With each passing day, the electric power grid continues the process of morphing into something new we haven’t seen before. This rapid transformation, with hundreds of thousands of assets already massing at the grid edge, will ultimately involve millions of Internet of Things (IoT) connected devices. Add that to the enormous level of intermittent renewable assets that have come on line and will in the years to come, and it implies a whole new way of managing our power grids. It will also require new ways of sorting, processing, visualizing and acting on all that data.
Hawaii and California signal what is to come
Some states haven’t witnessed much change yet, while others are in the midst of the transformation and signaling what is to come. The most extreme cases today are Hawaii and California.
In Hawaii, over 80,000 rooftop solar installations – totaling 700 MW – quietly pump surplus power into the grid during the middle of the day. The Hawaiian grid is now sated with rooftop solar and simply cannot absorb much more. Batteries will soon be installed in large quantities to help shift that flow of electrons to match more closely to when it is needed. And a huge amount of additional carbon-free resources will be needed to help the state meet its legislatively set 100% by 2045 target.
Meanwhile, California utilities already host over 850,000 net energy metered solar arrays exceeding 7,000 MW, and behind-the-meter batteries are emerging as one of the new frontiers there as well. SunRun reports in its latest quarterly earnings call that over 20% of new direct sales customers are adding batteries when they install solar panels. In parts of Southern California, that number is approaching 60%.
Numerous other states are pushing the regulatory envelope with goals for renewables, electric vehicle sales, or overall carbon reduction targets. At the same time, the technologies themselves keep improving and falling in cost, as global markets create continually more efficient supply chains.
A growing avalanche of data suggests new approaches will be necessary
All of this suggests that utilities, and especially grid operators who manage the bulk electric system, have a growing challenge on their hands: how to bring in the right data and present it in a way that humans can understand to help create real-time and actionable situational awareness. A company that has been working on addressing that specific problem is ResilientGrid.
ResilientGrid designs the interfaces between computers and people to increase the ability of grid operators to intelligently comprehend the data they are looking at, securely share what is relevant, and take the correct actions accordingly. In other words, the goal is to create improved ‘situational awareness’ so that we can create a more resilient electric power grid.
The company provides the tools – the human-computer interfaces – but also offers training and consulting and R&D services in ‘continuous human performance improvement.’ It focuses on the psychology and capabilities of the operators and helps them to better perform.
I first interviewed ResilientGrid’s current CEO – Mike Legatt – just over three years ago when he was working for the Electric Reliability Council of Texas (ERCOT – the Texas grid operator) as their Principal Human Factors Engineer. Essentially, he was the psychologist for ERCOT tasked with making their operators more effective by helping design better tools, based on their feedback.
Legatt departed from ERCOT in late 2016 and founded ResilientGrid to help other grid operators and utilities deal with these same issues. The company’s offerings have been well-received, and it is already operating on multiple continents and has been in part financed by utility holding company Ameren along with additional institutional investors.
I caught up with Legatt this summer, to get his viewpoints on the issues his company and its clients are dealing with.
Multiple vendors and invisible distributed assets – a difficult combination
Legatt sees a growing challenge of dealing with multiple vendors developing solutions at the grid edge – each of which has its ‘own version of secret sauce’ in terms of the way they secure and communicate their data. Part of the issue is that, to date, very few protocols have been established to marshal this information into common formats. So each EV charging company treats data differently, as does each smart thermostat company, and every other vendor. This creates a Tower of Babel. “Because it’s non-standard” Legatt notes, “you have difficulty getting best practices, and miss the potential capabilities these devices can have working in synergy with others for the benefit of the grid.”
Legatt observes that as you increase the level of connected assets, ‘then the criticality of the asset starts to ramp up.’ And since they are all connected in the IoT, they also become potential attack vectors for hackers.
At the same time, they are also invisible to grid operators, which is why one sees Independent System Operators in places like New England forecast for required generation ‘net of solar.’ They simply don’t have a very good sense of how many megawatts of distributed solar are out there, delivering electricity into the system on sunny afternoons.
Legatt comments that the situation is, in a word, “Painful. You knew there was equipment out there, but you couldn’t see it, and frankly you could not even model it.” And one could not communicate with the assets, either. That situation has become even more pronounced in the interim, with a deluge of new devices, and operators flying somewhat blind, unsure what’s changing due to demand, more efficient devices, or renewables.
Keeping people in the loop is more important than most people may think
we don’t have the math to model the incredibly complex interconnectivity between many systems across the grid, so instead operators overly rely on intuition, which is not a decent strategy for either technology or business outcomes.
In the meantime, human beings are sitting in controls rooms and constantly having to make critical decisions. “The complexity” he says “has exceeded our capacity for information processing. It really comes down to the question of how do we set people up to handle this new world.”
Legatt sees some companies starting to resort to Artificial Intelligence to address some of this complexity, but that has its own issues, as it potentially creates “out-of-loop syndrome” where the operator is now far removed from the context of the decision-making process. Once separated from that context, it becomes harder to re-insert oneself into the loop and re-integrate into the context of the information if human intervention is required (which is likely in more extreme circumstances or they wouldn’t have to be re-inserted into the process).
By contrast, ResilientGrid focuses on integrating legacy data and communications technology, such as the SCADA (supervisory control and data acquisition – the historical industrial language of the grid) with other, often higher-speed, feeds. However, the company does it in a way that keeps humans continuously in the loop, and designs better systems for their use. In that world, “AI and machines are still like co-pilots. They know their place and don’t want to make decisions for you. I think that’s the right balance.”
ResilientGrid’s job is to solicit input from operators, “versus telling them what their tools are,” and create better ways for people to interact with computer screens, AI, and the streams of relevant data. Not surprisingly, Legatt indicates that these issues are as much cultural as they are technical. Some companies, he comments, are open to change and developing new tools and approaches, “they get what we are doing, and we sit with system operators and ask about their challenges.” Other organizations, by contrast, “may focus on the fact that they haven’t yet had an incident.” Which means they may not fully understand the magnitude of challenges and risks in the new world they now inhabit. And they might not know what they don’t know.
Legatt sums up the challenge, noting that the energy industry was once a technology leader, but since then commercial electronics technologies have leapfrogged past everything. They are now affecting the grid and changing both the structure, and indeed our very notion of what the electric grid does and how it does it.
There was kind of this idea, electric power was a kind of cafeteria – you ordered what you needed and it would be there. Now we’re all on the same boat where everybody is rowing. Today, with all these new assets the grid is becoming part of an even larger system, and this has profound implications.”
In that boat, each new asset has the ability to stabilize and destabilize the grid. This newfound level of customer and device integration requires a change in the mindset of not just the grid operator, but the day-to-day consumer as well. Legatt, says we now have to think in new ways, for example, “I’m not just plugging in my car because it’s convenient for me. I’m now part of a bigger system.” That may take some getting used to, but it’s coming, one way or the other. It’s therefore incumbent on society to develop the best tools and determine how best to optimize interaction between man and machine.