Since the 1980s Model Predictive Control (MPC) has become the algorithm of choice for complex control problems in the oil and petrochemical industries. As MPC requires an optimal control problem to be solved in real time, the technique was suited primarily for situations where the process to be controlled is relatively slow and extensive computational resources are available. In the last decade we have developed various algorithms that allow MPC to be implemented on fast processes (with sampling times in the tens of nanoseconds) with modest computational requirements. I will describe the ideas behind these developments and the implementations on a range of industrial prototype systems in the electronics area. Show more
External linksSearch via SFX
PublisherNortheastern University Shenyang
Organisational unit03416 - Morari, Manfred (emeritus)
NotesLecture at the State Key Laboratory of Synthetical Automation for Process Industries, Northeastern University in Shenyang, China on 7 June 2012.
MoreShow all metadata