By S. Thomas Lavender and C. Richard Donnelly
Conventionally, most project developers considering a large hydroelectric development set a primary goal of optimizing the amount of energy derived from the site. The theory behind this focus is that there will not be opportunities for future expansion once the development is complete. Developers often believe that leaving some of the potential “on the table” would result in a significant long-term loss.
This traditional approach involves:
- Appraising the site’s geographic location and its total developable energy, based on hydrology and topography;
- Determination and arrangement of the project concepts in accordance with conventional engineering design criteria to utilize the full site potential;
- Dimensioning of the project components to realize the full site potential, including some engineering economic optimization of dam height and installed capacity;
- Preparation of a cost estimate for the concept;
- Calculation of the cost of energy from the proposed development; and
- Economic and financial evaluation of the proposed development.
In this approach, the project is scaled to provide the “engineering” economic optimum for both dam height and installed capacity. Following this step, the overall project economic and financial viability is determined. Therefore, the most significant factor in assessing the feasibility of the development is not established until some significant expenditures have been made.
Unfortunately, all too often for small- to medium-sized developments, the economic and financial characteristics of resources scaled in this way are found to fall short of acceptable values. Therefore, such projects are deferred for reconsideration at a later date when, perhaps, economic or political factors are deemed more favorable. However, performing the reassessment following the same traditional approach often leads to the same general conclusions. And the cycle repeats.
In some cases, the costs associated with such repeated studies may have been sufficient to have built a portion of a smaller-scaled hydro station. Equally disappointing is the reality that a potentially valuable renewable (i.e., “green”) energy resource has gone unused.
An alternative approach
To eliminate this unproductive and costly cycle of study and deferral, Hatch Energy developed an alternative procedure that can reduce upfront study costs and the time required to reach go/no-go decisions. This procedure involves initial planning based on optimizing the economic viability of a given site, even if this results in a decision to develop only part of the total site potential. The intention is to determine what, if any, development would be possible that would provide an acceptable return to a given developer. Thus, upfront costs can be minimized with the focus of the feasibility assessment based, initially, on return on investment rather than technical issues.
Figure 1 on page 36 shows the general logic of this procedure. Information used includes hydrology, a specified economic energy tariff, and investment parameters provided by the developer (rate of return on owner’s equity, investment leverage, interest rates, average cost of capital, tax rates, interest during construction, etc.). From this information, a domain of installed capacities per unit of head developed and costs per unit of installed capacity can be determined. These should result in a development that provides the specified (or better) rate of return. Because this process is largely an “on-paper” exercise, the costs associated with assessing non-viable options are reduced. The subsequent technical challenge is to engineer a site development that falls within the previously prescribed economically and financially viable “installed capacity-unit cost of capacity” domain.
Assessing the potential of the Pra River in West Africa
To assess the hydroelectric potential of the Hemang project site on the Pra River in Ghana, West Africa (see Figure 2 on page 38), Hatch Energy undertook a six-step procedure:
- 1) Description of the project concept developed in conventional studies;
- 2) Development of a spreadsheet financial model to aid analysis;
- 3) Financial analysis of the conventional development concept;
- 4) Use of the financial model to define the range of viable development scales;
- 5) Reconnaissance level layouts; and
- 6) Preliminary costing.
This led to the establishment of the parameters to be satisfied in order to achieve an economical development. Armed with these parameters, the engineers could focus on establishing solutions and economies that will allow development of the site.
The conventional development
The site subtends a drainage area of 22,200 square kilometers that yields a 58-yearaverage runoff of 184.6 cubic meters per second (cms). The total fall of the river through the study reach is about 30 meters over a river length of about 20 kilometers. Almost one-half of this fall is concentrated at a 5-kilometer-long series of small rapids.
Site development had been studied three times once at prefeasibility level and twice at feasibility level all using a basic concept that would maximize the hydroelectric potential of the site. This scheme comprises:
- A 42-meter-high dam;
- A three-unit, 93-mw close-coupled power station, with a maximum head of 30 meters and total station flow of 350 cms;
- A gated overflow spillway of 3,550 cms capacity; and
- One kilometer of transmission and substation interconnection.
The dam would create a reservoir of 39 square kilometers and total storage of 345 cubic millimeters, extending some 18 to 20 kilometers upstream. It would drown out significant arable land (2,600 hectares) and require relocation of part or all of some seven communities (6,400 persons). For this reason, the scheme had considerable environmental and social costs and, at best, the storage created would provide for weekly flow regulation and the station would be run-of-river. Its principal value would be to provide energy to the grid to reduce fuel expenditures at a complementary thermal plant in the system.
In 2000, the capital cost for the project was estimated to be US$205 million, yielding an all-in cost of US$2,204 per kilowatt of installed capacity. For the various financial parameters determined to be applicable by the developer, the internal rate of return (IRR) on the owner’s equity for the project, as conceived, was unacceptable.
