Compact genetic algorithms for the optimization of induction motor cascaded control

Recent research on compact GAs (cGAs) has proposed a number of evolutionary search methods with reduced memory requirements. The cGAs evolve a stochastic description of an hypothetical population processing a probability vector with update rules inspired to the typical selection and recombination operations of a GA. The cGAs well lend themselves to real-time implementations in constrained, low capacity microcontrollers, as they have reduced memory requirements and evenly distributed computational loads with respect to the standard, population-based GA. This paper considers the implementation of cGAs in the same microcontroller used to implement the cascaded control of an induction motor drive. We develop a real-valued version of a cGA that achieves final solutions of the same quality of those found by binary cGAs, with a significantly reduced computational cost. The potential of the proposed approach is assessed by means of an experimental study. The cascaded control system obtained through genetic search outperforms alternative schemes obtained with linear design techniques.