Negotiation and Learning in distributed MPC of Large Scale Systems

A key issue in distributed MPC control of Large Scale Systems (LSS) is how shared variables among the different MPC controller in charge of controlling each system partition (subsystems) are handled. When these connections represent control variables, the distributed control has to be consistent for both subsystems and the optimal value of these variables will have to accomplish a common goal. In order to achieve this, the present work combines ideas from Distributed Artificial Intelligence (DAI), Reinforcement Learning (RL) and Model Predictive Control (MPC) in order to provide an approach based on negotiation, cooperation and learning techniques. Results of the application of this approach to a small drinking water network show that the resulting trajectories of the levels in tanks (control variables) can be acceptable compared to the centralized solution. The application to a real network (the Barcelona case) is currently under development.

[1]  M. A. Brdys,et al.  Operational Control of Water Systems: Structures, Algorithms, and Applications , 1994 .

[2]  Alejandro Agostini,et al.  Feasible control of complex systems using automatic learning , 2005, ICINCO.

[3]  Bruce H. Krogh,et al.  Distributed model predictive control , 2001, Proceedings of the 2001 American Control Conference. (Cat. No.01CH37148).

[4]  Stephen J. Wright,et al.  Stability and optimality of distributed model predictive control , 2005, Proceedings of the 44th IEEE Conference on Decision and Control.

[5]  Richard S. Sutton,et al.  Reinforcement Learning: An Introduction , 1998, IEEE Trans. Neural Networks.

[6]  Riccardo Scattolini,et al.  Architectures for distributed and hierarchical Model Predictive Control - A review , 2009 .

[7]  Alberto Bemporad,et al.  Robust model predictive control: A survey , 1998, Robustness in Identification and Control.

[8]  Gabriela Cembrano,et al.  Optimal control of urban drainage systems. A case study , 2004 .

[9]  G. Bornard,et al.  Optimal control of complex irrigation systems via decomposition-coordination and the use of augmented Lagrangian , 1998, SMC'98 Conference Proceedings. 1998 IEEE International Conference on Systems, Man, and Cybernetics (Cat. No.98CH36218).

[10]  Gerald Tesauro,et al.  Extending Q-Learning to General Adaptive Multi-Agent Systems , 2003, NIPS.

[11]  G. Cembranoa,et al.  Optimal control of urban drainage systems . A case study , 2009 .

[12]  Carlos Ocampo-Martinez,et al.  Modelling and decentralized model predictive control of drinking water networks: the Barcelona case study , 2009 .

[13]  Valeria Javalera Rincón,et al.  Distributed MPC for large scale systems using agent-based reinforcement learning , 2010 .

[14]  Joseba Quevedo,et al.  Optimal control of a water distribution network in a supervisory control system , 2000 .

[15]  Vicenç Puig Cayuela,et al.  Development of a Simulation Environment for Water Drinking Networks: Application to the Validation of a Centralized MPC Controller for the Barcelona Case Study , 2009 .

[16]  Dragoslav D. Šiljak,et al.  Decentralized control of complex systems , 2012 .

[17]  S. Joe Qin,et al.  A survey of industrial model predictive control technology , 2003 .

[18]  Rudy R. Negenborn,et al.  Multi-agent model predictive control with applications to power networks , 2007 .

[19]  D. Ernst,et al.  Model predictive control and reinforcement learning as two complementary frameworks , 2006 .

[20]  James B. Rawlings,et al.  COORDINATING MULTIPLE OPTIMIZATION-BASED CONTROLLERS: NEW OPPORTUNITIES AND CHALLENGES , 2008 .