Research On Methods For Refined Dynamic Response Analysis And Damage Assessment Of FRP Strengthened RC Columns Subjected To Blast Loading

Strengthening the RC column with fiber reinforced polymer(FRP)can significantly improve its shear resistance,bending resistance and ductility,thus preventing its failure under blast loading.In this way,the progressive collapse probability of the building can be significantly reduced.In this study,the refined dynamic response analysis and damage assessment method of FRP strengthened RC columns under blast loading were systematically studied.Firstly,aiming at the problem that the existing finite element model can not accurately simulate the dynamic response of FRP strengthened RC column under blast loading,the dynamic mechanical properties of FRP materials and FRP-concrete interface and the constitutive model of concrete materials were studied systematically,and a refined finite element model that can accurately and reliably simulate the dynamic response of FRP strengthened RC column against blast loading was established.Secondly,in view of the lack of a simple,accurate and reliable damage assessment model of FRP strengthened RC column under blast loading,a large number of numerical simulations have been carried out,and a prediction model of residual axial load capacity for damage assessment of FRP strengthened RC column was established by using the genetic algorithm optimized neural network.Finally,in view of the uncertainty of dynamic response and damage degree of FRP strengthened RC column caused by the variation of blast load and material and size of RC columns,a performance-based blast fragility analysis of FRP strengthened RC column was proposed,and the failure probabilities of RC column strengthened with or without FRP were analyzed.The main research work and innovations are summarized as follows:(1)The dynamic mechanical properties of commonly used FRP materials were studied.Three kinds of FRP composites,i.e.,carbon fiber reinforced polymer(CFRP),basalt fiber reinforced polymer(BFRP)and glass fiber reinforced polymer(GFRP),were tested at different strain rates to study their dynamic tensile properties.The results show that Young’s modulus,tensile strength and failure strain of the FRP composites increase with the increase of strain rate,showing a significant strain rate effect.The strain rate effect of GFRP and BFRP is more significant than that of CFRP.The empirical formulas for predicting the dynamic increase factors(DIFs)of FRP mechanical properties at different strain rates were further proposed.(2)The dynamic mechanical properties of CFRP-concrete interface were studied.High-speed tensile tests were carried out on CFRP-concrete single shear specimens to study the dynamic mechanical properties of the interface under different slip rates.The results show that the ultimate bond strength,the maximum bond stress and the corresponding slip increase with the increase of slip rate,while the effective bond length is not affected by the slip rate.Furthermore,the empirical formulas for predicting ultimate bond strength,maximum bond stress and corresponding slip under different slip rates were fitted,and a dynamic bond-slip model of CFRP-concrete interface considering slip rate was established.(3)A refined finite element model is established to accurately and reliably simulate the FRP strengthened RC columns under blast loading.Based on the existing test data of mechanical properties of concrete under uniaxial and confining pressures,the strength surface,hardening and softening rule and flow rule and other parameters of KCC model were modified.By considering the strain rate effect of FRP and the dynamic bond-slip of FRP-concrete interface,a refined finite element model of FRP strengthened RC columns against blast loading was established.The results of numerical simulation are in good agreement with the experimental results.(4)A genetic algorithm optimized neural network model was established to accurately predict the residual axial load capacity of FRP strengthened RC column after the explosion.By simulating the dynamic response of more than 9000 unreinforced and FRP strengthened RC columns under blast loading,the effects of column size,stirrup ratio,longitudinal reinforcement ratio,axial load ratio,concrete strength and blast loads on their dynamic responses were studied.Based on the numerical simulation,a genetic algorithm optimized neural network model was established.Taking the above parameters as input variables,the residual axial load capacity of FRP strengthened RC columns can be accurately predicted by using the model.At the same time,another prediction model was established.The mid-height residual displacement of the RC column after the explosion is used to replace the blast load as the input variables of the model.The residual axial load capacity of FRP strengthened RC columns can also be accurately predicted by using the model.Only by measuring the mid-height residual displacement of the RC column after the explosion,the residual axial load capacity can be quickly predicted with this model.(5)A performance-based blast fragility analysis method for FRP strengthened RC column using the neural network prediction model was proposed.Firstly,the blast intensity,the anti-blast performance level and performance objectives of RC column are defined.Then,the Latin hypercube sampling method is used to generate the samples of RC column by considering the uncertainty of blast loading,material and geometry uncertainty of RC column.Finally,the residual axial load capacities of RC column samples under specific blast load are calculated by Monte Carlo simulation using the established neural network prediction model,and the probability distribution of different damage levels are counted,and then the fragility curve under specific damage level can be established.Typical RC columns strengthened with or without FRP strengthening are taken as examples to generate their fragility curves,and their damage probabilities were analyzed.