会议名称(中文)： 2015随机模型预测控制研讨会
会议名称(英文)： Stochastic Model Predictive Control Workshop
所属学科： 运筹学与控制论,自动控制理论与技术
开始日期： 2015-06-30
所在国家： 美国
所在城市： 美国
具体地点： Chicago, IL, USA
主办单位： American Automatic Control Council
[ 会务组联系方式 ]
联系人： Richard D. Braatz
会议背景介绍：
Model predictive control (MPC) is the leading paradigm for high-performance and cost-effective control of complex systems. Yet, the classical MPC formulation is incapable of systematically dealing with measurement noise, parametric uncertainties, and exogenous disturbances that are ubiquitous in complex dynamical systems. Uncertainties can lead to severe degradation of the performance of control systems, which is typically realized in the vicinity of safety, environmental, and quality constraints. Hence, robust control approaches that can effectively deal with uncertainties have significant economic and safety implications for operation of complex systems. Robust MPC using min-max or worst-case approaches are often used to deal with uncertainties described by bounded sets. Robust MPC guarantees robustness of stability, constraint satisfaction, etc. for all uncertainty realizations, but it can lead to conservative control inputs and reduced performance if the worst-case uncertainty realizations have a small probability of occurrence. Stochastic MPC is an emerging control approach that takes into account statistical descriptions of uncertainties, which can often be readily obtained during model development. SMPC thereby allows the tradeoff between robustness (i.e. probabilistic constraint satisfaction) and control performance. In this workshop, several approaches to (linear and nonlinear) SMPC will be covered, including randomized and sampling based approaches, polynomial chaos, adaptive constraint tightening, and methods based on polytopic tubes. Different approaches will be demonstrated using various examples including energy building systems, wind turbine control, ecosystem-based management, manufacturing systems, and chemical and pharmaceutical processes. |
