Capacity Identification and Operational Policy of the Proposed Broadlands Mini Hydroelectric Power Plant

Broadlands Hydropower Project (BHP) with an installed capacity of 35 MW and expected annual energy generation of 137 GWh will harness the last remaining hydropower potential of the Kehelgamu-Maskeli Oya (K-M) Complex located in Kitulgala area. The construction of the main dam of the BHP will directly affect White-Water Rafting (WWR) that takes place along the downstream of the said area. To sustain this sport activity which has become a tourist attraction, and to avoid conflicts that could arise among concerned parties, a certain quantity of water (yet to be decided) has to be released during day time from the main dam which would reduce the amount of annual energy that was originally expected to be generated from BHP. However, the possibility of recovering some of the lost energy by releasing the water through a mini hydroelectric power plant located downstream of Kelani River close to the BHP Main Dam cannot just be ignored. This research study therefore investigated the technical impact of releasing water under different operational policies through the main dam of the Broadlands Power Plant and the possibility of recovering the energy lost using the proposed mini hydroelectric power plant while catering to WWR. Finally, the economic impact of avoiding the cost of dispatching of thermal power plants under different operational scenarios of the composite power system as stated by the Ceylon Electricity Board (CEB) is presented.


Laxapana Complex
The Laxapana Complex set up under Kehelgamu-Maskeli Oya Development Project with five power stations operating in cascade, an installed capacity of 355 MW and an annual energy generation of 1563 GWh [5] is shown in Figure 1 in Appendix-A. Castlereagh and Maussakelle are the two main storage reservoirs feeding the complex.

Broadlands Hydro Power Project
Broadlands Hydro Power Project is the last large scale hydropower development project to be implemented for the Laxapana Complex. Its main work sites are located about 95 km away from Colombo. The 2x17.5 MW run-of-the-river type power plant currently under construction in Kithulgala area under the Project is expected to be completed in July 2019.
The Project consists of a weir, dam, water diversion tunnel, headrace tunnel and a power plant. The weir and dam are to be constructed in an area called Polpitiya in the Nuwara Eliya district in the Central Province, near the confluence of Maskeli-Oya and Kehelgamu-Oya, the two main upstream tributaries of Kelani River. Broadlands main dam built across Maskeli-Oya will receive directly the water released from the existing Polpitiya Power Plant (PP). A weir that will be built across Kehelgamu-Oya will divert the catchment inflow from Norton Pond to Broadlands Pond through a short non-pressure tunnel. From there, the water will be taken to the power plant situated in the Kalukohutenna area in the Kegalle District of the Sabaragamuwa Province through the headrace tunnel which will be nearly 3.5 km in length. Water will be released back to Kelani River thereafter through a tailrace canal. The proposed arrangement of the Laxapana Complex is shown in Figure 2 in Appendix-A.

White-Water Rafting (WWR)
Kitulgala, a rain forest area, which is fortunate to receive both monsoons, is one of the wettest places in Sri Lanka and is in the process of being developed as a tourist base mainly to facilitate WWR related activities. WWR is an adventurous and exciting sport which attracts both foreigners and locals equally. Figure 3 in Appendix-A shows some photographs of recent WWR activities .

Broadlands Mini Hydroelectric Plant
The main dam of BHP will have a direct impact on WWR related activities. To mitigate the impact, a certain quantity of water has to be released during day time which will then reduce the amount of annual energy that can be generated by BHP. However, the possibility of recovering some of the lost energy by releasing this water to a Mini Hydroelectric Plant (MHP) located downstream of Kelani River near the main dam of BHP needs to be explored.

Aims and Objectives
The

Methodology
The strategy adopted for this research study can be stated briefly as follows:  The algorithmic work flow chart is shown in Figure 2. The free open system led to the development of the proposed methodology shown in Figure 1.

BHP Water Resource and its Potential
The catchment area of the BHP, highlighted in yellow in Figure 4 in Appendix-A, consists of the area between Norton Pond and Kehelgamu-Oya Weir and the area between Laxapana Pond and Broadlands Main Pond. This was found to be 62.5 sq. km in area [1]. It is reasonable to assume that the rate of outflow of water at Polpitiya PP is equal to the rate of inflow of water from the upstream zones of Norton and Laxapana Catchment Areas.

