Journal article · Conference paper
A Reduced Dantzig-Wolfe Decomposition for a Suboptimal Linear MPC
Department of Applied Mathematics and Computer Science, Technical University of Denmark1
Scientific Computing, Department of Applied Mathematics and Computer Science, Technical University of Denmark2
Dynamical Systems, Department of Applied Mathematics and Computer Science, Technical University of Denmark3
Department of Chemical and Biochemical Engineering, Technical University of Denmark4
Center for Energy Resources Engineering, Centers, Technical University of Denmark5
CITIES - Centre for IT-Intelligent Energy Systems, Centers, Technical University of Denmark6
Linear Model Predictive Control (MPC) is an efficient control technique that repeatedly solves online constrained linear programs. In this work we propose an economic linear MPC strategy for operation of energy systems consisting of multiple and independent power units. These systems cooperate to meet the supply of power demand by minimizing production costs.
The control problem can be formulated as a linear program with block-angular structure. To speed-up the solution of the optimization control problem, we propose a reduced Dantzig-Wolfe decomposition. This decomposition algorithm computes a suboptimal solution to the economic linear MPC control problem and guarantees feasibility and stability.
Finally, six scenarios are performed to show the decrease in computation time in comparison with the classic Dantzig-Wolfe algorithm.
Language: | English |
---|---|
Publisher: | International Federation of Automatic Control |
Year: | 2014 |
Pages: | 2207-2212 |
Proceedings: | 19th World Congress of the International Federation of Automatic Control (IFAC 2014) |
Series: | I F a C Workshop Series |
ISSN: | 14746670 |
Types: | Journal article and Conference paper |
DOI: | 10.3182/20140824-6-ZA-1003.02357 |
ORCIDs: | Poulsen, Niels Kjølstad and Jørgensen, John Bagterp |