Elsevier

Water Research

Volume 183, 15 September 2020, 116072
Water Research

Laboratory experiments and simulation analysis to evaluate the application potential of pressure remote RTC in water distribution networks

https://doi.org/10.1016/j.watres.2020.116072Get rights and content

Highlights

  • Laboratory experiments to evaluate remote RTC for pressure control in water distribution systems.

  • Accurate pressure control under proper system calibration including gain parameter and control time.

  • Simulation analysis to compare model results and experiments.

  • Suitable use of steady-state modeling to describe real time pressure control for leakage reduction in WDNs.

Abstract

Experimental tests were performed to demonstrate applicability of remote real time control (RTC) of pressures for leakage reduction in water distribution networks (WDNs). The experimental tests were carried out in a laboratory pilot system equipped with a motorized plunger valve. A RTC system with the adoption of an integral-type control algorithm was implemented in order to adjust the valve on the basis of pressure measurements acquired in real time. A numerical model of the pilot system was used to verify the suitability of the hypothesis of steady-state conditions in simulating the laboratory tests. The results of the experiments show that, under appropriate calibration of the control algorithm, the RTC system is able to perform effective control of the pressure. Comparison between results of the simulations and experiments reveals that the steady-state model describes correctly the evolution of the pressure control processes observed in the laboratory pilot system, thus opening perspectives for testing remote RTC schemes for leakage management in real WDNs.

Introduction

Active control of pressure is considered as one of the most promising methodologies to reduce water leakages in water distribution networks (WDNs) (Araujo et al., 2006; Creaco and Walski, 2017). Evidence has amply proven that the increase in background losses (such as the leakage from pipe breaks, joints, etc.) is markedly correlated to the increase in operational pressure in the pipes of the distribution network (Thornton, 2002; Ciaponi et al., 2015).

Traditional pressure reduction methods generally make use of pressure control valves (PCVs). Normal practice is to install such devices at the inlets to the districts of the WDN and to set them properly in order to lower the piezometric level of the whole district (Thornton, 2002). Typically, PCVs are used for local control, with the objective of achieve and maintain the desired pressure set-point (locally) at the site of installation, immediately at the outlet of the valve. Pressure set-point can be constant or vary during the day according to prefixed time-scheduling. Most common types of PCVs include mechanically/hydraulically driven devices that need to be calibrated in situ (usually, by adjustment of a secondary screw-based pilot valve) as a function of pressure (Prescott and Ulanicki, 2008; Nicolini and Zovatto, 2009; AbdelMeguid et al., 2011). Alternatively, the valve can be adjusted on the basis of pressure measurements using electronic controllers (Janus and Ulanicki, 2018).

In principle, controlling PCVs according to local values of pressure does not assure proper control of the piezometric levels of the downstream WDN; in fact, such a type of control is significantly affected by the uncertainty in the estimation of the spatio-temporal distribution of nodal water demands, leakages, as well as energy losses in the WDN.

Recently, focusing on leakage reduction, various researchers (Campisano et al., 2016; Berardi et al., 2017; Page et al., 2017a) have explored the potential of methodologies based on the use of remote real time control (RTC) for improving pressure regulation in WDNs. Depending on the network characteristics, RTC systems can suitably replace traditional pressure control strategies by moving from local to remote control technologies (Creaco and Walski, 2018); indeed, remote RTC systems use (distributed) remote information about the current status of the WDN in order to improve the effectiveness of pressure (and thus leakage) control strategies. Typically, one or more pressure sensors are installed in the network, in nodes that are far (remote) from the valve site. Normally, such nodes are placed in the downstream part of the district, selected among those with low pressure (Campisano et al., 2010). Information obtained by the continuous acquisition of remote pressure measurements is transmitted through a specific communication infrastructure in real time (normally using GSM or dedicated radio lines), being finally used to adjust dynamically the upstream control valves (Campisano et al., 2010; Giustolisi et al., 2017). RTC is implemented by means of controllers, devices that, based on information received by the sensors in the network, calculate and provide commands to valve actuators in order to adjust the flow process and drive control node pressure to the set-point.

