سامانه همایش‌ها TSTA | ثبت‌نام و ارسال مقاله کنفرانس‌ها و کنگره‌های ملی

صفحه اصلی / سی و دومین کنفرانس ملی و دهمین کنفرانس بین المللی مهندسی زیست پزشکی ایران

ENHANCING POPULATION DIVERSITY AND OPTIMIZATION EFFICIENCY IN CAT SWARM OPTIMIZATION USING A FUZZY CONTROLLER

نویسندگان :

Seyede maryam Rezaei¹ Reza Boostani²

1- موسسه غیر انتفاعی آپادانا شیراز 2- دانشگاه شیراز

کلمات کلیدی :

Cat swarm optimization،Fuzzy inference system،Chaotic operators،optimization،metaheuristic algorithm

چکیده :

Optimization plays a critical role in artificial intelligence and various computational tasks. Metaheuristic algorithm such as GA, PSO, and CSO are widely used, yet CSO suffers from low exploration due to rapid convergence. In this study, we propose a fuzzy-based enhancement to CSO to increase population diversity and prevent premature convergence. A fuzzy controller dynamically adjusts agent velocities based on the execution stage and fitness relative to the global best. The proposed method was evaluated on ten complex benchmark functions, each executed 30 times to ensure statistical validity and compared to the standard CSO, Particle Swarm Optimization (PSO) and Genetic Algorithm (GA). Experimental results show that the fuzzy-based CSO significantly improves convergence accuracy and avoids local optima, achieving an average improvement of 12.4% in solution quality over the standard CSO and outperforming PSO and GA in 80% of the test cases. The proposed scheme can be applied in optimization of biomedical data such as learning of weights in deep learning structures and biomedical data analysis.