Abstract
In this paper, an approach for system identification and automatic control systems
design based on a combination of Particle Swarm Optimization (PSO) together with
Principal Components Analysis (PCA) is proposed. This methodology, referred to
as PSO-PCA optimization approach, incorporates the PCA concept on the PSO algorithm, which can be implemented off-line or even on-line. Different variants of the
PSO-PCA algorithm will be presented and compared with the classical PSO, in order
to evaluate the performance of the underlying algorithms, in simulations and in real
experiments.
design based on a combination of Particle Swarm Optimization (PSO) together with
Principal Components Analysis (PCA) is proposed. This methodology, referred to
as PSO-PCA optimization approach, incorporates the PCA concept on the PSO algorithm, which can be implemented off-line or even on-line. Different variants of the
PSO-PCA algorithm will be presented and compared with the classical PSO, in order
to evaluate the performance of the underlying algorithms, in simulations and in real
experiments.
| Original language | English |
|---|---|
| Title of host publication | Focus on Swarm Intelligence Research and Applications |
| Publisher | Nova Science Publishers |
| Pages | 177-192 |
| Number of pages | 16 |
| ISBN (Print) | 978-1-53612-452-1 |
| Publication status | Published - 2017 |
Keywords
- Constrained nonlinear optimization
- Control design
- Particle swarm optimization
- Principal components analysis
- System identification