milestone 4 summary report

Port Augusta Solar Thermal

Generation Feasibility Study

Milestone 4 Report

Final Balance of Study

July 2015

A project jointly funded by:

- Alinta Energy

- Australian Renewable Energy Agency, Emerging Renewables Program

- Government of South Australia, Enterprise Zone Fund

For more information:

www.alintaenergy.com.au/Port-Augusta-Solar-Thermal-Generation-Feasibility-Study

http://www.alintaenergy.com.au/Port-Augusta-Solar-Thermal-Generation-Feasibility-Study

105-RPT-006-Milestone_4_Summary_Report of 33

Table of Contents 1 Executive Summary ....................................................................................... 5

2 Introduction .................................................................................................... 6

2.1 Review of Assumptions ............................................................................................ 7 3 Analysis of Measure Progress ...................................................................... 9

3.1 Data Collection ......................................................................................................... 9

3.2 Balance of Study ....................................................................................................... 9

3.2.1 Scope of Works ................................................................................................................ 9

3.2.2 Methodology ................................................................................................................... 10

3.2.3 Uncertainties discovered ................................................................................................ 13

4 Financial Modelling and Assessment ......................................................... 15

4.1 Overview of financial modelling .............................................................................. 15

4.2 Methodology ........................................................................................................... 15

4.3 Input Data ............................................................................................................... 15

4.3.1 Capital Costs .................................................................................................................. 16

4.3.2 Operational Costs ........................................................................................................... 16

4.3.3 Electricity Generation ..................................................................................................... 16

4.3.4 Pricing ............................................................................................................................ 17

4.3.5 Other Assumptions ......................................................................................................... 19

4.4 Financial Modelling Outputs ................................................................................... 19

4.5 Cost Uncertainty Analysis ....................................................................................... 21

4.6 Parameter Sensitivity Analysis ............................................................................... 21

4.6.1 LCOE vs. Simple Payback Time .................................................................................... 21

4.6.2 Plant Configuration ......................................................................................................... 21

4.6.3 Market Forecast ............................................................................................................. 23

4.6.4 Time of Day Pricing vs DNI datasets .............................................................................. 24

4.6.5 Dispatch Methodology .................................................................................................... 25

4.7 Project Financial Viability ........................................................................................ 25

4.7.1 Financial parameter benchmarks required for project viability........................................ 26

4.7.2 LCOE Assessment ......................................................................................................... 30

4.8 Unexplored Concepts ............................................................................................. 31

4.8.1 Storage from grid power ................................................................................................. 31

4.8.2 Use of waste heat ........................................................................................................... 31

4.8.3 Hybridised solar .............................................................................................................. 31


4.8.4 Alternative location ......................................................................................................... 32

5 Near Commercial Technologies .................................................................. 32

5.1 Technical Maturity ................................................................................................... 32

5.2 Financial Potential .................................................................................................. 32 6 Further Information ...................................................................................... 33

7 Conclusion ................................................................................................... 33

Table of Figures Figure 1: AUD:USD exchange rates July 2014 – May 2015 ...................................................................... 11 Figure 2: Financial modelling methodology ................................................................................................ 15 Figure 3: Construction cost curve assumption ........................................................................................... 16 Figure 4: Forecast lifetime electricity production by quarter ....................................................................... 17 Figure 5: Forecast peak and off-peak electricity prices .............................................................................. 18 Figure 6: Forecast LGC price path ............................................................................................................. 18 Figure 7: Project lifetime cashflows ............................................................................................................ 20 Figure 8: LCOE as a function of IRR and CAPEX – Base Case/Reference Curve ................................... 27 Figure 9: LCOE as a function of IRR and CAPEX – Variation 2 ................................................................ 28

Table of Tables Table 4: Cost localisation factors ................................................................................................................ 10 Table 5: CAPEX cost estimate comparison: Options Study vs Balance of Study ...................................... 11 Table 6: Results of further industry consultation ........................................................................................ 12 Table 7: Solar resource data inputs for generation modelling .................................................................... 13 Table 8: Base case financial modelling key outputs ................................................................................... 20 Table 9: +/-30% financial modelling key outputs ........................................................................................ 21 Table 10: Alternate system configurations ................................................................................................. 22 Table 11: Alternate Systems – modelling outputs ...................................................................................... 22 Table 12: Base case vs. forward price curves ............................................................................................ 24 Table 13: IRR as a function of system design & forward price curve ......................................................... 24 Table 14: NPV @ 12% as a function of system design & forward price curve........................................... 24 Table 15: Minimum financial benchmarks for CSP investment at Port Augusta ........................................ 26 Table 16: Sensitivity of IRR & NPV to multiple variables ........................................................................... 29 Table 17: Contributions to LCOE by category and CAPEX vs OPEX ........................................................ 30 Table 18: LCOE contributions by category compared to SunShot targets ................................................. 30 Table 19: VAST Solar scaled base case vs. forward price curves ............................................................. 33


Acronyms

Alinta Alinta Energy group of companies

ARENA Australian Renewable Energy Agency

CAPEX Capital Expenditure

CPI Consumer Price Index

CSP Concentrating Solar Power

DNI Direct Normal Insolation

GHI Global Horizontal Insolation

IRR Internal Rate of Return

LCOE Levelised Cost of Energy

LRET Large scale Renewable Energy Target

LGC Large scale Generation Certificates

MW Megawatt

MWe Megawatt electric

NEM National Electricity Market

NPS Northern Power Station

NPV Net Present Value

OPEX Operational expenditure

PPA Power Purchase Agreement

SAM System Advisor Model

TMY Typical Mean Year

WEM Western Electricity Market


1 Executive Summary Alinta has completed the pre-feasibility investigations detailed in the Port Augusta Solar Thermal

Generation Feasibility Study. This is the fourth and final milestone report contributing to Stage One of the

study. This Milestone Four Report (Final Balance of Study Report) presents the findings of Alinta’s

extensive financial modelling and scenario analysis of the potential for a Concentrating Solar Thermal

plant in Port Augusta, South Australia. It builds upon the Draft Balance of Study Report previously

released.

Over the last several months Alinta has given further consideration to a range of additional possibilities

including five alternative CSP plant configurations, three additional forward price curves and adjustments

to capital cost, operating cost, revenue stream and capital grant funding. The most significant finding

presented in this report is that across the range of sensitivities, system types and forward curve

assumptions considered, there is no combination which returns a positive Net Present Value in the

financial model.

Further industry consultation has confirmed the suitability of the cost estimate assumptions made in the

Balance of Study Report.

Alinta’s internal analysis suggests that in order for a project of this type to attract private sector

investment at this time, the costs would need to be reduced by approximately 60%. If CAPEX and OPEX

were reduced by 60%, then the LCOE of this type of plant would drop from $201/MWh to $80/MWh. This

would be in line with the 2020 target for a drop in the cost of solar thermal technology that is being

pursued by the SunShot Initiative funded by the US Department of Energy.

