Skip to main content

Najib Kabir Dankadai

Improving torque control in switched reluctance motors.

Project title

Torque control of magnet-assisted switched reluctance motor (MSSRM)

Supervisors

Project description

Over the last decade, the Magnet-Assisted Switched Reluctance Motor (MSSRM) has gained interest in applications for:

  • Electric Vehicles and Hybrid Electric Vehicles (EVs/HEVs)
  • more-electric aircraft
  • industry
  • white goods

This is mainly because of its low-cost, fault-tolerance, simple structure and wide speed range of operation. But it inherently suffers from large torque ripple and machine nonlinearities.

This project proposed a new Direct Instantaneous Torque Control (DITC) of an MSSRM. Based on a new Adaptive Switching Quasi-Sliding-Mode (QSM) control, it minimises the torque ripple of the motor. In addition to the improved torque characteristics, the size, cost, reliability and robustness of the overall system are of great concern.

We need to ensure the reliability of the motor drive’s hardware and control algorithm. We have investigated the reliability and complexity of several widely used control methods for SRM drives. We based our invesigations on information flow complexity within the control technique. We designed the QSM control after considering the findings.

Furthermore, the power electronics needed to operate the drive increase its cost and volume. They also reduce its running reliability. Hence, we need to reduce the number of switches, diodes and sensors involved by improving the power converter topology. Thus, we propose a sensorless control of MSSRM using a single current sensor and Resistor-Dump (R-Dump) converter.

An R-Dump converter has a minimal power component compared to converters like an asymmetric half-bridge converter. We developed the sensorless control based on a new sliding mode observer. This reduces the feedback sensors by only using a single current measurement to estimate the rotor position of the MSSRM. We are comparing different methods, including:

  • Current Chopping Control (CCC)
  • Torque Sharing Function-based Direct Instantaneous Torque Control (TSF-DITC)
  • Direct Torque Control (DTC)
  • the proposed QSM control

Our results show that the proposed method achieved superior torque characteristics. It maintains good reliability, smaller size and lower cost.

Publications

Interests

Modelling and automatic control of nonlinear systems, machine learning

Qualifications

  • MEng (Hons) Mechatronics and Automatic Control, Universiti Teknologi Malaysia
  • BEng (Hons) Computer Engineering, Bayero University

Najib Dankadai Najib DankadaiLinkedIn

Programme information
Assessment and feedback
Student support