Sensitivity Analysis and Validation of the Intelligent Assembly Process for Permanent Magnet Rotors with the Balancing Grade G 2.5

Abstract

An increased range of electric vehicles not only sets high demand to the chosen concept for energy storage but also implies that electric motors with high power-to-weight ratio need to be employed. A possible approach for weight reduction in a permanent magnet synchronous machine is an intelligent assembly of the rotor components followed by a positive balancing process. This approach allows to predict and minimize the rotor unbalance, resulting in a reduction in size of balancing discs that can account for up to 10% of the rotor mass. A higher potential in reduction of the rotor mass lies in the ability to dispense entirely with balancing discs and the subsequent balancing process. This paper analyses whether a balancing grade of G 2.5 can be achieved through intelligent assembly. In a first step, the required measuring and mounting steps are presented. Possible measuring and mounting uncertainties are identified and quantified through literature, simulation and experimental results, allowing a sensitivity analysis of the entire process chain. The results of this analysis are used to identify the measuring and mounting steps with the highest influence on the deviation between an expected and measured rotor unbalance. For magnet assembly, a good model validity is achieved. Errors in perpendicularity between lamination stacks and shafts are found to have a high influence on the reachable rotor unbalance and therefore offer room for further improvement of the approach.