A chaotic salp swarm algorithm based on quadratic integrate and fire neural model for function optimization

Real-world problems generally do not possess mathematical features such as differentiability and convexity and thus require non-traditional approaches to find optimal solutions. SSA is a meta-heuristic optimization algorithm based on the swimming behaviour of salps. Though a novel idea, it suffers from a slow convergence rate to the optimal solution, due to lack of diversity in salp population. In order to improve its performance, chaotic oscillations generated from quadratic integrate and fire model have been augmented to SSA. This improves the balance between exploration and exploitation, generating diversity in the salp population, thus avoiding local entrapment. CSSA has been tested against twenty-two bench mark functions. Its performance has been compared with existing standard optimization algorithms, namely particle swarm optimization, ant–lion optimization and salp swarm algorithm. Statistical tests have been carried out to prove the superiority of chaotic salp swarm algorithm over the other three algorithms. Finally, chaotic SSA is applied on three engineering problems to demonstrate its practicability in real-life applications.