[5D2] Advanced non-linear modelling and prognostics of electromechanical interactions in electric transmissions
A Zippo, F Pellicano and M Molaie
University of Modena and Reggio Emilia, Italy
Electrified drivetrains introduce unique dynamic challenges due to the interplay between motor-generated forces and mechanical transmission behaviour. The transition from internal combustion engines to electric motors requires a deeper understanding of these interactions to enhance system efficiency, reliability and operational lifespan. A crucial aspect of this process is the mutual influence between electromagnetic excitations and mechanical vibrations, which can lead to instability, increased noise and performance degradation if not properly managed. This research presents a novel non-linear dynamic framework for electric drive multi-stage gear systems, incorporating both electromagnetic and mechanical forces. The model captures key non-linear phenomena, including torque ripple, gear meshing variations and transmission errors, to evaluate their impact on system stability. A neural network-based method is integrated to predict the remaining useful life (RUL) of plastic gears, enhancing prognostic capabilities for electric transmissions. The study utilises finite element analysis through Transmission3D Calyx software to refine gear mesh stiffness estimations and improve the accuracy of dynamic simulations. The governing equations are numerically solved to assess system responses under varying operational conditions. Through time-domain simulations, spectral analysis and phase space representations, the effects of electromagnetic forces on drivetrain vibrations are thoroughly examined. Results indicate that the interaction between motor excitations and gear transmission dynamics significantly influences system behaviour, amplifying vibration levels and affecting long-term durability. By incorporating artificial intelligence-based prognostics with advanced non-linear modelling, this study contributes to optimising electric transmission performance while extending the service life of plastic gears. The findings emphasise the necessity of integrating both electromechanical interactions and predictive maintenance strategies in the development of next-generation electric powertrains.
