Study on the relationship between assembly interference and deformation influence of axle hole assembly of electric vehicles motor core die based on BNPP

This study investigates the position deviation in the assembly of iron core dies for new energy electric motors due to interference fit. A mapping model is established using a back propagation neural network (BPNN) to correlate the ideal die axial hole interference fit clearance with position deviation. Utilizing the joint simulation of Abaqus and Isight, a comprehensive set of geometric models with varying interference clearances is developed, generating data on center deviations for different sizes and clearances. A BPNN is trained with the template hole diameter and insert outer diameter as inputs, and the x-direction and y-direction center deviations of the insert cylindrical axis as outputs. This model is then applied to predict the deformation effects of the interference fit on assembly. The results show that this method can accurately and efficiently predict the deformation in ideal axial hole interference fit assemblies.