SmartGrids are the Future: How AI and Machine Learning Are Revolutionizing the Power Grid

Google+ Pinterest LinkedIn Tumblr +

From national grids to individual homes, AI and predictive machine learning technology has the potential to optimize energy consumption autonomously — and launch a new generation of consumer and enterprise products in the process. NewtonX conducted a 40-person Expert Survey with executives at AI-powered home energy optimization companies such as Nest, as well as with executives at large-scale AI-powered grid energy providers including Siemens and Nnergix. The survey revealed three primary use cases for AI-powered energy optimization:

  1. Data Centers
  2. Grid optimization with intermittent renewables (wind and solar)
  3. Personal home optimization

Enterprises, cities, and homes have several incentives for adopting AI-powered energy optimization tools: for one, they provide major price reductions on the consumer and enterprise side. Additionally, as regulations increasingly provide tax incentives (small wind turbines qualify for a federal tax credit of 30% in the U.S) and renewable mandates, particularly in the E.U., adopting technologies that can autonomously optimize energy consumption is a smart strategic move for many.

Automating the Grid: How Companies Are Using AI to Automate City Energy Grids

The semi-recent addition of wind and solar energy to city grids has complicated balancing and optimizing energy. This has resulted in lost renewable energy, as well as power outages, such as the massive Northeast power outage of 2003, when 50 million people lost power for multiple days. In the U.K., National Grid, the infrastructure owner and provider of energy supply around the country, recently began introductory talks with Google’s DeepMind to develop a system to accurately predict demand patterns and balance the national energy system more efficiently with wind and solar energy sources. Similarly, in the U.S., the Department of Energy has teamed up with the National Accelerator Laboratory at Stanford, in order to develop AI and Machine Learning algorithms to process data from satellite imagery, utility operations, and other sources to build an autonomous grid.

One of the challenges that AI purports to fix is that when private users generate and use their own electricity from renewable sources, such as solar roof panels, this forces utility companies to buy excess energy from private users, who generate more electricity than they use, and send that excess back to the grid. As we recently wrote, solar-powered homes are becoming increasingly popular, which will only exacerbate this problem if we don’t reconfigure the grid.

Additionally, as it stands, when demand outpaces supply, utilities companies turn on fossil fuel-powered plants, called Peaker Plants, to avoid blackouts. This process is incredibly wasteful and expensive, both in terms of consumer cost and in terms of environmental impact. AI-powered systems mitigate the risk of this happening, as they can more accurately predict demand and supply, and can also tweak energy loads to increase efficiency.

Numerous countries, and even more cities are investing in autonomous grid technology. Siemens has released a software that can operate energy grids autonomously and tweak different energy loads to increase efficiency. When new energy sources become available (like a solar park or wind farm), the software adjusts responsively. Other companies such as Nnergix use machine learning to forecast atmospheric conditions to predict when and in what volume sustainable energy sources will be available. These tools will become increasingly important as solar and wind energy take up higher and higher percentages of grid energy.

Green Homes: The Market For Energy Optimization in Homes is Growing

Numerous technology companies have released products that allow consumers to optimize home energy consumption. For instance, Nest Labs, which sold to Google for $3.2B offers a thermostat which adjusts temperatures according to inhabitant occupation habits. This AI-powered thermostat helps customers save between 10 and 12 percent on heating bills and 15 percent on cooling. Nest also works with energy companies that offer customers deals for allowing Nest to make energy consumption adjustments to their thermostats when the utilities company needs to reduce demand. This saves customers money, reduces dependence on Peaker Plants, and helps account for adjustments needed due to reliance on renewable energy sources.

As Green living becomes increasingly popular, both in the U.S. and in Europe, the market for products like Nest will continue to grow. Tesla, for instance, put its bid in the market through solar roof panels as well as its Powerwall product, a rechargeable home battery system that stores energy from solar or from the grid and makes it available on demand.  

Optimizing Data Storage: Why Reducing Consumption in Data Centers is Saving Enterprises Millions

DeepMind’s optimization machine learning algorithms cut electricity usage at Google’s data centers by 15% last year. The AI predicted load on the data centers’ cooling systems, and then controlled equipment accordingly — which resulted in a 40% energy reduction for cooling systems. Not only was this a good move from an environmental perspective, but according to NewtonX experts, it also resulted in hundreds of millions of dollars in savings for Google.

It took DeepMind two years of modelling and adjusting 120 variables before the operation reached peak efficiency — but when that happened, the company ended up with a model that saved millions of dollars and cut energy consumption considerably. Other cloud providers and companies that require costly and energy-hungry datacenters are likely to follow suit in the coming years.

Saving the environment while saving money

Automating energy consumption through AI is a simple solution to the excess energy that businesses, homes, and cities waste on a daily basis. NewtonX experts cited lowering costs, tapping into tax incentives, and preparing for future environmental regulations as key reasons why the market for AI-powered energy optimization is likely to continue growing significantly over the next decade.


About Author

Comments are closed.