Adaptive Fuzzy Swarm-based Search Algorithm (AFSSA) for Complex Engineering Optimization

In recent years, swarm intelligence metaheuristic algorithms have emerged as powerful tools for solving real-world engineering optimization problems. However, their performance often degrades when applied to complex, high-dimensional problems. To address this limitation, we propose an Adaptive Fuzzy Swarm-based Search Algorithm (AFSSA), which incorporates a Fuzzy Dynamic Control Mechanism to dynamically adjust the optimization coefficients of swarm intelligence algorithms. AFSSA employs a Mamdani fuzzy inference system to enable smooth phase transitions during optimization, ensuring adaptability to the problem's unique characteristics. In this study, AFSSA is applied to enhance the acceleration coefficients of Particle Swarm Optimization (PSO) and Golden Search Optimization (GSO), resulting in AFSSA-PSO and AFSSA-GSO. The performance of these modified algorithms is evaluated on ۲۳ standard benchmark functions (with dimensions of ۳۰, ۱۰۰, and ۵۰۰) and the CEC۲۰۱۹ test suite, showing competitive results compared to other well-known optimization methods. Additionally, AFSSA is tested on data clustering problems, further demonstrating its versatility in handling complex real-world applications.

کلیدواژه ها:

Adaptive Fuzzy ، Swarm Intelligence ، Complex Optimization Problems ، High-Dimensional Optimization Problems