A distributed accelerated gradient algorithm for distributed model predictive control of a hydro power valley

Forskningsoutput: TidskriftsbidragArtikel i vetenskaplig tidskrift

Standard

A distributed accelerated gradient algorithm for distributed model predictive control of a hydro power valley. / Doan, Minh Dang; Giselsson, Pontus; Keviczky, Tamás; De Schutter, Bart; Rantzer, Anders.

I: Control Engineering Practice, Vol. 21, Nr. 11, 2013, s. 1594-1605.

Forskningsoutput: TidskriftsbidragArtikel i vetenskaplig tidskrift

Harvard

APA

CBE

MLA

Vancouver

Author

RIS

TY - JOUR

T1 - A distributed accelerated gradient algorithm for distributed model predictive control of a hydro power valley

AU - Doan, Minh Dang

AU - Giselsson, Pontus

AU - Keviczky, Tamás

AU - De Schutter, Bart

AU - Rantzer, Anders

PY - 2013

Y1 - 2013

N2 - A distributed model predictive control (DMPC) approach based on distributed optimization is applied to the power reference tracking problem of a hydro power valley (HPV) system. The applied optimization algorithm is based on accelerated gradient methods and achieves a convergence rate of O(1/k^2), where k is the iteration number. Major challenges in the control of the HPV include a nonlinear and large-scale model, non-smoothness in the power-production functions, and a globally coupled cost function that prevents distributed schemes to be applied directly. We propose a linearization and approximation approach that accommodates the proposed the DMPC framework and provides very similar performance compared to a centralized solution in simulations. The provided numerical studies also suggest that for the sparsely interconnected system at hand, the distributed algorithm we propose is faster than a centralized state-of-the-art solver such as CPLEX.

AB - A distributed model predictive control (DMPC) approach based on distributed optimization is applied to the power reference tracking problem of a hydro power valley (HPV) system. The applied optimization algorithm is based on accelerated gradient methods and achieves a convergence rate of O(1/k^2), where k is the iteration number. Major challenges in the control of the HPV include a nonlinear and large-scale model, non-smoothness in the power-production functions, and a globally coupled cost function that prevents distributed schemes to be applied directly. We propose a linearization and approximation approach that accommodates the proposed the DMPC framework and provides very similar performance compared to a centralized solution in simulations. The provided numerical studies also suggest that for the sparsely interconnected system at hand, the distributed algorithm we propose is faster than a centralized state-of-the-art solver such as CPLEX.

KW - Distributed optimization

KW - Hydro power control

KW - Accelerated gradient algorithm

KW - Distributed model predictive control

KW - Model predictive control

U2 - 10.1016/j.conengprac.2013.06.012

DO - 10.1016/j.conengprac.2013.06.012

M3 - Article

VL - 21

SP - 1594

EP - 1605

JO - Control Engineering Practice

JF - Control Engineering Practice

SN - 0967-0661

IS - 11

ER -