International Journal of Computational Methods

Steel fiber-reinforced ultra-high performance concrete (UHPC) material is prone to explosive spalling under elevated temperatures. With the addition of polypropylene (PP) fiber, thermal spalling of UHPC can be mitigated and its fire resistance can be improved. This research investigates the impact resistance of steel and PP fiber-reinforced UHPC slabs after exposure to elevated temperatures, and the structural behavior and damage were compared against normal strength concrete (NSC) slabs. Karagozian & Case concrete (KCC) model was adopted to simulate both NSC and UHPC materials. With consideration of thermal hazards, the material damage, equation of state and strain rate sensitivity were adapted. The validity of this numerical model was evaluated against available experimental results. The numerical model was used to investigate the impact resistance of the reinforced UHPC slabs after exposure to fire hazards. The effect of fire exposure time, impact velocity and impact mass on the resistance of the reinforced NSC and UHPC slabs were analyzed. The simulation results revealed that punching shear failure areas in the NSC slabs were 2.5 times, 3.4 times, 3.0 times and 1.2 times larger than the UHPC slabs after exposure to international standardization ISO-834 standard fire for 1h, 2h, 3h and 4h, respectively. After exposure to the standard fire ISO-834 for 2 h, the punching shear failure on the bottom side of NSC increased 90.9% with the increase in falling height from 1m to 7m, while for the UHPC slabs, the increment was around 67.9%. After exposure to the standard fire ISO-834 for 2h, the punching shear damage of the NSC slabs increased by 72.9% with the punch weight increased from 100kg to 700kg, whereas the damage in the UHPC slabs increased by 53.8%.

Steel fiber-reinforced self-compacting concrete (SCFRC) has been developed in recent decades to overcome the weak tensile performance of traditional concretes. As the flexural strength of SCFRC is dependent on the distribution of steel fibers, a numerical model based on Jeffery’s equation was developed in this study for investigating the effects of the concrete flow on the fiber orientation and distribution in SCFRC. This numerical method shows higher computational efficiency than available particle-based methods like SPH and LBM. The influence of casting parameters like casting method, formwork size and casting velocity on the fiber orientation is investigated from the perspective of the flow field of fresh concrete during casting. The simulation results show that the fiber orientation is largely dominated by the concrete flow during the casting process. Importantly, during casting SCFRC beam, fibers tend to be oriented in parallel along the longitudinal direction at the middle section, while they stick up at the end of the formwork due to the upward concrete flow. In addition, the results from parametric studies show that the formwork size and casting method could significantly affect the concrete flow during the casting process, ultimately the orientation of fibers in a SCFRC beam. Furthermore, it indicates that the casting speed needs to be carefully chosen in order to achieve the optimal fiber alignment.

This paper presents a numerical study on interfacial behavior between hybrid steel and polyvinyl-alcohol fiber-reinforced engineered cementitious composite (HSP-ECC) and concrete. Concrete damage plasticity (CDP) model was used to define the mechanical behavior of HSP-ECC and concrete materials. Interfacial behavior was simulated for as-cast and sandblasted surfaces by adopting traction-separation constitutive model. Key interface parameters were determined by calibration of finite element (FE) modeling results with experimental results. The FE modeling results illustrated the reliability and efficiency of the proposed numerical approach in the prediction of interface bond behaviors between HSP-ECC and concrete for different degrees of surface roughness.

Advances in testing of composite materials in recent years have mostly resulted in the ability to actively characterize the uniaxial response of composites while investigations of multiaxial load cases are rare due to various challenges. One of these challenges is the lack of a suitable specimen geometry for multiaxial testing without unwanted failure modes. Ideally, such geometry should be developed and assessed virtually by means of simulation before costly and time-consuming manufacturing and testing. Therefore, reliable, and efficient simulation methods are required that incorporate the evolution of damage in FRP composites.

This study investigates various geometric features virtually of cruciform specimens to test fiber-reinforced composites subjected to in-plane biaxial tensile loadings. The goal is to achieve uniform failure in the gauge region of the specimen due to biaxial stress states and to reduce any premature failure outside the gauge region. Efficient finite element simulation in the commercial software LS-DYNA is used to identify the optimal geometric features.

Data-driven calibration techniques, consisting of theory-guided feed-forward neural networks with long short-term memory, have previously been developed to find suitable input parameters for the finite element simulation of progressive damage in fibre-reinforced composites subjected to compact tension and compact compression tests. The results of these machine learning-assisted calibration approaches are assessed in a range of virtual open-hole strength tests under tensile and compressive loadings as well as in low velocity impact tests. It is demonstrated that the calibrated material models with bi-linear softening are able to simulate the structural response qualitatively and quantitatively with a maximum error of 9$\%$ with regards to experimentally measured open-hole strength values. Furthermore, the highly efficient models enable the virtual analysis of size effects as well as accurate force simulations in quasi-isotropic laminates under impact loading.

In this study, an extended finite element method (XFEM)-based numerical analysis procedure is developed as part of a framework for assessing damage slow growth behaviors of an adhesively bonded composite joint. This CFRP-CFRP single strap joint is stabilized with an aluminium honeycomb subjected to static compression through four-point bending. The adhesively bonded patch has a 140mm overlap length centered on its 440mm parent structure. The residual strengths are determined using an adhesive element failure criterion and a failure index based on energy release rate for disbonded and delaminated joints. XFEM is utilized to introduce pre-initiated cracks of various lengths and to calculate the energy release rates for each of four failure scenarios. Mesh convergence studies are conducted to acquire appropriate element sizes at crack tip. Additionally, multiple numerical benchmark models to experimental and numerical literatures are devised to validate individual components of the proposed finite element model. Results of the energy release rates for joints with cracks starting at the gap region suggest that a mixed-mode fracture occurs. At small crack lengths, mode-I is relatively low, and mode-II is high. As the crack length increases, mode-I increases, and mode-II gradually decreases. The energy release rates for joints with cracks starting in the taper end show that only a mode-II fracture exists. Finally, slow growth damage could be identified in each of the four damage scenarios. Joints with cracks initiated from the taper end exhibit substantially longer periods of slow growth damage when compared to joints with cracks initiated from the gap region.