Teaching Particle Swarm Optimization Through an Open-Loop System Identification Project PAULO MOURA OLIVEIRA, 1 DAMIR VRANC ˇ IC ´ , 2 J. BOAVENTURA CUNHA, 1 E. J. SOLTEIRO PIRES 3 1 CIDESD, Department of Engineering, University of Tra´s-os-Montes e Alto Douro (UTAD), 5001-801 Vila Real, Portugal 2 Department of Systems and Control, Jozˇef Stefan Institute, Jamova cesta 39, SI-1000 Ljubljana, Slovenia 3 CITAB, Department of Engineering, University of Tra´s-os-Montes e Alto Douro (UTAD), 5001-801 Vila Real, Portugal Received 15 February 2011; accepted 18 April 2011 ABSTRACT: The particle swarm optimization (PSO), one of the most successful natural inspired algor- ithms, is revisited in the context of a proposal for a new teaching experiment. The problem considered is the open-loop step identification procedure, which is studied as an optimization problem. The PSO canonical algorithm main issues addressed within the proposed open-loop step identification experience are: the swarm random initialization methodology, the population size variation, and the inertia weight selection. The teaching experience learning outcomes are stated, simulation results presented, and feedback results from students analyzed. ß 2011 Wiley Periodicals, Inc. Comput Appl Eng Educ; View this article online at wileyonlinelibrary. com/journal/cae; DOI 10.1002/cae.20549 Keywords: particle swarm optimization; control experiments; control education; open-loop identification INTRODUCTION A huge variety of engineering applications can benefit from the use of computational intelligence techniques, namely search and optimization problems. Well-known biological and natural inspired techniques, which proved successful in solving search and optimization problems are: simulated annealing (SA), genetic algorithms (GAs), population-based incremental learn- ing (PBIL), differential evolution (DE), ant colony optimization (ACO), and particle swarm optimization algorithm (PSO). The technique selected to be addressed in this teaching project is the PSO algorithm, originally proposed by Kennedy and Eber- hart [1]. This selection is based on two reasons: (i) the beauty and harmony associated with collective behavior of swarms, such as bird flocks and fish schools and (ii) the simplicity of the canonical PSO algorithm. Indeed, while other algorithms are also conceptually simple, they are not as simple for imple- mentation as the PSO. This makes the PSO algorithm a good candidate to be taught both in undergraduate and postgraduate courses. Since the pioneering work proposed by Kennedy and Eberhart [1], the PSO algorithm has been the subject of signifi- cant research efforts resulting in improvements [2–4] and com- prehension of its behavior [5], both for single and multiple- objective optimization [6] and it has been successfully applied in solving a myriad of optimization problems such as Refs. [7– 10]. Some software tools using the PSO have been proposed [11,12] applied to mathematical function optimization. The PSO has also been reported for teaching purposes [13], using a mobile robotics trajectory planning experiment and within a PID controller design experiment [14], in which it was com- pared with other modern-heuristics. The study of open-loop identification techniques based on step-response analysis is crucial in any undergraduate feedback control course. The step-response open-loop techniques are related to PID controller tuning methodologies, such as the well-known Ziegler–Nichols [15] and Cohen–Coon [16], as Correspondence to P. M. Oliveira (oliveira@utad.pt). ß 2011 Wiley Periodicals Inc. 1