Error Compensation Design for Optical Encoders

  • 蕭 侑晟

Student thesis: Master's Thesis

Abstract

For improving the tolerance of grating miniature optical encoders and meeting the requirement of robots CNC machines and various equipment needs ultra-high accuracy to the environmental noises in real-time this study uses three methods to reduce the measurement error caused by dirt vibration and component assemble misalignment for grating miniature optical encoders The main ideas behind these methods are using the measured signals which contain phase errors to remove the noises and recover original signals based on the parameters’ solution of a nonlinear system error model formulated by the relationship between input signals and output signals Based on system model three methods can be briefly expressed as: 1 Correcting phase errors via inversion method with the collecting input raw data 2 Using FFT method to calculate spectrum of the input signals a phase difference can be obtained for correcting the phase error and 3 Calculating the geometric relationship of Lissajous figure of two input signals based on Pascal’s theorem to search the optimal parameters for the nonlinear system error model to reduce the phase error effective For verifying the compensation performances these three methods are simulated in the well-known software platform: Matlab first The method based on Pascal’s theorem is selected for the realistic implementation due to its simple computational ability From the simulation results these three methods reveal almost the same compensation performances However the inversion method and FFT method suffer the consequence of computational burden due to their complicated process in practical Finally the Pascal’s theorem based method is realized practically
Date of Award2016 Sep 2
Original languageEnglish
SupervisorYung-Yu Chen (Supervisor)

Cite this

'