Info!
UPDATED 1 Sept: The EI library in London is temporarily closed to the public, as a precautionary measure in light of the ongoing COVID-19 situation. The Knowledge Service will still be answering email queries via email , or via live chats during working hours (09:15-17:00 GMT). Our e-library is always open for members here: eLibrary , for full-text access to over 200 e-books and millions of articles. Thank you for your patience.
New Energy World magazine logo
New Energy World magazine logo
ISSN 2753-7757 (Online)

Power demand forecasting using space-based data

30/10/2024

8 min read

Feature

AI generated image of rows of pylons with electricity lighting up along the transmission lines, set against a sunset sky Photo: (AI generated): Adobe Stock/Nisit
A new space-based initiative is looking to make electricity grids smarter and more resilient

Photo: (AI generated): Adobe Stock/Nisit

Ensuring that our energy systems can adapt to new challenges, shift demand patterns, produce efficiently and maintain stability is becoming increasingly important, writes Ben Moore, Senior EU Project Manager with Carr Communications. Enter the Respondent project, a European Union (EU) funded space-based initiative that is paving the way towards smarter and more resilient energy infrastructure through projects in power generation forecasting, predicting power demand, and smart grid timing and synchronisation.

Managed by the EU Agency for the Space Programme (EUSPA) on behalf of the European Commission, the multi-year project is only six months away from completing development of cutting-edge solutions to address energy challenges that confront the EU. Respondent and its modules integrate renewable energy sources while both safeguarding grid stability and improving efficiency.

 

Partners in the €2.7mn project are Spanish distribution systems operator Aněll, Carr Communications of Ireland, Spanish renewables developer Euskabea, Greenesco Energy of Greece, Spanish research institution IREC, Greek construction company Kiefer Tek and computer research centre Vicomtech. The project coordinator is Future Intelligence of Greece.

 

Power generation forecasting  
Despite the promising progress that has been made towards its integration into existing systems, solar and wind power’s inherent variability poses a significant challenge.

 

Power generation forecasting involves predicting how much electricity will be generated by renewable sources, like solar, within a specific timeframe. Without this data, grid operators would be hesitant to incorporate more renewable energy into their systems due to concerns about the stability of the grid. Respondent’s approach to power generation forecasting leverages weather patterns, mathematical models and historical data to provide forecasts.

 

While the idea of using power generation forecasting is not a novel concept, what sets Respondent apart from traditional methods is its use of space-based data from the Copernicus Earth Observation Programme. This data, which allows for precise weather and climate parameter estimations, is combined with in-situ weather information from Internet of Things (IoT) weather stations, which in turn feeds into artificial intelligence (AI) and machine learning (ML) algorithms.

 

These algorithms promise to generate forecasts with a higher accuracy than solutions that exist on the market today, by between 8–12%. Although that increase in forecasting accuracy may seem modest, even small margins are crucial in an industry where every percentage point in accuracy can mean the difference between a stable grid and widespread disruptions.

 

The EU-funded Respondent project and its modules integrate renewable energy sources while both safeguarding grid stability and improving efficiency.

 

Power demand forecasting  
Alongside the challenge of forecasting energy production from renewable sources is the equally important task of predicting energy demand to maintain a stable energy supply. Traditional demand forecasts often rely on historic consumption trends and weather predictions. On a very hot day, for example, there is a notable surge in air conditioner use that can drive-up energy consumption.

 

Respondent’s approach to power demand forecasting includes socio-economic data in demand forecasts, taking into account factors such as average income, population density, market conditions and the political stability of an area. By including these variables in its forecasting, Respondent aims to offer a more nuanced view of energy needs that is absent from traditional methods of forecasting. In addition, the Respondent forecasting method leverages AI/ML and multi-physics simulations.

 

Its enhanced predictions aim to help utilities and policymakers manage energy resources more efficiently, reduce operational risks and optimise investments by ensuring that demand aligns with generation capacity, as well as minimising power shortages or blackouts.

 

The results and insights from the first pilot demonstration of this module, held in mid-October 2024, are currently being analysed.

 

Smart grid timing and synchronisation  

As more renewable energy sources continue to be integrated into the grid, it will become increasingly important to guarantee precise timing and synchronisation across different parts of the network. Smart grids are dependent on accurate, time-stamped data to maintain their stability and efficiency. In modern power grids, real-time data from phasor measurement units (PMUs) is used to measure voltage and current across the network, with these devices relying on precise timing to ensure that grid operators have an accurate picture of the system’s state at any given moment.

 

Most PMUs today rely on the US’ global positioning system (GPS) signals for timing. This dependence on a singular system introduces potential vulnerabilities, such as signal interference, jamming, spoofing, and other issues that can have serious consequences for grid stability.