While the investigations and options reviewed as part of this study have been robust, thorough and

conclusive, there are several unexplored concepts which may have the potential to either reduce costs or

increase revenue. None of the concepts identified would have an impact large enough to be material to

the viability a CSP plant in Port Augusta at this time, however as the landscape of the electricity industry

changes over the next several years, these concepts may begin to play a role in project viability. There is

at least one local technology manufacturer which could be offering commercially competitive prices for

solar thermal generation in the near to medium term and there may be other investors better positioned in

a range of ways to further explore these possibilities.

At this time Alinta can definitively conclude that the construction of a 50 MW, molten salt power-tower

located in the town of Port Augusta is not an economically feasible option for Alinta.


2 Introduction This report represents the fourth of six milestones which comprise the Port Augusta Solar Thermal

Generation Feasibility Study. Milestone One, Project Definition Report, was submitted to ARENA in

January 2014 and Milestone Two, Options Study and Siting Study, was submitted to ARENA in May

2014. Milestone Three, Draft Balance of Study, was submitted to ARENA in January 2015. A public

version of all of these Milestone Reports has been posted on the Alinta Energy website:

www.alintaenergy.com.au/Port-Augusta-Solar-Thermal-Generation-Feasibility-Study

http://www.alintaenergy.com.au/Port-Augusta-Solar-Thermal-Generation-Feasibility-Study


2.1 Review of Assumptions

There were several high level assumptions made by Alinta which were inputs into the early stages of the Port Augusta Solar Thermal Feasibility Study.

The assumptions first presented in Milestone 1 (Project Definition Report) are shown in Table 1 below along with a current assessment of these

assumptions.

Table 1: Changes to Assumptions from Milestone 1 Report

Initial Assumption Previous changes Milestone 4

The location of the Augusta Power Station, and in the vicinity of the facility, is suitable for the siting and development of a solar thermal facility.

The site identified as optimal in the Siting Study is now known to be the subject of a Development Application by a third party. The proximity of the CSP plant to the Spencer Gulf raises potential corrosion issues due to salt water spray/deposition.

No change

Alinta Energy understands the current arrangements for land tenure permit the siting and development of a potential solar thermal facility on land within the control of Alinta Energy or adjacent to subject to the Sale / Lease arrangements between Flinders Power Partnership and the Government of South Australia.

No change No change

The life of the Leigh Creek Mine, which supplies coal to the Augusta Power Stations, would be extended through further investment by Alinta Energy.

No change In June 2015 Alinta Energy announced an intention to cease operating the Northern and Playford power stations and the Leigh Creek Coal Mine by March 2018.

The Augusta Power Stations would remain in operation, in their current form supplied by the Leigh Creek Coal Mine, until at least 2028 to 2032.

No change

The useable life of the Augusta Power Stations, including re-use of facility components, extends beyond the current expected life of the Leigh Creek Mine.

There are significant technical challenges to running NPS on only solar once the coal resource has been exhausted which would require extensive re-engineering of large parts of the plant.

No change


Initial Assumption Previous changes Milestone 4

The pre-measure activities and studies relied upon in the development of this study which detail the potential value and strength of the solar resource, the potential for hybrid solutions, and the potential utilisation of components from the Playford B Power Station is the best estimate and advice of the respective experts.

Use of components from Playford B was determined to be infeasible. Procurement of spares and replacement parts is extremely difficult. The entire facility would require upgrading in order to support the use of usable components.

No change

The range of project benefits, fuel diversity opportunities for South Australia, dispatchable energy potential, compatibility with South Australian energy system, network connection options, technology costs and acceptable technology types do not materially deviate from those understood at the commencement of this study.

Analysis by the Australian Energy Market Operator suggests that the grid in South Australia is oversupplied which leads to a disincentive for adding generation capacity of any kind.

The reduction in generating capacity due to the future closure of Northern and Playford is not large enough to negate the oversupply of capacity in the South Australian electricity market.

Progress beyond the study will depend on a number of factors outside the scope of this piece of work which have not been estimated or modelled at this point in time.

No change No change


3 Analysis of Measure Progress

3.1 Data Collection

On 5 July 2014 Measurement Engineering Australia (MEA) sent two representatives to the Northern

Power Station in Port Augusta to install a solar tracker, weather station and data collection equipment.

The equipment has since been recording the following variables at one minute intervals:

Weather Station

Global solar radiation

Air temperature

Humidity

Wind speed (min, max, ave)

Wind direction

Barometric pressure

Solar Data Station

Global solar radiation (min, max, ave)

Global diffuse radiation (min, max, ave)

Direct normal irradiance (DNI) (min, max, ave)

Temperature of shaded pyranometer

Temperature of un-shaded pyranometer

Temperature of pyrheliometer

Alinta has made a provision for validation of a full 12 month dataset which will allow for correlation

between satellite data and ground station records.

3.2 Balance of Study

3.2.1 Scope of Works

The scope of works for the Balance of Study is identified in the Agreement and is summarised below:

Details of the plant and its operation;

Capital and operating and maintenance costing at +/-30%;

Energy yield and generation profile;

Infrastructure requirements;

Environmental studies including land use, profile and identification of environmental issues;

Planning and development requirements for a Development Approval;

Network connection;


*Stakeholder consultation plan; and

*Preliminary financial evaluation.

*Sensitivity analysis on the effect on financial viability of a range of parameters

The items marked with “*” were determined to be within the core capability and expertise of Alinta

personnel and were completed internally. The remaining scope was completed by Parsons Brinkerhoff

(PB).

3.2.2 Methodology

The majority of the high level plant design was completed and reported in the Options Study Report

which comprised one of the deliverables for Milestone 2. Therefore only those aspects that have been

further refined or have changed or new challenges which have arisen will be discussed here.

Capital Cost

Project capital costs were estimated by a combination of scaling the detailed reference plant cost

estimate and generating estimates using commercial software and the internal estimating experience of

PB personnel. Alinta and PB discussed at length the value of the labour cost multiplier in the context of

the Port Augusta economy and local labour-force skills. Variation in this parameter could have a material

impact on both the capital and operations cost estimates. While there is a real potential for the labour

cost multiplier to be less than that used in the study, it would be well within the +/- 30% accuracy which

characterises the pre-feasibility stage.

Capital cost multipliers used in this report are show in Table 2 below.

Table 2: Cost localisation factors

Factor Value Comments

Labour cost multiplier 1.14 Ratio of Australian union labour rates to Californian union labour rated (from Thermoflow PEACE) further localised to Port Augusta

Material cost multiplier 1.34 Ratio of Australian to Californian material cost multiplier (from Thermoflow PEACE) further localised to Port Augusta

Currency exchange rate 1.1 AUD to USD

There is a significant reduction in estimated capital cost presented in the Balance of Study Report

compared to the cost estimate provided in the Options Study Report. Cost estimates in the Options

Study were based entirely on a literature review and desktop investigation. Alinta requested PB to

conduct a cost tightening exercise which would draw on direct industry current experience and

knowledge as well as consider the relative maturity of the solar thermal industry against well understood

technology price curves. The cost estimates provided in the Balance of Study report are informed by

conversations with leading industry players and consider more heavily the public knowledge around real

costs incurred by reference plants.

