Speed estimation of the three-phase induction motor with Volts/Hertz control using artificial intelligence techniques

Abstract

This paper deals with the design and development of a prototype for speed estimation of a three-phase squirrel cage induction motor using Volts/Hertz control. The speed estimation was using an embedded electronic device with an adaptive neuro-fuzzy inference system. In this way, it is expected to increase the reliability of the rotor speed measurement. For the estimation, the currents and voltages that feed the induction motor were taken as input data to training an adaptive neuro-fuzzy inference system. Additionally, a Volts/Hertz control technique was used to test the operation of the estimator at different operating speeds, comparing the estimation with a speed sensor coupled to the machine shaft. The estimation was supported on a Raspberry Pi 3B hardware. The algorithms were developed using Simulink, which allows a graphical block interface that facilitates the understanding of the system. For the power stage, a multilevel converter was developed, thus, ensures a low total harmonic distortion.