 

Aiming to enhance smart grid synchronisation, Respondent integrates high accuracy receivers from Europe’s Galileo satellite system into existing PMU devices. That eliminates the reliance on GPS and provides a more secure and precise timing system for smart grids.

 

This system ensures sub-microsecond-level accuracy. In addition to enhancing the reliability of grid operations, Respondent’s solution includes a cloud-based monitoring module that allows grid operators to remotely monitor and analyse time-stamped measurements from Galileo-enabled PMUs, further improving grid management.

 

With more accurate forecasts for both energy generation and demand, Respondent’s mission is to ensure that grid operators, policymakers and utilities have the information that they need to integrate renewable energy sources with greater confidence and efficiency. At the same time, its Galileo-enabled timing solution offers a more secure and precise system for managing smart grids, safeguarding Europe’s energy infrastructure against the vulnerabilities of other technologies that we are currently reliant on.

 

Cross-border interconnector project launched  

In October, Bayernwerk officially launched construction of a new Rottenburg 100 MV-A substation, one of 10 to be built under the Gabreta Smart Grids project between Germany and the Czech Republic.

 

This cross-border interconnector project aims to digitalise energy distribution networks in Bavaria, southern Germany, and the South Bohemia and South Moravia, by strengthening and modernising the electrical infrastructure (such as new medium-voltage substations and lines – 20, 22 and 110 kV), installing new communication equipment (such as optical fibre and power line communication) and smart components (such as monitoring and remote-control devices), and developing the ICT (information and communication technology) based control system and applications.

 

Over four years, the project will receive €100mn of EU funding under the Connecting Europe Facility (CEF) Energy programme, which amounts to half of the starting investment. The EU grant will contribute to increase grid hosting capacity, enable remote monitoring and control of MV grids, and improve grid observability and network planning.

 

‘Gabreta will be instrumental to modernise the distribution grid in the border region. These developments will increase the flexibility, security and quality of supply of the electrical grid, enabling the integration of decentralised renewables and contributing to the broader goals of the European Green Deal,’ said Olivier Silla from European Climate, Infrastructure and Environment Executive Agency CINEA, who participated in the opening event.

 

The implementation will be carried out by the respective network subsidiaries of the two companies, Bayernwerk Netz and the Czech network company EGD, in an area that covers almost 70,000 km2.

 

Gabreta Smart Grids is scheduled to run until the end of 2028.

 

Global review of smart grid technology innovation   

An International Energy Agency (IEA) report published in July 2024 analysed trends in smart grid technology innovation.

 

Smart grid technologies are pivotal for modernisation of a consistently overloaded grid. The report showcases where, when and in which subsectors innovation is occurring.

 

Apparently, 2011 saw a peak in smart grid innovation with 2,000 unique inventions produced, representing 11% of power sector innovations. Following a period of decline, the relative share of smart grid innovations increased to 13% in 2022.

 

In 2020, technologies related to monitoring or controlling equipment for energy generation units and supporting power network operation or management collectively accounted for 41% of total smart grid patent registrations. This increased share may be linked to the rising capital investment in power equipment during recent years.

 

East Asia (mainly Japan and China) has dominated smart grid innovation in the last two decades, accounting for over half of the total. Since 2007 it has consistently held the top position among regions. North America (mainly the US) and Western Europe (mainly Germany) together share the remaining smart grid inventions.

 

More than 40% of smart grid innovation is happening in 10 cities around the world. The top areas for innovation between 2000 and 2022 were Tokyo, Seoul, Beijing, Nagoya, Nuremberg and the San Francisco Bay area.

 

Europe emerges as a hub for smart grid technology specialisation, as shown by the OECD measure Revealed Technological Advantage (RTA). Conversely, Japan, the US and China exhibit lower RTAs, suggesting that despite their considerable innovation endeavours in smart grids, they lack specialisation in this particular field.

 

Although family size has decreased in smart grid inventions in more recent years, inventors have chosen to protect their inventions in fewer offices, and patent claims have increased since 2009.

 

  • Further reading: ‘Maintaining a stable electricity grid during the energy transition’. Renewable power sources impose many new demands on electricity grids built around large dispatchable point power generation sources, such as coal-fired power plants. Grid investment and expansion are not keeping pace with variable renewable energy deployment and the phase-out of dispatchable power capacity, creating associated risk to sensitive supply demand management needs, argues Debo Adams, Studies Manager, International Centre for Sustainable Carbon.
  • Global electricity demand is forecast to double by 2050, requiring extensive expansion of electricity grids and technological upgrades, according to a new report from DNV.