The overall capital cost estimate dropped approximately $200M following this exercise. The most

significant reduction came from the heliostat field. Early estimates using the labour, exchange and

material cost multipliers referenced above resulted in an estimated cost of $262/m2 for the heliostat field.

Following discussions with manufactures and installers of heliostats, a revised estimate of $150/m2 was

determined to be realistic.


Another significant reduction in the CAPEX cost estimate was realised by taking a weighted average of

contingency values applied to individual components rather than applying the same % contingency to the

final project cost. Table 3 below summarises the difference in CAPEX cost estimates by major

component.

Table 3: CAPEX cost estimate comparison: Options Study vs Balance of Study

Component

Options Study CAPEX

$M

Balance of Study CAPEX

$M

Difference $M

Site Improvements 32.3 8.2 -24.1

Heliostat Field 230.4 138.2 -92.1

Tower 26.0 21.5 -4.5

Receiver 90.4 86.0 -4.4

Thermal Energy Storage 79.3 83.4 4.1

Balance of Plant 129.5 118.1 -11.1

Contingency 117.6 67.9 -49.7

EPC & Owner Costs 81.8 53.4 -28.4

TOTAL 787.2 577.0 -210.2

Clearly the assumptions about foreign exchange rates and labour rates could have a material impact on

the CAPEX cost estimate. Preliminary analysis indicates that the CAPEX estimate has an exposure of

approximately 30-35% to the labour rate and an exposure in the vicinity of 50% to foreign exchange

rates. The current exchange rate is closer to 1.3 rather than 1.1 AUD:USD. This would have an effect of

adding ~10% or ~$60M on the CAPEX cost estimate.

Figure 1 below shows the movement of the Australian Dollar vs. the US Dollar over the previous 12

months.

Figure 1: AUD:USD exchange rates July 2014 – May 2015


Additional Industry Consultation

Alinta Energy contacted numerous industry participants in an effort to confirm the cost estimate

assumptions used by PB in the Balance of Study report. After this consultation Alinta has concluded that,

while there is still a moderate amount of variation in the values used by various suppliers, the values

used in the Balance of Study assessment conducted by PB are accurate within the scope of the Study.

The exception is the Power Block which appears to have been too conservatively estimated. Table 4

below summarises the information gathered through additional industry consultation and the percent

difference between the Balance of Study cost estimate and the recent response from industry.

Table 4: Results of further industry consultation

Cost Item Industry Response BoS est.

($M) Industry

Response Difference

Full Project

One organization that has previously constructed similar infrastructure undertook a high level review of a cost estimate for a CSP plant as specified in the Alinta study.

$577 ~$500 -13%

Another company experienced in construction and operation of solar thermal infrastructure replied to Alinta’s request with a statement that within the context of their experience and the recent study into the feasibility of CSP in Western Australia, the cost estimates in the Draft Balance of Study are of the right order of magnitude and therefore judged to be appropriate at this level of the study.

~ ~ ~

Heliostats

The BoS estimate was $150/m2 of aperture area. A

local South Australian manufacturer of heliostats has separately provided an indicative cost estimate for manufacture, supply and installation of heliostats at a rate of $155/m

2.

$138 ~$143 3%

Salt supply & melting

Parsons Brinkerhoff inquired to 2 European companies for the supply and melting of 20,000 cubic meters of thermal salts for a CSP plant in Port Augusta. The combined cost estimate came to ~$40M.

$48 ~$40 -17%

Power Block

A high quality engine manufacturer provided an indicative price for a 50MW steam turbine package of approximately $18M. The cost estimate used in the Balance of Study is ~$29M.

$29 ~$18 -38%

Six other companies did not respond in writing to Alinta’s request to provide comment on the cost

estimates contained in PB’s Balance of Study report.

Operational Cost

Operating and maintenance costs were also scaled from the reference plant in a manner similar to the

capital cost estimate. A burdened labour rate of 30% was assumed. Where there is a support function

that could be considered a corporate service, no labour cost was included. These roles, such as IT,

finance and human resources are assumed to be provided by existing Alinta Energy personnel.

With the decision to close Northern Power Station there will be an increase in annual estimated labour

cost as many of the resources that were assumed to be shared with existing operations would no longer

be shared. While this would be a material difference in cost it is not expected to be significant and would

certainly not stray from the +/- 30% accuracy band.


As with CAPEX, the OPEX cost estimate is also sensitive to labour rates, however less so to foreign

exchange rates. The ~$8M annual operational costs has an exposure of approximately 60% to the

labour rate.

Solar Resource Data

For a more accurate assessment of solar resource and generation profile of the plant, Alinta purchased a

typical mean year (TMY) data set synthesised from a 15 year record between 1999-2013. The report

provides a retrospective analysis of the past 15 years of solar irradiance, wind and temperature data.

The data used is comprised of hourly values, however the long-term average values are only calculated

using complete calendar years.

The difference between the dataset used during modelling for the Options Study and this purchased

dataset is shown in Table 5 below. The differences in these datasets does not make a material

difference on the selection of the plant configuration made during the Options Study. The overall

increase in solar resource expressed in the dataset purchased by Alinta for the Balance of Study

increases the modelled annual generation by the plant and therefore a nominal increase in modelled

revenue.

Table 5: Solar resource data inputs for generation modelling

Parameter Units Data for Options

Study Data for Balance of

Study Difference

Standard Error % 16% <9% -7%

Annual average DNI W/m2 235.5 279.2 +19%

Peak DNI W/m2 886 981 +11%

Summer average DNI MJ/day 24.7 28.5 +15%

Winter average DNI MJ/day 15.9 19.7 +24%

Annual average GHI W/m2 215 222.1 +3%

Network Connection

The obvious location to connect into the network is at the Davenport Substation just to the East of

Northern Power Station. There is currently a spare bay inside the substation that could potentially accept

the assets that would be required to connect the Augusta Solar Thermal plant. There are various options

in the definition of asset ownership and boundaries with ElectraNet, which owns and operates the

Davenport substation. The decision on what type of arrangement would suit best would be determined

by a commercial and contractual discussion that is detail beyond the scope of this report. The connection

of 50 MW of solar thermal power at Davenport substation would also not have a material impact on the

Marginal Loss Factor (MLF) of the network.

3.2.3 Uncertainties discovered

Plant Siting

Since the Siting Study was completed there has been a material change to the availability of land near

Northern Power Station. The land parcel identified as Option 1 has since been the subject of a

Development Application for a nearby operation and would no longer be available as a location for the


solar thermal plant. The next best location that was part of the siting study is approximately four

kilometres further south in the same direction from Northern Power station.

