基于硬件实时技术的陀螺仪性能分析系统

发布时间：2018-06-08 17:57

本文选题：动力调谐陀螺仪 + 性能分析　；参考：《天津大学》2014年硕士论文

【摘要】：陀螺仪是惯性导航系统核心测量元件之一。它通过敏感载体相对于惯性系的角速度,获取偏角信息。以动力调谐陀螺仪为例,它出现于第二代惯导技术时期,在之后几十年间得到广泛研究和应用。近年来,研究重点集中于对动力调谐陀螺仪系统整体性能的分析和改善,因而也对性能分析系统提出了更高要求。性能分析系统包括闭环控制回路和数据处理单元两部分,本文从系统实时性角度出发,尝试性地将硬件实时技术应用于陀螺仪控制回路的设计,从陀螺仪结构模型分析、开环系统辨识、实时控制回路设计等方面展开了研究。1、为了解陀螺仪基本特性,获取实时控制回路各项参数,选取开环辨识方法对陀螺仪表头进行建模分析。首先,介绍陀螺仪结构及其动力学方程,推导陀螺仪开环及闭环模型。介绍交叉耦合现象,并推导全解耦方法。然后,结合结构特征,分析陀螺仪噪声特点。最后,介绍最小二乘类辨识方法和频域辨识法,对频域辨识法进行改进,对比三种辨识方法的辨识结果。2、针对传统性能分析系统实时性不高的问题,提出实时性能分析系统,并将硬件实时技术应用于闭环控制回路的设计。实时控制回路采用Altera公司Cyclone系列FPGA为核心控制器,由信号调理单元、数据采集单元、数字信号处理单元、模数转换单元和功率放大单元组成。根据系统要求和辨识结果,使用硬件描述语言完成数据采集、解耦、校正、控制量输出等模块的划分和设计。3、搭建实验平台,对开环辨识方法和实时控制回路做仿真和实验。仿真及实验说明,相较于最小二乘类辨识方法和传统频域辨识法,改进后的频域辨识法拟合度更高,且辨识方法一致性好,辨识结果稳定。通过系统测试,验证了实时控制回路可以完成功能要求,并有效提高了系统采样率。由此表明,实时控制回路满足性能分析系统的控制功能,设计方案可行。最后,文章对研究、设计和实验进行总结,规划后期工作重点和可能出现的难点,提出进一步提高系统性能和测试精度的改进方案。
[Abstract]:Gyroscope is one of the core measuring components of inertial navigation system. The information of deflection angle is obtained by the angular velocity of the sensitive carrier relative to the inertial system. Taking dynamic tuned gyroscope as an example, it appeared in the second generation inertial navigation technology and was widely studied and applied in the following decades. In recent years, the research focuses on the analysis and improvement of the overall performance of the dynamic tuned gyroscope (DTG) system, and therefore puts forward higher requirements for the performance analysis system. The performance analysis system includes two parts: closed loop control loop and data processing unit. In this paper, hardware real time technology is applied to the design of gyroscope control loop from the point of view of system real time, and the structure model of gyroscope is analyzed. In order to understand the basic characteristics of gyroscope and obtain the parameters of real-time control loop, the open-loop identification method is selected to model and analyze the gyroscope head. Firstly, the structure of gyroscope and its dynamic equations are introduced, and the open loop and closed loop models of gyroscope are derived. The cross-coupling phenomenon is introduced, and the fully decoupling method is deduced. Then, the noise characteristics of gyroscope are analyzed. Finally, the least square identification method and frequency domain identification method are introduced, and the frequency domain identification method is improved. Comparing the identification results of the three identification methods, a real-time performance analysis system is proposed to solve the problem that the traditional performance analysis system is not high in real time. The hardware real-time technology is applied to the design of closed-loop control loop. The real-time control circuit is composed of signal conditioning unit, data acquisition unit, digital signal processing unit, A / D conversion unit and power amplifier unit. According to the system requirements and identification results, the data acquisition, decoupling, correction and control output modules are divided and designed by using hardware description language. The experimental platform is built, and the open-loop identification method and real-time control loop are simulated and experimented. The simulation and experiments show that compared with the least squares identification method and the traditional frequency domain identification method, the improved frequency domain identification method has higher fitting degree, better consistency and stable identification results. Through the system test, it is verified that the real-time control loop can fulfill the functional requirements and improve the sampling rate of the system effectively. It shows that the real-time control loop meets the control function of the performance analysis system and the design scheme is feasible. Finally, the paper summarizes the research, design and experiment, plans the later work focus and possible difficulties, and puts forward an improved scheme to further improve the system performance and test accuracy.
【学位授予单位】：天津大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN966

【参考文献】