Loss Minimization of an Electrical Vehicle Machine Considering Its Control and Iron Losses

In this paper, optimization of the control of an electrical machine allowing a minimization of its total losses is described. It is based on the use of an iron loss model [loss surface (LS) model] coupled to the electromagnetic finite-element simulations of the machine. The LS model is a scalar and dynamic hysteresis model developed many years ago at G2Elab and tested for iron loss prediction in several cases of electric machines. It is first characterized and improved in this paper for M330-35A SiFe sheets. Then, it is associated with a finite-element analysis to compute, in a post-processor mode, the local and global magnetic losses in the machine. For electric vehicle application, the whole torque-speed variation should be investigated. To do that, a quick response surface is constructed from a small number of simulations and the iron loss is determined. Then, a suitable optimization algorithm is developed. This approach is then illustrated by a case study and compared with classical optimization in which only the copper losses in conductors are considered. Gains of up to 50% reduction in the total losses of the machine in certain operating areas are observed.