This increases the cost of constructing a transmission line to the switchyard and increases the cost and

complexity of sharing any infrastructure or services between the solar thermal plant and the existing

activities at Northern Power station.

Another concern that has arisen is the potential for salt corrosion on the plant components. This is a

particular concern for the heliostats, however is relevant to all equipment and materials that would be

exposed to salt spray and deposition.

Cooling System

The proximity to the Spencer Gulf and the existing cooling water loops used for Northern Power Station

make a wet cooled condenser an obvious first option. However the expense of constructing adequate

pipework is prohibitive and therefore a cooling tower is proposed. The use of salt water in a cooling

tower introduces additional complications and has the potential to increase fouling and corrosion on

nearby infrastructure.

The logical conclusion is to move to an air cooled condenser which is also at additional cost.

These issues that have arisen with the plant siting have almost entirely removed the potential benefit that

was once thought to exist by co-locating the plant with Northern Power Station. It is Alinta’s opinion,

therefore, that the initial constraint of choosing a location in proximity to Northern is now not relevant. For

the purposes of the Study Alinta has continued to contemplate the location identified by PB in the Siting

Study.

Should a full feasibility study of a CSP plant in the vicinity of Port Augusta be undertaken by Alinta or

another party in the future, a new location in the region would likely be sought.

Time of Day Pricing

As part of the preliminary analysis undertaken for the Draft Balance of Study Report, Alinta investigated

the relationship between LCOE and simple payback time in the context of different CSP plant

configurations. This analysis led to the conclusion that there was the potential for systems with less

thermal storage, smaller heliostat field and therefore lower capital cost to have similar payback times to

the Base Case. When considering external investment and/or capital grant funding, a reduced CAPEX

would mean the same quantum of investment and/or grant would have a proportionally greater effect on

the viability of a project.

The value of Time of Day pricing becomes more significant as the storage capacity of the plant is

reduced, therefore it is important to understand the impact that ToD pricing has on the financial viability of

systems with less storage.

In the financial model, market prices are represented by peak and off-peak time-weighted average rates.

In order to determine the effect of hourly price granularity, an analysis was done comparing the revenue

generated by a system with the dispatch optimised to the time weighted average prices versus dispatch

optimised to an hourly price profile. The difference in recorded revenue between the two price curves

was less than 5%. While this is a material difference, it is not significant and is comfortably covered by

the hypothetical scenarios assessed by Alinta. The accuracy benefit gained by this level of modelling

does not justify the significant additional effort required. Alinta determined that the use of peak and off-

peak time-weighted average price signals is appropriate for this study.


Section 4.6 contains detailed results of further parameter sensitivity analysis and the impact on the

financial position of a CSP plant in Port Augusta.

4 Financial Modelling and Assessment

4.1 Overview of financial modelling The financial viability of the proposed 50 MW Solar Thermal Power Station in Port Augusta is a key factor

in assessing whether to continue to develop the project business case. Alinta has developed a financial

model to analyse the overall financial viability of the project. The evidence from the financial analysis is

that commercially the project is unviable. Even selecting for all of the most favourable and optimistic

assumptions, the modelled CSP plant does not achieve a 12% IRR.

The remainder of this section sets out the financial modelling undertaken and presents the detailed

outputs from the financial viability assessment.

4.2 Methodology

The financial modelling draws upon a number of information sources in order to calculate the ungeared

post tax nominal cash flows associated with the project. The financial modelling is undertaken on a

quarterly basis. The modelling methodology is set out in Figure 2:

Figure 2: Financial modelling methodology

4.3 Input Data

Input data is sourced from independent third party reports wherever possible. A number of the third party

reports have been specifically commissioned as part of the Port Augusta Solar Thermal Feasibility Study,

whilst others are commissioned by Alinta for use across within its broader business activities and have

been leveraged for use in this Study. In some instances, Alinta has made assumptions, drawing upon its

internal expertise where required.


4.3.1 Capital Costs

The $577M capital cost of the project used in the financial modelling is based upon the Balance of Study

Report prepared by PB. Alinta has supplemented the total capital cost estimate prepared by PB with a

construction cost curve, to forecast the potential actual expenditure over an assumed 2 year build period.

It is assumed, that construction starts on 1 January 2017. The resulting quarterly construction

expenditure curve is shown in Figure 3.

Figure 3: Construction cost curve assumption

4.3.2 Operational Costs

The $7.89M of annual operating costs in the financial modelling are based upon the Balance of Study

Report prepared by PB. Alinta has supplemented the estimated annual operating costs with the

assumption that the operating costs will escalate each July by CPI, and will be incurred equally across

the year.

4.3.3 Electricity Generation

The PB Balance of Study Report estimates the energy production capability of the project over its

expected life. Importantly, the financial model converts the information in the PB report into peak (7:00 to

22:00 workdays) and off-peak for each calendar quarter. Specifically, the model calculates:

Average proportion of peak and off-peak energy production for each month (based upon average daily production profiles in each month )

Average proportion of peak and off-peak energy production for each quarter (based upon total monthly production profiles )

Energy production during peak and off-peak times for each quarter for each forecast year. This utilises the forecast total annual production for each year of operation from the report (which


takes into account degradation) and applies the average quarterly peak and off-peak proportions (calculated in 2) above)

The total forecast peak and off-peak production by quarter is shown in Figure 4 below.

Figure 4: Forecast lifetime electricity production by quarter

4.3.4 Pricing

South Australian pool prices are based upon the forecasts contained in Acil Allen Consulting’s Australian

Energy Market, Analysis of the National Electricity Market (NEM), Western Electricity Market (WEM) and

Large scale Renewable Energy Target (LRET) report. The time weighted nominal quarterly forecast peak

and off-peak prices to Dec 2030 have been utilised. The price path is then interpolated between 2030

and 2035, and also 2035 to 2040. Beyond 2040, it is assumed that prices remain flat.


Figure 5: Forecast peak and off-peak electricity prices

Forecast large scale generating unit certificate (LGC) prices are also sourced from Acil Allen Consulting’s

Australian Energy Market, Analysis of the NEM, WEM and LRET report. Acil Allen’s reference case is

adopted for the base financial viability assessment. LGC prices are only forecast to 2030, when the

LRET scheme is currently scheduled to end. The forecast price path is shown in Figure 6 below and was

generated on the basis of the original RET target of 41,000 GWh. The outcome of the recent RET

Review was not known in time to be incorporated into this analysis. The expected effect would be a

small, downward pressure on LGCs.

Figure 6: Forecast LGC price path


4.3.5 Other Assumptions

A number of generic assumptions have been adopted in the financial modelling:

Tax assumptions:

o To calculate the post-tax cashflows of the project, we have adopted the current company tax rate of 30%

o Tax depreciation is based upon the diminishing value methodology, with a 200% multiplier and an assumed asset useful life of 25 years from construction completion

A number of costs and prices are assumed to escalate by CPI. Within the model it is assumed that CPI increases by 2.5% per annum, with adjustments occurring on 1 July in each of the forecast years.

