基于预测算法消除光电编码器抖动误差的研究

摘要

In order to reduce the count error of photoelectric encoder in highly-demanded real-time system such as CNC, which is caused by jitter, sequence forecast algorithm method is used based on GM (1,1) dynamic prediction model of grey system theory Modeling process and precision testing method have been elaborated after analysis the reason of miscount happening Finally, simulation researches and a improved GM(1,1) model are conducted Results show that this method can eliminate jitter error and improve the measurement precision effectively The size of the model sample is much smaller with a simply calculation model, the time-changed character of the system can be reflected dynamically. What' s more, the improved GM (1 ,1 ) model has better anti-jittering effect.%以减少数控机床等实时性要求高的机电系统内光电编码器因抖动引起的计数误差为研究目标,采用一种基于灰色理论新息GM(1,1)动态预测模型的数列预测算法.以错误计数发生的原因为基础阐述了该预测模型的建模过程和模型精度的检验方法.结合实例进行了仿真研究,并对普通GM(1,1)模型进行了改进.目前阶段的研究结果表明该方法能有效地消除光电编码器的抖动误差从而提高测量精度,并具有己知样本需求量小、计算简便、能动态地反映系统的时变性等优点,而且改进后的GM(1,1)模型具有更好的抗抖动效果.