基于集合卡尔曼滤波的实时校正方法

发布时间：2018-08-06 14:42

【摘要】：为减少非恒定水流计算中的不确定性,基于集合卡尔曼滤波提出多变量交替校正的方法。该方法通过交替校正水位和流量,避开了滤波过程中的大矩阵计算,实现了利用观测信息直接校正非恒定流状态的目的;同时,应用尺度转换方法提高水位滤波精度。数值试验重点考察了观测误差和水位变换系数对模型计算精度的影响。结果表明:观测误差越小,模型的计算精度越高;水位尺度变换系数能显著增强多变量交替校正方法的效果,变换系数越大,计算精度越高;基于集合卡尔曼滤波的多变量交替校正方法具有良好的校正性能,能显著提高河道水流的预报精度。
[Abstract]:In order to reduce the uncertainty in the calculation of unsteady flow, a multivariable alternative correction method based on set Kalman filter is proposed. By alternately correcting the water level and the flow rate, the method avoids the large matrix calculation in the filtering process and realizes the purpose of correcting the unsteady flow state directly by using the observation information. At the same time, the accuracy of the water level filtering is improved by using the scale conversion method. The effects of observation error and water level transformation coefficient on the accuracy of the model are investigated in numerical experiments. The results show that the smaller the observation error, the higher the accuracy of the model, and the higher the accuracy of the multivariable alternating correction method is. The multivariable alternative correction method based on set Kalman filter has good correction performance and can significantly improve the prediction accuracy of river flow.
【作者单位】：中国科学院南京地理与湖泊研究所;中国科学院大学资源与环境学院;
【基金】：水体污染控制与治理科技重大专项(2012ZX07101-011) 国家自然科学基金(51379059,51279047)
【分类号】：TV131.4

【相似文献】