Compensation of Speed Dependency in Sensorless Rotor Temperature Estimation for Permanent Magnet Synchronous Motor

Martin Ganchev (Speaker), Christian Kral, Thomas Wolbank

Research output: Chapter in Book or Conference ProceedingsConference Proceedings with Poster Presentationpeer-review

Abstract

The work proposes an improved method for estimating the magnet temperature in permanent magnet synchronous machines (PMSM´s). Originally the method implies an intermittent injection of a voltage pulse in the positive d-axis of the motor while keeping the load current zero. Thus, the resulting d-current response depends on both the initial value of the d-current itself and the actual magnetization level of the permanent magnets. Since the magnetization of the magnets depends on the temperature, different d-current slopes are associated with given temperature levels of the magnets. At higher speeds the current response is additionally affected by induced voltages of various sources which leads to erroneous estimation of the magnet temperature. By applying a voltage pulse in the positive and negative d-axis of the motor, symmetry of the induced voltages can be achieved in a manner that the difference of the current responses from the positive and negative pulse is no longer affected by the induced voltages.
Original languageEnglish
Title of host publicationProceedings of International Conference on Electric Machines, ICEM
Pages1-6
Number of pages6
Publication statusPublished - 2012
EventInternational Conference on Electric Machines, ICEM -
Duration: 2 Sept 20125 Sept 2012

Conference

ConferenceInternational Conference on Electric Machines, ICEM
Period2/09/125/09/12

Research Field

  • Not defined

Keywords

  • Condition monitoring
  • permanent magnet machines
  • rotor temperature
  • magnet temperature estimation
  • saturation effects.

Fingerprint

Dive into the research topics of 'Compensation of Speed Dependency in Sensorless Rotor Temperature Estimation for Permanent Magnet Synchronous Motor'. Together they form a unique fingerprint.

Cite this