Our blogs Blogs home
Energy in Transition

Energy in Transition

General

The Markets 3.0 question: integrated demand-side resources, market behavior, and market stability

This author no longer works for DNV GL.

Significant penetrations of price elastic load and dispatchable demand response can introduce complex dynamics in the day-ahead, hour-ahead, and real-time electric energy markets. This raises interesting questions about the conditions needed for market stability.


The “Markets 3.0” concept refers to the increased demand-side involvement in energy markets. A key driver of the integration of demand response resources into energy markets and operations, today, is the development of smart grid technologies and business models. The ability to manage demand at the end-use level in response to market price signals and controls holds promise to:

  • Save consumers money
  • Enhance grid operations
  • Make markets more efficient.

Indeed, many smart grid projects are financially viable only when energy market impacts are estimated and considered as part of the overall cost-benefit assessment.

Demand-side resource market interactions
Demand-side resources can interact with the market in several ways. Three key modes are as follows:

1. Dynamic pricing—an autonomous response to a market price signal
Smart appliances that can receive an energy price signal and control their on/off status or starting time are in development. More complex local controls could include a home automation system, which manages thermostat settings, air conditioner controls, and other loads based on energy price signals. Which price does the resource follow here: day-ahead, hour-ahead, or intra-hour (real time) prices? Anticipating the dynamic response of such autonomous price-sensitive loads becomes a new dimension in load forecasting for market operators. In recent years, the energy and utility industry and its stakeholders have been interested in estimating potential demand elasticity in relation to emerging smart grid projects and technologies.

2. Dispatchable demand response—a response to control / dispatch signals from the market and system operator or independent system operators (ISO)
Dispatchable demand response positions demand response to resemble a resource akin to conventional generation, as demand response has to participate in various energy and ancillary markets and is paid a market price for responding to dispatch. The Federal Energy Regulatory Commission (FERC) recently issued Order 745, which spells out principles for compensating demand response providers in the markets.

3. Self-optimizing customer—the customer self-optimizes energy usage over time in response to a schedule of market prices, as in the case of day-ahead hourly prices
This more complex level of interaction is an expected mode of microgrid operation. The microgrid operator looks at the published day-ahead hourly prices and then schedules the microgrid production, storage, and demand resources during the day to optimize the financial outcome. In another variant, the microgrid operator bids some of those resources into the market as production or dispatchable demand response resources. Various terms refer to this interaction, including virtual power plant, microgrid, and self-optimizing customer.

Portion of market dynamics model

Questions surrounding the integration of demand-side resources into market and grid operations
Market operators and their communities have several technical, economic, and behavioral questions in regard to the integration of dynamic pricing, dispatchable demand response, and self-optimizing customer into market and grid operations. From a market behavior perspective, these questions include the following:

  • What will market behavior be like with these new kinds of demand-side interactions?
  • How do we manage market and system behavior in the presence of these new elements?
  • What levels of demand-side penetration and which types of interaction will affect market behavior?

There are a number of on-going efforts underway to answer these questions. A few major conclusions can be made from the findings of these efforts:

  1. It is theoretically and practically possible for dynamic pricing resources at high penetration to adversely affect market stability under some conditions. Price oscillations can develop, which are self-sustaining. Under high penetrations, it is vital for market operations to understand dynamic pricing behavior and price elasticity and to include these in the market-clearing algorithms and results.
  2. Dispatchable demand response can be a useful resource, provided that the technical performance characteristics of its resources are matched to the market products they are supplying. Both dynamic pricing and dispatchable demand response can be problematic if this obvious principle is violated.
  3. Self-optimizing customers that operate autonomously in response to published day-ahead prices can have a destabilizing effect on the market and cause supply-demand mismatches as a result. A good argument can be made to get self-optimizing customer resources to be market participants, through information exchange with the ISO, and even through bid submission and full market participation.

Moving Markets 3.0 forward
Several states are exploring or have already committed to high levels of dynamic pricing participation, and many microgrids are in the planning stage. Incorporating information around these interactions in the markets is a sound approach. ISOs should begin to think about how they will integrate demand elasticity into their market-clearing load-forecasting algorithms, and how they will estimate demand elasticity on an ongoing basis. And since this will be critical to market pricing, how the ISOs achieve transparency in the elasticity estimation process is going to become an additional challenge.

Self-optimizing customers present a related challenge. There is a real potential for multiple self-optimizing-customer networks to destabilize the markets. It may not be realistic to require self-optimizing customers to bid into the market as conventional resources, given the time-shifting nature of their operations. Factoring them into market clearing will become critical—either via a bid offer process that leads to known schedules or a complex cross-elasticity process integrated with microgrid load and production forecasting. Either one requires the ISO to have model information of individual microgrid behavior.

Additional details and analysis of this Market 3.0 question is examined in a forthcoming DNV GL IEEE Smart Grid Innovative Technologies paper, “Markets 3.0 – The Impact on Market Behavior on the Integrated Demand Side Resources.” Download a draft version.

For additional insight into Markets 3.0, contact our expert:
Ralph Masiello, DNV GL Innovation Director

0 Comments Add your comment

Reply with your comment

Your email address will not be published. Required fields are marked *