Artificial Intelligence and the Future of the Power Grid

Artificial intelligence (AI) has become one of the fastest growing tech sectors, with over five billion dollars invested in AI startups.  Despite Elon Musk’s warnings about its dangers, AI is rapidly advancing and is expected to play a major role in our lives in transportation, healthcare, security, and other sectors.  Artificial intelligence—the ability of machines to perform cognitive functions normally associated with the human mind—has seen enormous advances in the past few years due to a type of AI called deep learning.  And the prevalence of artificial intelligence can already be seen in many everyday experiences; for example, when Facebook automatically recognizes faces in uploaded photos or when Apple’s Siri answers your question, AI is at work.

One of the industries where artificial intelligence is making important inroads is the electricity sector.  On the supply side, numerous companies are using AI to improve power production efficiency.  For example, earlier this year, GE announced AI related technology for wind turbines in Japan expected to increase power output by 5% and lower maintenance costs by 20%.  On the solar front, NEXTracker uses machine learning in its solar trackers, which can increase production by up to 6%.  And AI is not just for renewable resources: Siemens uses artificial intelligence algorithms to improve combustion efficiency, reduce emissions, and lower the wear on gas turbines.  UK-based EDF Energy is testing machine learning to predict demand for the next day more accurately than humans, resulting in energy saving in cogeneration plants up to 15%.  Finally, some coal-fired plants have used AI to increase efficiency and reduce emissions.  For example, Xcel Energy has implemented sophisticated artificial neural networks to make recommendations on how to adjust operations in order to reduce emissions in its Texas coal plants.  Clearly, AI is set to have a significant impact on how power plants operate in the future.

Artificial intelligence also has the potential of making a substantial difference in helping balance demand and supply of the electricity sector as well.  The recent rise of renewable energy, from both power plants and distributed generation, has caused its share of challenges to the power grid for producers, utilities, and consumers.  To help on the consumer side, in the town of Reidholz, Switzerland, forty homes are trying a new technology called Gridsense so that AI can improve how power is used within homes and helping ensure that “the power grid is always operating at optimal load” by adjusting customer energy consumption and coordinating with the photovoltaic generation in the neighborhood.  On a larger scale, Google’s DeepMind is in discussion with one of the UK’s energy providers, National Grid, to use artificial intelligence on their power grid to help balance supply and demand.  DeepMind has already used its program at Google’s data centers to cut electricity usage by 15%.

Another area where AI has the potential of making a big difference on the grid is in the control and operation of demand response.  This is where large consumers of electricity are rewarded when decreasing their energy requirements on short notice to help balance the grid, and this can be cheaper for the operators than turning on very expensive power plants.  Demand response programs have existed for some time now, and improving AI technology may provide significant benefits to consumers hoping to optimize their participation in the program.  As one source states, “Demand management is also seeing an explosion of AI activity with use cases covering areas such as demand response, building energy management systems, overall energy efficiency and DR game theory.”  One company, Upside, is using AI to manage a portfolio of storage assets to provide real-time energy reserves to relieve stress on the grid.  It has developed an Advanced Algorithmic Platform that manages demand response of different devices to be run in parallel.  Another company, Open Energi, uses AI to optimize companies’ assets to save energy and cut costs by choosing what time to run them based on supply and demand fluctuations in the power grid.

The use of Artificial Intelligence is already at work improving efficiency in the electricity sector for power plants, grid operators, and both large and small consumers.  Whatever lies in the future for the power industry, signs are promising that artificial intelligence will play an essential role in improving the overall efficiency on the grid.

Filed under: Artificial Intelligence, Energy Efficency, Power GridTagged with: , , , ,