摘要
针对非线性系统模型预测控制中预测模型容易失配且目标函数难以求解的问题,提出了一种基于在线支持向量机建模和遗传算法滚动优化的模型预测控制方法。该方法利用在线支持向量机建立被控对象的非线性模型,在线支持向量机是一种迭代学习的支持向量机训练算法,可以进行在线训练,从而实现模型在线自校正;并且通过遗传算法求解目标函数的最优控制量,完成滚动优化。对非线性系统的仿真研究结果表明,该方法有效且具有良好的自适性。
Aiming at the problems that the predictive model is often mismatching and is difficult to solve the nonlinear optimization function of nonlinear system model predictive control, an online support vector machine (OSVM) modeling and genetic algorithm (GA) rolling optimization based model predictive control method is proposed. This proposed method builds a nonlinear model for objects using OSVM, which is an iterative support vector regression learning algorithm and can be used for online training, hence the predictive model parameters could be adjusted online through online learning. Furthermore, the nonlinear optimization function is solved by GA optimization,and rolling optimization is reaiized in the system. The nonlinear case simulation results show that the system adaptability is improved by the proposed method.
出处
《系统工程与电子技术》
EI
CSCD
北大核心
2013年第6期1275-1280,共6页
Systems Engineering and Electronics
基金
国家自然科学基金(61273131)
江苏省高校优势学科建设工程资助课题
关键词
模型预测控制
在线支持向量机
遗传算法
滚动优化
model predictive control
online support vector machine (OSVM)
genetic algorithm (GA)
rolling optimization