Electromechanics

Optimization of Electric Machines: Optimization of machines for given specifications has noticeable implications across various industries and applications such as automotive and aeronautics sectors. By optimizing the geometry, materials, and winding configurations, machines can meet specifications and requirements using a multi-physics model (electromagnetic, mechanical, and thermal). Multi-objective optimization is often performed to obtain a Pareto Front representing a balance between different performances which can be conflicting such as efficiency and power density. Although time consuming, large-scale optimization using finite element models can be performed with powerful computers at Aalto University allowing the exploration of vast solutions in record time. Research on the optimization can make machines competitive and sustainable alternatives for green transports. 

High Frequency Losses in Windings: There are 4 sources of losses in windings: Joule Effect, Skin Effect (eddy currents induced by conductor’s own magnetic field), Proximity Effect (eddy currents induced by external magnet field), and Circulating Currents (uneven distribution of current between parallel strands). Traditionally, only the Joule Effect is included in the design process while high-frequency effects are considered in post processing. That is done by selecting the suitable configuration of windings and dimensions of conductors. However, at high speed, it is necessary to include high-frequency losses (commonly called AC Losses) in the design process and consider an adequate cooling technique. Although AC losses can be low (e.g., <10% of total losses), they can strongly affect the machine's cooling design. This becomes inevitable with the rise of PMSM as magnets strongly contribute to increasing losses due to proximity effect and circulating currents, especially with the increase of rotational speed of machines used in Electric Vehicles.

Read more: How to model high frequency losses in windings?

High-Speed Machines: High-speed machines are highly desirable in applications with space-constrained environments. Various high-frequency phenomena manifest in different components of the machine at various aspects. Electromagnetic challenges include eddy currents in the core and the windings. Thermal challenges include high losses density, necessitating innovative cooling techniques, and high frequency phenomena in the airgap. Mechanical challenges include increased stresses in the rotor and vibrations. Most of these phenomena can be finely modeled and included in the design and optimization process for given application.

Read more:

Why High-Speed Machines?High-Speed Machines: Economic Impact

Transport Electrification: Electric machines play a crucial role in propelling various electrified transportation systems, including automotive, aerospace, rail, and marine applications. Machines can be designed to meet various specifications dictated by industries, such as fault tolerance, maximal power, torque ripple, cogging torque, size, cost, and others. These specifications are essential factors considered during the design and optimization process. Two primary considerations for industries are enhancing efficiency and minimizing size, which can be achieved by proposing the appropriate machine type, exploring new materials, and performing optimizations.

Read more:

Why Electric Transports?

Axially Laminated Rotors: Designing high-speed axial-laminated synchronous reluctance machines poses significant challenges due to their intricate nature. This research delves into the intricate domains of electromagnetic and mechanical design methodologies while also considering cooling techniques with CFD analysis integration. Noteworthy focus is dedicated to comprehensive 3D analysis, precise rotor loss approximation methods, and the influence of electrical machine malfunctions and frequency converter operation on losses. 3D and 2D simulations are performed using supercomputers at Aalto University, employing cluster computing for efficient data processing and analysis. This work paves the way for advancements in high-speed synchronous reluctance machine technology.

Inverse Modeling of Core Losses: Accurate prediction and measurement of losses are vital for evaluating the machine's efficiency, temperature distribution, and cooling requirements. This research introduces a novel technique based on inverse modeling to estimate core losses and localized losses in the core regions of an electrical machine. The approach involves using short-term transient temperature measurements from the machine's core and validating them through a forward model. Two primary measurement methods were employed: thermal sensors embedded in a printed circuit board within the stator core, and surface thermographic measurements with an infrared camera. The results demonstrate the effectiveness of the inverse modeling technique in predicting core losses based on short-time transient temperature rise measurements.

Topology Optimization of Machines: The topology optimization is the design optimization to automatically solve the problem of placing material in the design domain to achieve best performances. In order to address the design requirements for high electromagnetic output torque, lightweight, and mechanical stiffness and strength, this research develops a topology optimization approach for the rotor design of synchronous reluctance machines (SynRMs). Considering the high dimensional constraints caused by the multi-physics performances, the augmented Lagrangian method based optimization framework is developed to solve the minimization problem. Effectiveness of the proposed method is verified by performing analysis and simulations of the topology optimization of SynRMs. Performances of the optimized structure are also verified with experiments on the prototype.

Magnetically Levitated Rotors: The control segment of a six-phase induction motor is studied considering various windings configurations. The machine is modeled using analytical model, magnetic equivalent circuit (MEC), and finite element model (FEM). Different aspects are studied: optimization of electromagnetic performances (torque and forces in the x-y directions), vibration modeling, and mitigation of torque and force ripples as well as vibrations. Models are validated by performing experiments on the induction machine prototype.

Condition Monitoring of Machines: Condition monitoring of electrical machines has witnessed significant advancements due to the deployment of data-driven machine learning models, addressing the increasing demand for reliable and efficient operation. This research project aims to explore the synergistic integration of machine learning and innovative data augmentation methods to enhance the accuracy of condition monitoring in electrical machines. The main objective is to develop precise and efficient data augmentation techniques by utilizing a reduced amount of measurement data and incorporating simulation data, thereby generating a large number of synthetic data. The ultimate goal is to develop novel data augmentation methods capable of replicating measured data by minimizing the deviation between measured and simulated data caused by noise and uncertainty.

Magneto-Mechanical Coupling: Thermodynamic couplings and in particular magneto-mechanical coupling in electromagnetic devices allows developing non-destructive testing methods. The model of magneto-elastic behaviour are proposed based on the writing of a Gibbs free energy, the terms of which have been determined using the theory of invariants. This project is focused on the formulation of the magneto-plastic coupling using the previous techniques and conducting an experimental campaign. This project aims also to describe the influence of plasticity on the performances of electrical steel sheets, which are subjected to multi-axial loading.