How Machine Learning Boosts Energy Production

– Data-Driven Energy #2

Many neighbourhoods or certain parts of cities use local cogeneration plants for electricity and heat generation. A recent project by the Siebel Energy Institute proved that machine learning algorithms can make energy production substantially more efficient.

“It is not just this “fast computation” aspect that is important. Machines are better at making predictions and handling the uncertainty inherent in any prediction because they can handle more variables simultaneously than a person can.” — Siebel Energy Institute

As it takes hours to turn water into steam, the system requires precise forecasting and planning in order to produce the right amount of energy. The machine learning tool is able to take account different variables like holidays and certain weather conditions when more people are more likely to be home.

What caught my eye in energy this week:

Giant batteries will soon become a reality.
The energy storage market is heating up thanks to new regulations and the fact that the price of batteries has fallen 40 percent since 2014. Now that the lenders are looking to finance large-scale energy-storage projects, it’s more likely that the giant batteries will soon become a reality. Large batteries would help utilities cope with the unpredictable output from wind and solar farms.

How the wind energy benefits from computing.
In larger wind farms, the turbines in the front pack are likely to block some of the wind from the turbines in the back. Smart sensors and wind farm optimization enables to adjust the angle and speed of the blades and rotors, and maximise the farm’s power output. In addition to improved optimization the computing technology can help wind energy with forecasting and wind farm maintenance.

About transactive energy, the Blockchain-based power trading.
Transactive energy could allow utilities to better leverage customer-sited resources to meet grid needs and help customers curate their power bills by deploying rooftop solar, distributed storage, or some another DER. And yet the utilities are sceptical about it.

The steel cylinder was lowered into the water to a depth of 11 meters. Photo courtesy: Microsoft.

Energy-efficient data centres to be built underwater.
Building the large data centres underwater and harvesting energy from the sea could help manage today’s phenomenal demand for new data centres in an environmentally sustainable way.

The great energy transition and its economics.
A great writing by David Bank where he discusses the economics of the great energy transition and why no one no longer calls renewables the alternative.

Thanks for reading! This was the second episode of a new blog called Data-Driven Energy. If you’re curious about renewable energy, the role of data in the energy industry and tips and tools for data-driven businesses, then I believe you will find it useful. We are sending Data-Driven Energy blog posts also via email once a week. Simply add yourself to the list and find the next issue in your mailbox.

Previous post → The One Thing Preventing Wind Energy Stakeholders From Earning Billions