With the application to water industry of the Industry 4.0 approach (i.e. the Water 4.0 approach), Real Time Control is gaining attention as a strategic tool for the management of Water Distribution Networks. Moreover, this approach favours the installation of a wired communication infrastructure, which guarantees new standards in communication between sensors, actuators and control units. This work focuses on the particular task of service pressure regulation, obtained by means of a closed-loop control based on Pressure Control Valves and frequency domain control algorithms. While the topics of the identification of proper control structures, and the analysis of their stability and robustness properties were covered in previous works, this paper is aimed at improving the design phase of such algorithms by formulating the regulator tuning problem as a bi-objective optimisation one. Its solution leads to a Pareto front that captures the typical design trade-off between regulation performances and control effort and robustness, and helps the designer selecting the favoured tunings to be tested on the plant. Simulations performed on a detailed model of two different WDNs highlight the benefits of the approach and the reliability of the obtained tunings.
Bi-objective optimisation based tuning of pressure control algorithms for water distribution networks
Galuppini G.
;Creaco E.;Magni L.
2020-01-01
Abstract
With the application to water industry of the Industry 4.0 approach (i.e. the Water 4.0 approach), Real Time Control is gaining attention as a strategic tool for the management of Water Distribution Networks. Moreover, this approach favours the installation of a wired communication infrastructure, which guarantees new standards in communication between sensors, actuators and control units. This work focuses on the particular task of service pressure regulation, obtained by means of a closed-loop control based on Pressure Control Valves and frequency domain control algorithms. While the topics of the identification of proper control structures, and the analysis of their stability and robustness properties were covered in previous works, this paper is aimed at improving the design phase of such algorithms by formulating the regulator tuning problem as a bi-objective optimisation one. Its solution leads to a Pareto front that captures the typical design trade-off between regulation performances and control effort and robustness, and helps the designer selecting the favoured tunings to be tested on the plant. Simulations performed on a detailed model of two different WDNs highlight the benefits of the approach and the reliability of the obtained tunings.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.