Artificial Intelligence Is Changing the Power-Delivery Grid

If Thomas Edison were alive today, he would be at home in today’s electric utility infrastructure. That saying is annoying. It’s a put-down to all the people working to make the grid better and more dependable. Heck, it’s like scratching fingernails down a chalkboard! I ran into a derivative of that saying the other day when I read an opinion from a pseudo-expert writing about the problems of the electric industry. This individual was upset with what he perceived as a lack of our industry keeping up with the latest technological innovations. That’s not only ridiculous, but it’s wrong. It really makes me wonder how someone can be so far off target in their thinking and consider themselves to be an expert.

As T&D World’s technical writer, I’m able to talk with the folks developing, deploying and operating the latest technologies being used by utilities around the world. I get to kick the tires and look under the hood on some of the most amazing digital toys ever deployed on the grid.

Earlier this year, I wrote a series of articles for T&D World on the innovations taking place with asset management (see "Integrated Asset Management" supplement, July 2017). Modern asset management systems have taken connectivity to a whole new level. They are using the Utility Internet of Things, intelligent digital components and cognitive computing to monitor and predict the performance of the power delivery system in general and specifically the health of individual pieces of equipment. I doubt Edison would feel very much at home with these cutting-edge technologies, but he would be interested, especially in the real-time monitoring and situational awareness of grid devices.

It’s Not Science Fiction

Some of us call situational awareness technology machine learning, but most of us use the more common term of artificial intelligence (AI). AI is being used on the grid right now and it is more widespread than you might think. Case in point: In June, the U.S. Department of Energy held its 2017 Transmission Reliability Program in Washington, D.C., and there were several AI presentations. The presentation that caught my attention was about using advanced machine learning with synchrophasor technology. Synchrophasor technology generates a great deal of data using phasor measurement units (PMUs) to measure voltage and current of the grid, which can be used to calculate parameters such as frequency and phase angle.

This PMU data can be used for real-time analysis of the grid to predict problems about to happen, but the huge amount of data makes this task impossible for an operator to spot trouble brewing. Imagine if AI-powered supercomputers were tied into the data flow. It would be possible to analyze what is taking place and to prevent catastrophic outages. That is not too farfetched; it is happening in other industries. Remember IBM’s supercomputer AI program called Watson? Watson caused a stir back in 2011 when it beat all the human Jeopardy champions and pocketed a cool million dollars. After Jeopardy, IBM introduced Watson to industries such as medicine.

Medicine is one of those industries that is awash with data, lots of data much like our power-delivery grid. It’s called big data, which means it is too large and complex for traditional methods of data processing. IBM estimates that 2.5 quintillion bytes of data are created every day. More shocking, IBM stated that 90% of the data in the world today was created in the last two years. The medical community’s big data stems from patient records, patient testing, drug testing, academic literature, research projects, conference papers and studies; you get the picture. In 2016, IBM and Siemens announced a partnership to help healthcare professionals navigate the massive medical databases. What does all of this have to do with the electric power industry in general and the transmission and distribution grid specifically?

Glad You Asked

Our industry has much of the same type of big data issues as the medical community. With the deployment of smart grid technology, almost every device on the grid is producing some type of real-time data. We are also trying to deal with off-line data sources such as equipment records, device maintenance records, testing records, academic studies, research projects and so on. And like the medical world, Watson is ready to apply its abilities to our industry.

Imagine PMUs, transformers, breakers and other devices sharing data with one another and Watson. Imagine Watson transforming that data into real-time information to spot aging trends, unearthing additional capacity, eliminating congestion and sniffing out potential failures before they become major issues. Well, it’s happening.

Recently, ABB and IBM announced they are forming a strategic collaboration in industrial AI to bring Watson’s cognitive insights to the transmission and distribution grid. AI promises to move beyond our current connected systems into the realm of predictive pattern matching and logical analysis.
AI programs like Watson will use data to understand, sense, reason and take actions that go beyond game-changing into the domain of ground-breaking. It’s going to be uncomfortable — even disruptive — but exciting.

As originally stated, Edison would be amazed at how the transmission and distribution grid has developed since his day. I think he would agree, our system is truly innovative, especially as AI is deployed and today’s impossible becomes tomorrow’s commonplace. ♦