PI controller tuned by bee swarm for level control systems

Level systems are commonly found in the industrial field, presenting many times, a reasonable difficulty to be tuned by classical control techniques. Showing that controller tuning is still very interesting and important, the Bee Swarm Optimization (BSO) comes as a way for solving such difficulties. The BSO is a metaheuristic method that guides the search process, which is inspired by the great organizational skills of bees to collect food. Applying it to control theory, this method uses the resulting control system of reference by a PI controller. Thus, this paper deals with determining the values of the proportional (K
_p
) and integral (Ki) gain of the controller for a level system by the BSO, considering Integral of Time-weighted Absolute Error (ITAE) as an objective function. The experimental results in a SMAR didactic plant were obtained to compare a BSO tuning method with a classical control method widely used in industry (Rivera IMC). Thus the effectiveness of BSO has been validated, showing it's an interesting technique to be used in the industrial field for this kind of process.