Transient state analysis of rehabilitated RC beams using finite element modelling and prediction using an artificial neural network

Abstract

The present research develops and verifies a simpler numerical approach for analyzing the thermal transient state of rehabilitated concrete beams reinforced with various types of FRP (fiber-reinforced polymer) subjected to high temperatures and specifically built as under-reinforced concrete beams. This approach offers a straightforward, efficient, and exact instrument for numerical analysis. The proposed analytical technique has been validated by load-displacement curves and cross-section temperature data, indicating its dependability and practicality. Subsequently, the validated approach was used to examine the impact of significant variables on the outcome and restoration of FRP-reinforced concrete beams at high temperatures. The methodology gives the Comparing conventional and CFRP, GFRP, AFRP reinforced beams using beam, truss, and shell elements. Thermal and UDL loads were applied, mesh at 25 mm × 25 mm. Transient analysis contrasts performance via displacement and temperature. The temperature versus displacement curve shows the FRP comparisons. Identifying the critical temperature before failure is crucial, emphasizing the curve’s significance in assessing structural performance and potential failure points. Nodal temperatures ranged 939 °C–963 °C (rehabilitated) versus 958 °C (conventional). 200 °C difference affects thermal boundary conditions for structural analysis and Conventional peaks at 320 °C, while AFRP, GFRP, and CFRP reach 358 °C, 385 °C, and 390 °C respectively. CFRP lasts 2400 min. Neural network models demonstrate effective generalizability, enabling satisfactory predictions of RC beam rehabilitation with CFRP laminates within the study’s parameter range.