Numerical simulation methods for reinforced concrete columns considering parameter stochasticity

Reinforced concrete column specimens with the same parameters often show differences in bearing capacity and deformation under the same loading path. Most of the existing studies attribute the causes to recording instrumentation errors or randomness in material and bond dimensions, but the method to improve the accuracy of numerical simulation of the reinforced concrete components is still unclear. For a given member, the randomness of material and geometry is an important cause of simulation error, which should be taken into account to improve the simulation accuracy. A weighted least squares estimation (WLS) method is used to fuse the test samples of reinforced concrete member section sizes, reinforcement and concrete with confidence factors as weights. The probability distribution functions of reinforced concrete member section sizes, reinforcement and concrete are calculated by a genetic algorithm method. Based on the existing seismic performance test data of reinforced concrete columns, a stochastic finite element model of reinforced concrete columns is developed for seismic performance analysis using Monte Carlo method through a mixture of two software programs, MATLAB and OpenSees. The numerical seismic performance indexes of eight types of reinforced concrete columns were compared with the test results to verify the importance of introducing stochastic parameters into the model. The results demonstrate the simulation accuracy of the finite element model considering parameter randomness is 88% higher than that of the finite element model without considering parameter randomness, which proves the effectiveness of the numerical simulation method for reinforced concrete columns considering parameter randomness.