WATMAN

Analysis and Design of distributed optimal control strategies applied to large-scale WATer systems MANagement, DPI2009-13744.

Project Description

 

The WATMAN Project deals with the global management of the hydric networks related to the urban water system (UWC).

 

In a situation of increasing awareness of water cycle management needs: to control scarce water resources more efficiently, to provide access to water to more regions, to control drinking water quality, to cope with extreme events such as draughts and flooding, to reduce the impact of used water on the receiving environment, the use of telemetry and telecontrol in UWC is increasing constantly.

 

The increasing availability of control hardware and information systems cannot, on its own, contribute to coping with these challenges. The key issue is the use of the information. Appropriate data validation, water system modelling and control strategies must be developed at a similar pace. Previous projects by the applicants have dealt with modelling and predictive/optimal control of water systems. Some important contributions thereof have been reported, especially in optimal control of drinking water systems and in sewer systems predictive optimal control for flood prevention and environmental impact reduction.

 

The context of optimal predictive control in water systems is generally one of large-scale problems, due to:

 

  • large spatially distributed networks,
  • time spans of predictive control horizons
  • modelling needs for nonlinear behaviours and pure delays, e.g. in river-flow equations.

In addition, water control systems are constantly being developed, expanded and/or partially closed for maintenance or renewal. There is an increasing need for fail-safe systems which must, at all cost provide uninterrupted service to consumers.


Barcelona UWC

In this context, de-centralized control is expected to provide a key contribution to applying optimal control schemes schemes:

  • incrementally to water systems under developing/expanding control implementation,
  • in a fail-safe mode when large networks are partially closed for maintenance
  • in a fault-tolerant mode when partial failures of the infrastructure, or the communications systems fail.

Thus, the objectives of  WATMAN project can be summarised as follows:

 

  • Modelling for hierarchical and distributed control/supervision
  • Model identification/partitioning algorithms for hierarchical and distributed control/supervision
  • Design of distributed optimal/predictive controllers for large-scale water systems
  • Development of coordination/cooperation mechanisms between optimal/predictive controllers using multi-agents systems.
  • Development of distributed supervision systems that allow validation, reposition and integration of data obtained from tele-measurement systems.
  • Software implementation of tools for distributed modelling, control and supervision of water networks of several cities.
Complex cases of study are waiting to be treated in WATMAN. The big and very complex sewer network of Barcelona, with strong nonlinear dynamics, hundred of constitutive elements and unpredictable weather as main source of disturbances, is one of the cases contemplated in WATMAN. Previous results reported by integrants of SAC already evidenced that this research line promises great improvements in the management of this sort of UWC networks. A second case of study consists in the drinking water network of Barcelona, which manages this appraised liquid from its collection in rivers and subterranean sources, and supplies it to the population of the metropolitan area of Barcelona. Despite this system is also a UWC network; it presents singularities, which do not allow to apply over it the previously developed approaches for sewage systems. So, all the research that will be made in the main topics of the whole project should be interpolated and new challenges and ways of research will be opened.

Most relevant publications

 

  • S. Tornil-Sin, C. Ocampo-Martínez, V. Puig and T. Escobet. Robust fault detection of non-linear systems using set-membership state estimation based on constraint satisfaction. Engineering Applications of Artificial Intelligence, 25(1): 1-10, 2012. <10.1016/j.engappai.2011.07.007>
  • C. Ocampo-Martínez, D. Barcelli, V. Puig and A. Bemporad. Hierarchical and decentralised model predictive control of drinking water networks: application to Barcelona case study. IET Control Theory and Applications, 6(1): 62–71, 2012. <doi: 10.1049/iet-cta.2010.0737>
  • S. Montes, V. Puig and J. Blesa. Robust fault detection based on adaptive threshold generation using interval LPV observers. International Journal of Adaptive Control and Signal Processing, 26(3): 258-283, 2011. <DOI: 10.1002/acs.1263>
  • V. Puig, C. Ocampo-Martínez, J. Romera, J. Quevedo, R. Negenborn, P. Rodríguez and S. de Campos. Model predictive control of combined irrigation and water supply systems: Application to the Guadiana river, 2012 IEEE International Conference on Networking, Sensing and Control, 2012, Beijing, China, pp. 85-90. <doi: 10.1109/ICNSC.2012.6204896>
  • J.M. Grosso, C. Ocampo-Martínez and V. Puig. Adaptive multilevel neuro-fuzzy model predictive control for drinking water networks, 2012 Mediterranean Conference on Control and Automation, 2012, Barcelona, pp. 1548-1553, IEEE. <doi: 10.1109/MED.2012.6265859>
  • D. Robles, V. Puig, C. Ocampo-Martínez and L.E. Garza Castañón. Methodology for actuator fault tolerance evaluation of linear constrained MPC: application to the Barcelona water network, 2012 Mediterranean Conference on Control and Automation, 2012, Barcelona, pp. 518-523, IEEE. <doi: 10.1109/MED.2012.6265690>
  • V. Puig, C. Ocampo-Martínez and S. Montes. Hierarchical temporal multi-layer decentralised MPC strategy for drinking water networks: application to the Barcelona case study, 2012 Mediterranean Conference on Control and Automation, 2012, Barcelona, pp. 734-739, IEEE. <doi: 10.1109/MED.2012.6265726>
  • F J.M. Grosso, C. Ocampo-Martínez and V. Puig. A service reliability model predictive control with dynamic safety stocks and actuators health monitoring for drinking water networks, 2012 IEEE Conference on Decision and Control, 2012,  IEEE, to appear.
  • R. Sarrate, F. Nejjari and A. Rosich. Sensor placement for fault diagnosis performance maximization in distribution networks, 2012 Mediterranean Conference on Control and Automation, 2012, Barcelona, pp. 110-115, IEEE.<doi: 10.1109/MED.2012.6265623>