基于向量相似度的多模型局部建模方法研究

针对一类仅使用大批历史数据结构未知的非线性工业过程,根据数据驱动及局部建模的基本思想,提出一种基于局部模型算法的在线多模型辨识策略。从向量相似的角度提出了一种新的选择数据信息(即建模邻域的确定)的方法,有效提高了获得当前时刻系统最佳局部模型的数据精确度。给出了权值选定的适合度标准及带宽h选择的快速方法。最后对算法进行了特性分析及仿真研究,并与其他局部建模算法的计算结果进行比较,验证了本文辨识算法的有效性。

Local multi-model method based on similarity of vector

This paper proposed an on-line multiple models identification strategy based on data-driven and local-modeling for a class of unknown-structure nonlinear systems on the basis of numerous historical database. And it proposed a new approach to determine data information (determine the neighborhood), it could improve precision of the data which used to get optimal local model. Then gave the goodness of fit criterion and rapid selection method to determine bandwidth. Finally, it did some characteristic analysis and simulation studies, and the results illustrate the validity of this approach.