Although water conduit systems around the world could generate significant amounts of hydropower, developers have yet to largely tap that potential. Using examples in South Africa, a planning tool was created to help guide growth in this exploitable sector.
By Ione Loots, Marco van Dijk, Stefanus van Vuuren and Jay Bhagwan
Rapid industrial growth in South Africa has created an energy crisis that came to a peak in 2008, when rolling power cuts started to be implemented around the country. Since the crisis in 2008, load shedding has been implemented numerous times, resulting in many businesses, factories and mines closing down between 2008 and 2015. The addition of electricity generating capacity from all available and feasible renewable sources – including hydropower – is therefore vital.
Currently, hydro contributes about 17% of global energy production, and Africa is the most underdeveloped continent with regard to hydro generation, with only 5% of its estimated potential exploited.
The energy found in water supply conduits is often disregarded as a source of electricity, but numerous transfer schemes and distribution systems offer potential. Pressure-reducing stations are installed to dissipate excess energy along a conduit, as well as upstream of water treatment plants and reservoirs. The energy dissipated could instead be used to generate electricity through the installation of hydro turbines in the conduit.
|Data loggers were installed to take pressure and flow measurements at Waterkloof Reservoir’s pressure-reducing station to project its hydro generating potential.|
A scoping investigation performed in 2010 indicated that there is substantial potential for pico (defined as up to 20 kW), micro (21 kW to 100 kW) and mini (101 kW to 1 MW) hydro installations in water distribution and transfer systems in South Africa.1 The country has 278 municipalities and various water supply utilities, almost all of which have pressure-dissipating stations in their water distribution systems.
Currently there is no substantial development in South Africa’s conduit hydro market, despite its significant potential, due largely to a lack of knowledge and technical understanding. It is difficult to quantify this potential, however, as there is no database with all the flow and pressure data for South Africa. Rather, the need for this research was to empower utilities to discover their potential, thus the need for a user-friendly system to guide municipalities and potential power producers through the process of developing conduit hydropower.
The University of Pretoria, with support from the Water Research Commission, engaged in a project to investigate the potential of extracting energy from existing and newly-installed supply and distribution systems. One of the products of this research was the creation of the Conduit Hydropower Decision Support System (CHDSS), which was designed to help facilitate the conduit hydropower development process in South Africa.
The Conduit Hydropower Decision Support System
A systematic approach must be followed when assessing hydropower potential in a distribution network to ensure that all relevant factors are considered. The procedure for determining potential, developed by Ione Loots in 2013, was summarized in a series of flow diagrams, while a tool developed in Microsoft Excel facilitated the calculation of all factors that needed consideration.2
Editor’s Note: The full report with examples of the researchers’ flow diagrams can be downloaded here: http://bit.ly/HRWConduits
The development procedure consists of three stages:
- Phase 1: Pre-feasibility Investigation: This pre-feasibility study comprises various first-order analyses and studies to rapidly determine whether more in-depth studies would be worthwhile. The input required for this phase is limited, so that a first-order assessment can be done without spending excessive capital. Important information includes average values for flow rate and pressure head, as well as determining where the electricity would be used and how much is needed. Phase 1 analysis can also determine a net present value (NPV), internal rate of return (IRR) and payback period.
- Phase 2: Feasibility Study: If Phase 1 indicates project viability, a more in-depth investigation can be done, using as inputs measured flow and pressure records. An initial turbine selection, as well as a more detailed economic analysis to obtain an NPV, IRR and payback period can also be determined during this phase. In addition, Phase 2 encourages environmental and regulatory studies.
- Phase 3: Detailed Design: If the previous stages demonstrate viability, a detailed design for the hydropower plant can be completed.
Testing the CHDSS
This section illustrates the application of the procedural method in a municipal context. The research project was conducted in Tshwane, with all analysis performed on sites within the city’s Metropolitan Municipality Bulk Water Services distribution network.
A significant portion of Tshwane’s water demand is supplied by Rand Water. The water gravitates from Rand Water’s storage facilities in Johannesburg to the relatively lower-lying hills in Tshwane. The city’s water distribution network consists of 165 reservoirs, 39 water towers, more than 10,000 km of pipe and more than 280 pressure-reducing stations. Consequently, there are many sites within Tshwane that may have exploitable conduit hydropower potential.
