摘要
RBF核函数的表达式较复杂,用于模型预测控制时滚动优化求解困难。该文建立一种基于RBF核SVR的预测控制模型,利用多智能体粒子群(MAPSO)算法求解模型预测控制中的滚动优化,推导出最优控制律,对其进行Matlab仿真并与其他方法进行比较。选取一个非线性系统及连续搅拌槽式反应器进行测试分析,结果表明:该预测模型能迅速抑制干扰,较快返回预设跟踪轨迹,展现了良好的抗噪、抗干扰能力。
Although the RBF is the most widely used SVR kernel function,the expression complexity of RBF makes it difficult to get analytical formula of rolling optimization in model predictive control.A model of predictive control based on RBF-SVR is established,the multi-agent particle swarm optimization (MAPSO) algorithm is used to obtain the optimal control inputs.For nonlinear systems,Simulation results show that the proposed algorithm has demonstrated excellent adaptive ability and robustness and can be effectively applied to the nonlinear system.
出处
《实验技术与管理》
CAS
北大核心
2016年第6期103-107,116,共6页
Experimental Technology and Management
基金
重庆市研究生教育教学改革研究项目(yjg143061)
重庆邮电大学校级教改项目(XJG1523)
关键词
模型预测控制
支持向量回归
仿真实验
控制系统
RBF
model predictive control
support vector regression
simulation experiment
control system
RBF