The Acil Allen reference curve assumes the current policy that there is no price on carbon.

4.4 Financial Modelling Outputs

During an internal audit of the financial model, Alinta discovered that nominal values had been used for

forecast revenue instead of real values. This had the effect of applying CPI twice to the forecast revenue

and thereby over-estimating the revenue that would be generated during the lifetime of the modelled

plant. The updated results of the model are presented below.

The forecast annual cashflows of the project are shown in the chart below. Figure 7 shows that once the

facility is operational, it would be profitable. However, based upon the metrics presented in Table 6

below, the level of expected future profitability does not justify the large capital investment required to

build the facility.


Figure 7: Project lifetime cashflows

Table 6: Base case financial modelling key outputs

Metric Value Comments

Net Present Value -$359.8M

Based upon a target post tax cashflow discount rate of 12% (ungeared). Represents the size of the shortfall of the project from being commercially viable.

Internal Rate of Return 1% Indicates that the project does not generate a return on the capital invested in the project, as it does not meet a hurdle rate of 12%.

Levelised Cost of

Energy1

$201/MWh Represents the revenue that would be required per MWh for the project to achieve the required return metrics.

Realised revenue2 $96/MWh

Represents the revenue that is forecast to be realised per MWh produced.

1 Levelised Cost of Energy is calculated as [NPV of Capital Costs and Operating Costs (excluding any tax payments)] divided by the

[NPV of Demand escalated CPI]

2 Realised Revenue is calculated as [NPV of Total Revenue] divided by the [NPV of Demand escalated CPI]


4.5 Cost Uncertainty Analysis

The scope of this (Milestone 4 report) is to identify the cost of the project to within a tolerance of +/- 30%.

The financial evaluation has been based upon the expected costs estimated by PB to within this level of

accuracy. Given the level of accuracy with in the costs, it is prudent to undertake a sensitivity analysis to

confirm whether, across the plausible range of cost estimates (i.e. +/- 30%), the project could be

considered to be commercially viable. As the metrics in the table below indicate, even if the capital costs

have been over-estimated by 30%, the project would still be unviable, with a Net Present Value (NPV) of

-$180M, and an Internal Rate of Return (IRR) of 3.4%. It is on this basis that it is considered that the

project is unlikely to be commercial under any plausible cost estimate.

Table 7: +/-30% financial modelling key outputs

Metric -30% +30%

Capital costs $403.9 $750.1M

Net Present Value (@ 12% IRR) -$182.3M -$515.6

Internal Rate of Return 3.4% -0.4%

Levelised Cost of Energy $149/MWh $253/MWh

Realised revenue $96/MWh $96/MWh

4.6 Parameter Sensitivity Analysis

4.6.1 LCOE vs. Simple Payback Time

During the early phases of the Study, Alinta and Parsons Brinkerhoff selected LCOE as the defining

metric for use in optimising storage hours and solar multiple of each system under consideration. Use of

this metric is ubiquitous and fundamental to all investment decisions in generation infrastructure. Alinta

uses LCOE to rank the internal portfolio of potential projects in order of investment priority. Some

analysis undertaken for Milestone 3 suggested that LCOE may not be the most accurate metric for

evaluating a CSP plant against competing technology options.

The lowest LCOE system with 15 hours of storage effectively performs as a baseload plant. This leads to

electricity generation at times of lowest prices as well as highest prices. Systems with between 1-7 hours

of storage offer a small but material improvement in simple payback time. This is due to the ability of

these systems to regularly generate during the higher priced, late-afternoon/early-evening peak periods

while avoiding generation during the late-night/early-morning trough.

Reduction in hours of storage and a proportional amount of heliostats translates to a significant reduction

in CAPEX. A plant with 4 hours storage reduces CAPEX to ~65% of the cost of the reference plant with

15 hours storage.

4.6.2 Plant Configuration

In order to explore more fully the potential impact of plant configuration on financial viability, five

additional plant configurations were interrogated. The selection of storage hours and solar multiple was


recommended by IT Power to cover the entire range of systems that were thought to have the potential

for better economic performance than the Base Case.

In order to align with the boundary condition set by Alinta at the outset of this study, the size of the power

block was held at 50MW. The hours of thermal storage and the size of the heliostat field were varied as

presented in Table 8 below.

Table 8: Alternate system configurations

Alternate System

Power Block

Storage hours

Solar Multiple

Base case 50 MW 15 3.5

AS 1 50 MW 4 1.8

AS 2 50 MW 6 1.7

AS 3 50 MW 6 2.2

AS 4 50 MW 6 2.5

AS 5 50 MW 8 2.4

Each of these systems was created in SAM and the output simulated against the same TMY file that was

used when modelling the base case. Hourly generation profiles were created out of SAM and each

generation profile was inserted into Alinta’s financial model.

Effect on LCOE & IRR

Key metrics for each of the alternate systems are presented in Table 9 below.

Table 9: Alternate Systems – modelling outputs

Alternate System

CAPEX $M

Annual GWh

% peak % off peak

LCOE (10%)

IRR

Base case $577 301 %50 %50 $201 % 1.1

AS 1 $357 167 %72 %28 $230 % 0.0

AS 2 $362 166 %72 %28 $235 % -0.2

AS 3 $402 199 %70 %30 $215 % 0.7

AS 4 $431 205 %70 %30 $224 % 0.4

AS 5 $439 220 %66 %34 $214 % 0.6

The single metric which will determine the viability of an investment is the IRR. In Table 9 one can see

the IRR of the alternate systems is less in each case than the IRR of the base case. This indicates that,

once all factors are introduced into the calculation determining the economic payback, LCOE (ultimately


driven by capacity factor) is still an appropriate metric to use for ranking the economic performance of

CSP systems in Port Augusta.

4.6.3 Market Forecast

In order to better understand the potential effects of forecast prices on the financial viability of a CSP

plant of this type, Alinta commissioned Acil Allen to create three forward curves for the spot price in the

South Australian market. Each of these forward curves incorporated different assumptions about

variables that were considered likely to impact the market: Renewable Energy Target policy, future

carbon price and rate of uptake of domestic PV.

The impact that these variables have on the expected future price of electricity in South Australia is

complex, however it is heavily dominated by assumptions about a future price on carbon.

Forecast price curves

The forward curve used in the reference case for the financial model in the Draft Balance of Study report

was commissioned by the Alinta Energy Wholesale department as part of strategy and planning for the

coming years. The alternative forward curves used in the modelling for this study are based on the

reference case with changes to one or more key assumptions that were thought to have the potential to

have a material impact on the financial viability of a CSP plant in Port Augusta.

The three alternative forward curves are described below.

Forward Curve 1 – Assumes a price on carbon is reintroduced into the market from July 2020 with

prices based on the current forward curve for emissions abatement permits in the EU.

