Modification of Fish Swarm Algorithm Based on Lévy Flight and Firefly Behavior

Area of Science:

• Computational Intelligence
• Optimization Algorithms
• Swarm Intelligence

Background:

• Artificial fish swarm algorithm (AFSA) often converges to local optima in complex optimization problems.
• Multidimensional and multi-extreme value functions present significant challenges for traditional AFSA.

Purpose of the Study:

• To propose a novel artificial fish swarm algorithm, LFFSA, to overcome the local optimum convergence issue.
• To enhance the global optimization capabilities of fish swarm algorithms.

Main Methods:

• Incorporating firefly algorithm's moving strategies into AFSA's chasing and preying behaviors.
• Introducing Lévy flight for improved searching capabilities.
• Utilizing nonlinear view and dynamic parameters for step size to limit the search band.

Main Results:

• LFFSA demonstrated superior performance compared to other tested algorithms.
• The proposed algorithm showed enhanced convergence speed.
• Optimization accuracy was significantly improved by LFFSA.

Conclusions:

• LFFSA effectively addresses the local optimum convergence problem in AFSA.
• The integration of Lévy flight and firefly behavior yields a more robust optimization algorithm.
• LFFSA offers a promising solution for global optimization of complex functions.