Let’s stop thinking of the distributed, unpredictable world of the grid edge as a wild frontier to be subdued via utility command and control. Instead, let’s start thinking of it as an internet-enabled free market, where customer devices and grid systems can barter over the proper way to solve their mutual problems, and settle on the proper price for their services, in close to real time.
That’s one way to describe “transactive energy,” a concept for organizing the proliferating world of independent energy agents in a way that won’t crash the one-way, centrally controlled grid to which they’re being connected. While most of the work on this front is still being done in research papers, smart grid conferences and computer simulations, a few pilot projects around the world are trying it out in the real world -- tentatively at first, but with an eye on making it a part of grid deployments to come.
Now the industry has a document that can help speed that task: the GridWise Transactive Energy Framework (PDF). Released at the Smart Grid Interoperability Panel (SGIP) inaugural conference this month, it's something like a dictionary for the language of transactive energy, as well as a guide for how various pieces of today’s energy infrastructure might translate into the transactive model.
Last week, I spoke to Pacific Northwest National Laboratory’s Ron Melton to learn some key terms from the 65-page document. Melton serves as administrator of the GridWise Architecture Council and also directs the Pacific Northwest Demonstration Project, a DOE-funded project putting transactive energy concepts to the test at a regional scale.
The core issue at hand, Melton said, is that the physical infrastructure being added to the grid -- distribution grid-connected solar panels, transmission grid-connected wind turbines, customer-owned microgrid systems and energy storage assets, demand response-enabled thermostats and smart appliances -- is nearly impossible for existing utility and grid operator networks and control systems to manage. First of all, much of it is in the hands of customers, not utilities. Second, many of these systems act too quickly, and in too great a volume, to actually monitor and manage in real time.
At the same time, utilities can’t throw out their existing platforms and start over, he said. But with the right design in place, distributed systems, organized in a hierarchy of control layers but all sharing commonly understood sets of data, could create what the framework describes as “a loosely coupled set of controls with just enough information exchange to allow for stability and global optimization through local action.”
“This more distributed approach, which brings together the combination of grid economics and grid controls into an engineering and economics framework, is one of the tools that the utility needs to accommodate this change in an orderly way,” Melton said. That means that a transactive energy system can’t just tackle the technical details of multi-device, multi-party data exchange. It also has to handle the economic relations between those parties.
“We wanted to leverage the power of the economic systems to optimize the balance of the system,” he said. To get there, GWAC’s planners have settled on the core defining characteristic of “value,” as laid out in the document’s definition of transactive energy as “a set of economic and control mechanisms that allows the dynamic balance of supply and demand across the entire electrical infrastructure using value as a key operational parameter.”
Defining “Value” in Transactive Energy Terms
So what does “value” mean here? In simplest terms, it equates to a price -- whether it’s the market price of generating, transporting and consuming power across the system, or the price of maintaining, protecting and optimizing the power lines and transformers, demand response and energy storage assets, and other equipment that aids in that task, he said.
GWAC’s transactive energy framework lays out several “attributes” that get at this value-defining proposition. One is the “value discovery mechanism,” or the “means of establishing the economic or engineering value (such as profit or performance) that is associated with a transaction.”
This includes “considerations of economic incentive compatibility and acceptable behavior,” or, in plain language, finding ways to prevent people from gaming the system. “Our economic expert had us put it in there,” Melton noted, given that there’s always potential for people to manipulate a fast-acting, complex transactional system for their own gain.
Another attribute, dubbed “assignment of value,” takes a crack at the tricky issue of putting a number on variables that aren’t usually defined as having value. “We recognize that, to your grandmother, unless she just happens to be a grid economist -- to most people, when they talk about value relative to their use of electrical power, they’re not necessarily thinking in terms of dollars and cents,” he said. “They’re valuing comfort, or they’re valuing light, or they’re valuing being cool in the summertime, or pumping the water in their swimming pool.”
In other words, “they may have values that are not articulated quantitatively, but are articulated through their choices.” To get at that challenge, the GWAC framework lays out the “assignment of value” attribute, for those “sub-elements” that haven’t traditionally had the price they pay for energy directly connected to its cost. This can also apply to assigning a value to carbon emission reductions, energy efficiency improvements, better integration of mandated levels of renewable energy resources, or any number of benefits that don’t get reflected in strict economic terms.
