Performance Enhancement for Systems With Disturbances and Measurement Noise by Modified Equivalent-Input-Disturbance Approach

This article presents a modified equivalent-input-disturbance (MEID) approach for enhancing the control performance of a system with disturbances and measurement noise. In this approach, the measured output of the system is filtered by a newly added low-pass filter to reduce the influence of the measurement noise. Regarding the low-pass filter and the system as a new plant, an MEID estimator is devised for the new plant to estimate and compensate for disturbances. Since the newly added filter reduces the effect of measurement noise, an all-pass filter is designed in the MEID estimator instead of a low-pass filter in the conventional EID estimator to ensure causality. The all-pass filter causes a smaller phase lag between the estimated disturbance and the compensation signal than that caused by the low-pass filter. Thus, the MEID approach improves the disturbance-rejection performance compared to the conventional EID approach. Experimental results on a permanent-magnet synchronous motor confirm the validity and superiority of the MEID approach.