Structured mixed-sensitivity H∞ design using Particle Swarm Optimization

Abstract

In this paper, a new structured mixed-sensitivity ℋ∞ design approach, using Particle Swarm Optimization (PSO) technique, is proposed. The case study of an electrical DC drive benchmark is adopted to illustrate the efficiency and viability of the proposed control approach. The optimization based synthesis problem is formulated and solved by a constrained PSO algorithm. In the proposed control strategy, a PID controller's structure is adopted. Simulations and experimental results show the advantages of simple structure, lower order and robustness of the proposed controller. A comparison to another similar evolutionary algorithm, such as Genetic Algorithm (GA), shows the superiority of the PSO-based method to solve the formulated optimization problem.