Solar Generation Curve Shifting for Revenue Optimization

The contribution of renewable energy to a country’s energy mix has increased substantially in the last decade or so. This is primarily because of the reduction in the Levelised Cost of Energy (LCOE) that renewables have achieved over the years. The question is, does the reduced cost of energy due to increased renewables get reflected in the wholesale market energy price? The answer changes depending on how a country estimates its energy prices based on supply and demand. In this post, we will go through the price curve of two such energy markets i.e., GB and Australia, and try to find the driver behind the price curve. The aim of analysing the price curve is to find out, can the generation curve from a solar plant be modified to take advantage of these price curves.

How is a country’s wholesale price calculated?

In GB, the trading period for electricity is half-hourly based. During each trading period electricity generators bid for price along with the capacity it can supply. These bids are then stacked in ascending order i.e., the cheapest first and this is continued till the demand is met. However, the price of the electricity is set based on the most expensive electrical unit purchased and hence the wholesale price generally gets defined by electricity supplied by the power plants that use fossil fuel as an energy source which has a much higher LCOE compared to renewable plants. This method of pricing is called marginal cost.

The Australian market works on similar marginal cost pricing except, instead of half-hourly based pricing Australia in 2021 moved to a 5-minute trading period and the electricity market in Australia is divided into zones while the whole of the GB market operates on single wholesale electricity price. Although based on a similar principle, the price curve observed in the Australian market varies massively from the GB market, primarily due to the difference in renewables (especially solar) contribution to the respective grids.

Impact of renewables on the price curve

Since the wholesale price in Australia changes with zones, for the sake of comparison we will use Queensland wholesale price and try to draw a parallel with the GB price curve. One of the reasons for selecting Queensland is because it has one of the highest renewable contributions of 26% into its daily energy mix, dominated by residential solar which contributes 10.6% (data between June 2022 – June 2023) followed by utility-scale solar contributing 8.4% of the total energy mix. The total solar contribution of 19% in Queensland is much higher than in GB where the solar contribution as of 2023 sits at 4.5%. The primary focus of this article will be on the impact of solar on the price curve.?

Fig. 1 compares the currency adjusted monthly wholesale electricity price from 2023 for GB and Australia. From the price curves, it can be observed that there is a clear price dip during the day hours in Australia which is primarily linked to access energy available in the system dominated by solar compared to the demand, and most of the residential demand is met locally by residential solar. At the same time, the GB price curve does not dip as much primarily because the wholesale price is dominated by fossil fuel generating sources (primarily Natural Gas) and the solar penetration into the grid is still quite low.

This eventually pushes one to think that with the increase in solar contribution, the GB price curve will probably start shaping like the Australian one and the latter may get even worse with more solar. A subsequent question that arises with this is can the solar generation curve be shifted to take advantage of the evening price rise i.e. design the solar plant for revenue maximization and that’s what we look to explore in the next section.?

Solar curve shifting

Almost all fixed tilt structures installed in the Northern Hemisphere (NH) are designed to face true south as it helps in maximizing the yield from the solar farm and vice versa for projects in the Southern Hemisphere (SH). This design philosophy has been followed for years now as the objective is to maximize the project generation but if the objective is changed to maximizing the revenue, then this might not hold true in today’s time. From the previous section, we observed that with the increase in solar contribution, the day price has been on a constant fall, which means the projected revenue can be maximized if the solar generation curve can be pushed more towards late afternoon hours to capture the price increase.?

For solar projects this can be achieved by changing the project azimuth, i.e. instead of the solar modules oriented to face south (in NH) we can shift it to face towards south-west which would help in capturing more energy during late afternoons, on the other hand this will reduce the energy generated during morning hours. Fig. 2 shows that how the energy curve changes over the months with the change in azimuth for a project located in GB.

From Fig. 2 it is clear that the months between April to September show some evident increase in generation during evening periods while for the other months, it is not that evident. The overall project generation would fall due to the change in azimuth, but this exercise aims to see if there is any substantial revenue gain from this, which is what we explore in the next section.

Project revenue analysis

We will try to understand and analyse the impact of change in azimuth on project revenue by forming three cases which are,

• Case 1 – In this case, the project is located in GB and the revenue is calculated using the GB price curve
• Case 2 – For this case, the project is located in Queensland and the respective zonal price cure is applied
• Case 3 – This case helps in understanding revenue for a GB located project superimposed with a price curve observed in Queensland driven by solar

Several insights can be drawn from the above three cases, a few of which are,

• From Case 1 it can be observed that with a change in azimuth, the yield reduces and the project revenue reduces as well. This is primarily because the price curve is highly dominated by fossil fuel energy sources
• In Case 2 as expected an increase in project revenue is observed as the orientation moves more toward the west primarily because of the jump in price observed during late afternoons
• The other important point to take note of in Case 2 is the absolute revenue from a project located in Australia is almost half of that of a project in GB even though the generation is almost twice as high. This is because of the price drop observed during the solar generation hours. This makes a strong case for using co-located BESS for an Australian project where energy arbitrage would give a huge boost to project revenue given the price volatility
• In Case 3, the Australian price curve is super-imposed on the GB located project, ignoring the absolute revenue values and concentrating on the percentage gain from the base case (i.e., 0° azimuth) it can be seen that a substantial gain of 5% in revenue is observed for a 30° azimuth. This happens to be the sweet spot as the projected revenue decreases beyond this point with the increase in azimuth. The objective of this case is to show that as we move towards increased solar contribution in GB, we can expect to observe a skewed price curve and design changes will need to be done to maximize project revenues

Key takeaways

The article aims to elucidate how the wholesale price curve changes with an increase in renewables (especially solar) contributions to the grid. It is fascinating to understand how several countries around the world estimate energy prices based on supply and demand and how historic pricing methods still ensure that the price curve is dominated by fossil fuel energy sources. Looking at the Australian grid price helps us understand how increasing distributed renewable energy sources shape the price curve and once the GB market reaches that stage how we can expect the price curve to be shaped.

The article challenges the conventional azimuth assumptions for solar projects and provides an innovative approach to the problem where the azimuth is moved more towards the west for revenue optimization. This approach ensures that the project takes advantage of the late afternoon price rise. Re-orientating the solar structure is just one such approach, there could be a lot of other ways to tackle this issue, like programming trackers to capture more of the afternoon peak, etc. The solution can be region-specific as each region has its specific energy mix and hence the price curve varies accordingly.