Technology revolution: How US energy suppliers are benefitting from AI 

Traditionally, the energy supply industry has been a bit slow on the uptake of new technology, but Plug and Play are hoping to change all that. A Silicon Valley based company, Plug and Play specialise in connecting large energy companies with start-ups who are offering solutions to problems they have encountered.

Their recent spring summit saw 15 entrepreneurs pitching their businesses to energy companies with a range of ideas designed to solve problems in the energy supply industry. Over half of the pitches focussed on artificial intelligence (AI) and machine learning.

Some of the issues being tackled with AI included reducing energy use in buildings, monitoring of power lines and pipelines as well as making intelligent predictions about when network elements could fail. Maintenance and prediction of performance were huge themes at the summit. Founder and leader of Plug and Play, Wade Bitraf, spoke to GreenBiz , saying:

"AI and machine learning are tools that can apply across the board for the energy sector"

In some ways, the US is a few steps aheadof the UK in terms of AI in energy supply. Although there are a handful ofexamples of AI being used in practice, such as Cornwall’slocal energy market , they are few and far between.

Of course, some of the issues across thepond are unique to them… wildfires and extreme heat, for instance, aren’tfrequent occurrences in the UK. However, the issue of extreme weather eventsand flooding is certainly pertinent to UK suppliers too. Some of the ways inwhich companies in the US are already leveraging the power of AI include:

• Autonomous grid : With grids now havingto manage energy from a variety of different sources, the operation of thesesystems has become far more complex. Siemens have developed a software package called Active Network Management (ANM) whichtracks how the grid interacts with different energy loads, tweaking parts tooptimise it for efficiency.
• Failure management : AI is being used toobserve equipment and detect failure before they happen. SparkCognition are one provider ofthis type of technology, combining sensors, analytics and operational data to predictwhen infrastructure could fail.
• Powerline and substation monitoring :Being able to keep an eye on critical infrastructure components can improvesafety and reduce downtime. Korean company Alchera use cameras, both regular and thermal ones, to monitor substations andpowerlines for things like smoke, fallen trees and intruders. Added to this, a USbased start-up called VIA works topredict when transformers could be at risk during a weather event.
• Maintenance cycle predictions : Accordingto Ensemble Energy , US energycompanies lose $27bn each year because they perform maintenance on a reactiverather than a proactive basis. Their company combines data sources with machinelearning to predict when hardware will need maintenance, enabling supplycompanies to get it done before it causes expensive problems.

There are many other applications for artificialintelligence in the energy supply market, from helping end users cutconsumption to exploring for new fossil fuel reserves. However, one of the mostexciting applications for the future of our industry is in the renewable energysector.

The renewable energy boom is a great opportunity both for the economy and for the planet. But the intermittency of technologies such as wind and solar makes it hard to manage. Add into this the potential risks of downtime or network interference delivered by extreme weather events and you’ve got a highly volatile energy supply with an irritating level of unpredictability.

The key to making renewables work is going to be energy storage. Adding in redundancy, resiliency and energy efficiency through energy storage has the potential to make a low carbon grid work for everyone, from the producers right through to the end users.

But to make energy storage really work well, it needs to be intelligent, working proactively to manage the intermittent nature of renewable energy towards a continual supply. Intelligent energy storage (IES) has the potential to deliver new value streams for renewable energy projects, by providing more control and flexibility in the face of changing rates.

AI’s ability to decipher sets of data as well as to use patterns to make predictions poses some exciting possibilities for the future of energy networks. How energy supply businesses embrace these innovations remains to be seen, but it’s certainly an interesting time for the energy market worldwide.