限定记忆法辨识带钢热镀锌退火炉模型参数  

Identification of strip steel galvanized annealing furnace model parameters using fixed memory method

作　　者：李训艳[1] 周玉国[1] 逯倩倩[1] 

机构地区：[1]青岛理工大学自动化工程学院,山东青岛266033

出　　处：《计算机时代》2012年第1期21-23,共3页Computer Era

摘　　要：最小二乘参数估计方法可用于线性的或非线性的系统参数辨识,包括动态的、静态的和参数的或非参数的模型辨识,其递推算法更是收敛可靠,简单实用。但是随着数据的不断增长,最小二乘的递推算法将失去修正能力,出现数据饱和现象,限定记忆最小二乘法解决了这一问题,并能得到无偏、一致、有效估计。以已建立的连续带钢热镀锌退火炉数学模型为实例,用限定记忆最小二乘法辨识连续带钢热镀锌退火炉模型参数。通过对限定记忆最小二乘法的研究,进行模型参数辨识,并给出辨识结果和分析,结果证明了该方法的优越性。The least square method can be used for the parameters identification of linear or nonlinear system,including dynamic,static and parametric or nonparametric,its recursive algorithm is more converging and reliable,simple and practicable.But with data increasing,the least square recursive algorithm will lose its correction ability,and data saturation phenomenon will appear.The fixed memory least square method has solved this problem,and can give unbiased,consistent and effective estimates.Taking the mathematical model built for continuous strip steel galvanized annealing furnace as an example,the fixed memory least square method is used to identify continuous strip steel galvanized anneal furnace model parameters.Through studying the fixed memory least square method,the model parameter identification is carried out,and the identification results and analysis are given,the results prove the superiority of this method.

关 键 词：热镀锌退火炉 参数辨识 限定记忆最小二乘法 递推算法 仿真 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]