非线性动态模型的迭代滤波及Bayes预测  

The Iteration Fillering and Bayesian Forecasting of Nonlinear Dynamic Models

作　　者：葛颜祥[1] 张承进[1] 张孝令[1] 

机构地区：[1]山东矿业学院应用数学与软件工程系

出　　处：《山东矿业学院学报》1993年第2期181-185,共5页Journal of Shandong University of Science and Technology(Natural Science)

摘　　要：就观测误差方差V_t已知和V_t=V常数未知两种情况,对于非线性动态模型,利用新获得的观测值y_t及滤波公式重新对t—1时刻的状态进行估计,改进了均值与随机状态向量的接近程度,然后重新对非线性函数进行线性化,之后再进行滤波,直到均值和方差没有显著的改善为止,由此再进行下一步的递推和预测。In this paper, authors use the new observation y, and filtering formulae to reestimate the state at time t—1 of noulinear dynamic models with regard to the variance V, of observing error known or V_t=V unknown, then linearize the nonlinear functions and fil- tering again, till there are no remarkable changes about m_(t-1)^(i), C_(t-1)^(i). Thus we can go on the next recurrence and forecasting

关 键 词：非线性 动态模型 迭代滤波 预测 

分 类 号：O211.64[理学—概率论与数理统计]