Non-intrusive efficiency estimation of induction machines under various power supplies
Master Thesis
2013
Permanent link to this Item
Authors
Journal Title
Link to Journal
Journal ISSN
Volume Title
Publisher
Publisher
University of Cape Town
Department
License
Series
Abstract
Considering that 45% of the world's generated electricity is consumed by induction machines, determining an induction motors efficiency non-intrusively is of great importance in that it enables the machine to operate productively whilst ensuring that the energy consumed by the machine is utilized efficiently. International efficiency testing methods such as the IEEE 112-B can determine a motors efficiency accurately at the cost of hindering the machines productivity. Alternatively, various methods used to determine a machines efficiency in-situ do so at the cost of accuracy. This research proposes a method that determines an induction machines efficiency over a range of load conditions from tests conducted and centered around one thermally stable load point in the least intrusive manner possible. Coupled with vibration sensors used to determine a motor's speed, measured input voltages and currents are used to deduce a machine efficiency-load profile through the use of a modified evolutionary algorithm, the Non-Intrusive Efficiency Estimation using Population-Based Incremental Learning(NIEE-PBIL) algorithm. Five temporal load measurements are taken, centered around one thermally stable load point, to determine the machines efficiency profile from two equivalent circuit implementations; the Standard Circuit NIEE-PBIL and the Iron-Loss NIEE-PBIL.
Description
Keywords
Reference:
Gajjar, C. 2013. Non-intrusive efficiency estimation of induction machines under various power supplies. University of Cape Town.