Numerical and experimental modal analysis applied to e-axle systems

The project addresses the definition of a methodology for the modal characterization of e-axle systems of electric vehicles. The project is developed in collaboration with SKF industrie S.p.a, automotive unit, located in Airasca (TO), in particular with the RTD engineering team. The studied e-axle is composed by the electric motor (rotor and stator), the power electronics (inverter) and the gearbox (reducer and differential). The whole study can be split into two workflows: The first part is the experimental modal analysis. Experimental tests were conducted by means of the modal hammer testing with the e-axle in a free-free boundary condition. Input and output signals have been acquired, then processed and analysed in Matlab in order to obtain the frequency response functions (FRF's). The quality and repeatability of the measurements are evaluated through the coherence. FRF's have been used to extract the modal parameters: natural frequencies, damping ratios and mode shapes. Two modal extraction methods have been used to robustly extract the aforementioned structure's properties: Least squares complex exponential (LCSE) method and Rational fraction polynomial in the Z-domain. The outcome of the different tests have been analysed and compared by means of the modal assurance criterion (MAC). The parallel workflow is devoted to the construction of a suitable numerical model of the e-axle. The scanned files of the different parts of the e-axle represent the starting point for the assembly generation. Geometry is imported in ANSYS Mechanical in order to build the FEM model. The numerical modal analysis is performed to find natural frequencies and modes. The final stage of the work is related to the procedure of validation and alignment of the numerical results with respect to the experimental data.