Abstract
With the ongoing advancement of transportation electrification, electric machines (EMs) are increasingly deployed in safety-critical applications, particularly within the aerospace sector, where high power density and operational reliability are equally essential design requirements. The drive toward higher power density, however, subjects electrical insulation systems (EIS) to additional stressors, including thermal overload, reduced air pressure, and pronounced thermo-mechanical effects. Current IEC standards incorporate thermal lifetime as a key metric for EM reliability assessment, typically evaluated under multiple levels of constant temperature stress. In contrast, electric machines operating in real transportation scenarios are predominantly exposed to cyclic thermal loading, the underlying degradation mechanisms of which remain insufficiently understood.These knowledge gaps give rise to several critical challenges: (a) quantifying the additional lifetime consumption induced by cyclic thermal stress; (b) characterizing the resulting lifetime reduction mechanisms; and (c) developing a lifetime model capable of accurately fitting and extrapolating temperature-cycling behavior under complex mission profiles. Beyond cyclic thermal loading, the influence of low air density environments on insulation reliability has yet to be fully resolved. Furthermore, reliability-oriented design strategies derived from the proposed lifetime model must be validated and demonstrated through practical electric machine case studies.
To address these challenges, this Thesis advocates the development of novel physics-of-failure (PoF)–based methodologies for evaluating lifetime consumption and degradation of EIS in low-voltage electric machines. The objective is to embed reliability considerations into the early design phase, thereby reducing dependence on conservative safety margins and avoiding unnecessary over-engineering. The central focus of this work is thermal stress, which is identified as the dominant aging mechanism in low-voltage EIS, followed by a critical examination of the shortcomings of conventional lifetime models. A comprehensive review of the state of the art in low voltage motor insulation systems is first presented, including a clear distinction between Type I insulation for low-voltage random-wound windings and Type II insulation for medium- and high-voltage form-wound windings. Within this framework, TEAM stresses, namely thermal, electrical, ambient, and mechanical stresses, which are identified as the principal contributors to insulation degradation. Building upon this foundation, a series of targeted experimental investigations is conducted using stator motorette specimens that replicate the electromagnetic and thermal characteristics of permanent-magnet synchronous machines. Breakdown voltage is adopted as the end-of-life criterion to quantify the effects of constant temperature exposure, thermal cycling with varying mean temperatures, amplitudes, and periods, as well as low air-pressure conditions on EIS lifetime. The results reveal additional lifetime degradation induced by thermo-mechanical stress and demonstrate a nonlinear reduction in insulation lifetime under reduced pressure environments.
To improve prediction accuracy, the Thesis develops the Arrhenius–Miner Basquin model, which integrates thermal-chemical degradation (via the Arrhenius Miner framework) and thermomechanical fatigue (via a temperature-corrected Basquin-type stress–life formulation). Experimental validation shows this model reduces prediction errors to 4–19% compared to Arrhenius-only approaches, and a phenomenological form is derived to link lifetime directly to measurable parameters (average temperature, fluctuation, cycle period) for engineering usability. The model is further applied to practical scenarios: offline lifetime prediction for both aerospace and automotive EMs; online monitoring; and preventive maintenance.
This work enables EM designers to prioritize reliability from the start, shortens certification timelines, and avoids over-engineering. It provides key theoretical and technical support for EIS design, lifetime assessment, and reliability assurance of low voltage EMs in aerospace and transportation, addressing the gap between modern EM operating demands and traditional assessment methods.
| Date of Award | 15 Jul 2026 |
|---|---|
| Original language | English |
| Awarding Institution |
|
| Supervisor | Weiduo Zhao (Supervisor) & Salman Ijaz (Supervisor) |