Implementation of the Model Reference Adaptive System Algorithm in Arduino

Abstract

This paper describes a study and experimental verification on the implementation of algorithm Adaptive Reference Model System in the Arduino/DUE platform. The chosen algorithm allows to estimate both the angular velocity and the electrical torque of an induction squirrel cage motor which allows adapting an adjustable model to the reference model, where both models have as input the voltages and power currents of an induction machine and outputs of the models the vector of the rotor flow. The adjustment process requires a PID family controller to obtain null error in product point of the outputs of said models. The system is tested via simulation on the Matlab/Simulink platform to verify the operation of the algorithm. In addition, the system is implemented on the Arduino/DUE platform and the results obtained in the laboratory are shown to verify the proper functioning of the inference system. Finally, the results obtained from laboratory, simulation and algorithmic model are compared.