Speed regulation of DC motors based on on-line optimum asynchronous controller tuning and differential evolution

The use of bio-inspired algorithms into controller tuning is one of the main topics in the last decades. One crucial issue when using bio-inspired algorithms to on-line tune the control system (adaptive controller tuning) is the computational time. In this work, an Asynchronous Adaptive Controller Tuning (AACT) approach is proposed to reduce the frequent activation of the tuning process, and hence, the computational cost could be reduced. This approach is based on a proposed event function, which determines the update time instant of the control parameters through the Differential Evolution algorithm. Comparative results with a Synchronous Adaptive Controller Tuning (SACT) approach are included in the study case of the speed regulation of the DC motor under the effects of uncertainties in the load. The SACT approach periodically updates the controller parameter at each sampling time. The comparative analysis indicates that the proposed AACT significantly reduces the controller parameter update without considerably increase the regulation error.