HomeSmart Grid T&DEnergy EfficiencyManaging electricity demand a 'negawatt network' in action in the US

Managing electricity demand a ‘negawatt network’ in action in the US

Operators of on-site generation plant are just one of many power consumers who have joined forces to cut electricity demand peaks in one part of the US by innovative management of existing assets. Gregg Dixon describes how demand-side management is making a significant impact on grid reliability.

When grid system planners think about how to ensure that supply meets electrical demand now and in the future, they rarely consider the fact that demand is actually part of the equation that can be controlled. As is often the case, planners forecast demand growth, model scenarios associated with weather and coincident peak demand, factor economic impacts, plug in a reserve margin and, voilàƒ , the output tells them how many gigawatts of generating capacity will be needed to reliably meet demand.

Staff at EnerNOC’s Network Operations Center monitor customer assets and markets around the clock in real time
Click here to enlarge image

Largely due to market regulation, de-regulation and re-regulation, the challenge over the past 10 years has been to actually put the iron and wires in the ground to support load growth. ‘Not in my back yard’ (NIMBY) and ‘build absolutely nothing anywhere near anyone’ (BANANA) are watchwords that we are all well aware of and it’s not entirely clear anymore who is responsible for ensuring adequate capital investment to keep the grid healthy. Political, economic, environmental and sociological issues are presenting greater and greater challenges to satisfying the requirements of the electricity market and system plans in a timely fashion.

Traditional supply infrastructure (such as central plants, transmission lines and distribution centres) costs more that expected, takes more time to site than scheduled, and is required to meet increasingly strict guidelines for environmental performance. Unfortunately traditional system planning methods haven’t caught on to the fact that the demand side of the equation is becoming increasingly flexible and able to act much like traditional infrastructure.

Demand side of the equation

What is the demand side of the equation? A diverse set of assets make up the demand side including motors, lights, central processing units (CPUs), distributed generators, building management systems and meters. All these assets can be interconnected through today’s ubiquitous and inexpensive broadband telecommunications network and controlled remotely – in one fashion or another – to react to various market signals from price signals to grid emergency warnings.

Many marketing phrases have been invented to describe this demand-side network including ‘smart grid’, ‘future grid’ and ‘virtual power plant’. Regardless of what you call it, there is little doubt that, if properly developed, this resource can provide incredible value to the grid-level system plan. This is highlighted using a real-world example in this article. For our purposes, we’ll call this resource the ‘Negawatt Network.’

What happens if the system planner creates a forecast that requires hundreds of megawatts of capacity to be sited over the next 24 months to meet growing demand in a load control area but no new megawatts are due to come to fruition during this period? Just such a scenario has occurred many times, and one instance in particular has served as a real-world laboratory for determining the exact value of this Negawatt Network.

In 2003, the Independent System Operator of New England (ISO-NE) recognized a serious deficiency in the south-west Connecticut control area. There was no incentive for power plant developers to rush in to build new capacity in this highly populated area, which is one of the most congested electrical control areas in the US.

As a result of this recognized deficiency, ISO-NE developed a Request for a Proposal (RFP) that sought up to 300 MW of quick-start capacity which could be called upon to meet peak demand and to preserve ISO-NE’s reliability standard for delivery of electricity. ISO-NE chose a handful of the 26 proposals submitted that expected to deliver 266 MW of quick-start capacity within a two-year period.

Characteristics of a Negawatt Network

Since awarding this initial set of contracts in early 2004, the state of Connecticut has taken ISO-NE’s lead and encouraged even more of these demand-side resources to come to market. This has resulted in nearly 500 MW of technology-enabled demand-side resources that act as a quick-start Negawatt Network.

What are the characteristics of this Negawatt Network (Figure 1) that make it so valuable and a model for other grid operators to follow?

  • The entire 500 MW is as diverse a resource as possible. It ranges from the control of centralized air conditioners in homes to dimming lights in grocery stores to turning on back-up generators at multi-megawatt data centres.
  • The entire Negawatt Network can be turned on within 30 minutes of notification, with a majority of the resource ramped up within minutes, performing as well or better than traditional infrastructure. In effect, ISO-NE has an ‘easy button’ for dropping a massive amount of power from its grid without the need for rolling blackouts.
  • The 500 MW Negawatt Network actually has the effect of removing the need for nearly 550 MW of power plant production. This amount of demand requires approximately 550 MW of central plant power that experiences anywhere from 8% to 10% transformation losses before reaching end-users. In addition, the 500 MW that is removed offsets a portion of ancillary services (such as reactive power production) needed to support the demand.
  • The 500 MW represents approximately 7% of peak demand for the Connecticut load control area without having to build a single central plant, transmission line or distribution centre to deploy it. It is expected that 10% of Connecticut’s peak load will eventually be enabled in this Negawatt Network.
  • The 500 MW Negawatt Network was deployed in stages, with as much as 100 MW being enabled in one month. This illustrates just how quickly demand-side resources can be integrated into a system plan when traditional resources cannot be sited quickly enough.
  • The 500 MW Negawatt Network is very environmentally friendly. As we say: ‘The greenest kilowatt-hour is the one never used’. While many distributed generation plants and back-up generators are part of this network, much of the network consists of loads that can be curtailed to simply reduce the load required from power production.
  • ISO-NE’s Negawatt Network is monitored in real-time, just like traditional infrastructure. Thousands of metering points provide five-minute interval readings in real-time through network operation centres such as EnerNOC’s, directly to ISO-NE’s control room.

