Hybrid particle swarm optimisation with adaptively coordinated local searches for multimodal optimisation
Summary: Particle swarm optimisation (PSO) is a population-based stochastic search algorithm. Two common criticisms exist. First, PSO suffers premature convergence. Second, several existing PSO variants are designed for a specific search space thus an algorithm performing well on a diverse set of problems is lacking. In this paper, we propose a hybrid particle swarm optimisation with adaptively coordinated local searches, called NMRM-PSO, to make up the above demerits. These local search algorithms are the Nelder mead algorithm and the Rosenbrock method. NMRM-PSO has two alternative phases: the exploration phase realised by PSO and the exploitation phase completed by two adaptively coordinated local searches. Experiment results show that NMRM-PSO outperforms all of the tested PSO algorithms on most of multimodal functions in terms of solution quality, convergence speed and success rate.
- Particle swarm optimization with adaptive mutation for multimodal optimization
- An improved adaptive particle swarm optimization approach for multi-modal function optimization
- A novel particle swarm optimisation with hybrid strategies
- Multi-objective particle swarm optimization algorithm with local optimal particles search method
This page was built for publication: Hybrid particle swarm optimisation with adaptively coordinated local searches for multimodal optimisation
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2224023)
