Detection and Identification of Rotor Faults in Axial Flux Permanent Magnet Synchronous Motors Due to Manufacturing and Assembly Imperfections

In this paper, a method is proposed to detect and identify the types of rotor faults in single stator single rotor axial flux permanent magnet synchronous motors, due to manufacturing and assembly errors. Three types of faults are considered: A shift in the rotor geometry along the axial direction, a parallel shift of the rotor geometry axis with respect to the stator axis, and inclined rotor. The proposed method is based on using the motor voltages <inline-formula><tex-math notation="LaTeX">$(V_d$</tex-math></inline-formula> and <inline-formula><tex-math notation="LaTeX">$V_q)$</tex-math></inline-formula>, extracted from the controller, as features for detecting the type of the faults and estimating it's severity. The change of the point <inline-formula><tex-math notation="LaTeX">$(V_d,V_q)$</tex-math></inline-formula> in the <inline-formula><tex-math notation="LaTeX">$V_d$</tex-math></inline-formula>–<inline-formula><tex-math notation="LaTeX">$V_q$</tex-math></inline-formula> plane is used as a fault indicator. The shift direction is used to detect the fault's type, and the amount of the shift is used to estimate the severity. A quasi-3D computation is used to model the axial flux motor under healthy operation and the three types of faults. A three dimensional (3D) Finite Element Analysis and experimental tests are performed on a 6 slots, 8 poles axial flux motor to validate the quasi-3D computation. Finally, an algorithm is proposed to detect and discriminate between the different faults under different operating conditions.