Forward Curve 2 – Assumes a price on carbon is reintroduced by July 2020. Prices are based on the

assumption that there is a global agreement to reduce emissions by 2050 by 80 per cent on 2000 levels.

This scenario also assumes that the Large Scale Renewable Energy Target is unchanged from the

original target with a target of 41,000 GWh by 2020.

Forward Curve 3 – Assumes the same carbon prices as forward curve 2 but also assumes the RET is

changed so that the target equals 30% of demand by 2030 with the scheme expiring in 2040. Forward

curve 3 also assumes a stronger uptake in rooftop PV in line with the Rapid scenario presented in the

2014 National Electricity Forecasting Report published by AEMO.

Each of the Alternate System was modelled against these three additional forward curves to calculate

potential revenue

Effect on IRR

Unsurprisingly, the variable that has the most significant impact on the future market price of electricity is

the price of carbon. This is seen in the analysis where all scenarios have a greater IRR when run against

Forward Curve 2 and Forward Curve 3 where carbon prices drive a national target to reduce carbon

emissions to -80% of 2000 levels by 2050. This is much more ambitious than the current policy target of

-5% by 2020 and beyond most expectations of any near-term policy. Table 10 below presents the

financial metrics of the base case when the different forward curves are assumed. As the forward curve

affects only revenue, not expenses, there is no change to the LCOE of the system ($201/MWh).


Table 10: Base case vs. forward price curves

Forward Curve

IRR NPV (12%)

$M

Reference curve

1.07% -$360

FC 1 0.80% -$352

FC 2 2.10% -$336

FC 3 1.96% -$337

When the various alternate system configurations are also considered in the context of the different

forward curves there are two things that become apparent:

1. Shown by the data in Table 11, the capacity factor at which a CSP plant is run has more effect on the IRR than altering system design to maximise the ratio of on-peak to off-peak generation. As the storage component of the system increases, so does the capacity factor and the IRR.

2. Shown by the data in Table 12, the most significant factor driving the NPV of a CSP system in this study is the capital cost. The Base Case has the highest CAPEX. CAPEX reduces from AS 5 down to the lowest for AS 1.

Table 11: IRR as a function of system design & forward price curve

Forward demand curve

System configuration

Base Case AS 1 AS 2 AS 3 AS 4 AS 5

Reference curve 1.07% 0.00% -0.16% 0.66% 0.35% 0.55%

FC 1 0.80% -0.37% -0.53% 0.34% 0.02% 0.22%

FC 2 2.10% 1.05% 0.87% 1.72% 1.41% 1.64%

FC 3 1.96% 0.88% 0.70% 1.56% 1.25% 1.48%

Table 12: NPV @ 12% as a function of system design & forward price curve

Forward demand curve

System configuration

Base Case AS 1 AS 2 AS 3 AS 4 AS 5

Reference curve - $ 359,792 - $ 239,732 - $ 245,552 - $ 258,191 - $ 282,880 - $ 284,082

FC 1 - $ 351,537 - $ 234,767 - $ 240,670 - $ 252,303 -$ 276,648 - $ 277,495

FC 2 - $ 335,773 - $ 224,322 - $ 230,289 - $ 240,546 - $ 264,028 - $ 264,377

FC 3 - $ 337,413 - $ 225,638 - $ 231,626 - $ 241,965 - $ 265,544 - $ 265,889

4.6.4 Time of Day Pricing vs DNI datasets

The revenue stream in the financial model assumes that the CSP plant would receive spot prices for all

electricity delivered to the grid during the lifetime of the plant. As both spot price and DNI are variables

which can move dramatically over short periods, Alinta commissioned an analysis of the correlation

between spot price and DNI. The significance of a correlation would be seen in an (assumed) increase in


revenue when running historical spot price record against historical DNI data as opposed to a TMY file of

DNI data.

The results of this investigation show that:

There does not appear to be a real time correlation between spot price and DNI

There is a correlation between cumulative daily DNI and ambient temperature

There is a correlation between ambient temperature and spot price

Therefore, while DNI has a direct impact on the temperature of the day and the temperature of the day

does correlate to the spot price for electricity, at least during summer months, the short term fluctuations

in DNI which will be different between a TMY file and a real data file are not material when considering

their impact on the behaviour of the spot market.

Therefore a TMY file is appropriate for modelling the performance of a CSP system and potential revenue

streams in a dynamic electricity market.

4.6.5 Dispatch Methodology

It is not possible to implement sophisticated plant dispatch routines within SAM. When modelling a CSP

system that operates in a base-load pattern this is not a significant constraint to the model. However,

when the system is not designed as a base load and more closely resembles a peaking plant, the ability

to optimise the dispatch can have a significant impact on the financial performance by enabling a better

match between the electricity sent out and peaks in the market price.

Alinta engaged Solar Reserve in order to understand better the potential gains in revenue that could be

possible through optimising the plant dispatch. A CSP plant of the same configuration and capacity as

the Alinta base case was run through a dispatching protocol with perfect foresight of the spot market

prices. The revenue forecast through the perfect foresight model was approximately 18% higher than the

revenue forecast with the SAM dispatch regime.

There are two key assumptions that inform this analysis and must be noted here:

1. All electricity exported can be sold at the market rate; and

2. Dispatch of the CSP plant is done with full knowledge of all future market prices.

Both of these assumptions are unrealistically optimistic, however this exercise leads to the conclusion

that there is a material gain to be made by optimising the dispatch of a CSP plant, which cannot be

effectively done in SAM. It would be reasonable to assume that half (~10%) of this gain could be

captured by virtue of detailed market knowledge and understanding without having perfect foresight of

future market prices.

4.7 Project Financial Viability

All organisations will select investments based on different benchmarks and therefore an investment that

may not be attractive to one organisation may be considered a worthy investment by another. Some

variables that are key to a company’s decision whether to invest include, but are not limited to:

Requirement for minimum return on investment (Internal Rate of Return - IRR);

Organisational risk profile;

Access to and cost of capital; and

Logical synergies with existing assets and/or business strategies.


These metrics and others will typically act as a filter and those potential projects which meet the minimum

criteria will then be ranked against each other to create a priority listing. Depending on the capital and

other resources which may be available, access to finance and other potential constraints. One or more

of the potential projects may be pursued. Therefore it may be the case that a contemplated project is

thought to be profitable but there are logistical or other constraints that prevent an investment being

made by a particular organisation.

4.7.1 Financial parameter benchmarks required for project viability

Any potential project investment will be determined by many and varied factors which are prioritised and

valued differently by different organisations. There are several different perspectives that can be taken on

how to make a CSP plant at Port Augusta a financially viable investment. The simplest approach is to

define the decrease in CAPEX that would be required to give the project a minimum IRR of 10%. This

level of IRR does not represent a benchmark for Alinta’s investment decisions, rather it represents the

lowest benchmark IRR that some companies may contemplate when shortlisting potential investments.

Holding all other parameters fixed, the reduction in CAPEX would need to be approximately 60% as

shown in Table 13 below.

