The disadvantages of particle swarm optimization (PSO) algorithm are that it is easy to fall into local optimum in high-dimensional space and has a low convergence rate in the iterative process.
Dear Assen,
please see these papers on PSO
Jumping Particle Swarm Optimization, DOI:10.1007/978-981-16-2380-6_65, In book: Proceedings of Sixth International Congress on Information and Communication Technology
2-Particle Swarm Optimization: A Comprehensive Survey, DOI:
- 10.1109/ACCESS.2022.3142859.
3-Strategy Dynamics Particle Swarm Optimizer,DOI:10.1016/j.ins.2021.10.028
With kind
Assen Beshr
I recommend the following steps:
1. Create a 'population' of agents (particles) that are evenly scattered over X.
2. Determine the position of each particle in relation to the objective function ( say the below function).
3. Update a particle if its current location is better than its prior best position.
In addition, have a look at this:
1. Article Improvement of Particle Swarm Optimization: Application of t...
2. https://www.sciencedirect.com/topics/computer-science/particle-swarm-optimization
3. Article Particle Swarm Optimisation: A Historical Review Up to the C...
4. https://mdpi-res.com/d_attachment/remotesensing/remotesensing-13-02514/article_deploy/remotesensing-13-02514-v2.pdf
Assen Beshr : what does your problem look like? Are you sure you need a metaheuristic?
I've had good experience with Differential Evolution (https://en.wikipedia.org/wiki/Differential_evolution).