The only option for attaining financial viability would be to obtain a wide range of reduced financial expectations and concessions by negotiating:
- A tax concession, from 30 percent to 10 percent;
- A lower interest rate, from 15 percent to 5 percent;
- A lower equity requirement of 20 percent versus 33.3 percent;
- A longer amortization period (i.e. plant life), extending from 25 years to 50 years; and
- A subsidized or more favorable “green” energy tariff, from $0.0550 per kilowatt-hour to $0.0608 per kilowatt-hour.
These significant requirements resulted in repeated deferral of the project, despite the fact that the country needed energy.
Re-assessment based on financial considerations
To quickly assess a range of alternatives, Hatch Energy developed a financial model that defined scales of development that would be financially viable for a specified unit cost of installed capacity. In this way, a “domain” of scales of development (i.e., installed capacity) could be defined. The methodology used is based on a simplified, pre-computer-age, flow duration curve analysis. The use of spreadsheets to affect analyses allows the consideration of a large number of alternatives (e.g., different dam height/storage volume configurations), in short facilitating the development of relationships such as that illustrated in Figure 3 on page 39.
The flow duration curve analysis is executed on a per-unit head basis to simplify the calculations. By this method, the hydrologic potential of the site (megawatts and megawatt-hours [mwh] per meter of head developed) can be evaluated on the basis of the hydrologic record only, with a particular head being decided subsequently on the basis of topography and other practical considerations (e.g., minimum operating head for a particular type of turbine). This method, while perhaps “crude” compared to more elegant computer-age models, is adequate for the purpose of site reconnaissance. This analysis is not intended to be definitive for proceeding with design and financing. Rather, it should quickly and inexpensively establish the scale of development that is likely to be financially viable before undertaking the necessary costly field and design studies.
For the selected financial parameters and assumptions made for this project, Figure 3 depicts a range of installed capacities that could be developed that would satisfy the specified financial hurdles.
In reviewing these results, it is noted that the negative slope of the relationship is a manifestation of the fact that the site has a fixed total potential energy (i.e., average flow rate times head) and, therefore, fixed total potential revenue from energy sales. As a consequence, as investment in any non-revenue-generating, but necessary, components (such as transmission lines, dams, canals, spillways, access roads, powerhouse civil works, etc.) increases, there must be a corresponding decrease in the funds available for investment in turbine-generator capacity.
The non-linearity of the relationship is a consequence of the non-linearity of the hydrology. That is, an additional increment of installed capacity will not have the same degree of utilization and, therefore, will not generate the same amount of revenue as the previous increment because of the natural variability of flow in the river. It also is noteworthy that the conventionally conceived scheme provides financial returns that are significantly less than the desired hurdle. Therefore, the challenge was to determine if the reduced capital and social costs resulting in a smaller scaled project could be sufficient to provide the desired IRR of 25 percent.
Of course, the results would change depending on the alternative being considered, particularly if a given alternative would provide significantly more storage. This would allow for more generation during peak periods and, therefore, more revenue. In the present analysis, the screening of the alternatives is made purely on energy revenues, without any attempt to differentiate between off- and on-peak revenues. In general, for small to medium-sized projects that are “run of river,” the simplification is acceptable. For projects that have storage potential, the spreadsheet analysis can be modified to account for peaking.
As a first step in evaluating an alternative financially viable scheme, the distribution of costs of the conventional project, as shown in Table 1 on page 40, were examined as a tool for identifying where costs might be reduced for a smaller scale development.
The two categories that show the least scope for cost reduction are the general works and generating works. The general works are relatively small to begin with, and the “dimensions” of the mechanical and electrical equipment are to some degree site-independent, as the type of equipment (although not the capacity) had been previously optimized. While there are cost reductions associated with selecting a smaller turbine, as well as the potential for cost reductions through strategic location and design simplification, such savings generally would be insignificant compared to the total cost reductions that were required to make this project viable.
On the other hand, the water control works were, in this case and in general, very dependent on the project scale and nature of the site. To achieve the cost reductions needed, engineering designs needed to be focused on defining innovative alternativesthat would take advantage of site conditions using the smallest possible civil works while still developing the available head and safely passing the project design flood. Such designs also would result in automatic reductions in the engineering and management and contingencies items.
In addition, more efficient, compact designs could significantly reduce social and environmental mitigation costs. In particular, a revised project layout that confined the headpond area upstream of the flow diversion works within the normal river flood plain would largely avoid both resettlement costs and loss of arable land for continued agricultural production.
Costing the alternative development
For the site of the previously conceived scheme, we chose three possible project arrangements that would achieve the goal of confining the resulting headpond within the normal flood levels of the river (see Table 2 on page 40). All of these sites were chosen to take advantage of head gained from different steep sections of the river.
