基于卡尔曼滤波理论的质量数据评估方法研究

Research on Quality Data Evaluation Method Based on Kalman Filter Theory

针对质量评估中遇到的数据不确定性、测量过程存在误差的特点,论文提出运用卡尔曼滤波理论评估数据的方法,基于系统产生的观测数据,经过一个不断的"预测—修正"的递推过程,得到当前系统数据的线性最小方差无偏估计,最后通过Matlab编程进行了仿真验证。结果表明,针对产生线性变化数据的系统,卡尔曼滤波能够很好地评估和拟合数据,减少数据产生和测量过程中带来的误差和不确定因素,为进一步利用数据进行质量评估和技术状态分析打下良好基础,表明了卡尔曼滤波理论在质量数据评估方向的研究价值。

In view of the uncertainty of data in quality assessment and error in measurement process,this paper proposes amethod of using the kalman filter theory evaluation data,based on the observation data system,through a continuous＂predict-fixed＂recursive process,of linear minimum variance unbiased estimation of the current system data,finally has carried on the simu-lation by MATLAB program. Results show that the system for linear change data,kalman filter can be very good assessment and fit-ting data,reduce the data and measurement error and uncertainty in the process,for further utilization of data quality assessmentand technical analysis to lay a good foundation,suggests that the theory of kalman filter in the direction of the data quality assess-ment research value.