Learning-Based Tuning of Supervisory Model Predictive Control for Drinking Water Networks

Publication journal: Engineering Applications of Artificial Intelligence. Volume 26, Issue 7, August 2013, Pages 1741–1750

Authors:  J.M. Grosso , C. Ocampo-Martínez, V. Puig

This paper presents a constrained Model Predictive Control (MPC) strategy enriched with soft-control techniques as neural networks and fuzzy logic, to incorporate self-tuning capabilities and reliability aspects for the management of drinking water networks (DWNs). The control system architecture consists in a multilayer controller with three hierarchical layers: learning and planning layer, supervision and adaptation layer, and feedback control layer. Results of applying the proposed approach to the Barcelona DWN show that the quasi-explicit nature of the proposed adaptive predictive controller leads to improve the computational time, especially when the complexity of the problemstructure can vary while tuning the receding horizons.

Read more: Learning-Based Tuning of Supervisory Model Predictive Control for Drinking Water Networks