Markets 3.0: Integrating the Demand Side

By Ralph D. Masiello

As demand response mechanisms allowing for rapid and automatic responses to market conditions become ubiquitous in grids—a development dubbed Markets 3.0—there will be wide implications for both technology and procedures. For example, the smart meters being rolled out today may not be quite smart enough; measurement and validation tasks used in settlements will have to be rethought; and there will be demand for appliances capable of acting and reacting autonomously.

There is a palpable surge of interest in bringing more and more demand resources into the day ahead and real-time markets for energy and ancillary services. The main driver is the growing need for system flexibility, so higher levels of variable energy resources—basically renewables—can be integrated into markets and system operations. In some regions prospective retirements of older conventional plants is an additional driving factor.

Increasingly, demand response (DR) is expected to be less expensive than the alternatives of keeping older conventional plants available or increasing procurement of ancillary services from conventional plants. This represents a big shift from an older perception of DR as not much more than a peak shaving resource that might be useful a few days a year.

Meanwhile, all sorts of retailers, aggregators and vertically integrated end user organizations, from vehicle charging providers to the U.S. Department of Defense, are beginning to recognize they can benefit not only from participating in energy markets, but also from offering ancillaries, which could bring large new revenue streams. The general belief is that provision of grid services, such as managed charging as a dynamic DR resource, will boost revenue while shaving costs.

There are several dimensions to this anticipated wave that are “disruptive” in that they require changes in the way we think about smart grid and utility-driven technology adoption as well as serious market design questions. What does this mean for the industry? Based on DNV KEMA's recent engagements with various market operators, state agencies, technology innovators and potential market participants, the following points can safely be made:

(1) The Internet disaggregates everything. Once a business moves to the Internet as a business model, intermediaries in the value chain – retailers energy supply companies and demand response aggregators—have to adapt or die, if they are to escape the fate suffered by so many travel agents. Making a profit on marking up a product, whether airline tickets or energy, no longer works. Rather, the profit is made off the information value added to the customer or the product supplier. Utilities are ill equipped to compete in this space due to regulatory constraints, not to mention culture and business practices.

Specifically, the following challenges will have to be met head-on:

  • Market operators will have to deal with orders of magnitude more participants, most of them much smaller than today’s wholesale access customers.
  • In the future, the lines will further blur between retail and wholesale points of demarcation. “Take out points” used for settling energy consumption may need to change from today’s most commonly used approach of zonal versus nodal.
  • Many of today’s smart meters may not be smart enough for the emerging world of Markets 3.0. Meters will need to be multi-channel (to capture different end uses), capable of interval metering at real-time market time periods (which are five minutes today but could shorten) and web enabled to simultaneously talk to the market, the utility, the aggregator and various end use devices.

Interestingly, some commercially available sub meters have such capabilities. Meters that cannot be adapted to this functionality could be a barrier to Markets 3.0.

(2) Bringing demand response into markets inevitably brings up the question of measurement and validation (M&V): What would the load have been had the DR instruction not been issued? This question has to be applied to each participant. When the market operator has direct visibility of an end use and direct control of it, this issue does not arise. But as more and smaller end users are enlisted, the M&V question will loom large. Settlements is always where all the difficulties in market design come home to roost, and basing large scale settlements on M&V will be a challenge because:

  • It will get market operators into big data, as they increasingly perform statistical analytics on large volumes of end user energy data with "real money" at stake.
  • Market participants (aggregators especially) will have to invest in the same analytics to perform shadow settlements (a term used by participants to describe the process they conduct to mimic the ISO settlements calculation and to prepare their own calculations of what they are owed or owe to the market as a double check).
  • Maintaining transparency and protocol-driven settlements based on statistical analytics will require additional investments in analytics to validate the analytics. Calculations must be explicitly rules-based; there should be no room for human judgment or arbitrary decision making in the financial statement calculation.

(3) Less contentious and hopefully less expensive than the approach explained above are ones that simply rely on autonomous price responses—which let end use loads see the real time prices and react. Appliances are coming to market with that capability nominally built into the software.

A new element in market operator thinking will be required when price is a control signal. That is, it will be important to (a) understand end use price elasticity and factor that into market clearing; (b) be aware of the time dynamics in the responses of different end uses to prices (how fast they change and what the “payback” dynamics of thermal or work-related end uses are); and (c) factor those time dynamics into market and system operations.

Analyses we have done have shown that because load price response is faster than most conventional generation response to signals today (generators typically have five minutes to respond to dispatch signals), the conditions for instability in a control system sense are all in place at high penetrations of price responsive load. This is unless the elasticity and dynamic response are factored into market and controls. All this requires visibility, and visibility will have to come via non-traditional sensor and communication technologies, not from utility SCADA and AMI. Use of ubiquitous low-cost communications will be mandatory.

The bottom line is this: The genie is out of the bottle—technology advances and reliability economics are leading customers to invest in energy resources in increasing numbers. The markets and operations are going to have to adapt to integrate these resources, and that integration is going to have to take advantage of communications and IT technology, especially the Internet, to give customers the direct participation and benefits they will expect and demand.




Dr. Ralph D. Masiello, a member of the National Academy of Engineering, an IEEE Smart Grid Technical Expert and an IEEE Life Fellow, is DNV GL’s Innovation Director and Senior Vice President. He has served as chairman of the IEEE Power Systems Engineering Committee and serves now on the editorial board of the IEEE Power and Energy Magazine. In 2009 he received the IEEE PES Charles Concordia Power Systems Engineering Award. He earned his B.S., M.S. and Ph.D. in electrical engineering from the Massachusetts Institute of Technology, where he worked on the very early applications of modern control and estimation theory of power systems.