The work presented in this thesis aims at the modelling and optimisation of electrical machine for an automotive application.The first part shows the constraints required to electric/hybrid automotive specifications. A clustering method which allows to reduce evaluations number of the operating points is described. Next, an optimal pre-sizing of the machine is presented and designed in order to respect this optimal pre-sizing.In what follows an accurate and fast analytical electromagnetic modelling of the machine is performed. Well, the analytical modelling developed is related to a genetic algorithm. Two solutions of permanent magnet synchronous machines (PMSM) designed to automotive application are finally showed and analysed.
