Stochastic Creep and Shrinkage Effect of Steel-Concrete Composite Beam

Abstract

Creep and shrinkage of concrete may induce time-variant structural responses of steel-concrete composite beam. Since the creep and shrinkage are the most uncertain material properties of concrete, stochastic analysis is performed to obtain the long-term random responses of steel-concrete composite beam. MC90 model is chosen to describe the creep and shrinkage of concrete, and age-adjusted effective modulus method is adopted for creep analysis. The randomness about creep and shrinkage model, mean compressive strength and elastic modulus of concrete, ambient humidity, magnitude of sustained load, stiffness of shear connector, and so on are concerned. Two random analysis methods, i.e., Latin hypercube sampling and a hybrid approach combining with Monte Carlo sampling and response surface method, are adopted. The stochastic responses of deflection and stress of steel girder and concrete slab are studied. Meanwhile, with sensitivity analysis, the statistical influences of random variables on long-term responses are investigated.