What is LCOE and how to use it in photovoltaic projects?

The Levelized Cost of Energy, or simply LCOE (Levelized Cost of Energy), is a term that has been widely used in the photovoltaic solar energy market.

This parameter is seen as the new metric, or the most effective metric, for evaluating the viability of a photovoltaic system, to the detriment of the system's traditional cost per watt.

How is LCOE calculated and how should it be used? Is it purely a marketing metric, used by equipment manufacturers to argue that their solution is better than another?

Is it something that designers should look at to choose the best solutions for their project? Is it something that investors should look at to make the best decisions?

In this article, we will discuss what LCOE is, how it is calculated, and how it is used. Some limitations and analysis errors associated with this metric will also be shown.

It is worth noting that the objective of the article is to present in a clear, direct and simplified way what the LCOE is, not including analyzes with more specific financial indicators such as IRR, NPV, ARR, etc.

We will save the exploration of these concepts in LCOE analyzes for future articles. At the end, we will show a case study applied to a 5 MW solar photovoltaic plant in DG (distributed generation), evaluating the most viable solution among the various solutions presented.

What is LCOE?

At its core, LCOE was created and designed to compare the relative cost of energy produced by different energy generation sources. The purpose of its creation was to understand which energy source would be most competitive in a given generation project: hydro, thermal, wind or solar, for example.

LCOE calculation

LCOE is defined as the division of total project costs, including not only invested capital (Capex) but also operating costs (Opex), by the energy generated throughout the plant's entire operation.

The calculation must also include any residual costs such as, for example, the value of the equipment at the end of the project's life. In a simplified way, we can equate LCOE as follows:

LCOE = CT / EP

Where:

• CT = Total cost of the plant, including Capex, Opex and residual [R$];
• EP = Total energy produced over the useful life of the plant [kWh].

The total plant cost (CT) can be calculated as:

CT = Capex + Opex – Residual

Where:

• Capex = Cost of building the plant [R$];
• Opex = Cost of operating the plant throughout its useful life [R$];
• Residual = Value of equipment at the end of its useful life [R$].

It is possible to identify that the LCOE returns a value in R$/kWh, which is a more tangible value for the energy consumer, who is used to seeing this value on their energy bill every month.

Furthermore, it represents real value for this consumer who is investing his money to save on energy bills, and not to have a more or less powerful solar park.

LCOE in solar energy

As mentioned above, LCOE was born from an initial need to compare different energy sources, but if we have already defined that solar energy technology is already the chosen one, does it still make sense to use this indicator?

The answer is yes, it makes perfect sense. The LCOE will allow you to compare different products, technologies, architectures and solutions and conclude which of them deliver the lowest cost for energy. Some factors that impact the LCOE of a photovoltaic system:

• Climatic conditions (temperature, solar radiation, wind incidence, etc.);
• Site conditions (property cost, terrain irregularities, proximity to substations, infrastructure, exposure to dirt deposition, environmental licenses);
• Photovoltaic modules (monocrystalline or polycrystalline, bifacial or monofacial, PERC, MBB, type N or type P);
• Inverters (string or central inverter, with multiple MPPTs or not, with forced or natural ventilation, conventional inverters or optimizers, overload);
• Structures (fixed or mobile – solar trackers, single-post or double-post, construction material, type of surface treatment);
• Operational issues (maintenance frequency, required labor, access to the plant, etc.)
• Miscellaneous costs (park monitoring, insurance, etc.).

Using LCOE helps identify the best opportunities for a given application, allowing you to specifically assess whether a given architecture, concept or component change has a beneficial cost implication or not.

The LCOE calculation allows, for example, to evaluate whether it is worth using photovoltaic modules with greater power and efficiency, but with a higher price in R$/Wp, or purchasing modules with a lower rate of degradation, but with a higher price.

Or even test the feasibility of applying solar trackers or adopting a fixed axis solution. It is worth mentioning that in some situations, a change may affect another choice.

In this case, all changes must be taken into . For example, suppose that in a given case the project team is asked by the plant owner whether it is worth using a central or distributed inverter topology.

Choosing one or the other will not only impact the cost of the inverters, but also cables and electrical connections, freight, labor for installation, operation and maintenance, and more. This way, all the changes that each option carries need to be taken into to calculate the LCOE.

Limitations of LCOE

A common misconception is that the project with the lowest LCOE is always the objective to pursue. LCOE is a good tool for studying options within a project and guiding decisions from a macro perspective.

However, it is not always the most useful metric when making decisions about projects, especially in specific cases. For example, in an LCOE analysis, a specific solution from a specific supplier may appear to be the most competitive, with the lowest R$/kWh ratio.

However, this solution may not be a completely technically validated solution, or this supplier may not have a solid market share, with the risk of it ceasing to operate in the local market due to a strategic decision and putting the service for possible failures at risk. provided for under warranty.

