams. Adaptive Network-based Fuzzy Inference System (ANFIS) and Artificial Neural Network (ANN) were applied in this study as modelling tools to predict deflection and strain in tie section for HSSCC deep beams. The complexities and difficulties to predict the accurate behaviour of deep beams can be overcome by the use of these latest tools in the area of artificial intelligence and computational intelligence. Finite element modelling was used to determine the strain and neutral axis depth variation at the height of mid span section. The results clearly show that the distribution of horizontal strains and the stresses in deep beams are nonlinear and completely different from the linear distribution in normal beams. At the ultimate limit state, the stress distribution in concrete surface located in mid-span is no longer parabolic as in normal beams. Deep beams develop several neutral axes before ultimate failure is reached. The number of neutral axis decreases as load increases and reduces to one at failure load. The failure of deep beams with longitudinal tensile steel reinforcement less than that suggested by ACI codes is flexural with large deflections and no inclined cracks. As the longitudinal tensile steel reinforcement increases, failure due to crushing of concrete at nodal zones was clearly observed. The appearance of first inclined crack in compression strut trajectory is independent of tensile and web reinforcement ratio variations. The modulus of rupture for HSSCC deep beams is close to the ACI 318-95 code. The first crack appeared at a load over 13 percent of the ultimate load in deep beams. ACI code provisions for the prediction of shear capacity of reinforced HSSCC deep beams are conservative. The Artificial neural network (ANN) displayed a superior ability in predicting mid span deflection with 10-10-4-1 and predicting strain in tie section of deep beam with 10-11-10-1 architectures. The suitability of ANN and ANFIS techniques in deflection prediction and tie strain prediction was evident when compared to Linear Regression (LR). ANN is more flexibility in training data compared to ANFIS while in term of incomplete data, ANFIS has proper response to unseen data. Finite element modeling shows a very good agrement with experimental data analysis and confirms the strain and neutral axes depth variation at the mid span section. Based on finite element modeling, the compressive strains were much less than 0.0025 in extereme compression fiber and less than 0.002 along the inclined compression struts.

    Item Type: Thesis (PhD)
    Additional Information: Thesis (PhD) - Faculty of Engineering, University of Malaya, 2012.
    Uncontrolled Keywords: Fuzzy Inference System; Web strains; Tensile bars strains; Neutral axis depth variation
    Subjects: T Technology > T Technology (General)
    T Technology > TA Engineering (General). Civil engineering (General)
    Divisions: Faculty of Engineering
    Depositing User: Mr Prabhakaran Balachandran
    Date Deposited: 27 Dec 2019 04:04
    Last Modified: 18 Jan 2020 10:01

    Actions (For repository staff only : Login required)

    View Item