Much of the available studies on RTC (e.g., Campisano et al., 2010; Giustolisi et al., 2017; Page et al., 2017b; Creaco et al., 2018) have mainly invoked use of simulation approaches to demonstrate advantages of remote control as compared to local control. Some approaches (Janus and Ulanicki, 2018; Galuppini et al., 2019) consider use of sophisticated methods proper of control engineering with a focus on system design in order to control pressure signals in WDNs including high-frequency components. Such type of approach aims at accurate control of pressure signals determined by pulsed demands in the WDN. However, tracking such signals requires control frequency that may be incompatible with the valve operation, or increase malfunctioning/failure of the valve itself. Conversely, other approaches focus on the hydraulic objective of reducing water leakages in the WDN. To achieve this objective, the control of the (low-frequency) pressure component associated to the daily pattern of the water demands in the network has been recognized as sufficiently accurate (Campisano et al., 2016; Creaco et al., 2019).

In such a context, simulations have been carried out to test performances of various control algorithms, that include hydraulics-based algorithms (AbdelMeguid et al., 2011; Giustolisi et al., 2017) and algorithms based on the use of PID (Proportional-Integral-Derivative) logic. A complete review of available algorithms for remote RTC of pressure in WDNs can be found in the recent work by Creaco et al. (2019).

PID is probably the most widespread type of control logic in embedded programmable logic controllers (PLCs), owing to its conceptual simplicity, and the opportunity for the continuous adaptation to the controlled system (Åström and Hägglund, 1984). This type of control has been used in various modalities in the field of water engineering for the control of both urban drainage systems (Campisano et al., 2000; Schuetze et al., 2004; Pothof and Kooij, 2007) and water distribution systems. Results from the literature show that effective valve control in WDNs can be obtained by adopting algorithms which do not necessary include all the three terms of the complete standard PID. In any case, accurate calibration of the used algorithm is required (Campisano et al., 2012; Ziegler and Nichols, 1942). Creaco and Franchini (2013), Creaco (2017), Page and Creaco (2019) report improvements in the control action with algorithms that include use of simultaneous measurements of pressure and flow discharge (although at higher costs and complexity of implementation).

Although the potential for application of remote RTC in WDNs was explored by various studies, much of these studies were carried out using numerical approaches. Conversely, up to now, only few investigations have concerned the development of experiments to evaluate the RTC potential at the laboratory and/or field scales. An example of laboratory application is provided by Fontana et al. (2018a) which developed an integral (I) control algorithm to adjust settings of a pilot valve-operated PCV. The same authors tested with success the developed algorithm in a district of the WDN of the city of Benevento, Italy (Fontana et al., 2018b), confirming suitable pressure control and water leakage reduction in the network. However, the control system adopted by the authors was not fully implemented as a scheme of remote RTC; in fact, such control system adjusts the outlet pressure of the valve by using an empirical model of the network to estimate head losses between the site of the PCV and the remote control node. A very similar approach was used by Bakker et al. (2014) for the dynamic control of flows and pressures in Poznań, Poland through performing real time adjustment of pumps settings. Also in this case, a pressure-demand model of the network was calibrated offline and used for determining the settings of the pumps. Therefore, the approaches adopted in previous papers are not based on the use of pressure measurements transmitted in real time from the network to the control site. With regard to the used control, such approaches make use of a closed-loop scheme for controlling local pressure at the valve site while relying on the use of an open-loop for the control of pressure at the remote node. In principle, if the relationship between local and remote pressure is not modeled properly, even a good regulation of local pressure might not assure effective control at the remote node. In these cases, large network portions may exhibit pressure excess/deficit during the day as compared to set-point values.

The objective of this paper is twofold. Novel experiments were carried out in a pilot rig at the laboratory scale to verify applicability and effectiveness of remote RTC for pressure control with reference to leakage reduction in WDNs. Remote RTC was implemented by adjusting in real time the opening degree of a motorized plunger valve. Differently from previous literature studies, the valve control is based exclusively on pressure measurements acquired in remote (and transmitted) in real time using a closed-loop control scheme. As further objective, the paper aims at demonstrating the suitability of the hypothesis of steady-state conditions to simulate the RTC laboratory experiments. To achieve this objective, an already available numerical hydraulic model of the pilot system was upgraded with a control module and used for both the preliminary tuning of the controller and the successive simulations.

