基于改进灰色预测模型的半球谐振陀螺在线故障检测

On-Line Fault Detection of HRG Based on an Improved Gray Prediction Model

针对基于半球谐振陀螺的导航系统故障检测的实时性要求高、数据变化缓慢等特点,提出了一种改进型的灰色预测模型,将移动窗口初值优化的灰色预测模型和最小二乘结合,对该类惯性传感器的故障进行实时在线检测。对处置优化的灰色预测模型的残差信号进行建模,提高了预测的精度,从而实现了利用少量的历史数据对下一时刻数据的准确预测,达到对半球谐振陀螺实时故障检测的目的。详细描述了改进型灰色预测模型的建模方法和步骤,并针对半球谐振陀螺的3种故障形式,与普通灰色预测方法进行了对比仿真研究。结果表明该方法可以准确、有效地进行在线故障检测。

An improved gray prediction model is proposed to perform on-line fault detection of the navigation system with hemispherical resonator gyro(HRG).The improved model combines the moving window gray prediction with the least square theory,which can satisfy the characteristic of the high real-time demand and slowchanging data of the navigation system.The residual signal of the optimized gray prediction model is modeled to improve the prediction accuracy,so as to realize the accurate prediction of the next time data with a small amount of historical data,to achieve the purpose of real-time fault detection of HRG.The modeling theory and procedures are introduced in detail.Finally,simulations of the three kinds of HRG fault patterns are studied comparing with the normal gray predictive method,and the results show the correctness and effectiveness of the proposed method.