Two case studies were used to test the applicability of the CHDSS. The sites were selected to represent a variety of circumstances that would enable comprehensive testing of the procedural method and are therefore not necessarily two of the sites with the highest generating potential in Tshwane.
Water to Garsfontein Reservoir is supplied from Rand Water sources and the Rietvlei Dam. Garsfontein is a bulk reservoir in the distribution network, supplying a significant portion of the eastern Tshwane metropolitan area.
Phase 1: The information used was obtained from the city’s infrastructure management query station (IMQS) database. Suitable flow rates and pressure heads for hydropower generation could not be guaranteed at this site for future scenarios. It was assumed that parallel pipes would be installed in the future so that the final conditions at the site would have twice the current flow but the same corresponding pressure heads.
However, as this is only a rough estimate, future upgrades were not allowed for in the economic analysis and a design life of only 15 years was selected to determine economic feasibility. If future conditions prove to be positive, an additional feasibility study for the expansion could be done in years to come. This could almost be seen as a separate project and would therefore not have an impact on the feasibility of the project already analyzed. With an IRR of 22% and NPV of more than ZAR25 million (US$1.92 million), the Phase 1 analysis indicated a full feasibility test should be undertaken.
|An existing pico hydropower turbine installed at the Pierre van Ryneveld Reservoir near Tshwane was used by researchers as a case study for small conduit hydropower.|
Phase 2: It was necessary to visit the site and assess the practicability of a hydropower plant there. Considered aspects included: space for the plant, safety of the turbine and other equipment from theft or vandalism, noise impact on the surroundings, and accessibility to the site during construction. Once site practicality was established, instrumentation was installed to measure flow and pressure in the system.
A potential challenge with bulk supply systems is that in some cases, they supply multiple reservoirs and therefore are not just a simple system with one inflow and one outflow. In a simple system, there will be a specific inverse relationship between flow and pressure head, with a flow rate always associated with the same head. However, in a complex system with various independent inflow and outflow points, as in the case of Garsfontein Reservoir, a specific correlation will not be found between flow rate and pressure head. It was therefore necessary to carefully select the design pressure head in this system.
The next step was to perform an economic evaluation. The CHDSS was used to determine an NPV of almost ZAR27 million (US$2 million) and an IRR of 21.5%. Therefore, economic analysis indicated a detailed design was warranted.
Phase 3: The effect of system optimization had to be considered. Flow in the pipe was normally controlled at 3.6 m3/h or 5.1 m3/h until the reservoir was full, at which stage the flow in the pipe became almost zero. However, hours with high power potential did not typically correlate well with hours of high electricity values. Therefore, operational changes to ensure better correlation between the peak generating periods and the higher income potential peak electricity periods by using the available reservoir storage would produce higher revenue.
|A primary consideration is determining whether a site could accomodate a hydropower plant. Open space at Garsfontein Reservoir, which supplies the eastern Tshwane metropolitan area, helps make it attractive.|
However, as Garsfontein Reservoir serves various distribution zones and other reservoirs in Tshwane, it was not advisable to adjust the operational philosophy of the reservoir significantly to obtain a more constant flow, as this might have a detrimental effect on downstream water supply.
According to the IMQS, the maximum future average annual daily demand (AADD) at Garsfontein Reservoir will be about 176,000 kl/day, which is double the current AADD of 85,000 kl/day. For this analysis, it was assumed that parallel pipes would be installed in the future so that final conditions at the site would meet the future AADD, but with the same corresponding pressure heads.
A detailed economic evaluation was conducted with obtained costs, where applicable. A sensitivity analysis was also conducted to determine the sensitivity of its feasibility when considering alternative inflation rates. The analysis showed there is economically exploitable potential at the site, with an expected NPV of ZAR295 million (US$22.6 million) and an IRR of 22%. It is proposed that a 730 kW grid-connected Francis turbine or pump-as-turbine be installed based on the current prevailing conditions, with space allowed for duplication of the capacity for future extensions.
The major difference between this site and Garsfontein is that Waterkloof Reservoir only supplies potable water to a single water distribution zone. Operational changes were therefore considered for this site, and the ensuing discussion will focus on this aspect.
Phase 1: This reservoir supplies water to a developed residential area and little change in future scenarios is expected. Therefore, a design life of 30 years was selected. Even though the first phase analysis only indicated an IRR of 12%, a full feasibility study was done because operational changes could have a positive impact on the economic feasibility of the project.
