Daniele Novara
Pumps As Turbines (shortened as “PAT”) consist of regular water pumps running in reverse as turbines, and therefore generating power from a stream of pressurized water. How do these devices operate when installed within a water network in parallel to a Pressure Reducing Valve (PRV)? And how to design a system able to cope with sudden variations in water pressure and flow rate?
Whether you are a regular visitor of the Dŵr Uisce website or just someone who occasionally checks the project updates, you may be already familiar with the concept of using “Pump as Turbines” (in short, “PAT”) to generate power from water networks. In short, these devices consist of regular water pumps which are utilized in reverse to generate electricity from a pressurized water stream as an alternative to conventional (and expensive) custom-made water turbines. These devices can be used to produce carbon-free, clean and renewable power with a low installation cost and ease of maintenance and they are particularly suitable for integration with existing water infrastructures.
In fact, most water transport or distribution systems such as drinking water, irrigation or industrial cooling networks will typically have nodes at which pressure must be reduced in order to avoid leaks and pipe bursts. This is normally achieved via a Pressure Reducing Valve (PRV) which dissipates the excess water pressure as heat and noise. However, a Pump as Turbine can be inserted in parallel to such PRV and recover a portion of the dissipated pressure as useful electricity. As a consequence, the power generated by the PAT will offset the electricity consumed by the whole water network in a circular economy approach.
The Dŵr Uisce research team over the recent years has focused extensively on several aspects of the PAT technology which were previously unknown, helping the scientific community as well as the general public to learn more about this class of devices and their application. These efforts culminated in the construction of a hydraulic test rig at Trinity College Dublin to test the performance of centrifugal PATs and eventually to the installation of two devices in Ireland and Wales. However, both these pilot plants are located at sites which offer a “conventional” hydropower layout where a portion of the water flow of a river is diverted into a pipeline and eventually across the turbine. Therefore, neither sites are located in a fully pressurized water network in parallel to a PRV as mentioned in the previous paragraph, which would pose additional challenges in the system design. Among these challenges, the main one is to ensure at any time that the water flow is never disrupted by the presence of the PAT even under exceptional circumstances. This translates into the need of a bypass pipeline which diverts from the turbine the water that can’t be processed by it. However, at the same time it is also important to minimize at all times the amount of water that bypasses the PAT since this results in an energy loss.Despite the fact that over the last decade there has been a number of scientific publications investigating the behaviour of PATs in water networks, most of them either did not include experimental data or utilized a sophisticated computer-operated bypass valve which greatly increased the complexity of the system. As opposed to this approach, what if a conventional PRV could itself be used as the bypass valve without the need of an additional element? How well can a PRV act as a turbine bypass valve, and how finely is it possible to tune it?
In order to answer these questions, the existing hydraulic test rig at Trinity College Dublin has recently been upgraded with a PRV in parallel to a PAT (Figure 2) and tests are ongoing to evaluate the interactions between the two devices under varying flow and pressure conditions.
Figure 2: Upgraded test rig with PAT in parallel to a PRV
