Reducing Calorimetric Loss Measurement Time for Electric Machines through Optimized Operating Point Sequencing
Year: 2025
Type: conference-paper
Abstract: Calorimetric loss measurements in high-efficiency electric machines are accurate but time-intensive due to lengthy stabilization times at each operating point. This work proposes an optimized sequencing method to minimize testing time. Using a hybrid thermal model, stabilization times between operating points are estimated to construct a distance matrix, and the sequence is optimized via a Traveling Salesman Problem, starting from an idle state. Applied to an Interior Permanent Magnet Synchronous Machine for thirteen operating points, this method reduces total testing time to 5.7 hours, a 23% improvement over random sequences (7.4 hours) and 16% over power-grouped sequences (6.8 hours) typically used in the literature. Results highlight that gradual temperature changes, particularly in the magnets, are crucial for minimizing stabilization times. Additionally, the optimized sequence follows a trend of increasing losses, suggesting that loss-based ordering could serve as a useful heuristic. This heuristic reduces run time to 6.2 hours, though still 8% longer than the optimized sequence. These findings demonstrate that the proposed methodology enhances the practicality of calorimetric loss measurements for future testing campaigns.
Institutions: Ghent University, University of Antwerp
Language: English
Subfield: Electrical and Electronic Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W4412021526