Experimental Research and Numerical Analysis on Fire Resistance of Prefabricated Concrete Composite Beams

Objective This study develops a new method to investigate the fire resistance limit of T-shaped precast reinforced concrete stacked beams under fire conditions to understand the unique performance differences of T-shaped precast reinforced concrete stacked beams under fire conditions and to examine the damage development process and fire resistance limit. Methods This study designed eight test beams with detailed arrangements of dimensions, reinforcement, and temperature measurement points. The test beams were divided into two groups based on specific conditions. The specimen numbering rules and the mechanical properties of the reinforcement are provided. Pre-fire pre-compression was conducted by applying a constant jack load of 38.47 and 57.70 kN at three equal points on the top of the laminated beams using a hydraulic jack. The fire test was performed once the cracks had fully developed. The crack width was measured using the Concordia Crack Width Gauge. The furnace heating curve from the ISO 834 standard heating curve was applied to simulate a realistic fire scenario. The vertical displacement of the specimen and the relative slip of the laminated surface were measured. A finite element model considering the effect of cracks was developed using ABAQUS finite element analysis software. The simulation was conducted using sequential thermal coupling, and temperature rise curves were plotted using the measured furnace temperature data. The complex contact between cast-in-place and precast slabs and precast beams in stacked beams under fire conditions was accurately simulated using the Coulomb-cohesion hybrid model. A calculation method for determining the fire resistance limit of stacked beams was proposed based on the experimental and simulation results. The relationship equation between the fire resistance limit and each parameter was established using SPSS regression software, and the experimental, simulated, and formula-based values were compared to verify the accuracy of the equations. Results and Discussions The following observations were made by comparing the temperature, crack characteristics, and deflection-time development curves of the experimental and simulated values: 1) The error in temperature comparison between the experimental and simulated values was within 6%. After 60 min, the simulated values were slightly lower than the experimental values. 2) The simulated crack development depth was slightly smaller than the actual crack depth; however, the development trend remained consistent. During the first 60 min of ignition, the deflection growth curves were nearly identical. After 60 min, the simulated values were lower than the experimental values, with the error maintained within 10%. 3) The simulated fire resistance limits of all specimens were compared to the test values, and the errors remained within 8%. The accuracy of the formulas fitted using SPSS software was above 0.985, with the errors among the experimental, simulated, and formula-derived values remaining within 11%. The results indicated that the temperature distribution calculated by numerical simulation accurately reflects the actual temperature distribution of the cross-section. In addition, the simulated crack development curves and deflection-time development curves exhibit similar trends to those observed in the actual tests, verifying the accuracy of the finite element simulation and ensuring the reliability of the research findings. Based on these results, subsequent parametric analyses can be conducted. Conclusions The load-holding level is one of the critical factors influencing the fire resistance limit of stacked beams and a higher load-holding level results in a lower fire resistance limit. The influence of stacking parameters on the fire resistance limit is relatively minor but still warrants consideration. The span-height ratio also exerts a specific impact on the fire resistance limit of stacked beams, with the fire resistance limit tending to decrease as the span-height ratio increases. An increase in the thickness of the concrete protective layer leads to a linear improvement in the fire resistance limit of the stacked beams. A comparison between simulation and experimental results indicates that the fire resistance performance of stacked beams can be more accurately modeled using the Coulomb-cohesion hybrid model. The proposed method for calculating the fire resistance limit demonstrates high accuracy, with the error between the calculated value and the test value ranging from approximately 3% to 11%. This provides a reliable theoretical foundation for the subsequent design of the fire resistance performance of stacked beams, as well as an effective guideline and reference for damage assessment, reinforcement, and repair of T-shape precast assembled concrete stacked beams after fire exposure. This study holds substantial significance for enhancing the fire resistance and safety of stacked beams and provides a valuable reference for academic research and engineering applications in related fields. © 2025 Sichuan University. All rights reserved.