Position sensorless speed estimation in switched reluctance motor drive with direct torque control-inductance vector angle based approach

Fuat Kucuk, Hiroki Goto, Hai Jiao Guo, Osamu Ichinokura

Research output: Contribution to journalArticlepeer-review

Abstract

Feedback signals of rotor speed and motor torque are essential in most of Switched Reluctance (SR) motor control applications. An SR motor has highly nonlinear characteristic that does not allow to be modeled by simple equations. In Direct Torque Control (DTC) drive, which enables easy control of torque ripple in the SR motor, position sensor is employed to obtain the feedback signals. Position sensor causes DTC drive not only less reliable but also more expensive. Estimation of feedback signals is required in order to eliminate position sensor. This paper concerns about sensorless speed estimation under the DTC condition and presents a simple method. Simple sensorless speed estimation is proposed based on inductance vector angle. The inductance vector angle is obtained by applying α - β transformation to the phase inductances. A relay triggers a speed calculation circuit according to its band limits and the inductance vector angle. Inside the circuit, triggering time is kept in a memory until the next triggering. Rotor pole pitch is divided by the time difference between two consecutive triggerings. Finally the estimation circuit outputs the rotor speed. Sensorless speed estimation is simulated and verified experimentally to show its validity.

Original languageEnglish
Pages (from-to)5+533-538
JournalIEEJ Transactions on Fundamentals and Materials
Volume128
Issue number8
Publication statusPublished - 2008 Dec 1

Keywords

  • Direct torque control
  • Inductance vector angle
  • Sensorless speed estimation
  • Switched reluctance motor

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'Position sensorless speed estimation in switched reluctance motor drive with direct torque control-inductance vector angle based approach'. Together they form a unique fingerprint.

Cite this