Due to the recent growth of the electric and hybrid traction market, the automotive industry is asking electric machines manufacturers to ever increase the power density, in order to have lightweight, more compact and less expensive powertrains. As a result, there is an increasing trend towards designing electrical machines for higher fundamental frequencies (that can reach above 1 kHz) and higher electric and magnetic loadings which result in higher losses. Therefore, whether the need is to increase the power of the machine keeping the same size, or the goal is to reduce the machine's size at the same power, the result is that the surface available to extract losses becomes less and less.
The drive for higher power density, therefore, will produce two main consequences. The first one, the easier to spot, is that research will be directed towards more effective ways to extract heat from electric machines. The second one consists in the fact that - with the machine performances approaching the limits - the electromagnetic design and the thermal design will become tightly interconnected.
The proposed research aims at addressing both the previously described issues with specific applications to Axial Flux Permanent Magnet (AFPM) Synchronous Machines.
Using a water cooled AFPM machine as a baseline and keeping the same active parts, within this project a modified prototype will be built equipped with the proposed direct oil-cooling for end windings system. It is expected that the new cooling system will be more efficient in extracting heat from the machine, allowing a significant increase in power density. The fully modified AFPM machine prototype will be tested experimentally using image-based velocimetry techniques and modeled using CFD analysis.
The proposed research brings in an intrinsic innovation content, since it proposes a technical solution with direct oil cooling of end windings for a Torus-type AFPM machine, that has never been presented and implemented before. By allowing a more efficient heat extraction, it is expected that it will be possible to increase the nominal current of the machine (with respect to the water cooled counterpart) and thus increase the mechanical power for the same size. The following figures can help evaluate the potential advancement over the state of the art. The ¿Electrical and Electronics Technical Team Roadmap¿, published in 2017 by the U.S. DRIVE Partnership among the U.S. Department of Energy and several Automotive OEM and energy companies, defines power density targets in the order of 5.7 kW/l by 2020 and 50 kW/l by 2025 [1]. The AFPM water-cooled prototype already shows a remarkable power density of approximately 9.4 kW/l. Some preliminary calculations indicate that the oil cooled prototype could almost double the power density, reaching values in the order of 18 kW/l. If these results will be confirmed by the proposed research, it will indeed prove to be a big step forward towards the 2025 target.
Another important advancement over the state of the art is expected in relation to the design procedure for electrical machines in general and for AFPM machines in particular. As it has already been pointed out, the drive for power density increase is changing the design of an electric machine into an intrinsic multiphysics problem. Therefore, the traditional procedures which analyze one physics at a time will need to be replaced by new ones. It is expected that this project will enable a step towards the forecast multiphysics approach, by allowing an integrated analysis of the electromagnetic and the thermal problems. Of course, this will be just a step along a path which is expected to include also the mechanical problem into the picture, together with all the mutual relationships between different physics.
[1] U.S. DRIVE partnership, "Electrical and electronics technical team roadmap," Accessed: Jun. 21, 2019. [Online]. Available: https://www.energy.gov/sites/prod/files/2017/11/f39/EETT%20Roadmap%2010-...