AleaSoft Energy Forecasting, November 25, 2024. Interview by Irene Garrido Domínguez, from Revista Energética, with Antonio Delgado Rigal, PhD in Artificial Intelligence, founder and CEO of AleaSoft Energy Forecasting.
Currently, there is no regulation in Spain to promote energy storage, which limits investment in this area. From your perspective, what kind of regulatory incentives would be necessary for stand‑alone BESS and hybridisation to be developed on a large scale in the Spanish market? What lessons could we learn from other countries in this respect?
Currently, Spain has some regulatory basis, but there is still a lack of experience in the development of energy storage. The existing regulation is only a first step. A more robust and detailed framework is necessary, which unifies criteria for project assessment and is flexible enough to adapt and improve as the sector gains experience and barriers are identified. For example, one possible improvement would be allowing more flexibility for hybrid plants, so that they do not lose priority in dispatch by including batteries.
In addition, incentives are still lacking, which limits the profitability of storage projects. The sector is awaiting the announced capacity market, scheduled for early next year, although there is no certainty yet about when and how it will be implemented. For investors, this certainty is crucial.
An example of how a capacity market can support storage development is the case of the United Kingdom, where 3.5 GW of installed capacity were reached in 2023. There, participation in ancillary services has also been an important source of revenue, although over time this revenue has declined as competition has increased. This is a lesson we could learn from this market: revenues from ancillary services will tend to decline as new projects come online.
Italy and Australia offer specific incentives and capacity remuneration schemes to support storage projects and respond to peak demand. In California, specific auctions have been set up and tax credits and subsidies are provided to facilitate investment in energy storage.
Considering current costs and market dynamics, do you think it is feasible in the short term to integrate storage batteries to manage the intermittency of renewable energies and avoid curtailments? What business models could make this investment profitable in the coming years?
The integration of battery storage systems to manage the intermittency of renewable energies and avoid curtailments is an increasingly viable solution in Spain. Although the initial costs are still high, the price of batteries has been decreasing, falling by 50% in the last year, which has helped projects with batteries to become profitable. Currently, this is more relevant in hybrid systems that combine renewable energies and storage and that already have access to the grid and balance of plant, which reduces the CAPEX associated with the battery. This model also helps to avoid curtailments and maximises the use of the energy produced.
A significant part of the revenue from battery storage will come from price arbitrage. Batteries can be charged when electricity prices are low and feed that energy into the grid when prices are high, taking advantage of fluctuations in market prices. In addition, participation in ancillary services offers another source of revenue, although, as commented above, this revenue may be more limited in time.
Although green hydrogen in Spain is currently in a practically experimental phase, what structural changes, both in terms of infrastructure and policies, do you consider necessary for green hydrogen to become a competitive and widely adopted energy vector in our country?
Firstly, infrastructure for the production, transport and storage of green hydrogen is essential, and it is important to start planning and developing these projects well in advance.
In addition, it is essential to support the development of green hydrogen through specific aids and incentivise its use in industries such as steel, cement, long‑distance transport and aviation, which are difficult sectors to electrify. It is also important to promote its use as a primary carrier to produce other renewable fuels, such as methanol or ammonia.
Finally, it is important to incentivise the use of green hydrogen for seasonal energy storage. Its production at times of high renewable energy generation would not only reduce curtailments but would offer a large‑scale storage solution, which can be used at times of the year when renewable energy production is not sufficient to meet demand.
With the growth of renewable energy installed capacity, especially solar energy, we have seen an increase in price cannibalisation, particularly in the spring months. What tools or market reforms do you consider appropriate to mitigate this phenomenon? Does a marginalist market still make sense in these circumstances?
To reduce price cannibalisation, it is essential to develop storage, increase demand and make it more flexible. Storage allows energy generated during low demand hours to be shifted to peak hours, which helps to balance supply and demand, reduces cannibalisation and stabilises prices.
Regaining the demand lost in recent years, in part due to the COVID crisis and the subsequent energy crisis, and continuing to drive its growth will be essential measures to mitigate the fall in prices at peak renewable energy production hours. One way to achieve this is through Data Centres, electrification of industrial demand and electric transport. In addition, incentivising consumption during these hours will help to reduce the impact of renewable energy generation on market prices.
On the other hand, long‑term power purchase agreements (PPA) offer a way to ensure stable revenues for renewable energy producers and reduce their exposure to spot market prices, thus mitigating the impact of cannibalisation. In the mid- and long‑term, increasing the capacity of international interconnections will allow electricity to be exported during the hours of high renewable energy production, thus avoiding a sustained fall in prices.
At AleaSoft we believe that the marginalist market is still appropriate, as it is the one that provides the necessary signals for investors. In the more than 25 years of operation of the marginalist market in Spain, the electricity system has undergone a great transformation, from a high dependence on coal in its early years, to the development of combined cycle gas turbines, wind energy, and in recent years, photovoltaic energy. During all this time, the market has worked correctly and has maintained its equilibrium.
