The harmonic error component under dynamic measurement is one of the main reasons that restricts the application of high precision and high resolution time-grating angular displacement sensor in the field of dynamic measurement. To address the problem of harmonic suppression in time-grating angular displacement sensor under dynamic measurement, the system model of time-grating angular displacement sensor is firstly briefly described. Secondly, the dynamic error mathematical model of the time-grating angular displacement sensor is formulated. Then, the mechanism of dynamic error of the sensor is explained, and the basic principle of adaptive Kalman filter (AKF) is elaborated the dynamic error suppression model of time-grating angular displacement sensor adapted to Kalman filter. The simulation results show that the dynamic error of the time-grating angular displacement sensor is reduced by about 70% after the adaptive Kalman filter under the condition of constant speed and variable speed, and with the increase of the sensor speed, the more obvious the suppression effect of harmonic error is. In the experimental application, the filtering algorithm has a good real-time prediction of the measured value of the time-grating angular displacement sensor, the sensor can run more quickly and stably, and the measurement error is reduced by about 80% at the speed of 100 r/ min. The results show that the adaptive Kalman filter has a significant role in suppressing the dynamic harmonic error of the time-grating angular displacement sensor, and can greatly improve the dynamic measurement accuracy of the sensor.