多变量系统的辅助模型辨识方法的收敛性分析

丁锋，谢新民[1]假设协方差阵P0-1（t）（s）的最大特征值与最小特征值之比（即条件数）有界，和系统噪声{w（t）}方差有界，证明了辅助模型辨识其法参数估计的一致收致性．本文将分别放松这两个条件，即ⅰ）条件数无界，ⅱ）系统噪声为非平稳噪声，且方差无界，探讨了辅助模型算法的收敛性．分析表明在这种较弱的假设下，辅助模型算法保持很强的鲁棒性，参数估计是一致收敛的．

基于辅助模型的量化控制系统辨识方法

针对具有通信约束的量化控制系统模型,在采用随机重复性试验测量信息的技术上,提出了基于辅助模型的量化系统参数辨识方法.首先分析了在随机重复性试验方法下量化系统的模型特征并给出了分两步辨识的策略.分析表明,在上述模型里系统具有时变的估计误差,推导了进行参数辨识所满足的持续激励条件,并给出了基于辅助模型的多新息量化辨识递推算法.接着研究了所给出辨识算法的收敛性分析,得到了系统参数估计误差上界的计算式,最后将方法推广到一类Hammerstein非线性系统量化辨识问题上.数字仿真验证了该算法及结论的有效性.

MAPS方法在航空发动机性能寻优控制中的应用

利用模型辅助模式搜索方法(MAPS方法),研究了某型涡扇发动机基于部件级非线性实时数学模型的性能寻优控制。该方法是一种直接搜索方法,不需要计算发动机非线性模型对控制量的偏导数,并通过运用kriging方法构造非线性代理模型来减少发动机模型的调用次数,进而减少了计算量,提高了优化速度。应用MAPS方法对航空发动机最低油耗、最大推力和加力最大推力3种控制模式进行了优化计算,论文给出了优化计算结果。

Models for Teaching Culture in Computer-Assisted Language Learning

CALL （computer-assisted language learning） has tremendously transformed the teaching of language, with its wide application in many aspects of language teaching. However, how to integrate CALL into the teaching of culture is still rarely discussed. The purpose of the paper is to explore feasible models for teaching culture in CALL and their effects on students＇ acquisition process. Based on two fundamental pedagogical approaches （participatory pedagogy and multiliteracies pedagogy） for teaching culture, this paper proposes three pedagogical models （problem-posing model, web-quest model, and computer-supported collaborative learning model） for teaching culture in CALL. In the end, this paper illustrates a combination of the three models in real CALL setting by a cultural teaching case. A questionnaire survey and interviews are conducted to reflect on students＇ feedback, which gives an insight into possible adjustments in teaching models and the paper also proposes future possibilities in applying these models into teaching culture