Table 13: Minimum financial benchmarks for CSP investment at Port Augusta

Parameter Modelled Value Required Value Difference

LCOE $201 / MWh $80 / MWh

- 60%

Installed cost $10.5M / MW gross $4.2M / MW gross

CAPEX $577M $231M

CAPEX / kWh / yr $1.86 $0.745

This is presented graphically in Figure 8 below. Each of the three coloured lines represents a given IRR

(10, 12 & 15%). The green line represents the levelised revenue while the LCOE is determined by the

point on the IRR curve corresponding to the CAPEX of the project. The difference between the LCOE

and the levelised revenue is the shortfall which would need to be recovered in some form in order for the

investment to appear economic.


Figure 8: LCOE as a function of IRR and CAPEX – Base Case/Reference Curve


Running multiple sensitivity analyses on the variables under consideration shows that, while there could

be a significant difference in the shortfall depending on assumptions about cost and revenue, no

combination of assumptions considered makes the project appear economic. Making the most optimistic

assumption on all variables on the base case is represented graphically in Figure 9. The assumptions

behind this scenario include:

CAPEX is 30% less than estimated;

OPEX is 30% less than estimated;

An up-front grant of $100M is applied to CAPEX; and

Dispatch optimisation provides a revenue uplift of 10% over modelled value;

Figure 9: LCOE as a function of IRR and CAPEX – Variation 2

Table 14 on the following page contains the results of a range of sensitivity analyses that were run using

different combinations of cost assumptions and forward curves. It should be noted that there is equal

probability of the actual cost being 30% more than the estimate as being 30% less than the estimate. No

scenario with an increased capital cost has been explored.

In this case, no combination of assumptions about system design, forward curve and CAPEX/OPEX

combinations result in a system that is a viable investment.


Table 14: Sensitivity of IRR & NPV to multiple variables

Note: LCOE is calculated on a 10% IRR

ScenarioForward

CurveSystem

Capital Grant

$M

Revenue

upliftIRR

LCOE

$/MWh

NPV $M

(12%)

Reference Reference Base case 577 - 577 -$ 7.9 - 7.9 0% 1.1% 201$ -$360

Variation 1 Reference Base case 577 -30% 404 50$ 7.9 -30% 5.5 10% 6.1% 125$ -$134

Variation 2 Reference Base case 577 -30% 404 100$ 7.9 -30% 5.5 10% 7.4% 110$ -$93

Variation 3 Reference AS 1 357 -15% 303 50$ 6.0 -15% 5.1 0% 2.5% 169$ -$141


Variation 5 FC 1 AS 1 357 -15% 303 50$ 6.0 -15% 5.1 0% 2.3% 169$ -$137

Variation 6 FC 1 AS 1 357 -30% 250 100$ 6.0 -30% 4.2 10% 8.4% 107$ -$35

Variation 7 FC 2 AS 1 357 -15% 303 50$ 6.0 -15% 5.1 0% 3.7% 169$ -$128

Variation 8 FC 2 AS 1 357 -30% 250 100$ 6.0 -30% 4.2 10% 9.7% 107$ -$25



Variation 11 FC 1 AS 2 362 -15% 308 50$ 6.0 -15% 5.1 0% 2.1% 172$ -$142

Variation 12 FC 1 AS 2 362 -30% 253 100$ 6.0 -30% 4.2 10% 8.0% 110$ -$39

Variation 13 FC 2 AS 2 362 -15% 308 50$ 6.0 -15% 5.1 0% 3.5% 172$ -$133

Variation 14 FC 2 AS 2 362 -30% 253 100$ 6.0 -30% 4.2 10% 9.3% 110$ -$29



Variation 17 FC 1 AS 3 402 -15% 342 50$ 6.3 -15% 5.4 0% 2.9% 160$ -$150

Variation 18 FC 1 AS 3 402 -30% 281 100$ 6.3 -30% 4.4 10% 8.5% 105$ -$40

Variation 19 FC 2 AS 3 402 -15% 342 50$ 6.3 -15% 5.4 0% 4.3% 160$ -$139

Variation 20 FC 2 AS 3 402 -30% 281 100$ 6.3 -30% 4.4 10% 9.8% 105$ -$28



Variation 23 FC 1 AS 4 431 -15% 366 50$ 6.7 -15% 5.7 0% 2.5% 168$ -$169

Variation 24 FC 1 AS 4 431 -30% 302 100$ 6.7 -30% 4.7 10% 7.7% 113$ -$55

Variation 25 FC 2 AS 4 431 -15% 366 50$ 6.7 -15% 5.7 0% 3.9% 168$ -$158

Variation 26 FC 2 AS 4 431 -30% 302 100$ 6.7 -30% 4.7 10% 9.0% 113$ -$42



Variation 29 FC 1 AS 5 439 -15% 373 50$ 7.2 -15% 6.1 0% 2.7% 162$ -$169

Variation 30 FC 1 AS 5 439 -30% 307 100$ 7.2 -30% 5.0 10% 8.0% 109$ -$53

Variation 31 FC 2 AS 5 439 -15% 373 50$ 7.2 -15% 6.1 0% 4.1% 162$ -$157

Variation 32 FC 2 AS 5 439 -30% 307 100$ 7.2 -30% 5.0 10% 9.3% 109$ -$39

CAPEX

$M

OPEX

$M/yr


4.7.2 LCOE Assessment

When breaking down the component contributions to LCOE as determined by Alinta’s financial model, it

is apparent that the findings are well aligned with existing literature. Alinta found that, in the context of

the base case, capital costs contribute approximately two-thirds to LCOE. As expected, the heliostat field

makes the single largest contribution in both CAPEX and OPEX. Table 15 below shows the contribution

the OPEX and CAPEX costs make to the LCOE of the base case plant modelled at Port Augusta.

Table 15: Contributions to LCOE by category and CAPEX vs OPEX

Category CAPEX

contribution OPEX

contribution Total $/MWh

Solar Field 43 32 75

Receiver / HTF 31 7 39

Power Plant 35 23 57

Thermal Energy Storage 24 6 30

TOTAL $/MWh 133 68 201

Table 16 identifies the contributions that the major component categories make to the total LCOE

modelled. The four categories were chosen to align with the metrics use by the SunShot Initiative which

is being run by the Department of Energy in the United States. The SunShot Initiative has the aim of

reducing LCOE of CSP technologies by approximately 70% between 2010 and 2020.

Also in Table 16 is the 2013 technology costs achieved by SunShot and the 2020 target costs. The sum

of CAPEX expenses that were not clearly associated with one of these categories was allocated

according to the percentage of the known CAPEX costs for each category. General OPEX costs were

divided evenly among the four categories.

