| RC (Reinforced concrete) D-region components are a class of components thatthe cross section strain do not meet the plane section assumption, their stressdistributions are very complex, and the beam theory is not applicable. New Chinesecode recommonds stress methods based on elastic or elastoplastic analysis fordesigning these components, but no clear formulas exist. Mostly, RC D-regioncomponents are designed by rule of thumb based on the experimental analysis. Themechanical concept of these methods is ambiguous, it is hard to understand andmaster. It may lead to the design conservative or unsafe. However, the extensiveapplication of the RC D-region components highlights the blind spots of domesticdesign code. In recent decades, the foreign researchers have reasonably explained thestress of the RC D-region components by using the strut-and-tie model. This modelclearly express the stress transfer path and can be easily understood. It has beenapplied in the design code of concrete structures of some countries. Many domesticresearchers have used the strut-and-tie model design method to study the RC D-regioncomponents, but there is still no unified conclusion. Drawing on the strut-and-tiemodel design method and putting forward a unified design method for the RCD-region components in China is an arduous and long-term task. A further study ofthe RC D-region components is of great significance.In general, the strut-and-tie model is established on the basis of structure realstress distributionis, which is on behalf of the mechanism action of the force, it is areasonable design method. However, it is very difficult to get an accurate strut-and-tiemodel in the design. Liu X and YI W J put forward a new kind of topologyoptimization methods called genetic evolutionary structural optimization algorithm(ie GESO algorithm), they combined the Genetic algorithm and ESO algorithm. Thisalgorithm can quickly find the optimal model, and we can construct the strut-and-tiemodel refers to it. According to the optimal model, deep beams with opening andZ-beams were constructed into strut-and-tie model, and then be designed refers to theBuilding Code Requirements for Structural Concrete ACI318-08. As a contrast,specimens which not optimized were designed in the Code for Design of ConcreteStructures. Based on the existing topology optimization model, the following contentswere studied: (1) Compared the crack, deformation and final failure mode of each specimendesigned in the Code for Design of Concrete Structures GB50010-2010with thatoptimized by GESO algorithm and designed in the Building Code Requirements forStructural Concrete ACI318-08Appendix A. Analyzed the factors of the differentfailure modes, some improvements were proposed for the design;(2) By analysing the strain data of reinforcement and concrete collected in theexperiment, the strain along cross section of the specimen does not meet the planesection assumption. Meanwhile, the reasonableness of GESO algorithm resulting instrut-and-tie model was evaluated;(3) Used the large-scale nonlinear finite element software ABAQUS to set upfinite model of the test specimens, and by comparing the results of numericalsimulation with the test results, the accuracy of the ABAQUS finite elementcalculation was verified. |
