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 ,表明了该方法的有效性 Aiming at the complicated nonlinear system with unknown structure, a multi-model modeling method is proposed. Based on the idea of ​​rational modeling, the data set that matches the current mode of the system is found out from a large amount of system input and output data. Then, the local polynomial fitting (L PF) algorithm is used to establish the local model of the system. State changes to establish a number of system models to achieve accurate modeling of global nonlinear systems. The method is simple and easy to implement. Simulation and comparison with other methods show that the method is effective