Regulación de velocidad con torque constante, basada en un controlador directo con redes neuronales artificiales, para un motor BLDC

Torque ripple occurs in Brushless Direct Current Motor (BLDC), and it is due to nontrapezoidal flux density distribution around the air-gap and variable magnetic reluctance of the air-gap as a result of stator slots distribution. These torque ripples periodically change with rotor position and are shown by speed variations, which degrade the BLDC motor performance particularly at low speeds because of the low inertial filtering. In this manuscript, a Direct Controller based on Artificial Neural Networks is proposed with the aim of reducing the speed ripple in a BLDC motor, thus allowing for smoother operation. Experimental evaluation is performed on a microcontroller evaluation platform and the speed variations are registered to compare its performance to the PID controller. Test results validate the effectiveness of the proposed controller in reducing the torque ripple, achieving a 50% reduction in speed range variation comparing to PID controller.