输入非均匀采样广义输出误差模型的递推贝叶斯参数辨识算法  

RECURSIVE BAYESIAN PARAMETER IDENTIFICATION ALGORITHM FOR GENERAL OUTPUT ERROR MODELS WITH NON-UNIFORMLY SAMPLED INPUT DADA

作　　者：景绍学[1,2] Jing Shaoxue(School of Electrical and Information Engineering, Jiangsu University ,Zhenjiang 212013 ,Jiangsu,China;Department of Electrical Engineering, Huaian College o f Information and Technology ,Huaian 223003 ,Jiangsu,China)

机构地区：[1]淮安信息职业技术学院电气工程系,江苏淮安223003 [2]江苏大学电气信息工程学院,江苏镇江212013

出　　处：《计算机应用与软件》2016年第6期329-333,共5页Computer Applications and Software

基　　金：江苏省研究生培养创新工程项目(CX-LX12_0648)

摘　　要：针对传统最小二乘算法在辨识过程中没有考虑噪声的协方差和参数的先验概率密度的问题,提出一种递推贝叶斯算法。该算法以最大化参数的后验概率密度函数为准则进行参数估计。实验结果证明所提算法可以获得更高精度的参数估计值。收敛性分析表明,该算法给出的参数估计值收敛于参数真值。该算法综合考虑了噪声方差、数据的先验概率分布和参数的先验概率分布,可以获得比最小二乘法更高的精度的估计值。In light of that traditional least squares method does not take into account the covariance of noise and the priori probability density of parameters in the process of identification, we proposed a recursive Bayesian parameter identification algorithm. The algorithm uses the posterior probability density function of maximised parameters as the criterion to estimate parameters. Experimental result proved that the proposed algorithm could acquire the estimates of parameters in higher accuracy. Convergence analysis indicated that the estimates of parameters provide by the proposed algorithm converged to their true values. The algorithm comprehensively considers the noise variance and the priori probability distributions of data and parameters, it is able to obtain the estimates with higher accuracy than the least-squares.

关 键 词：输入非均匀采样系统 广义输出误差模型 递推贝叶斯算法 参数估计 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]