基于多速率卡尔曼滤波方法的位移和加速度数据融合

为了尽可能地利用加速度传感器和位移传感器中的冗余信息、提高振动实测数据的精度,将结构同一位置的加速度和位移观测值进行数据融合处理,提出了一种基于维纳过程加速度模型的多速率卡尔曼滤波数据融合方法。首先给出了基于维纳过程的多速率卡尔曼滤波数融合模型;然后在加速度和位移观测值的基础上,通过多速率卡尔曼滤波数据融合得到位移、速度以及加速度的的最优估计,并对融合结果进行卡尔曼平滑处理,进一步提高了状态估计准确性;最后数值分析了不同噪声水平和采样频率比对融合算法的影响,而且与相关文献的结果作了对比分析。结果表明,本文方法合理有效且具有一定的噪声鲁棒性。

Data Fusion of Displacement and Acceleration Measurements Based on Multi-rate Kalman Filtering Technique

In order to exploit the inherent redundancy in the displacement transducer and accelerometer information,the collocated displacement and acceleration measurements are integrated,which may improve the accuracy of the vibration test data.An multi-rate Kalman filtering data fusion method for displacement and acceleration measurements based on Wiener process acceleration model is investigated.At the beginning,the multi-rate Kaman filtering data fusion model is introduced,then the optimal estimation for the displacement,velocity as well as acceleration are obtained based on the displacement and acceleration measurements.In addition,Kalman filtering smoothing step could yield the improved estimation accuracy.In the end,the effects of noise level and sampling rate ratio are investigated with comparison to the reference 1,which shows the effectiveness and noise robustness of the method.