Intelligent optimization method for complex supertall inclined CFT frame-core tube-outrigger truss structures
摘要
Complex super high-rise buildings are widely used in public and landmark projects, assuming important social and economic roles. The structural design of such buildings often suffers from low efficiency, long design cycles, and heavy labor and computational costs. In this paper, an intelligent optimization algorithm based on an improved full stress criterion and integrated with the Grey Wolf Optimization algorithm is proposed for the automated and intelligent structural design to provide economy and efficiency for design work. A typical actual complex super high-rise structure is used to investigate the effectiveness of the proposed algorithm, and an additional CFST frame–RC core tube structure with moderate complexity is introduced to further verify the applicability across different structural scales. The results show that the optimized structure can reduce the material cost by 23.63%, compared to the original structural scheme, and can effectively improve optimization efficiency by more than five times relative to traditional heuristic algorithms, compared to other algorithms. After optimization, both the stiffness constraints at the structural level and strength constraints at the component level satisfy the code requirements. Finally, a generalized stiffness-based structural importance analysis is utilized to assess the contribution and corresponding change of the substructures to the structural force transferring system. The importance results show that the importance of interior core tube increases, with that for the exterior wall increasing and the interior wall decreasing, and the importance of the perimeter frame decreases, with that for the columns increasing and the beams decreasing. These findings reveal a redistribution of stiffness and force-resisting mechanisms, suggesting that strengthening the core tube enhances seismic performance while enabling a more economical design of the perimeter frame. Overall, the proposed algorithm demonstrates consistent performance in both moderately complex and highly complex structures, highlighting its general applicability and providing practical guidance for the structural design of complex super high-rise buildings.