基于PSO-BP神经网络光电编码器误差补偿研究

发布时间：2018-06-15 01:06

本文选题：绝对式光电编码器 + BP神经网络　；参考：《传感技术学报》2017年08期

【摘要】：考虑到高精度绝对式光电编码器应用广泛,其角度测量精度对整个系统精度影响较大,但由于角度传感器生产安装过程中产生的误差等原因,使得传感器在实际应用中存在一定的误差。而使用传统误差补偿方法难以得到较好的补偿效果,本文使用一种基于PSO的BP神经网络作为角度传感器误差补偿系统的算法。通过实验验证,该种算法能够对角度传感器误差进行较好的补偿,与补偿前相比,其标准偏差提高了12.5倍,最大误差和平均误差降低到9.6%和8.5%,提高了传感器检测精度。与使用了基于传统BP神经网络和基于多项式拟合算法的误差补偿系统进行对比实验,结果表明,其补偿效果亦优于这两种算法。
[Abstract]:Considering the wide application of high precision absolute photoelectric encoder, its angle measurement accuracy has a great influence on the accuracy of the whole system, but due to the errors in the production and installation of angle sensor, The sensor has some errors in practical application. However, it is difficult to get a better compensation effect by using the traditional error compensation method. In this paper, a PSO-based BP neural network is used as the error compensation system of the angle sensor. The experimental results show that the algorithm can compensate the error of angle sensor better. Compared with before compensation, the standard deviation is increased 12.5 times, the maximum error and average error are reduced to 9.6% and 8.5%, and the precision of sensor detection is improved. Compared with the error compensation system based on traditional BP neural network and polynomial fitting algorithm, the results show that the compensation effect is better than these two algorithms.
【作者单位】：辽宁工程技术大学机械工程学院;
【基金】：国家自然基金项目(51404132)
【分类号】：TN762;TP183

【相似文献】