The Altener Committee of the European Union and the European Small Hydro Power Association (esha) are keen to promote investment in small run-of-river hydro power schemes. Such schemes can provide a reliable source of energy, especially in remote areas, provided that the hydrological conditions are suitable. However, in such remote locations, the availability of reliable measured flow records is severely limited.

To address this problem a software package has been developed by the Institute of Hydrology on behalf of ESHA, providing an automated procedure for assessing the hydrological potential of a site. Once the hydrological characteristics have been determined, the feasibility of developing small hydro schemes in these areas can be evaluated. The software is currently operational in the UK and Spain and a version covering Italy will be available in 1998.

Called the Hydro Power Atlas (HydrA), the software is a menu-driven package which incorporates regional flow estimation models and databases of climate and hydrogeological response characteristics. The automatic estimation procedures are ideal for the rapid hydrological assessment of sites for regional planning and for hydropower assessment at gauged and ungauged locations. The program enables the user to:

•Compare the power output for different turbine types at different locations.

•Suggest preliminary designs for the sites, including the most suitable turbine and capacity.

•Select key sites for more detailed investigation.

•Potentially reduce expenditure on unnecessary hydrological surveys.

Various hydrological principles are incorporated into HydrA. The hydrology of a catchment is an important factor when considering whether a site is a suitable location for a small hydro power scheme — there must be a reliable source of water to drive the turbines. Within a catchment, the rainfall and evaporative losses are important in determining the total available water resource, while the soils and underlying geology will influence the way in which the catchment responds.

A generally accepted method for characterising the hydrological regime is to determine the cumulative frequency distribution of flows expressed as a percentage of time that the specified flows are equalled or exceeded. This relationship is more commonly referred to as the flow duration curve and, although it does not convey any information about the sequence of flows, it graphically represents the complete range of flows from low to flood.

The flow duration curve can be derived directly from daily flow data by assigning daily flow values to class intervals and counting the number of days within each class interval. The proportion of the total number of days above the lower limit of any class interval is then calculated and plotted against the lower limit of the interval. If the flows are normally distributed, plotting the points using a normal probability scale for the frequency axis and a logarithmic scale for the flow axis, will define the flow duration curves as a straight line. In addition, plotting the flow axis as a percentage of the mean flow enables catchments of different sizes and with different climatic conditions to be compared.

In the areas that might be suitable for hydro power development, such as remote upland areas, gauged flow data is not available for deriving the flow duration curve – alternative techniques are required.

The shape of the flow duration curve is a function of the catchment hydrogeology. This observation has been used to form the basis for developing an estimation procedure at ungauged locations.

Previous studies have identified that the key flow statistics required for the estimation of the flow duration curve at an ungauged location are the mean flow and a standardised low flow statistic (expressed as a percentage of the mean flow). In the UK, the Q95 flow (the flow equalled or exceeded for 95% of the time) is the low flow statistic most commonly used by the water industry. However, in drier countries such as Spain and Italy, the Q90 provides a more reliable low flow statistic (the Q95 in these countries is often zero).

Ungauged locations

In order to be able to implement the design procedures in ungauged locations, it is first necessary to identify the extent of the catchment, by defining a catchment boundary, and to determine the physical characteristics of the catchment. The catchment boundary is overlain onto maps of the relevant catchment characteristics. Average values of rainfall and evaporation can easily be calculated and the fractional extent of each soil class within the catchment can be determined. The soil classes can then be grouped into hydrogeological units, based on similar characteristics and similar low flow responses.

At ungauged locations, the long-term mean flow can be estimated directly using a simple relationship incorporating the annual average rainfall and the potential evaporation, scaled by the catchment area. A simple linear model relates the standardised low flow statistic to the fractional extent of each hydrogeological unit.

Since the shape of the flow duration curve is controlled by the hydrogeological response of the catchment, by grouping together flow duration curves derived for gauged locations with similar hydrogeological characteristics, a family of flow duration curves can be derived. These curves represent the typical response of catchments with different geological characteristics. Therefore, at an ungauged location, the selection of appropriate curves can be determined based on the magnitude of the standardised low flow statistic.

