DocumentCode
1027889
Title
Parameter estimation for induction machines based on sensitivity analysis
Author
Ansuj, Somchai ; Shokooh, Farrokh ; Schinzinger, Roland
Author_Institution
Federal Univ. of Santa Maria, Brazil
Volume
25
Issue
6
fYear
1989
Firstpage
1035
Lastpage
1040
Abstract
Induction machine parameters supplied by manufacturers are usually sufficient for short-circuit analysis only. System studies that involve transient simulation of machines require additional parameters that are not readily available, but nonetheless are essential for an accurate modeling of the machines. A mathematical method for estimating the equivalent circuit parameters of induction machines from the most readily available performance characteristics is presented. In this method, machine equations are used to estimate the parameters and sensitivity analyses are then performed with respect to the circuit parameters to match the given performance characteristics. An example is included to illustrate how well the performance of the estimated model matches that of the actual machine. The example is a three-phase, 1000 HP, 4.16 kV, 60 Hz induction motor. The example indicates that, with a restricted set of input data, it is possible to find the elements of a machine model that duplicates the given performance to a high degree of accuracy. Without sensitivity analysis, accuracies on the order of 10% are obtained, but after sensitivity analysis an accuracy of 0.1% or better can be achieved
Keywords
equivalent circuits; induction motors; parameter estimation; sensitivity analysis; transients; 1000 hp; 4.16 kV; 60 Hz; equivalent circuit parameters; induction machines; mathematical method; parameter estimation; sensitivity analysis; short-circuit analysis; three-phase induction motor; transient simulation; Bars; Equivalent circuits; Induction machines; Manufacturing; Parameter estimation; Rotors; Sensitivity analysis; Stators; Steady-state; Torque;
fLanguage
English
Journal_Title
Industry Applications, IEEE Transactions on
Publisher
ieee
ISSN
0093-9994
Type
jour
DOI
10.1109/28.44239
Filename
44239
Link To Document