Ultimate Bearing Capacity Analysis And Parameter Optimization Of Prestressed Steel-concrete Composite Beams

In recent years,steel-concrete composite structures were increasingly used in practical engineering.However,in the actual design and application process,the calculation of the bearing capacity was conservative.In order to improve the utilization rate of the materialsandreduce thecosts,this paper presents a method of optimizing the design of prestressed steel-concrete composite beams.It was uses the extreme value search ability of the particle swarm algorithm to search the specified sample space for the optimal combination of design parameters.This paper focuses on the application of artificial neural network and particle swarm optimization in the optimization of bearing capacity of prestressed steel-concrete composite beams.The main contents include:?1?According to the existing setup model's method ofcompositebeam,the three-dimensional finite element model of steel-concrete composite beam and prestressed steel-concrete composite beam was established.Thenthe standard finite element model was obtainedby the modification ofthe existing experimental results.?2?In according withthe standard finite element model,the sensitivity of steel tensile strength,concrete axial compressive strength,concrete slab thickness,initial tension prestress and stud diameter to the ultimate bearing capacity of prestressed steel-concrete composite beams were analyzed.The sensitivity levels were 0.66,0.17,0.25,0.13 and0.31 respectively.The average values were:1.002,0.783,0.819 and 0.83 respectively.?3?The BP neural network prediction model was set upto predict the ultimate bearing capacity of prestressed composite beams.Then10 sets of test samples were randomly choiced in the existing research.The ratios of ultimate bearing capacity and test value of prestressed steel-concrete composite beams obtained by neural network prediction,design specification,Nie Jianguo and Hu Shaowei are compared.The average values were:1.002,0.783,0.819,0.83.The neural network predictionwasapplied inthe Hushang Bridge in Shaoxing andthe applicabilityof the method was verifiedin pratical engineering.?4?This paper took the CO₂ emission as the objective function,and the particle swarm optimization algorithm was used to search for the parameters that minimizes the CO₂emission of the composite beam in the constrained space.The optimized composite beam was analyzed through finite element modeland BP neural network prediction,the ultimate bearing capacity obtained fromthe finite element analysis is reduced by 15.6%,and the BP neural network prediction value is reduced by 9.4%.The optimized composite beamstill meets the code requirements.The stiffness optimized composite beamreduced by3.2%,andCO₂ consumption reduced by 44.79%.The results show that the method is feasible in the optimization design of prestressed steel-concrete composite beams.