The principles of hydro power estimation

Once the overall flow conditions have been determined for the site of interest, either from gauged data or estimated from catchment characteristics, the designer of a small hydro plant must choose the turbine types that would be appropriate for the site, then identify the range of flows within which the turbines can be operated.

Manufacturers produce operating envelopes for different turbine types, which indicate the range of head and flow conditions within which the turbines will operate. For the proposed site, the selection of the appropriate turbine types can be determined from the operating envelopes based on the head and flow conditions.

The range of operational flows for turbines is determined by an environmental minimum flow (which must be maintained in the river to sustain fish life) and the requirements of downstream users. In addition, the minimum operating flow of the turbine needs to be taken into account. No turbine can operate efficiently in all conditions — below a certain percentage of the rated flow, the turbine efficiency decreases rapidly as the discharge increases. The rated flow and minimum turbine or environmental operating flow (whichever is greatest), when marked on the flow duration curve, define the usable part of the flow range. A first approximation of the average annual energy output and power potential can be determined by integrating the usable area under the flow duration curve.

The way in which this can be achieved is to divide the usable part of the flow duration curve into vertical incremental strips, for example at 5% intervals on the probability scale. The gross energy for each strip can be calculated, incorporating the flow, the specified head and the known efficiency characteristics of the turbine and the individual components (ie gearbox, generator and transformer). The gross annual average energy output is then the sum of the energy contributions from the individual strips.

However, the net energy output will be lower, to take into account periods when the turbines will not be operational, due to maintenance and repairs. A similar approach can be adopted to calculate the annual average power.

By comparing the energy, power output and capacity, the designer can then determine the optimum turbine type and capacity for the site.

HydrA has been designed to run on a PC with Microsoft Windows. Within the software, the estimation of catchment characteristics, the derivation of the flow duration curve, the selection of the turbine types and the calculation of the power potential are undertaken separately through a series of modules.

The software incorporates raster databases of rainfall, potential evaporation and the hydrological response of soils, plus regionally derived models, which enable the physical and hydrological characteristics of the catchments to be identified automatically. This capability is of particular benefit to designers with little hydrological expertise.

The only pre-requisite for the estimation at an ungauged site is for the user to define the catchment boundary from topographic maps and input the co-ordinate pairs into the software. An example of the estimation of catchment average rainfall is illustrated in the diagram above.

The software also incorporates recognised operational envelopes and efficiency criteria for eight turbine types: cross flow; Francis open flume; Francis spiral case; Kaplan; semi Kaplan; Pelton; propellor; and Turgo.

Automatic estimation procedures can be applied to all user-defined sites. Once the catchment boundary has been defined, each of the modules can be run sequentially. However, where gauged climate statistics or flow data are available, the catchment or flow analysis modules can be bypassed and the next module can be run using the available data. Each module can be accessed through icons from a toolbar or from pull-down menus.

Data can be entered and edited interactively by the user through a series of dialogue boxes and templates, or it can be retrieved from an existing file.

The results from the separate analysis modules are displayed on the screen and can be printed or written to a file for use in the other modules and applications. Such a flexible approach means that the results can be transferred to different software packages for further investigation, such as exporting the power results for economic analysis.

The output is presented such that it is possible to compare the energy, power and capacity output for each of the selected turbine types, thus enabling the designer to determine the optimum turbine type and capacity for the site. This means that the initial assessment of the potential of selected sites for small hydro power schemes can be achieved rapidly and efficiently.

HydrA was formally launched in March 1997, and since then has been used by consultants, regulatory bodies and universities throughout the UK and Spain. Regional models for use in Italy were completed during 1997 and have been incorporated into the program. Documentation for the software is currently in preparation, and it is anticipated that the new package will be released later this year.

Similar applications

Due to the general nature of the design models, similar applications can be developed for any country. Within the current phase of the project the coverage of the Hydro Power Atlas is being extended to include Austria, the Republic of Ireland, Belgium and Portugal.

Additionally, current plans include development of the software in India and Nepal, and interest has been shown by organisations in Albania, Slovakia, Fiji and the Philippines.