By Groupe Réaction Inc.

Cost modeling calculation of CSP projects is a very important feature for all players involved in the sector. Project developers and EPC companies use cost modeling to assess the economic viability of a project, whilst policymakers, regulators and utilities use it to assist them in establishing incentive and support schemes. Cost modeling is a very powerful tool that allows for estimation of the LCOE (Levelized Cost of Energy) and other economic metrics based on project costs, financial parameters and energy yield.

Under the current global financial crisis cost modeling is an important tool that can mean the difference between the successful attraction of investors and banks or a project’s abandonment.

Although different methodologies can be used, the final goal of any cost-modeling tool is to calculate the required tariff at which electricity generated by the CSP plant has to be sold in order for the project to cover all expenses, debt and its equity investors’ required rate of return. To do so, the cost model needs to balance simplicity, accuracy, precision and representativeness. Thus, the complexity, time requirement and effort associated with the cost modeling process regarding input data and calculations will depend on the intended use of the results from a simple initial feasibility analysis to PPA (Power Purchase Agreements) negotiations or policy making decisions.

The following figure shows the structure of a typical cost modeling tool. As in any other modeling tools, the accuracy and uncertainty of the results depends directly on the quality of the input data. In later stages accurate results are required and the cost modeling tool should be fed with real and precise input data while estimations could be enough for the early stages of a project. However the model complexity remains the same since the model is usually developed once (or sold to a third party) and used for different projects. Therefore, it is mainly the input data that determines the accuracy and uncertainty of the results of the cost modeling process.

Figure 1: Cost modeling process

From initial pre-feasibility studies until reaching financial close, cost-modeling is a continuous and iterative process that is performed in almost every stage of the development process in order to bring the project to success. 

In CSP project development, as in any other type of renewable energy project, the development process is normally broken down into the following stages:

·         Market Study (Prospecting)

·         Pre-feasibility study

·         Feasibility study

·         Development

·         Detailed Design and Construction

During the prospection phase, a potential market and site is assessed to find the suitability to develop a CSP plant. Right after this opportunity is identified, the first cost-modeling calculations take place in the so-called “pre-feasibility study” of the project.

Even though the project is in its early stage phase, the pre-feasibility study should be (already) a high-level review of the main aspects of the project, including solar resource assessment, energy yield calculation, construction cost and LCOE estimation, as well as the grid connection availability and site suitability.

Cost modeling & LCOE calculation

The purpose of the pre-feasibility phase of a project is to assess if the project is worth taking forward without committing significant effort and expenditure from investors.  Cost modeling plays a very important role during this phase as it performs the key task in   the techno-economic optimization of the CSP plant that is required to find the optimal plant configuration that leads to a minimum LCOE. This parameter, as previously explained, is commonly used to compare the competitiveness of the project among other technology alternatives or even other investment options.

One of the most common methodologies used for cost modeling to calculate the economic performance of CSP projects is the discounted cash flow (DCF) method (used for example by NREL´s SAM (System Advisor Model) and other specialized financial models). This type of cost-modeling tool provides estimations of the projects revenues, operating costs and expenses, tax, debt and capital repayment obligations. A DCF cost model is a very versatile tool for calculating after tax cash flows that are discounted in order to calculate, among others, the net present value (NPV) and the internal rate of return (IRR) of the project. These financial outputs are used as indicators to investors to decide whether a project is attractive or not.

How to do it?

In terms of cost-modeling as shown in Figure 1, the first action to consider is the definition of the input data: energy yield, cost data and financial assumptions.

Firstly, with technical configuration of the plant and a specific set of solar radiation and ambient conditions representative to the site (defined by a Typical Meteorological Year, - TMY), a performance model (such as SAM) is used to calculate the energy yield which is one of the inputs for the cost modeling tool.

Secondly, the total investment costs (CAPEX) and operating costs (OPEX) need to be determined. As this is a first assessment, and developers are trying to keep expenditure low, the estimated costs are most likely to be based on indicative quotes or comparison with similar projects or benchmarking.  However, estimating the investment cost of a CSP plant has proven to be one of the most sensible parameters when assessing the feasibility of a CSP Project. As this is a major issue, most time and efforts have to be allocated to this task and several sources of data (publications, papers, journals, etc.) and industry feedback have to be analyzed in order to find and set an average and actual cost according to the market conditions and technology selected.

Table 1 and 2 give brief breakdowns of the CAPEX and OPEX costs to be considered when estimating the overall cost of a CSP plant.

Table 1: CAPEX Breakdown Structure

Table 2: OPEX Breakdown Structure

Thirdly, as shown in Figure 1, the financial assumptions are defined.

Once the complete set of inputs is defined, the cost modeling tool calculates the economic metrics of the projects, such as the LCOE, NPV (Net Present Value) and IRR (Internal Rate of Return) which are usually the reference metric used to assess the feasibility and attractiveness of the project. The accuracy of the results will directly depend on the accuracy of the input data.  Usually at this stage of development (pre-feasibility), project development and construction cost uncertainties remain preliminary (they could change by up to ±30%) and still, the final project needs to be optimized and clearly defined in order to make it bankable.

A number of factors need to be taken into account when calculating the potential LCOE of a plant. The below equations take into account a typical project finance scheme with a combination of equity and bank loan(s), where the weighted average cost of capital (WACC) is used as the discount rate according to the following formula:

The assumptions taken for the WACC calculation must be carefully evaluated since it can influence the investor decision towards one option or another.

The formula used for calculating the LCOE is:

(Source: CSP Parabolic Trough Report: Cost, Performance and Key Trends)

Where,

CAPEX in the initial investment

OPEX_t is the O&M costs in year t

n is the lifetime of the project in years

E_t is the electricity generated in year t

r is the discount rate (as explained, due to the project finance structure assumed as a reference, WACC has been used as the discount rate)

If the outcome of the cost modelingperformed during the pre-feasibility study is favorable, a detailed feasibility study should be carried out. This consists of a significantly more detailed assessment of all aspects of the project where the feasibility of the project is evaluated with enough detail for the interested parties and investors to make a commitment to proceed with its development.

The approach taken at this stage follows an iteration exercise, where cost modeling results are updated in order to increase its accuracy since the results of the previous cost modeling performed in the prefeasibility study are based on initial inputs (solar resource, energy yield and costs) that usually come from estimations, assumptions and market-alike projects. For that purpose, real quotes from manufacturers, accuracy on the solar resources assessment and energy yield calculation need to be pursued. Once again, cost-modeling assessment is used to successfully arrive at a reliable outcome of the feasibility study and reduce its uncertainty to a minimum.

After the feasibility study, the development stage is reached. The development phase takes the project from the feasibility phase to the financial close. This includes the gathering of all permits needed and also the outline design and selection of contractors. On this side, based on the latest cost-modeling outcome, the developer already has a good idea of the expected price an EPC contractor would ask for.

When doing proper cost-modeling, the developer could even make a project more attractive by setting a public tender for constructing the plant. This approach has been adopted lately in different tenders worldwide, where the offering company sets a fixed budget to the project that is normally based on the results of an internal cost-modeling optimization task.

Finally, the key systems and structures will be designed in detail by the EPC company and after this, the project will start construction.

To comment on this article write to the editor, Jennifer Muirhead