强干扰力矩环境下磁悬浮陀螺全站仪转子完备性检验及异常信息处理方法研究

To obtain pure gyro rotor azimuth information from the complex disturbance torque environment is a key problem to improve the maglev gyroscope orientation precision and the stability of north-seeking. The study found that in a state of suspension of the gyro rotor dynamic system has a high sensitivity. Under the strong disturbance torque, rotor observed data sequence will show the different frequencies of periodic response and significant nonlinear forced movement characteristics, and accompanied by some anomalies such as occasional drift and jump, thus observation data produced noise, and restricted the maglev gyroscope north-seeking precision. In order to solve this problem, first of all, this project adopt experiment and simulation technology to classify different types of disturbance torque, reveal physical impact mechanism of rotor forced movement and the basic relationship between all kinds of typical data and dynamic instrument structure; Secondly, establish the standards of rotor completeness detection and maintenance, determine the conditions of rotor completeness detection, complete the design of real-time self-inspection program; Finally, based on the modern time series and neural network theory, take into account two-position sampling data characteristics of maglev gyroscope, construct the abnormal data filtering denoising and smoothing processing method to adapt to all kinds of typical disturbance torque conditions, to improve the maglev gyroscope observation precision, stability and environmental adaptability.

如何从复杂的干扰力矩环境中获取纯净的陀螺转子方位信息是提高磁悬浮陀螺仪定向精度及寻北稳定性的关键问题。研究发现，处于悬浮状态的陀螺转子动态系统具有极高的敏感性，在强干扰力矩影响下，转子观测数据序列会表现出不同频率的周期性响应及显著的非线性受迫运动特征，并伴有偶发的漂移、跳变等异常现象，从而产生观测噪声，制约了磁悬浮陀螺仪寻北精度的提高。针对这一问题，本项目首先采用实验仿真技术对不同类型的干扰力矩展开系统深入的分类研究，揭示转子受迫运动的物理影响机制以及各种典型数据现象与动态寻北过程间的基本关系；其次建立转子完备性检测方法与标准，明确转子完备性检测条件，完成转子完备性实时自检验程序设计；最后在综合现代时间序列、神经网络等理论的基础上，顾及磁悬浮陀螺双位置采样数据特征，构建适应各种典型干扰力矩环境下仪器异常信息的滤波去噪及样本平滑处理方法，以显著提高磁悬浮陀螺仪的观测精度、稳定性与环境适应性。