Choosing suitable variants of Differential Evolution and Particle Swarm Optimization for the optimization of a PI cascade control

Robust control is an intensively studied field in control theory where analytical solutions for robust control problems often lead to complicated control structures. On the other hand, for simple structures like the PI cascade control l here are no analytical solutions which leads to the need of easy-to-use optimization algorithms. Differential evolution and particle swarm optimization are well suited for this problem because of the real-valued representation of solutions, fast convergence behavior and ease of use. However, seven al variants exist for both algorithms, and from literature if does not become clear which one performs best. Therefore, in this paper several strategies of differential evolution an d different neighborhood topologies for particle swarm optimization are applied for the optimization of a PI cascade control. A performance comparison shows that both algorithms are able to solve this optimization problem but especially using differential evolution the quality of the best solution varies for different strategies.