Table 16: LCOE contributions by category compared to SunShot targets

Category Port

Augusta Sunshot

2013* Sunshot

target 2020*

Solar Field 75 68 27

Receiver / HTF 39 27 14

Power Plant 57 55 27

Thermal Energy Storage 30 27 14

TOTAL $/MWh 201 177 82

*converted from 2010 USD to 2014 AUD at exchange rate of $0.8 USD:AUD and inflation at 1.09


If the SunShot Initiative is successful in reducing the technology costs in line with the targets summarised

above, a CSP plant as modelled in this study has the potential to be independently economic not long

after the year 2020.

4.8 Unexplored Concepts

Through internal workshops and through discussions with other industry participants, several concepts

have been identified that would be worthy of further investigation in Stage Two of the Study. These

concepts are not considered to be significant enough to overcome the financial viability gaps identified

thus far in the Study.

4.8.1 Storage from grid power

It would be possible to build in the capability to heat the thermal storage tanks with power imported from

the grid. This would give the flexibility to purchase electricity at low or negative prices and return

electricity to the grid during higher prices. There are several factors which would limit the benefit of this

concept:

Upgrade of the interconnection to Victoria is likely to flatten market fluctuations therefore raising the minimum market prices and reducing peak prices;

The round trip efficiency of this process will be limited by the efficiency of the power block which would be on the order of 40%;

As the thermal storage capacity of a CSP plant increases, the window for profitable purchase and re-selling through this mechanism decreases;

The complexity that is introduced by this functionality could be very high. A deeper understanding of the

details would be required before knowing whether this idea would actually be of benefit.

4.8.2 Use of waste heat

In many settings, the economics of a CSP plant are enhanced because the useful energy is not just the

electricity that is generated, but the low grade waste heat from the system is also beneficial to some other

end user or industrial process. This is the case with the smaller solar thermal plant that is being

constructed by Sundrop Farms in Port Augusta.

In the context of the CSP plant modeled by Alinta at Port Augusta there is no obvious beneficial use for

waste heat from the power block.

4.8.3 Hybridised solar

Linear Fresnel is a cheaper technology than the central receiver with molten salt. The potential for using

a Linear Fresnel field to pre-heat the heat transfer fluid prior to entering the main receiver at the top of the

tower could be considered. This may reduce the overall area of the heliostats required.

There have been CSP plants recently commissioned which include a PV component. These plants are

cheaper per unit of installed capacity and therefore also in terms of LCOE. As discovered earlier,

capacity factor appears to drive LCOE more strongly than all other variables. Therefore, adding PV

capacity may reduce the average LCOE of a hybrid PV/CSP plant but presumably only if there were

additional CAPEX invested in adding PV to the design. The addition of PV capacity would not

economically replace thermal capacity, but would augment it.


4.8.4 Alternative location

While Port Augusta is a convenient site in terms of infrastructure availability and solar resource, there are

other locations in South Australia which offer a better solar resource. Depending on the additional cost to

install a substation and/or high voltage transmission line, an alternative location could increase the output

of an identical CSP plant by a material amount.

5 Near Commercial Technologies Of the inquiries made by Alinta into technologies that are expected to be commercial in the near term,

most were met with a reluctance to reveal information. Any technology that is under an advanced stage

of development would be expected to have an impact on the market and competitive influence of the

company developing the technology. This may be the motivation for the lack of information sharing in

this area.

The only company to respond with any information of substance was Vast Solar.

5.1 Technical Maturity

Alinta visited the Vast Solar demonstration facility in Jemmalong NSW in March 2015. This facility, once

operational, will have a 1.1 MWe capacity and is intended to prove the integrated operation of the

hardware and software developed by Vast Solar. While Alinta understands that the individual

components of Vast’s technology have all been trialed and tested, it is the successful operation of the

entire system that will encourage energy companies and developers to invest in Vast Solar.

Without making a thorough engineering analysis of the Vast Solar CSP System, Alinta’s view is that the

key items which will require rigorous testing and data records to verify performance will be the operation

of the heliostats, the material durability of the mirrored surfaces, the heat transfer properties/efficiency of

the receiver and storage system and the use of sodium as a heat transfer fluid.

Vast Solar technology has not yet proven itself to a stage that would make Alinta comfortable investing

substantially in the company, however Alinta believes there is promise in the path Vast Solar is taking.

5.2 Financial Potential

The cost estimates provided by Vast Solar for the purposes of this study have not been developed for a

CSP plant matching the Base Case in the study. Rather, Vast Solar have taken a cost estimation, which

they believe to be +/-10%, developed for a 30MW plant with 4 hours storage and scaled those cost

estimates to a CSP plant that is equivalent to Alinta’s Base Case.

Based on the assumption that the output from a Vast Solar plant with the same sized power block and

thermal storage would deliver the same electricity generation profile as the power tower in Alinta’s Base

Case, the revenue from a Vast Solar plant would be the same.

Table 17 below presents the results of Alinta’s financial model when the CAPEX and OPEX provided by

Vast Solar are used as primary inputs. Based on these results, if costs prove to be as projected, the Vast

Solar technology appears to be close to commercial.


Table 17: VAST Solar scaled base case vs. forward price curves

Forward Curve

IRR NPV (12%)

$M

Reference case

9.7% -$35.8

FC 1 10.0% -$30.5

FC 2 11.1% -$15.2

FC 3 10.9% -$17.1

6 Further Information A dedicated public webpage has been established on Alinta Energy’s website:

http://alintaenergy.com.au/about-us/power-generation/port-augusta-solar-thermal.

All milestone reports and media releases have been publicly-available on this website through the course

of the study.

Also free to the public will be the data that Alinta has collected since the installation of the solar tracker

and weather station in early June, 2014. Monthly files containing all solar data and weather station data

will be freely available to the public upon request.

On Thursday 23 April 2015 Alinta held a public information session in Port Augusta. There were

approximately 50 attendees which included local and federal politicians, representatives of NGOs and

South Australian Government agencies as well as interested individuals. At this session Alinta presented

a comprehensive summary of the progress and findings of the study to date. The session was well

attended and attracted significant media. The presentation given on the day is available on the website

referenced above.

7 Conclusion The cost of Stage One has been approximately ~$950,000. The forecast cost to undertake Stage Two of

the feasibility study is ~$1,160,000.

The outcome of Stage Two would be better certainty about the present day cost for the construction of a

CSP plant in Port Augusta. In Table 14 Alinta presented the results of thorough sensitivity analysis

including scenarios where all assumptions were at the extreme optimistic margin. Even in these

scenarios this potential project does not meet Alinta’s minimum benchmarks for further investment

consideration.

Alinta has an expectation on investment returns that is likely more aggressive than many other

organisations. Another proponent with a different appetite for risk, lower minimum investment returns

and potentially other strategic interests would be better placed to pursue a full feasibility stage of this

study or a similar study.

Based on the information compiled over the last 18 months of Stage 1 of the Port Augusta Solar Thermal

Generation Feasibility Study and the anticipated costs and benefits of refining the current level of

knowledge, Alinta has decided not continue into Stage Two.

http://alintaenergy.com.au/about-us/power-generation/port-augusta-solar-thermal

milestone 4 summary report

Documents