Polpitiya Water Outflow Rates
A significant factor that has to be considered when comparing different types of turbines is their relative efficiencies both at their design points and at reduced flows [3]. Therefore, the turbine efficiency curve plays an important role by indicating the power output for a particular discharge flow-rate of water.
Since no turbine efficiency curve applicable for a broad spectrum of operations (specially for part and low loads) was available for the Polpitiya PP, the tailrace water level and the velocity of the discharged water had to be measured for each active power load (Table-1). Daily rainfall data of the three hydrological stations and other required details over a period of about 20 years were obtained from the Meteorology Department.

Spill Data and Rainfall Data
Since the K-M complex is cascaded, the details of spill data at Norton and Laxapana Ponds very important for the computation of water flow-rate, had to be obtained.

Figure 3 -Spilling of Laxapana Pond
The daily rainfall data obtained were assigned/ divided into drainage areas of the corresponding rainfall gauging stations as shown in Table 2. Drainage area (m²)

Inflow due to Rainfall
The average inflow rates resulting from the rainfall in the catchments between Norton Pond and Kehelgamu Weir and between Laxapana Pond and Broadlands Main Dam are graphically presented in Figure 4.   Table 3 presents the corresponding discharge water flow rates . The tailrace water levels along with the velocity of tail water discharged for each active power load at the Polpitiya Power Plant had to be measured and hourly data from the System Control Centre and Laxapana Power Plant of the Ceylon Electricity Board for a twenty-year study period had to be collected as described in Section 3.2. The average out-flow rate at the Polpitiya Power Plant is shown in Figure 6.  The inflow pattern at Kehelgamu Oya Weir is presented in Figure 7.

Inflow at the BHP Main Dam
Inflow at the Broadlands Main Dam can be expressed as follows:  Figure 7.

Representation of Main Dam Levels
The hydrograph in Figure 9 shows the level variation of BHP Main Dam throughout the year and for any particular value of the level, there are months during which the level would exceed that value.

Generation Pattern of BHP
BHP is identified as a committed run-of-theriver type power plant with a maximum discharge of 70 m³/s and an installed capacity of 2x17.5 MW [4].

Broadlands Pond The significant features of the Broadlands Main
Pond are shown in Table 4.

Reservoir Elevation-Storage Curve
One of the most important physical characteristics of a reservoir is its elevation-storage curve. Elevation "y" (m) and volume "x" (m³) of a reservoir have a physical relationship which is dependent on the topography of the surrounding area. A mathematical function for the elevation-storage curve for the reservoir can be established. In most cases, linear simplification is applied, and y is approximated using the linear relationship, , where y is the elevation of the water surface above a given reference level. It is piece-wise linear and can be computed using elevation-storage data [4].
Reservoir elevation-storage relationship between max. operation water level (121.0 m) and min. operation water level (111.0 m) was obtained using the following equations developed using the curves shown in Figure 5 in Appendix-A.   Figure 11. The expected annual firm energy yield at 98% probability is 76.4 GWh and the expected annual secondary energy is 61.6 GWh.

Energy Loss due to Water Released for Rafting Activities
The amount of water to be released for WWR was used as the input variable. The outputs were the annual amount of energy expected to be generated by the Broadlands PP and the amount of energy lost by the release of water for WWR. Two proposals were prepared in this regard. Water discharge was taken as a variable and the period during which water was released was taken as 8 hrs. Proposal-1 presumed that water was released for WWR throughout the year although in reality, WWR is not possible throughout the year, i.e., during very dry periods and during floods.
Proposal-2 was based on this fact.

Proposal-1
Simulation studies were carried out on the basis that the amount of water released for WWR throughout the year including the dry period varied from 6 m³/s to 20 m³/s from 9.00 a.m. to 5.00 p.m daily. The expected amount of annual energy generated by the Broadlands PP and the amount of energy lost due to the water released for rafting are shown in Table 5.

Proposal-2
To identify the dry season of the year, the percentage contribution made by Maskeli Oya, Kehelgamu Oya and Polptiya PP discharge to the inflow of Broadlands Pond were considered. The detailed percentage contribution made to the energy generated by the Laxapana Complex during the 20 year period considered is shown in Figure 12.   Figure 11.

Energy Loss due to Water Released for Rafting Activities
The amount of water to be released for WWR was used as the input variable. The outputs were the annual amount of energy expected to be generated by the Broadlands PP and the amount of energy lost by the release of water for WWR. Two proposals were prepared in this regard. Water discharge was taken as a variable and the period during which water was released was taken as 8 hrs . Proposal-1 presumed that water was released for WWR throughout the year although in reality, WWR is not possible throughout the year, i.e., during very dry periods and during floods. Proposal-2 was based on this fact.