Of course, “assuring stability” of the system is also a key attribute. As the GWAC framework notes, in a system that relies on automated call-and-response from lots of devices at speeds beyond the real-time supervision of grid operators, “minor errors can build on each other, and sometimes spiral out of control.”
Real-World Examples of Transactive Energy in Practice
To demonstrate how all of this might look in the real world, GWAC’s transactive energy framework includes two case studies. The first is the Pacific Northwest Demonstration Project, which is directing $178 million in DOE stimulus grants to connect eleven utilities and 60,000 metered customers in a framework that spans the states Idaho, Montana, Oregon, Washington and Wyoming.
The project has linked up vendors IBM, 3TIER, QualityLogic, Alstom Grid and Netezza to build a system that can pinpoint and forecast real-time electric loads across an array of “nodes,” or points on the system from generation to end use. The wind power-rich Northwest region is already facing challenges in balancing energy supply and demand across the Bonneville Power Administration’s territory, and states in the region have set aggressive renewable energy and efficiency standards expected to increase pressure on the system.
As for value, the project uses “transactive incentive signals” to represent real and projected costs of electricity being delivered at each point in the system. While “we haven’t implemented it as a pure economic transaction, it is cost-based -- and we’ve implemented it so that it could be an economic transaction,” Melton said.
The second project, a partnership between AEP Ohio’s DOE-funded GridSMART program and Battelle, the institute that manages PNNL, tackles transactive energy on a more local scale. It also tackles the translation of two very different measures of value: the price that customers are willing to pay for air conditioning and the cost that utilities face to replace or renovate distribution circuits subjected to excess demand for power.
That’s done by taking utility measurements of how close the pilot project’s distribution feeder line and transformer are to being overloaded, and translating that into to five-minute variable prices being sent to smart thermostats in several hundred homes served by that feeder, Jason Black, grid systems research leader at Battelle, said in an interview last month. Five-minute price changes were picked to correspond to the intervals that grid operator PJM uses to dispatch generation resources.
Those overload conditions tend to happen only during a handful of the hottest hours of the year, exactly the same time that air conditioners tend to be running full-blast. To avoid sticking those customers with punitive high prices, Battelle has developed algorithms that allow the smart thermostats to predict a coming overload condition and precool their homes, then lay off when the peak hits, all in five-minute increments that should smooth out the overall effect on temperatures.
Utilities could deploy systems like this to avoid expensive grid upgrades merely to manage the brief moments when demand exceeds capacity on those feeder lines, he noted. The avoided costs of that upgrade could, in turn, play a role in offering participating customers incentives to participate -- a feedback loop that’s made possible by translating utility costs into customer prices.
Black noted that Battelle has licensed the underlying technology to Calico Energy, one of several players working with utilities to connect customers to more direct participation in grid-centered energy pricing programs. Of course, the task of getting regulators and customers on board with such novel, and potentially disruptive, real-time pricing programs could remain bigger challenges than getting the technology itself to work.
It’s possible to see a lot of other uses for transactive energy in managing complicated challenges on the frontiers of the grid. One example is energy storage, where the value of new technologies rely as much on how they’re able to make money by solving a multiple set of grid and customer needs as they do on the price of the storage itself.
Another example is the struggle between utilities and renewable power industry advocates over how to split the costs and benefits of enabling grid-balancing functions for smart inverters, which can mitigate some of the disruptive influences of solar PV and other distributed energy resources on the grid.
Melton noted that there will doubtless be many different approaches to applying transactive energy principles, as well as plenty of challenges to overcome in terms of ensuring stability, security and a level playing field for all the new participants in the game. But the end result could well provide utilities and their regulators a much better way to future-proof their investments, both those already made and those being planned for the future, he said.
“The real point is, many times today we make an investment in a particular type of technology based on a single expression of value -- we will get a single benefit from this,” he said. “What we’d like to see with a transactive energy approach is that you’re defining all the possible value streams from this particular approach -- both in the way it interacts with the rest of the system, but also in the way the system interacts with it.”