To avoid curtailments and make more efficient use of renewable energy, increasing electricity demand would be key. Which sectors do you see as priorities to foster this demand? What policies or incentives could be implemented to accelerate the electrification of the economy, especially in areas such as electric mobility, heating or industrial use of electricity?
To boost electricity demand, it is essential to promote electrification in key sectors such as industry, transport, and new areas such as Data Centres or the production of green hydrogen and other renewable fuels.
Incentivising consumption at times of peak renewable energy generation could be achieved through a tariff that sends favourable price signals for consumption at these times.
In addition, developing transmission and distribution grids is essential to manage the increase in demand and renewable electricity production. Expanding charging infrastructure is also important to accelerate the adoption of electric vehicles.
Finally, policies such as tax incentives and subsidies can facilitate the electrification of industrial processes and transport, supporting a transition towards a more electrified and sustainable economy.
For years, the manufacture of renewable energy equipment (photovoltaic modules, solar inverters, battery cells, etc.) has been dominated by Asian companies. How could Europe encourage the local manufacture of this equipment in order to gain strategic independence? How do you assess the current policies promoted by the EU in this regard? Are they viable?
The manufacture of renewable energy equipment has been dominated by Asian companies, especially Chinese, which have saturated the global market with products at very competitive prices. While the implementation of protectionist measures could help to protect and incentivise local industry, these policies could also increase the costs of the energy transition and slow down the process if not properly managed.
Current EU policies, such as the Green Deal Industrial Plan, the Net‑Zero Industry Act and the Critical Raw Materials Act, are positive steps towards a more independent industry. However, the viability of these measures will largely depend on their ability to make European products competitive in price and quality with Asian alternatives.
AleaSoft uses artificial intelligence in its energy forecasting methodology, how is this technology used and what types of algorithms do you consider more suitable for predictive analysis in volatile energy markets? What specific advantages does AI bring in terms of accuracy and adaptation to climate change?
The use of Artificial Intelligence (AI) in AleaSoft’s AleaModel models allows us to analyse large volumes of data to identify patterns and do accurate forecasts. Thanks to AI, our models adapt quickly to changes in variables, including changes in weather patterns resulting from climate change, which is critical to predict prices and other variables of energy markets.
In particular, recurrent neural networks are especially useful to work with time series, such as market data, and are effective in predicting price, demand and production patterns in the energy sector.
After 25 years of experience, what is the average margin of error of AleaSoft’s forecasting models, and how have you managed to reduce it over time? Can you describe any specific cases where your forecasts have made a significant difference to your clients’ strategic decisions?
The basis of our AleaModel methodology has been maintained since the foundation of AleaSoft, and although today Artificial Intelligence (AI) is booming, at AleaSoft we were pioneers in its use to do highly accurate forecasts for the energy sector. Constant investment in R&D&i has been fundamental to the evolution of our models, which have been improved through continuous research and innovation.
The average margin of error of our forecasts varies depending on the time horizon and the analysed variable. For example, for short‑term national electricity demand forecasts, the margin of error is typically less than 2%. However, for market price forecasts, this margin can be higher due to the intrinsic complexity and volatility of energy markets.
In long‑term price curve forecasting, the main objective is to capture the market equilibrium and project it into the future. An example we often show is a long‑term price forecast we carried out in October 2010, whose accuracy has been maintained over the years. At that time, the generation mix was very different from today, with a predominance of combined cycle gas turbines, a significant presence of coal and less interconnection capacity with France.
Since then, the market has changed considerably, with the growth of photovoltaic energy and also wind energy, which was already present in 2010 but has continued to develop, the doubling in interconnection capacity with France compared to when we made the forecast, and the virtual disappearance of coal. However, our forecasts have remained accurate, as our models were able to capture the market balance in the historical data and project it into the future.
In the context of an energy transition towards a decarbonised and digitalised economy, what role do you believe AleaSoft will play in developing a more resilient and flexible electricity system globally? What technological or methodological innovations do you plan to implement in the coming years to meet these challenges?
In our more than 25 years of experience, we have witnessed the evolution of the Spanish electricity market practically since its origin, and we are motivated to continue accompanying this process in the next 25 years when Europe aims to achieve net zero greenhouse gas emissions. Our clients will be key players in the energy transition, facing the challenge of continuing to deploy renewable energies and developing new vectors such as energy storage. At the same time, large consumers will be challenged to decarbonise their processes, and grid operators and distributors will have to plan and ensure the stability of a system with a higher penetration of intermittent energy sources.
In this context, we aim to continue supporting our clients in planning and optimising their operations through our forecasts, analyses and knowledge of energy markets. To achieve this, we will continue to invest in R&D&i, improve our models and services, and adapt to the changes the sector will experience in the coming years.