Another good example concerns the reliability of the system and its components. A given solution may have better LCOE, but may have greater susceptibility to failures. LCOE does not take into the reliability of the energy produced by a project.

In short, although LCOE is valuable in many situations, given the variability of situations and the complexity of the energy sector in general, LCOE is just one among many factors that must be considered when making decisions on photovoltaic projects.

Case study: LCOE analysis of a 5 MW minigeneration plant

In the following example we will explore some common issues in minigeneration solar energy projects:

• Optimal overload for inverters;
• Use of higher or lower power modules.

We emphasize that the purpose of the examples is not to say that one option is better than the other, but rather to direct readers on how to carry out LCOE analyses.

It should also be noted that the values ​​used may not represent reality at a given moment, given the dynamics of market prices, which vary according to availability, exchange rate situation, economic policies (such as ex-tariffs), among other things. For both examples, we will use the cost composition and values ​​found in Figure 1 as a reference.

The plant analyzed in this example is located in the north of Minas Gerais, uses trackers and has 5 MW of AC power. Adding the annual operating costs of this plant, we can create a table listing items that have a cost proportional to the number of modules, costs that are fixed and recurring costs:

To obtain the LCOE value, the PVSyst software will be used, which takes into the total generation of the system, modeled losses and the decline in yield of the modules and inverters over the 25 years of operation of the plant.

Example 1: 330 W polycrystalline or 440 W mono-PERC modules?

The considered costs of photovoltaic s are shown below:

The 330 Wp module appears to be a good solution, as it has a lower total cost, however, the costs of items that are proportional to the number of modules, such as structure, labor, area rental and O&M, must also be taken into . (operation and maintenance). With this we have:

As LCOE directly depends on the amount of energy generated, it is important to highlight the role of generation losses for both modules. The 330 Wp module has a temperature power loss coefficient of -0,41 %/oC, while the 440 Wp mono-PERC module has a coefficient of -0,35 %/oC.

If we consider a module operating temperature of 60 oC, the 330 Wp would lose 10,3% of its power, while the 440 Wp would only lose 8,75%. This difference in losses due to temperature will be directly reflected in the amount of energy produced and consequently in the LCOE value.

Completing the data from the PVSyst economic analysis module, we then obtain:

The 440 Wp solution, even with the higher module cost and total construction cost, produces significantly more energy than the solution with 330 Wp modules, which brings the final energy cost to the customer, the LCOE, of 254 R$/MWh, an advantage of 16 R$/MWh compared to the first solution.

This difference, which to someone less attentive may appear to be small, represents a gain or loss (depending on the choice made) of more than R$5 million over the course of the plant's operation.

Example 2: Inverter overload

In this example, 9 variations of the same system will be compared, with the same 440 Wp module, but with inverter overload varying between 120% and 150%. The cost estimation methodology will be the same as in the previous example.

The overloaded inverter works with module power above its nominal input power, something more than common in photovoltaic projects. The total cost of the work increases with the addition of modules, but generation does not increase in the same proportion.

This is due to the energy clipping effect, when the inverter limits the power extracted from the modules to its output power at the AC point. The figures below show the behavior of Capex (cost of the work) and energy generation as a function of the inverter loading rate.

Because of this asymmetry in gains, the LCOE curve will not be fully linear with the amount invested in the work. This means that there is a range of optimal overload values ​​for this plant, as we see in the following figure.

The best LCOE range in Figure 8 is approximately between 132% and 137% of loading, with 135% being the optimal value – that is, the loading value that provides the lowest energy cost in R$/MWh.

It is worth mentioning that the optimal overloading result is valid for a specific situation: the plant used in this example. For other regions, inverter technologies, plant topologies and technical conditions the result may be different.

Conclusion

LCOE (levelized cost of energy) is an important parameter in analyzing the viability of photovoltaic projects, although it is not the only aspect to be considered. LCOE helps in making decisions regarding several variables in a project, such as the type of module to be used, the type of inverter and the inverter loading rate, among other things.

More important than the cost of a solar plant (R$/Wp), the LCOE reveals to the customer or investor the cost of the energy generated by that plant (in R$/kWh or R$/MWh), becoming a important indicator of economic return.

In this article, two examples were analyzed, considering two very common variables in most projects: type of module used and inverter overload.

In the first case, it was shown that high-efficiency, more expensive 440 Wp modules provide better LCOE than the solution with 330 Wp modules. In the second case, it was shown that it is advantageous to overload the inverter in a 5 MW plant.

In the example analyzed, the best LCOE was obtained with a loading index of 135%, that is, the power of the set of photovoltaic modules is 35% above the nominal power of the inverter.

The decision to use fewer or more modules in a solar plant is not obvious and LCOE assists in this decision making, ensuring the best configuration for the solar plant at the lowest cost of the energy produced.