Modeling and Compensation of Eccentricity Errors for Accurate Speed Acquisition in Nonstationary Conditions

Instantaneous angular speed (IAS) signals are used in motor control and fault detection. However, encoders inevitably suffer from eccentricity errors during manufacturing and installation, which reduces angle measurement accuracy and leads to significant noise in the IAS signal. Existing eccentricity error compensation methods have limitations such as low efficiency and application implementation. On the other hand, the eccentricity error coupled with the engraving shape error will cause the double helix structure in the amplitude modulation function of IAS signal. An eccentricity error model (considering the engraved shape error) as well as an error compensation technique are proposed for improving the goniometric accuracy and eliminating the noise in the IAS signals caused by the coupling of the eccentricity and engraving shape errors. The eccentricity error model and the compensation method are verified by simulation analysis, experiments on the rotary vector reducer test rig and motor test rig. Experimental results show that 71.6% of the goniometric error has been compensated, the noise in the IAS signal has been significantly eliminated and the vibration performance of the motor is improved at high speeds.