引用本文:史忠科.模型在线辨识方法及其应用[J].控制理论与应用,1995,12(6):787~791.[点击复制]
SHI Zhongke.On-Line Model Identification Method and Its Application[J].Control Theory and Technology,1995,12(6):787~791.[点击复制]
模型在线辨识方法及其应用
On-Line Model Identification Method and Its Application
摘要点击 2532  全文点击 455  投稿时间:1994-04-02  修订日期:1994-12-14
查看全文  查看/发表评论  下载PDF阅读器
DOI编号  
  1995,12(6):787-791
中文关键词  模型辨识  在线估计  非线性系统  飞行动力学  飞行试验
英文关键词  model identification  on-line estimation  nonlinear system  flight dynamics  flight test
基金项目  
作者单位
史忠科 西北工业大学自动控制系 
中文摘要
      本文提出了一种有效的非线性模型和参数在线估计方法。为了实现模型在线辨识,本文根据误差性能指标,给出了模型判据及计算式。根据递推加权最小二乘算法和优选判据,导出了模型和参数同时在线估计的有效算法。为了提高计算效率和数值稳定性,模型辨识和参数辨识均采用了U-D分解方法。新方法可用于飞行器非线性气动模型和参数的实时估计。实际应用结果表明,使用该方法可以有效地确定多项式、样条函数模型结构,参数辨识的结果满足工程要求。
英文摘要
      An efficient on-line identification method for both model and parameter for nonlinear system is presented in this paper. To achieve on-line model identification, a new algorithm is developed which is based on the cost function of error statistics. By means of recursive least square method and model choosing criterion, an efficient on-line method for both model and parameter identification is obtained. To get high numerical stability and computational efficiency, U-D factorization is used in both model and parameter estimation. The new method has been used in the on-line estimation of aerodynamic coefficient of an aircraft. The results of applications show that the models represented by polynominal and spling functions can be determined accurately.