| With the development in economy,science and technology and the widespread application of lighter structural systems and higher strength materials in recent years,the super high-rise buildings have become taller and constructed with more irregular geometric shapes.Wind load plays a controlling role in the structural design for such slender and taller buildings,which will cause safety and serviceability design problems.How to balance the safety,serviceability and economy in structural design of tall buildings has always been an important issue for structural engineers and wind engineers.However,the existing research works on wind-resistant optimization design are mainly based on the Optimal Criteria（OC）algorithm,which need to explicitly express the constraints in terms of design variables through complex formulas and is easier to trap into local optimal solution when dealing with complex engineering problems.Moreover,the gradient-based OC algorithm cannot handle problems including discrete variables.With the advantages of strong universality and global optimization ability,as well as avoiding the tedious iterative format derivation of OC algorithm,the modern intelligent algorithms have good application prospects in complex engineering problems and they are easier to develop into universal optimization design software for engineering practice.Based on the Genetic Algorithm（GA）and Particle Swarm Optimization（PSO）as representatives of intelligent algorithms,this thesis presents an integrated finite element analysis and optimal design procedure for the wind-resistant design of super-tall buildings,which aims to minimize the total weight of the designed building subject to the constraints of top acceleration and lateral drifts.The major research is summarized as follows:（1）A mathematical model for the wind-resistant optimization design of high-rise buildings is established,which takes the sectional dimensions of structural members（beams,columns and shear walls）as design variables,the wind-induced responses of the building（the lateral top displacement,the inter-storey drift ratio and the wind-induced acceleration）as the constraints,and the total weight of the building as the objective function.（2）The interface of the joint simulation using MATLAB and ANSYS is built.Based on the equivalent static wind loading calculation method,the random dynamic wind load is converted into the equivalent static wind load.It calls ANSYS through the MATLAB main program while allowing for instantaneous updating of the wind loads on buildings during the optimal process.The analysis of structural dynamic properties and wind-induced displacements are performed by ANSYS-APDL Batch operation.The improved intelligent algorithm is programmed in MATLAB to find the optimal solution efficiently.（3）It avoids the complex explicit expression of the constraint conditions using the design variables.For objective functions violated the constraint condition,a certain penalty function is applied to make them much larger than the required target values and thus their corresponding fitness in the algorithm will become small and be eliminated gradually under the action of optimization operators.（4）Genetic algorithm and particle swarm optimization algorithm are used to study the wind-resistant optimal design of super-tall building.The principle of the algorithms and the values of control parameters are introduced in detail,and the algorithms are appropriately improved.The test functions are applied to illustrate the viability of the algorithms and to set the control parameters properly.The CAARC standard tall building is adopted to verify the effectiveness of the presented method.The example demonstrates that the proposed optimal design procedure is not only able to redistribute the lateral stiffness of tall buildings to satisfy the constrains of the wind-induced responses including top displacements,inter-storey drift ratios and atop acceleration,but also is able to reduce the wind loads on buildings especially in the across-wind direction. |
