Fast torque estimation of in-wheel parallel flux switching machines for hybrid trucks.

Purpose – Transient torque calculations of the parallel flux switching machines, both cogging and electromagnetic, require a long simulation time for transient analyses. This paper seeks to present an optimization method for the accurate but time consuming transient models. Design/methodology/approach – A superposition principle is used to optimize the simulation time of the machine model. Finite element method (FEM) is chosen as the example machine model, since it is widely used among researchers for its accuracy. The machine geometry is simplified by reducing the number of rotor teeth, because these parts are re-meshed with each transient step. Torque results are compared to the full machine model to find the best representation. Findings – Among compared simplified machine geometries, the two teeth model gives the most accurate results. Research limitations/implications – The superposition method requires a modelling method such as FEM. The method offers a geometrical simplification of the machine, not a complete model. Practical implications – Parallel flux switching machines should be considered as promising candidates for hybrid and electrical truck applications due to their high power density. For these kind of applications, a fast torque estimation tool helps greatly in investigating noise related mechanical problems, which have a direct effect in passenger comfort. Originality/value – Whereas researchers in this area mainly focus on accurate but time-consuming modeling of this nonlinear machine, this research shows an optimization of these methods to speed-up them. The proposed optimization method can be integrated with any analytical or numerical machine model. [ABSTRACT FROM AUTHOR]