The paper is structured in sections. First, the experimental methods are presented, including the description of the laboratory pilot system, the implemented RTC system architecture, the framework of the experiments, as well as the used numerical model. Secondly, results of the different experiments are presented with specific emphasis on the pressure control performance of the RTC system. Third, simulation results are discussed and compared with results of the experiments to demonstrate the suitability of the steady-state hypothesis for the simulation of real time pressure control for leakage reduction. Finally, the discussion concerning the analysis of the results is reported including potential and limitations of the used approach, followed by main conclusions of the research and future perspectives.

Section snippets

Description of the laboratory pilot system

The pilot system was installed at the Laboratory of Hydraulics of the Department of Civil Engineering and Architecture of the University of Catania, Italy. The system mainly consists of 4 sub-systems: i) an upstream head tank equipped with a pump (outside the laboratory room); ii) the adduction pipe section from the pump to the laboratory room; iii) the pipe section of interest for the experiments; and iv) the recirculation system to convey back flow to the upstream tank. The upstream tank has

RTC tests with the ball valve fully open

The experimental results concerning the RTC tests carried out with K = 1 are reported in the graph of Fig. 3. The figure shows measured pressures at both US and CN during the whole experiment. The pressure set-point value is reported in the figure for comparison. The figure also reports values of the flow discharge Q, and of the opening degree a of the plunger valve during the test. Results show that the control algorithm, in this case, allows to achieve the set-point in approximate way since

Discussion

The obtained results demonstrate the potential of application of closed-loop remote RTC schemes for pressure control in the pilot system. The larger costs of installation/operation of closed-loop remote RTC systems in WDNs would be balanced by an expected increase in pressure control performance due to the real time use of pressure measurements. In this concern, closed-loop schemes may take also advantage of use of models that may allow improved tuning of the regulator and design of the

Conclusions

In this paper, the results of experiments carried out in a laboratory pilot system were analyzed in order to show potential for application of remote RTC in controlling pressure in WDNs.

Remote RTC was applied to control the pressure in the pilot system by adjusting in real time the opening/closure of a motorized plunger valve based on remote measurements of pressure acquired in real time. An integral-type algorithm was considered for the implementation of the closed-loop control of the valve. A

Funding

This research was partially supported by the Department of Civil Engineering and Architecture of the University of Catania, under the Research Programme “Piano Triennale della Ricerca 2016/2018”.

CRediT authorship contribution statement

Alberto Campisano: Conceptualization, Investigation, Methodology, Writing - review & editing. Carlo Modica: Conceptualization, Investigation, Methodology, Writing - review & editing. Fabrizio Musmeci: Investigation, Software. Camillo Bosco: Software. Aurora Gullotta: Writing - review & editing.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

References (38)

  • E.F. Brater et al.

    Handbook of Hydraulics for the Solution of Hydraulic Engineering Problems

    (1976)
  • A. Campisano et al.

    Potential and limitations of modern equipment for real time control of urban wastewater systems

    Urban Water J.

    (2015)
  • A. Campisano et al.

    RTC of valves for leakage reduction in water supply networks

    J. Water Resour. Plann. Manag.

    (2010)
  • A. Campisano et al.

    Field-oriented methodology for real-time pressure control to reduce leakage in water distribution networks

    J. Water Resour. Plann. Manag.

    (2016)
  • A. Campisano et al.

    Calibration of proportional controllers for the RTC of pressures to reduce leakage in water distribution networks

    J. Water Resour. Plann. Manag.

    (2012)
  • C. Ciaponi et al.

    Procedure for defining a pressure-outflow relationship regarding indoor demands in pressure-driven analysis of water distribution networks

    Water Resour. Manag.

    (2015)
  • E. Creaco

    Exploring numerically the benefits of water discharge prediction for the remote RTC of WDNs

    Water

    (2017)
  • E. Creaco et al.

    Unsteady flow modeling of pressure real-time control in water distribution networks

    J. Water Resour. Plann. Manag.

    (2017)
  • E. Creaco et al.

    Testing behavior and effects of PRVs and RTC valves during hydrant activation scenarios

    Urban Water J.

    (2018)
  • Cited by (0)

    View full text