Snowy 2.0 Pumped Hydro-Electric Storage feasibility
Annad veldi ehf Iceland
Snowy 2.0 feasibility
10.8.2018 / SJ skuli@veldi.is
Snowy 2.0 Pumped Hydro-Electric Storage feasibility The Snowy 2.0 project is an expansion of the Snowy Mountains Hydroelectric Scheme in Australia. A pumped-hydro expansion, Snowy 2.0 will increase the production capacity of the existing power generation with improved storage capacity and help to stabilize delivered electricity at the power markets. This has proved important in recent times, as intermittent electricity from wind and solar power stations are constantly increasing. Assumptions The scheme in table 1 shows the balanced situation that defines the production cost of electricity. It is characterized by financial performance where earnings from operation and fixed cost weigh each other up, payback period is equal to defined lifetime of 60 years and internal rate of return is equal to defined interest rate of 6%. Table 1. Snowy 2.0 Base Case Scenario
It is necessary to have in mind that excluding cost of pumping as shown in line 29 is insufficient and cost of pumping must be included to make the power plant operational as is done in line 30.
Annad veldi ehf Iceland
Snowy 2.0 feasibility
10.8.2018 / SJ skuli@veldi.is
Full storage of the upper reservoir of Snowy 2.0 will last for 175 hours, at a full 2000 MW run in the power plant.
Natural inflow to the upper reservoir is not taken into account.
76% efficiency is assumed, meaning that for every 100 MWh for pumping, 76 GWh could be produced later in the power plant.
The year is divided into 19,9 time cycles each of 405 hours, 230 hours for pumping and 175 hours for power production. This leads to 8075 hours/year activity in the production/pumping power plant and 8760-8075=685 hours or 28,5 days for maintenance. At full capacity, 4588 hours/year (9176 GWh/year) will then be needed for pumping and 3487 hours/year (6974 GWh/year) will be available for power production.
Capital Costs of Snowy 2.0 is assumed 4500 MAuD and cost of transmission 2000 MAuD. With currency rate of 0,73 USD/AuD the resulting total cost is then estimated 4745 USD.
Results
Cost of Production will be 88 USD/MWh with financial costs and fixed cost of operation and maintenance included and also variable cost of purchasing power at off-peak prices for pumping. The inclusion of cost of pumping is crucial as mentioned above.
Available at all times when needed will be: off-peak market energy for pumping at price of 30 USD/MWh and peak energy market to offset produced energy at price of 88 USD/MWh, equal to power production cost.
This will then lead to Internal Rate of Return of 6% and Payback Period of 60 years.
Figure 1 shows how Internal Rate of Return and Payback period varies with market price of sold power production. Figure 1. Snowy 2.0 Financial Performance Purchased off-peak power at 30 USD/MWh
Figure 1 shows that if 10% rate of return is required, the price of sold electricity should be 114,50 USD/MWh and payback period then will be 15,7 years.
Annad veldi ehf Iceland
Snowy 2.0 feasibility
10.8.2018 / SJ skuli@veldi.is
Discussion:
Figure 2 explains an example of market price behaviour for one day and purchasing price for pumping and selling price for produced energy. The model behaviour is explained by the black curve, which then sometimes: overestimates the purchasing price for pumping underestimates the selling price of power production overestimates both the period of pumping and the period of power production Figure 2. Market behaviour
Market prices [USD/MWh]
180 160 140 120 100 88: Production price 80 60 40 30: Pumping price 20 0 -20 -40 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Time [hours]
Purchasing
Selling
Market
Model
An interesting aspect is what actions will be taken when market price is between the pumping price of 30 USD/MWh and the production price of 88 USD/MWh? What will happen if the market price is in that interval most of the time?
Sensitivity Analysis
If interest rate per year is varied from 0% to 10%, then cost of production, with variable cost of pumping included, would be as shown in table 2.
Table 2. Sensitivity Analysis
Interest Production Cost
Rate
[USD/MWh]
0%
1%
2%
3%
4%
5%
6%
7%
8%
9%
10%
Of the production cost, 39 USD/MWh comes from cost of pumping, same result for all values of interest rate, see also table 1.
Annad veldi ehf Iceland
Snowy 2.0 feasibility
10.8.2018 / SJ skuli@veldi.is
And finally
A better result would be obtained by computer simulation of the power plant working with the surrounding power system. Thereby one would be able to simulate the power plant pumping and production depending on market prices at all times. It could be complicated to predict the effect Snowy 2.0 itself would have on the market price.
The results presented in this desktop analysis are not exact but should reveal a ballpark value of more extensive research. In many cases it is important to run a simplified version of a project in parallel with wider simulation and optimization studies.
Battery storage comparison
Table 3 shows a comparable desktop analysis of 100 MW battery storage, resulting in production cost of 144 USD/MWh.
Table 3. Battery Storage Scenario
Skuli Johannsson, Reykjavik, Iceland, www.veldi.is 4