Subdivided Error Correction Method for Photoelectric Axis Angular Displacement Encoder Based on Particle Swarm Optimization

作     者：Gao, Xu Li, Shuhang Ma, Qinglin 

作者机构：Changchun Univ Sci & Technol Sch Photoelect Engn Changchun 130022 Peoples R China 

出 版 物：《IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT》 (IEEE试验设备与测量杂志)

年 卷 期：2020年第69卷第10期

页      面：8372-8382页

核心收录：

学科分类：0808[工学-电气工程] 08[工学] 0804[工学-仪器科学与技术] 

基　　金：National Natural Science Foundation of China Outstanding Youth Fund of the Jilin Science and Technology Department, the "Thirteenth Five-Year" Science and Technology Project of the Department of Education of Jilin Province [JJKH20190544KJ] China Postdoctoral Science Foundation [2018M641778] 

主　　题：Gratings Signal processing algorithms Shafts Harmonic analysis Error correction Particle swarm optimization Displacement measurement Error correction grating fringe signal optimization algorithm photoelectric encoder subdivision precision 

摘      要：To improve the subdivision accuracy of the absolute photoelectric shaft encoder, a particle swarm optimization (PSO) algorithm is proposed for error correction. The error is generated by the subdivision of the grating Moire fringe signal, and a Moire fringe photoelectric signal data acquisition system is built based on a digital signal processor. First, based on the principle of measuring the angular displacement with grating diffraction interference, a mathematical model for the Moire fringe photoelectric signal is established, and then the influence of signal deviation on the subdivision error is noted. Second, the compensation model of the signal subdivision error is established. Then, the PSO algorithm is introduced, and the inertial weight and speed factor in the algorithm are optimized. Seven undetermined parameters in the signal model are identified using an optimization algorithm. Finally, according to the subdivision error compensation model, the data processing system can automatically correct the subdivision error caused by the direct current (dc) offset, orthogonality error, and sinusoidal error. Taking the 23-bit photoelectric encoder as the experimental object, the actual static subdivision error can be reduced by at least 0.30-0.46. In the encoder operation, if the correction preprocessing is performed on all the grating Moire signals, then the subdivision accuracy of the encoder can be improved, and its spatial adaptability and angular reliability will also be improved.