Nonlinear error compensation algorithm for signal resolution of chromatic confocal measurements

• Published in: Measurement : journal of the International Measurement Confederation Vol. 226; p. 114091
• Main Authors: Dai, Jiacheng, Zeng, Wenhan, Zhong, Wenbin, Jiang, Xiangqian (Jane), Lu, Wenlong
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 28.02.2024
• Subjects: Chromatic confocal; Error compensation; Regression tree model; Chromatic confocal; Regression tree model; Error compensation
• ISSN: 0263-2241; 1873-412X
• DOI: 10.1016/j.measurement.2023.114091

•Nonlinear errors severely interfere with the accuracy of chromatic confocal measurements at the submicron scale.•Nonlinear error stems from the deviation between the wavelength-distance mapping and the polynomial model.•The proposed method can effectively record the nonlinear error in calibration and compensate it in measurement.•The results show that the method can achieve a 40 % accuracy improvement over a 4 mm range. The mechanisms of the nonlinear error, primarily stemming from the deviation of the polynomial model fitting the wavelength-distance mapping relationship in chromatic confocal measurement, have been elucidated. A compensation method is proposed for the nonlinear error, based on a regression tree model. Through calibration experiments using our chromatic confocal sensor, an impressive improvement of over 40 % in precision could be achieved by our proposed compensation method, which advances from ±1 μm to ±0.6 μm within a 4 mm measurement range.