Researchers have introduced a more efficient approach to designing wind-resistant high-rise buildings by combining particle swarm optimization (PSO) with an enhanced penalty function to dynamically update wind loads during the design process.
Study: An improved particle swarm optimization for wind resistance performance design of high-rise buildings. Image Credit: Ken Felepchuk/Shutterstock.com
Published in Advances in Wind Engineering, the study proposes a method that uses sectional dimensions of structural members as design variables, minimizes total structural weight, and treats wind-induced responses as critical constraints—all within an optimization framework tailored for tall buildings.
Background
As modern skyscrapers grow taller and more flexible with the use of high-strength materials, they become increasingly vulnerable to wind-induced motion. Ensuring these structures remain safe, comfortable, and cost-effective under wind loads poses a significant challenge for engineers. The design process must account for a range of complex factors—non-linear behaviors, non-convex and discontinuous variables, and geometric sensitivities—all of which make structural optimization especially demanding.
To manage this complexity, researchers are turning to advanced computational methods. Among these, PSO has gained traction for its multi-point search ability, gradient-free approach, and fast convergence, making it well-suited for civil engineering applications involving large-scale structures.
Methodology
In this study, the researchers developed a mathematical model that defines sectional dimensions as discrete design variables, with the goal of minimizing total building weight. The optimization was subject to constraints on wind-induced displacement, acceleration, and available cross-sections.
The PSO algorithm updated its position vectors by comparing each iteration’s values to the closest feasible discrete variable. An improved penalty function was introduced to better handle constraints related to wind-induced responses.
To support this, the team integrated ANSYS (for simulating structural dynamics and wind responses) with MATLAB (to run the PSO algorithm). This hybrid setup allowed the system to instantly update equivalent static wind loads (ESWLs) throughout the optimization process.
The method was tested on a 60-story rectangular tower, and the full computation—run on a high-performance computer—took approximately 40.6 hours.
Results and Discussion
The optimized design led to a modest 3.11 % increase in structural weight to meet minimum natural frequency requirements. Initially, the objective function rose sharply, then gradually declined and stabilized after roughly 160 iterations.
Notably, natural frequencies improved by 17.64 % in the X-direction and 15.38 % in the Y-direction. Since the Y-direction consistently exceeded the target, the X-direction governed the wind-induced acceleration constraints, highlighting the need to prioritize across-wind considerations in design.
Shear forces were higher along the wind direction, but the algorithm proved especially effective in mitigating across-wind loads. Inter-story drift ratios began above acceptable levels, indicating low initial stiffness, but quickly dropped within limits in early iterations and fluctuated near the target as the process converged.
Shear wall thicknesses adapted throughout the design. Floors 41–60 started with conservative dimensions that were trimmed in early iterations. Floors 21–40 saw ongoing fluctuations, while the walls on floors 1–20 required a sharp initial increase to meet stiffness needs. These adjustments confirmed the penalty function’s effectiveness in managing displacement constraints.
Conclusion
This study demonstrates that the improved PSO algorithm offers a practical and efficient solution for optimizing high-rise buildings against wind loads. By accommodating discrete variables and complex constraints, and enabling real-time load updates, the method delivers reliable performance outcomes.
Tested on a 60-story reinforced concrete frame-tube structure, the approach showed strong potential for broader application in civil engineering. Its use of natural frequencies as a basis for optimization suggests it could be extended to a variety of tall and slender structures.
Journal Reference
Li, Y., Zhou, J., Chen, F., & Sun, M. (2025). An improved particle swarm optimization for wind resistance performance design of high-rise buildings. Advances in Wind Engineering, 2(2), 100053. DOI: 10.1016/j.awe.2025.100053, https://www.sciencedirect.com/science/article/pii/S295060182500024
Disclaimer: The views expressed here are those of the author expressed in their private capacity and do not necessarily represent the views of AZoM.com Limited T/A AZoNetwork the owner and operator of this website. This disclaimer forms part of the Terms and conditions of use of this website.