切换非线性系统全局优化运行的经济预测控制  

Economic Model Predictive Control for Global Optimal Operation of Nonlinear Switching Systems

作　　者：杨亚茹[1,2] 李少远[1,2] 

机构地区：[1]上海交通大学自动化系,上海200240 [2]系统控制与信息处理教育部重点实验室,上海200240

出　　处：《自动化学报》2017年第6期1017-1027,共11页Acta Automatica Sinica

基　　金：国家重点基础研究发展计划(973计划)(2013CB035500);国家自然科学基金(61233004;61590924)资助~~

摘　　要：切换非线性系统在不同模式间平稳切换和经济切换是全局优化运行的主要需求.针对不同模式有限时域下控制算法可行域未必存在交集的系统,提出了对应的经济预测控制算法(Economic model predictive control,EMPC)及切换策略.切换发生时,该方法在实时优化层求解和更新可行中间点,并构造基于耗散的局部EMPC辅助性能指标,在考虑中间点稳定性问题上使其尽可能逼近原经济性能.在先进控制层,利用局部EMPC将状态逐次稳定至中间点,同时利用中间点问题得到的最优轨迹保证模式间的经济切换.最后,分析了切换过程的暂态经济性.该方法实际可操作性强,仿真结果说明了方法的有效性.In this work, an economic model predictive control（EMPC） and corresponding switching strategy for nonlinear switching systems are presented to handle the unreachable problem between mode transitions. An open-loop optimal transition for systems to prepare for upcoming switching is proposed. The corresponding calculation is done in the RTO layer, and optimal mid points and trajectories are updated in due time. Systems states are driven to the mid points respectively in the preparation phase and transition to the succeeding mode is complished in the switching phase by a switching scheme. Both optimal steady-states and mid points′asymptotic stability in the related mode are insured by a dissipativity based EMPC with auxiliary controlling index, and simultaneously a computational upper bound is established for best improving the transition economic performance. Transient performance analysis and simulation result have verified the validity of the method.

关 键 词：非线性系统 模式切换 经济预测控制 不可达 耗散 

分 类 号：O231[理学—运筹学与控制论]