Proposal-1
Simulation studies were carried out on the basis that the amount of water released .for WWR throughout the year including the dry period varied from 6 m³/s to 20 m³/s from 9.00 a.m. to 5.00 p.m daily. The expected amount of annual energy generated by the Broadlands PP and the amount of energy lost due to the water released for rafting are shown in Table 5. The detailed percentage contribution made to the energy generated by the Laxapana Complex during the 20 year period considered is shown in Figure 12.

Figure 12 -Maskeli Oya, Kehelgamu Oya inflows and Polpitiya discharge
As can be seen from Figure 12, the major contribution to the Broadlands PP has been As can be seen from Figure 12, the major contribution to the Broadlands PP has been made by the Polpitiya discharge. Therefore, the pattern of the water discharged downstream had to be further studied using the hourly active power readings of the Polpitiya PP taken during the 20-year period. The active power duration curve of the Polpitiya PP is shown in Figure 13. It is clear from Figure 13 that WWR is not possible during certain periods due to lack of water in the stream. This can be identified from the Polpitiya PP loading configuration for the corresponding minimum water discharge ( Table 6) . The graphical interpretation showed that rafting was not possible during approximately 57 days of the year. Simulation was then carried out considering the dry period and the values obtained for the expected amounts of annual energy generated/loss are shown in Table 7.

Impact on Environmental Release
According to the Environmental Impact Assessment (EIA) of BHP, a continuous discharge of 0.2 m3/s has to be present to cater to water requirements downstream. At the time of preparation of the EIA, WWR was not a significant activity. However, at the time of commencing the project, WWR had developed well and the CEB decided to release water to maintain it. In this scenario, it was worth studying the impact of minimum environmental release (E-flow) to the BHP energy yield. This study concentrated on all possibilities and the results are presented in Table 8.

Economic Impact
The economic impact of the water released for WWR was evaluated based on the following: I. Economic loss without Broadlands MHP II. Economic loss with Broadlands MHP Table 9 presents the expected daily loss of power and energy from the Broadlands PP during 8 hrs with the amount of water released varying from 6 m³ to 20 m³.  Table 10, in Scenario-II,the rated output of the Broadlands MHP would vary when the quantity of water that was discharged varied from 6 m³ to 20 m³.  The above four scenarios followed by three sensitivity cases were considered for assessing the economic impact as described below:

Scenario-I (without mini hydro PP)
• For prevailing unit cost of generation • +5% to the prevailing unit cost values • -5% to the prevailing unit cost values For this purpose, the amount of annual energy expected to be generated by the Broadlands MHP required was computed using Proposal-1 and Proposal-2 as explained in Sections 5.1 and 5.2.
The results are presented in Table 11.

Conclusion
Downstream flow pattern of the Polpitiya PP was established using the inflow data of the concerned catchment areas during the past twenty years and the regulated outflow of the Polpitiya PP. It is revealed that the downstream water flow for WWR is not viable for about 57 days every year due to very low water levels caused by extremely dry weather .

8.1
Environmental Release The maximum impact of the E-flow on BHP generation is found to be about 1% which is negligible compared to the energy loss expected due to WWR.

8.2
Daily Energy Losses The daily energy losses computed for with and without Broadlands MHP are shown in Table  12. It is noted that the percentage of the daily energy loss is not dependant on the Broadlands Pond water value and WWR discharge water quantity values.

Annual Energy Losses
The annual energy losses computed are tabulated in Table 13. Assessment was done for the following four cases.

Water Value of the Broadlands Pond
The average water value of the Broadlands Pond for both cases, i.e., with and without Broadlands MHP are tabulated in Table 16 and Table 17 respectively. The sensitivity of the water value to the avoided unit cost of thermal power generation is also presented.

Recommendations
 Study and determine a better platform for optimising BHP energy while releasing water for WWR thereby enabling policy/ decision makers to determine the impact of the duration and amount of water to be released for WWR.
 Identify the extremely dry period of Kelani River to minimize the energy generation loss of BHP, since in reality WWR is not possible during the dry season due to lack of water.
 In this study, 19 m is used as the rated effective head for Broadlands MHP. Investigate whether the rated effective head can be further increased since it is a very important factor when recovering at least to some extent the energy lost. Appendix-A