基于高斯和粒子滤波的动态称重数据处理

Study of Dynamic Weighing Filtering Method Based on the Gaussian Sum Particle Filter

由于动态称重过程中的噪声干扰,导致动态称重信号处理中存在数据处理速度慢与精度低等不足。为了提高动态称重的快速性与准确性,本文将高斯和粒子滤波算法应用于动态称重数据处理。在对动态称重系统建立状态空间模型的基础上,引进高斯和粒子滤波算法,利用高斯和逼近状态的后验密度,提高了对状态分布估计的精确性。实验结果证明,高斯和粒子滤波方法有效地提高了动态称重的速度与精度,比较实验结果说明本文方法优于传统的扩展卡尔曼滤波和粒子滤波效果。

In the dynamic weighing signal process,the slow speed and weak accuracy of the data process is caused by the noise interference in the dynamic weighing.To improve the speed and the accuracy of the dynamic weighing,the Gaussian sum particle filer algorithm is presented to process the dynamic weighing data.To improve the estimation accuracy of the state distribution,this novel algorithm and the Gaussian sum distribution is used to approximate the poster density,based on the state-space model of the system.The simulation shows this algorithm improves effectively the speed and the accuracy,and the comparison results show this algorithm can outperform the traditional EKF and the PF.