Axial Flux Switched Reluctance Machine Modelling by a Quasi-3D Reluctance Network

Switched reluctance machines find applications in numerous sectors such as electric vehicles, mechanical presses, pumps, conveyors, compressors, fans, washing machines… and can replace classical induction machines in situations where operating conditions do not suit them. The axial flux version of these machines is characterized by its high-speed capability and high power density making it especially suited for traction applications. Design of electrical machines could be highly time consuming for structures having 3D main magnetic flux as the axial flux machines and especially when using 3D finite element method which is not suitable for the early stages of the design process. In this paper, an alternative quasi-3D reluctance network (Q3DRN) modelling method is proposed combining both an acceptable precision and a reduced computation time in one package, and it is applied to a 24-pole segmented rotor, 36-slot per stator topology of an axial-field double stator switched reluctance machine (AFDSSRM) fed with a 3-phase current. The proposed modeling approach is validated by comparing obtained results to the 3D finite elements ones for the magnetic flux density in the air gap and the electromagnetic torque in nonlinear permeability conditions. The proposed method is 20 times faster than 3D finite elements with an error not exceeding 8% on total torque.