基于粒子滤波及其改进算法的磁悬浮陀螺数据处理研究

发布时间：2019-05-13 08:00

【摘要】：随着国民经济的飞速发展,各类关乎国计民生的地下工程越来越多,如高铁地铁隧道、输水隧洞、矿山巷道等。这些隧洞的长度越来越长,贯通精度要求也越来越高。陀螺全站仪由于具有全天时、全天候自主寻北定向能力,使得其具有独特优势应用于诸多大型工程中。GAT磁悬浮陀螺全站仪作为众多陀螺定向测量仪器的一种,因其独特的磁悬浮支承体系、力矩反馈闭环测量技术构架及寻北过程数据实时下载与事后滤波处理功能,使得该系统优势彰显并脱颖而出。本文主要针对GAT磁悬浮陀螺全站仪定向测量所采集到的转子、定子电流数据优化处理作一系列研究工作,在充分分析定转子电流数据基本特征基础之上,采用粒子滤波及其改进算法对定转子电流数据进行滤波处理研究。主要内容及创新点如下:1.讨论分析粒子滤波原理、算法及优劣。粒子滤波算法是一种基于动态系统的空间模型对非线性系统有着较好滤波效果的算法,但其在具有优越性的同时也存在着自身样本粒子退化和样本粒子多样性匮乏的问题,为了规避其劣势之处,针对转子电流数据的相关性提出了基于观测路径相似性的粒子滤波算法。2.在分析研究陀螺数据误差源基础之上,明确了定转子电流的数据误差主要来自于仪器系统误差和数据采集过程中周围环境对其影响的偶然误差,并针对影响过程提出了一系列措施以降低各种误差混入采集的数据中,从而保证数据的质量。3.通过对定转子电流数据的特征、粗差探测和工程实例分析研究,提出了采用基于外部方位先验信息对陀螺数据进行预处理方法,研究结果表明:此方法能有效提高数据的稳定性和最终计算成果的可靠度。4.通过比对标准粒子滤波算法、自适应卡尔曼滤波算法和基于转子电流相关性的特征提出的基于观测路径相似性粒子滤波改进算法,通过在转子电流数据中滤波应用,表明该改进算法可更好的改善优化数据质量。论文研究成果通过实例验证,对优化陀螺数据质量,改善和提高陀螺精度具有重要意义。
[Abstract]:With the rapid development of national economy, there are more and more underground projects related to national economy and people's livelihood, such as high-speed railway subway tunnel, water transportation tunnel, mine tunnel and so on. The length of these tunnels is longer and longer, and the accuracy of these tunnels is higher and higher. The gyro total station has its unique advantages in many large projects because of its all-day and all-weather ability to find north. GAT maglev gyro total station is one of many gyro directional measuring instruments. Because of its unique maglev bearing system, moment feedback closed-loop measurement technology framework, real-time download and post-filter processing function of data in north-seeking process, the advantages of the system show and stand out from the others. In this paper, a series of research work is carried out on the optimal processing of rotor and stator current data collected by GAT maglev gyro total station, on the basis of fully analyzing the basic characteristics of stator and rotor current data. Particle filter and its improved algorithm are used to filter the stator and rotor current data. The main contents and innovations are as follows: 1. The principle, algorithm and advantages and disadvantages of particle filter are discussed. Particle filter algorithm is a kind of algorithm which has good filtering effect on nonlinear system based on spatial model of dynamic system, but it also has the problem of degradation of sample particles and lack of diversity of sample particles. In order to avoid its disadvantages, a particle filter algorithm based on observation path similarity is proposed for the correlation of rotor current data. 2. On the basis of analyzing and studying the data error source of gyro, it is clear that the data error of stator and rotor current mainly comes from the system error of instrument and the accidental error of the influence of the surrounding environment on it in the process of data acquisition. In view of the influence process, a series of measures are put forward to reduce all kinds of errors and mix into the collected data, so as to ensure the quality of the data. Based on the characteristics of stator and rotor current data, gross error detection and engineering example analysis, a method of pre-processing gyro data based on prior information of external azimuth is proposed. The results show that this method can effectively improve the stability of the data and the reliability of the final calculation results. 4. By comparing the standard particle filter algorithm, the adaptive Kalman filter algorithm and the improved particle filter algorithm based on observation path similarity based on the characteristics of rotor current correlation, the improved particle filter algorithm based on observation path similarity is applied in rotor current data. The results show that the improved algorithm can improve the quality of optimized data better. The research results of this paper are verified by an example, which is of great significance to optimize the data quality of gyro and improve the accuracy of gyro.
【学位授予单位】：长安大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TB22

【相似文献】