What is a smart grid?

By | 3 minute read | April 17, 2018

Image by Mikael Kristenson via Unsplash.

Our new Martian friend just landed on Earth and is excited to learn about the latest developments in human technology. In this Q&A series, IBM experts explain complicated topics to a Martian (and you).

On Mars, we don’t have smart grids and our power supply is limited. Sure, we have solar power but our electrostatic dust storms were way overblown by The Martian.

I was curious to find out more, so I asked Jeff Katz, the Chief Technology Officer of Energy and Utilities at IBM, to tell me more about smart grids.

What is a smart grid?

It’s an intelligent overlay to the existing electrical grid infrastructure. It’s highly interconnected and uses sensors, meters, digital controls and analytic tools to automate, monitor, and control the two-way flow of energy.

Why is that useful?

Utilities can leverage new data from smart grids to improve their existing grids.

This is similar to electronic toll collection systems, which have improved road use and reduced congestion. Governments automated a manual toll system and added devices to vehicles, then collected and analyzed data on cars passing through toll lanes for new insights. They didn’t rebuild roads—they made them smarter.

Does this data help consumers?

By installing smart meters on homes and businesses worldwide, utilities companies can accurately bill consumers for the energy they actually use, and provide consumers with a better idea of their average daily usage in hourly increments versus monthly.

Utilities can be proactive with this information. They can anonymously share with a consumer if they are their neighborhood’s “energy hog,” then offer consumers a free energy audit to help assess why they use so much power.

All this data provides a feedback loop to consumers about their behavior, helping them become smarter—and greener—about their energy usage.

What about power outages? The global dust storms on Mars tend to limit our solar power.

People understand that power outages happen, but don’t like figuratively being in the dark about when power will come back on.

Smart meters can quickly communicate that a home’s power is out so crews can be dispatched to fix it.

Utilities can also text consumers that their—or their families’—power is down and provide an estimate of when it will be restored. So people can go out to dinner—or rush home to save the food in their fridge.

Where does AI come in?

AI can help companies understand the vast amounts of data about power failures, pinpointing what happened just before and after a failure. AI can correlate the unstructured human data stored in files, such as maintenance reports, with structured info from smart grids to discover patterns they may not have noticed before.

By analyzing this data, AI can find the root cause of a failure and quickly tell a technician, “the last ten times you saw this event, here’s what your colleagues had to fix.”

How can companies prepare for a smart grid future?

Data is one of the most difficult parts of smart grid projects so utilities should first start with analytics. Before buying any new sensors that accumulate more data, utilities need to understand the data they already have. I call this “smart grid version zero.”

The real value of past data is that it helps predict the future. It also provides a better clue as to where to spend money on new sensors to fill in missing data so that the grid can be smarter.

Anything else?

Yes, Martians are green so this may appeal to you.

You know the quickest way to my three hearts—talking about me. Please continue.

The smart grid is critical for making the best use of green or renewable energy added to the grid. Renewable energy is cleaner but intermittent—you can’t say, “We need more wind or sun.”

But we can more dynamically optimize the grid to keep everything in balance. And utilities can use new insights, like weather data, to predict when it will be sunny for solar energy or windy for wind farms.

Very enlightening. Thanks, Jeff!