摘要
光纤传感器测量表面形貌系统的精度会受系统中噪声的影响.为了提高系统的精度和可靠性,提出了一种基于关系矩阵的统计加权数据融合算法.将该算法和基于卡尔曼滤波的均值融合算法应用于表面形貌测量数据的处理中,并对融合后的数据进行了对比分析.结果表明:基于关系矩阵的统计加权的数据融合算法对固定测量点的标准差为0.040 1,变异系数为0.084;基于卡尔曼滤波的均值融合算法对固定测量点的标准差为0.034 1,变异系数为0.073.基于卡尔曼滤波的均值融合算法比基于关系矩阵的统计加权数据融合算法能更为精准有效地还原表面形貌.
The accuracy of surface topography measurement system based on optical fiber sensor will be affected by noise in the system.In order to improve the accuracy and reliability of the system,a statistical weighted data fusion algorithm based on relational matrix is proposed.This algorithm and the mean fusion algorithm based on Kalman filter were applied to the processing of surface topography measurement data,and the fused data were compared and analyzed.The results show the standard deviation and variation coefficient of the statistically weighted data fusion algorithm based on relational matrix are0.040 1 and 0.084 for fixed measurement points,and the standard deviation and variation coefficient of the mean fusion algorithm based on Kalman filter are 0.034 1 and 0.073 for fixed measurement points.Compared with the statistical weighted data fusion algorithm based on relational matrix,the Kalman filter-based mean fusion algorithm can restore surface topography more accurately and effective.
作者
杨瑞峰
杨睿
郭晨霞
吴耀
YANG Rui-feng;YANG Rui;GUO Chen-xia;WU Yao(School of Instrument and Electronics,North University of China,Taiyuan 030051,China;Automatic Test Equipment and System Engineering Research Center of Shanxi Province,Taiyuan 030051,China)
出处
《中北大学学报（自然科学版）》
CAS
2020年第1期91-96,共6页
Journal of North University of China(Natural Science Edition)
基金
山西省科技攻关资助项目（201703D121028-2)
关键词
关系矩阵
统计加权
卡尔曼滤波
数据融合
表面形貌
relationship matrix
statistical weighting
Kalman filter
data fusion
surface topography
