TBM导向系统中倾角仪测量误差补偿

发布时间：2018-06-24 09:12

本文选题：TBM倾角仪小波滤波边界效应 + 　；参考：《华中科技大学》2015年硕士论文

【摘要】：激光标靶导向系统用于对TBM(Tunnel Boring Machine)进行位姿测量和自动导向。在该系统中,倾角仪起着测量TBM滚角和俯仰角的重要作用。在硬岩掘进过程中,TBM的振动会给倾角仪测量引入很大的随机误差,使其精度不能满足系统要求。基于对倾角仪测量误差补偿和非平稳性随机信号处理方法的研究,本文提出了一种基于小波阈值滤波的方法,对倾角仪测量结果进行误差补偿。本文在小波分析理论的基础上,提出了一种结合信号拓展和Mallat算法中近似系数正交多项式边界延拓的小波阈值滤波方法,对倾角仪测量结果进行误差补偿。首先分析了小波阈值滤波边界效应产生的根本原因,据此提出利用信号延拓和近似系数延拓两种方法减小边界效应。通过MATLAB平台仿真,确定了信号正交多项式延拓和近似系数正交多项式延拓的具体参数,以及实时滤波窗口的长度。然后,结合不同的信号拓展方法和近似系数拓展方法,对多种模拟信号进行滤波仿真。通过仿真结果对比,确定了与信号特征相对应的滤波参数,得到了最终的综合滤波方法。最后,在模拟TBM振动条件下,完成了倾角仪测量实验。实验数据处理结果表明,本文提出的滤波方法比现场使用的均值滤波方法和传统小波滤波方法更有效,而且它能够保证误差补偿精度小于1mrad。
[Abstract]:The laser target guidance system is used to measure the position and orientation of TBM (Tunnel boring Machine). In this system, the tiltmeter plays an important role in measuring the roll angle and pitch angle of TBM. The vibration of TBM in the process of hard rock excavation will introduce a large random error to the dipmeter measurement, which makes the accuracy of TBM not meet the requirements of the system. Based on the research of measurement error compensation and non-stationary random signal processing method, a wavelet threshold filtering method is proposed to compensate the measurement results of the obliquity meter. Based on the theory of wavelet analysis, this paper presents a wavelet threshold filtering method based on signal expansion and boundary extension of approximate coefficients orthogonal polynomial in Mallat algorithm, which compensates the error of the measured results of the obliquity meter. Firstly, the fundamental cause of the boundary effect of wavelet threshold filtering is analyzed. Based on this, two methods, signal continuation and approximate coefficient continuation, are proposed to reduce the boundary effect. The parameters of signal orthogonal polynomial extension and approximate coefficient orthogonal polynomial continuation and the length of real-time filtering window are determined by MATLAB simulation. Then, different signal expansion methods and approximate coefficient expansion methods are used to filter various analog signals. By comparing the simulation results, the filtering parameters corresponding to the signal characteristics are determined, and the final integrated filtering method is obtained. Finally, under the condition of simulating the vibration of TBM, the measurement experiment of dipmeter is completed. Experimental data processing results show that the proposed filtering method is more effective than the mean filtering method and the traditional wavelet filtering method, and the error compensation accuracy is less than 1mrad.
【学位授予单位】：华中科技大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：U455.31

【相似文献】