摘要
针对结构未知的复杂非线性系统 ,提出一种多模型建模方法。从理性建模的思想出发 ,在积累的大量系统输入输出数据中找出与系统当前模态相匹配的数据集合 ,再用局部多项式拟合 (L PF)算法建立系统的局部模型 ,根据系统模态的变化建立系统的多个模型 ,实现对全局非线性系统的准确建模。该方法运算简单 ,易于实施。仿真并与其它方法比较 。
Based on model-on-demand idea, a multi-model approach to identification for unknown-structure nonlinear systems is presented. The modeling idea is to find out, from vast historical system input-output data sets, some ones matching with the current working point, then to develop a local model using local polynomial fitting (LPF) algorithm. With the change of working points, multiple local models are built, which can realize the exact modeling for the global system. Simulation results show fine performances of the method′s simple, effective and reliable estimation.
出处
《控制与决策》
EI
CSCD
北大核心
2002年第1期11-14,共4页
Control and Decision
基金
国家自然科学基金项目 (6 9934 0 2 0
6 0 0 740 0 4)
