基于自相关特性的经验模态分解微机械陀螺去噪方法  

Micro mechanical gyro denoising method based on EMD autocorrelation

作　　者：刘晓光[1,2] 郝沙沙 王光磊[1,2] 娄存广[1,2] 刘秀玲[1,2] LIU Xiao-guang HAO Sha-sha WANG Guang-lei LOU Cun-guang LIU Xiu-ling(College of Electronic and Information Engineering, Hebei University, Baoding 071002, China Key Laboratory of Digital Medical Engineering of Hebei Province, Baoding 071002, China)

机构地区：[1]河北大学电子信息工程学院,保定071002 [2]河北省数字医疗工程重点实验室,保定071002

出　　处：《中国惯性技术学报》2016年第4期537-541,共5页Journal of Chinese Inertial Technology

基　　金：河北省教育厅项目(QN2015071);保定市科技局项目(15ZG010)

摘　　要：陀螺的噪声是影响组合导航系统精度的重要因素之一。以农机多传感器组合导航系统为研究背景,在分析经验模态分解去噪和小波去噪的基础上,提出了一种基于自相关特性的经验模态分解去噪方法。该方法根据本证模态函数分量的自相关函数特性,提出了一种含噪本证模态函数筛选策略。该方法能够自适应地确定主要含噪的本证模态函数分量,避免了需要人为确定的不足;同时,结合改进小波阈值去噪的优势,避免了将混叠在噪声中的有效信号完全消除,使其具有一定的自适应性。为了验证方法的有效性,利用农机组合导航系统中微机械陀螺的实际输出数据,分别采用改进阈值小波去噪方法、经验模态分解去噪和改进的经验模态分解去噪方法进行了对比试验。结果表明,改进经验模态分解去噪方法的效果要优于前者,在一定程度上能够改善农机多传感器组合导航系统的定位精度。The gyro noises could significantly influence the precision of an integrated navigation system. Based on the analysis of conventional EMD and wavelet denoising, a denoising method based on empirical mode decomposition（EMD） autocorrelation is proposed by using multi-sensor integrated navigation system for agricultural machinery. The method proposes an intrinsic mode function（IMF） screening strategy based on autocorrelation function characteristics of the IMF, and can adaptively determine the main noise components of IMF to avoid the human intervention. Further more, the method combines the advantage of wavelet threshold denoising, avoiding the complete elimination of useful signals in the noises. In order to verify the validity of the method, a contrast test is made by using the improved wavelet threshold denoising, EMD denoising and the improved EMD denoising method, respectively, based on the actual output data of micro mechanical gyro in the multi-sensor integrated navigation system for agricultural machinery. Test results show that the improved EMD method is superior to the former, and it can, to some extent, improve the positioning accuracy of the multi-sensor integrated navigation system in agricultural machinery.

关 键 词：经验模态分解 微机械陀螺 去噪 数据处理 

分 类 号：U666.1[交通运输工程—船舶及航道工程]