To determine the relative financial merits of these alternatives, a costing model was developed with project characteristics for Alternative 1 represented. The model uses a regression equation for the relationship that was established in Figure 3 and a prescribed gross head to determine the scale of development (i.e., installed capacity) that will yield the specified rate of return on owner’s equity. The analysis clearly showed that Alternative 1 provided the best returns, with the optimum scale being 7.53 mw at a unit cost of $1,464.14 per kilowatt. In the other two alternatives, the cost of the longer canals overwhelmed the benefits, rendering them financially unviable. This desktop study therefore allowed one of the design “criteria” for the development to be clearly defined. Layouts must minimize or, preferably, eliminate the requirements for canals.
Using the model, it was established that the project should be developed with an installed capacity of 7.18 mw for the specified unit construction cost of $1,464.14 per kilowatt. Using these results, it is possible to outline a “design guideline” for the engineering team that will direct their conceptual level assessments during a visit to the site in the ongoing studies.
Guidelines for developing conceptual layout
Following are the assumptions and parameters used in developing a financially viable scheme:
- 1) Installed capacity = ±7 mw
- 2) Head = ±10 meters
- 3) Turbine design flow = ±80 cms
- 4) Canal length <200 meters
- 5) Spillway capacity = ±3,550 cubic meters (previously noted from earlier hydrologic studies)
6) Total unit construction cost <$1,450/kw (i.e., <$10.5 million) possibly broken down as:
- Generating works <$700/kw (i.e., <$5.25 million),
- Civil works <$500/kw (i.e., <$3.75 million),
- Other costs <$200/kw (i.e., <$1.5 million).
The ongoing challenges
At this point, the developer has a good idea of an attractive scale of development at this site. The developer also knows the amount of equity and financing that must be arranged and the limits on the terms of any arrangements for these and can make the decision, or not, to continue with the project. The ongoing challenge, then, is mobilizing the required financing and “putting in place” all required permits and approvals.
In addition, this approach provides the engineeringdesign team details of the necessary size of structures(powerhouse and water control works) required before the team enters into the field investigation and preliminary design phases. Their challenge will be to visit the site to identify a general arrangement for the project components that best utilizes the natural site features to minimize the size and number of structures to capture the flow and develop the required head. The ongoing engineeringchallenge will be to realize component designs within a general arrangement that result in constructioncosts equal to or less than those indicated in the guidelines.
Using conventional means of evaluation, development of the full potential of this hydropowersite proved uneconomic and financially unviable in repeated studies. Thus, all of the potential of the site, first identified in studies almost 20 years ago, has remained unutilized. If a 7.18-mw project had been developed 20 years ago, as our alternative analysis method indicates would be economically viable, the project would have yielded 960,000 mwh of economic green energy to date with an attractive financial return.
This is a rather drastic example due to financial thresholds that are quite high with respect to parameters that would be used in North America or Europe. Of course, relaxing these criteria would result in a progressively larger installedcapacity. However, in this example, in the absence of such concessions, there would be no point in executing expensive field programs for a larger scale of development. One of the values of the model is that it can be used to try out the “what-ifs?” What if:
- Tax concessions can be obtained for clean energy development?
- More favorable tariffs can be realized for green energy?
- Financing can be obtained with less owner equity?
- A public-private partnership or other ways of obtaining “grant or cheap money” to build some of the infrastructure that could have other economic benefits can be found?
This stepwise approach can be used to assess the likely range of potential developments that could be justified financially, allowing subsequent field investigations and studies to focus on appropriate scales of development. While such extreme changes in scale at other sites around the world are unlikely, using this approach to reconsider sites previously considered to be uneconomic might allow additional green energy to be generated in areas that often have a real need for clean power.
It is of note that this example of a small-scale scheme derived on the basis of financial scaling is reminiscent of the small-scale watermills developed to provide local facility for grinding of grain before the advent of electrical power. In fact, many of these old mill sites, long abandoned, have been and are being rehabilitated and retrofitted to provide economic green energy to the electrical power grid system with attractive financial returns. s
Messrs. Lavender and Donnelly may be reached at Hatch Energy, 4342 Queen Street, P.O. Box 1001, Niagara Falls, Ontario L2E 6W1 Canada; (1) 905-374-0701, extension 5204 (Lavender) or extension 5303 (Donnelly); E-mail: [email protected] or [email protected].
Tom Lavender, formerly senior project manager in the planning division of Hatch Energy (a member of the Hatch group of companies), developed the methodology discussed in the article. Rick Donnelly, P.E., director for water and wind power at Hatch Energy, assisted in the documentation of the methodology.
This article has been evaluated and edited in accordance with reviews conducted by two or more professionals who have relevant expertise. These peer reviewers judge manuscripts for technical accuracy, usefulness, and overall importance within the hydroelectric industry.