基于微粒群算法的非线性系统建模方法研究

Study of nonlinear system modeling based on particle swarm optimizer

针对非线性系统多模型自适应控制中的模型覆盖问题,提出一种基于微粒群算法的多模型建模方法.首先,对非线性系统定义了基于混合逻辑模型的多模型描述,建立了非线性系统的混合线性多模型;然后,基于微粒群优化算法对非线性系统进行优化建模,在保证建模准确性的同时采用最少的子模型逼近非线性系统;最后,通过一个仿真算例表明了该建模方法的有效性.

A particle swarm optimizer（PSO） based multiple-model modeling method is introduced to deal with the model covering problem in the multiple model,adaptive control of nonlinear system. Firstly, the nonlinear system is described as a mixed logic model based multiple models. And a mixed logic linear model is constructed accordingly. Then according to the mixed logic linear model, optimal modeling is realized based on the PSO, which employs the least sub-model to approximate the nonlinear system under the condition of accuracy. Finally, a simulation example shows the effectiveness of the proposed modeling method.