Experimental characterization of two Pumps As Turbines for hydropower generation


Pumps As Turbines (PATs) are effective for energy recovery in Water Networks.

Literature is fairly lacking in predicting PAT performance curves.

Experiments were carried out to assess the validity of existing relationships.

Data allowed to obtain more general relationships than existing ones.


In recent years, the use of turbines or Pumps operating As Turbines (PATs) has been proven to be a sustainable alternative for managing Water Distribution Networks (WDNs), by coupling pressure control and leakage reduction with hydropower generation.

Pumps running in reverse mode can be an effective alternative to using turbines for energy production in WDNs. Many commercial models are readily available on the market and a number of economic and technical advantages for installation, operation and maintenance can be found. Theoretical and experimental criteria for predicting pump performance in turbine mode and for the optimal installation of a PAT in WDNs can be found in the literature. Nevertheless, the prediction of PAT characteristic curves is still an unresolved issue, because of the lack of information provided by manufacturers and the few laboratory campaigns that focus on the topic.

For this purpose, the laboratory results in the present study aim to assess the performance of pumps operating in reverse mode. Two centrifugal pumps were investigated: a centrifugal horizontal single-stage pump and a vertical multi-stage pump. Experiments were compared with theoretical models available in the literature, in order to assess their reliability in predicting PAT performance when data are lacking.


  • Pump As Turbine;
  • Hydropower;
  • Experimental data;
  • Water Distribution Network;
  • Centrifugal pump;
  • Best Efficiency Point


  • PAT, Pump As Turbine;
  • WDN, Water Distribution Network;
  • PRV, Pressure Reducing Valve;
  • BEP, Best Efficiency Point;
  • CFD, Computational Fluid Dynamics;
  • PLC, Programmable Logic Controller;
  • SCADA, System Control And Data Acquisition

1. Introduction

Hydropower generation is a relatively recent topic in the field of Best Management Practices of Water Distribution Networks. It defines a relevant category in the operative approach of leakage control, focused on the recovery of the excess pressure [1], by generating electric power [2].

Pressure Reducing Valves (PRVs) are often used in WDNs to prevent the downstream hydraulic grade from exceeding a set value. Nevertheless, the excess head can be exploited for hydropower generation by using turbines and/or Pumps As Turbines (PATs) [3]. PATs are pumps running in reverse mode, by inverting flow direction and using the electric motor as a generator [4]. The possibility of using pumps operating in turbine mode has been widely accepted since the third decade of XX Century [5] and [6] but only in recent years it was pointed out the benefit of their use in WDNs [7], [8], [9], [10], [11] and [12]. Jain and Patel [13] provided a comprehensive review of the state-of-the-art of PATs, summarizing the main researches carried out.

PATs generally exhibit worst performance against reaction micro-turbines but, at the same time, investment and maintenance costs are largely lower [14]. PATs also show the benefit arising from the wide set of pump models commercially available, with easier installation, maintenance activities and availability of spare parts [15]. Various researchers indicated the single-stage centrifugal pumps, operating in the range of low to medium head, as the most convenient, from both technical and economic standpoint [16], [17], [18] and [19].

On the other hand, the range of flow rates over which a single PAT unit can operate is much smaller than in a conventional turbine. This issue, critical in WDNs, can be overtaken by coupling a by-pass line to the main power generation line where the PAT is installed for energy production. On both lines, a PRV is installed, for pressure and flow regulation [10] and [14]. At the same time, higher operative adaptability can be obtained by installing a frequency converter to regulate the rotational speed at varying inflow discharge [9].

From a technical viewpoint, the main issue of selecting PATs is the lack of information, as performance curves are rarely made available from manufacturers. In order to obtain this information, some authors developed experimental and theoretical models to assess PAT performances. Almost all these models were introduced as function of the Best Efficiency Point (BEP), which represents the operating condition with maximum efficiency.

By using a semi-empirical one-dimensional approach, some authors [20], [21], [22], [23], [24] and [25] developed models to predict discharge and head drop ratios in turbine and pump mode, as a function of the efficiency ηp of the machine in direct mode. At a generic operating point, the efficiency η can be calculated, for pumps and turbines (or PATs), respectively as:





with γ specific gravity, Q flow discharge, H hydraulic head, P used or generated power. The subscripts p and t refer to the pump mode and turbine mode, respectively.

Hancock [26] correlated discharge and head drop ratios with the efficiency in turbine mode ηt. Schmiedl [27] correlated discharge and head drop ratios to the hydraulic efficiency of the machine ηhp, which was calculated as follows:


View the MathML source

View the MathML source

Be the first to comment

Leave a Reply

Your email address will not be published.