Low speed sensorless model predictive current control of a three phase induction motor from a single phase supply

Abstract

The three phase induction motor (3φ IM) is arguably most popular owing to its rugged and robust nature. However in single phase systems, it becomes almost mandatory to use motors other than the 3φ IM. To operate it from a single phase supply, a 3φ IM conventionally requires AC to DC conversion, followed by conversion from DC to three phase AC, which calls for energy storage in the DC link, resulting in a larger footprint of such 1φ–3φ motor drives. An effective solution for running a 3φ IM from a single phase supply, can be the use of single phase to three phase Matrix Converter (1φ–3φ MC). This paper proposes a predictive current control strategy that employs Finite Control Set Model Predictive Control (FCS-MPC) for the modulation of 1φ–3φ MC and subsequent control of 3φ IM. Predictive model of the motor is also used in estimating the motor speed. Sensor requirements are significantly reduced as stator currents and voltages are appropriately predicted. The entire control algorithm uses only one voltage sensor to sense the 10 input voltage. The converter delivers a superior quality of motor current as compared to its thyristor based counterpart, i.e the 1φ–3φ cyclo-converter. However best results are achieved at low output to input voltage ratios and low frequency, hence making it particularly suitable for low speed applications such as home appliances and treadmills. The converter being inherently bi-directional can be of particular interest in lift systems.