학술논문

Lifetime estimation of type I random-wound electrical machines under active thermal cycling
Document Type
Conference
Source
2022 IEEE Electrical Insulation Conference (EIC) Electrical Insulation Conference (EIC), 2022 IEEE. :294-298 Jun, 2022
Subject
Engineered Materials, Dielectrics and Plasmas
Power, Energy and Industry Applications
Industries
Insulation
Thermal expansion
Accelerated aging
Thermomechanical processes
Thermal loading
Reliability
Ageing
Motorette
Thermal cycling
Heat losses
PDIV
Language
ISSN
2576-6791
Abstract
Safety-critical industries such as automotive, or aircraft industries need reliable electrical machines, with predictable lifetime. Estimating the lifetime of electrical machines can be done by running accelerated ageing tests. Such test involves the application of one or more ageing stresses above their nominal in-service level. For realistic lifetime assessment, the standards invokes superimposing stress factors including thermal, electric, ambient and mechanical constraints (known as TEAM, that contribute to ageing). However, thermally induced mechanical stress is seldom accounted for. Thermo-mechanical stress arises due to the thermal gradients and the thermal expansion mismatch between neighbouring materials. Indeed, thermal cycles such as on/off duty cycles, load or speed changes induce repeated thermo-mechanical stress on the insulation components, causing fatigue.In this work, a thermal cycling campaign has been implemented on motorettes representative of type I electric motor insulation. PDIV decrease was observed at the phase-to-phase and phase-to-ground insulation throughout ageing. A failure criterion based on PDIV was chosen to enable a Weibull lifetime estimation for each. Eventually, the times to failure were used to calculate the parameters of a Coffin-Manson fatigue law.