Use of double correlation techniques for the improvement of rotation speed measurement based on electrostatic sensors

Abstract

Electrostatic sensing technology using correlation signal processing offers an approach to the measurement of rotational speed in the automatic control system of large generators and centrifugal machines. In this article, a double autocorrelation method was proposed to improve the accuracy and robustness of the measurement on the designed test rig. An electrostatic sensor was used to obtain signals from the rotor surface. Then, the rotational speed was determined from the period of rotational motion calculated from a double autocorrelation method. At the same time, experiments with sampling rates of 2ksps (kilo samples per second), 5ksps, 10ksps, 20ksps were carried out on a laboratory-scale test rig under a rotational speed range from 400 r min−1 to 4200 r min−1. The results show that the double autocorrelation method improves the accuracy and robustness. The measurement accuracy also improves with the sampling rate-the relative errors using 2ksps, 5ksps, 10ksps, and 20ksps are within 1.5%, 1%, 0.4%, and 0.3% respectively. The linearity of them is 1.47%, 0.61%, 0.28%, 0.17% correspondingly. The experiments also reveal that the measurement error has a tendency to increase with the rotational speed.