Phase 2: Instrumentation was installed to measure flow and pressure in the system. Flow and pressure data was collected at the pressure- reducing valve in the valve chamber upstream of the reservoir.
It was noticed that changing the operational procedure to maintain the average flow rate would generate significantly more power on an annual basis, even though a smaller turbine was required. Research showed an 84 kW turbine at a 98% load factor could produce 720 MWh annually, whereas a 125 kW turbine at a 40% load factor could only produce 506 MWh annually.
|Conduit generating potential is being noticed by municipalities around the world, some of which have already begin installing generating equipment like this system, owned by the city of Logan in Utah, U.S.|
With an NPV of almost ZAR17 million (US$1.3 million) and an IRR of 28%, the Phase 2 analysis indicated that a detailed design was warranted. It should be noted that this phase indicated a significantly shorter payback period – seven years – compared to the value calculated in Phase 1, or 15 years. The main reason for this is that operational changes to the system resulted in a better load factor about 98% of the time, allowing more electricity to be generated annually.
Phase 3: Flow to the reservoir is normally controlled at about 780 m3/h until the reservoir is full, at which stage the flow in the pipe becomes almost zero. Hours with high power potential do not typically correlate well with hours of high electricity value. Therefore, operational changes to ensure better correlation between generation and the more lucrative peak hours would result in a 17% increase in annual electricity generation.
As Waterkloof Reservoir serves only one distribution zone, the operational philosophy could be adjusted to obtain a more constant flow, with higher flow values at periods of peak electricity demand, all the while ensuring the reservoir levels remain within an acceptable range to prevent the potential risk of non-supply. The potential analysis was therefore done for constant flow rates.
Based on information in the IMQS, the maximum future AADD at Waterkloof Reservoir would be about 12,000 kl/d, which was roughly 30% more than the current AADD of 9,000 kl/d. Operation with the current flow could be manipulated so that an above-energy current flow (equal to the future average flow) could be maintained during peak energy tariff hours, with a slightly below-average flow in off-peak times. In this way, a turbine with a capacity to match future flow rates could be installed now and future expansion would not be necessary. A flow range including the current and future average flows was therefore used and a Pelton turbine was recommended.
A detailed economic evaluation was conducted with obtained costs, where applicable. The future flow scenario was applied from 2029, or 13 years from the project commencement.
A sensitivity analysis was also conducted to determine the sensitivity of project feasibility when considering alternative inflation rates. The analysis of hydropower at Waterkloof Reservoir showed there is economically exploitable potential at this site, especially if operational changes are made.
Conclusion and recommendations
Hydropower represents a nexus of water and energy. A technically feasible scheme assists in reducing the higher operating costs, mainly due to energy increases, and provides a sustainable solution while having a positive environmental impact.
Guidelines detailing the decision support system for energy generation have been compiled to assist municipalities in South Africa to understand the process involved in developing conduit hydropower facilities in their water supply systems. The aim is to provide the necessary guidance in the development process and efficient operating of conduit hydropower schemes.
The guidelines are now available via the Water Research Commission’s website, www.wrc.org.za, and various municipalities and utilities have begun implementing the reccomendations.
Already, utility Bloemwater has installed a 96 kW, and Tshwane has installed a 150 kW plant and begun tendering for a 2.4 MW plant. Rand Water and Johannesburg Water have also put out tenders for 12 MW and 5 MW projects, respectively.
Internationally, these principles can be applied anywhere where there is excess pressure in a pipeline. Similar plants have been installed in places like Switzerland, Austria, Italy, Spain and the United States.
. 1Van Vuuren, S.J. “A High Level Scoping Investigation into the Potential of Energy Saving and Production/Generation in the Supply of Water through Pressurized Conduits,” WRC Report No. KV 238/10, Water Research Commission, 2010.
. 2Van Vuuren, S.J., Van Dijk, M., Loots, I., Barta, B. and Scharetter, B.G. “Conduit Hydropower Development Guide,” WRC Report No. TT597/14, Water Research Commission, 2014.
Ione Loots, Marco van Dijk, Stefanus van Vuuren are professional engineers and members of the faculty at the University of Pretoria. Jay Bhagwan is a civil engineer and the executive manager of Water Use & Waste Water Use for the Water Research Commission.