Open AccessPower loss segregation in electrical machines through calorimetry and inverse thermal modelling
Author(s) -
Nair Devi Geetha,
Arkkio Antero,
Haavisto Ari
Publication year - 2020
Publication title -
iet electric power applications
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.815
H-Index - 97
eISSN - 1751-8679
pISSN - 1751-8660
DOI - 10.1049/iet-epa.2019.0217
Subject(s) - inverse , power loss , thermal , power (physics) , induction motor , inverse problem , control theory (sociology) , calorimetry , computer science , materials science , mathematics , engineering , thermodynamics , electrical engineering , voltage , mathematical analysis , physics , geometry , control (management) , artificial intelligence
This study combines calorimetric measurement of power loss of an induction motor with inverse thermal mapping of its temperatures to segregate the net power loss into its constituent components. Constrained least square fitting is the tool used to achieve the inverse mapping. The lumped thermal network of the motor serves as the forward thermal model, establishing the physical relationship between the motor's power losses and its resultant temperature rises. The objective of the iterative minimisation is to find the actual loss components that correspond to the temperatures measured, and which sum up to the total loss of the motor that is measured with a calorimeter. Data ranges and constraints are imposed to attain a reasonable, unique solution. Each loss component predicted by the inverse mapping inherently includes the contribution from stray load losses as it is fitted to the measured temperatures of the motor. Given the challenges, this inverse thermal routine achieves the loss segregation in a fairly good fashion.