| Calibration is an important way to improve the accuracy of sensors.Two different calibration methods,comparative calibration and self calibration,can be determined by the difference of the need of using other high-precision external benchmarks to obtain measuring errors.Due to the need of external high-precision reference and the influence of its accuracy,the calibration cost of comparative calibration is usually very high and it also has high difficulty in execution.Therefore,the problems of comparative calibration can be solved by making the sensor itself as a benchmark and calibrating sensor by itself.Self-calibration method with multiple reading heads and self-calibration with single reading head are universal for calibrating the displacement sensor,which have little difference in calibration accuracy generally.But for the linear time grating displacement sensor,using the self-calibration method of the single reading head requires higher uniform speed motion of the moving ruler in the calibration process.Accordingly,a self-calibration method based on the two reading heads is proposed.Two reading heads are used to construct a function of measuring values at a certain distance measurement,and then error function of the sensor can be calculated through the function.Finally and self-calibration can be carried out based on the frequency spectrum of the error function.A self-calibration system is built based on the self-calibration model established before,and the accuracy experiment of self-calibration system is carried out to prove the validity and feasibility of self-calibration method.The experimental results show that the self-calibration method can make use of the measured data of the sensor to realize the self-calibration.Moreover,after using this self-calibration method and the traditional calibration method to calibrate the sensor,the comparison error of both in comparison with the grating sensor are within ?2?m.Higher work efficiency and lower work cost can be achieved by adopting the method proposed,though the calibration effect of the method proposed is similar to that of the traditional calibration method with the help of the external benchmarks. |