Figure 1. Elements of the Negawatt Network
Click here to enlarge image


Action over theory – how it works

The ISO-NE Negawatt Network is a case of an ounce of action being worth a ton of theory. Much of what academics and pundits have been drawing up and demanding in the so-called ‘intelligent grid’ has been deployed in this 500 MW real-world deployment. It is valuable to provide examples of the diversity of these assets and the underlying technology to showcase the fact that every load control area has a similar opportunity here and now.

  • Hundreds of grocery stores participate in this network by allowing EnerNOC to remotely meter and control lighting circuits, building management systems and back-up generators. Each store has the ability to curtail, on average, 75 kW of load from the grid. Various forms of the broadband network are deployed to support these customers, from the use of the customer’s existing local area network (LAN) to installing inexpensive broadband wireless devices.
  • Large industrial customers with on-site cogeneration systems are able to shift process work and export excess power to the grid. For example, one customer has a 12 MW cogeneration system that provides approximately 80% of the plant’s power needs. When the Negawatt Network is called upon by EnerNOC at ISO-NE’s request, the plant is ramped down and the power is exported to the grid to support local loads.
  • Hundreds of telecommunications switching centres transfer power to their back-up generation systems to alleviate grid stress when their own network operations centre receives a signal. Typically these generators are already being exercised each week to ensure equipment integrity. An ISO-NE demand-response event simply acts as an exercise event for these customers, alleviating the need to exercise the units that week. It also enhances their operations as they receive advanced warnings of issues on the grid to proactively switch to their auxiliary power source.
  • EnerNOC has many manufacturing customers with excess emergency power. These emergency generators are remotely controlled by EnerNOC to support the entire load of the manufacturing process. Because of the over-sizing typical in emergency generation installations, excess power is exported to the grid to provide further grid support

Extraordinary benefits

The Negawatt Network provides tremendous value to all grid stakeholders in the region. The benefits are not readily apparent to rate payers and end-users because the assets required by the Negawatt Network already exist physically, but just need to be enabled by EnerNOC’s technologies and methodologies. Unlike large, traditional power plants, consumers might never know that this latent power plant has been fired up. However, the value of unlocking the Negawatt Network is extraordinary and comes in many forms. Specific to ISO-NE’s Negawatt Network, the following benefits are clear:

  • Grid reliability is the major benefit. Where traditional infrastructure was unable to be sited quickly to meet growing demands, the Negawatt Network was able to be deployed quickly to mitigate potentially disastrous power losses. According to a 2005 ISO New England Demand Response Program Evaluation, using a conservative assumption that the programme reduced the ‘Loss of Load Probability’ (LOLP) by only 10%, the reliability benefits alone exceeded the costs of the programme.
  • Because 500 MW of capacity have been enabled, there is a reduced need for traditional infrastructure. The cost savings resulting from deferred generation, transmission and distribution investment are significant at an estimated $20-45 million per year.
  • Wholesale market price mitigation is substantial, especially when demand is peaking. The 500 MW Negawatt Network can act to dramatically reduce high wholesale, real-time power prices when deployed. Although only 7% of peak demand is reduced when the Negawatt Network is deployed, wholesale power prices have been reduced during demand-response events called by ISO-NE, reducing consumer costs in the ISO New England territory by an estimated $2 million in 2005 alone. This reduction applies not just to the energy that does not need to be purchased on the wholesale market, but it reduces the cost of every MWh being purchased during the hours that the Negawatt Network is deployed. In 2005 alone, ISO New England estimated that the 500 MW Negawatt Network reduced capacity costs by $30 million.
  • End-users who participate in the Negawatt Network receive monthly capacity payments and energy payments for the energy they remove or supply to the grid during demand-response events. They also receive the reduced cost associated with curtailed load. Although all rate payers benefit, those end-users who provide the negawatts receive substantial economic value for their direct participation. Total electricity bill savings for an end-user can amount to as much as 10% for participation in ISO-NE’s programme.
  • End-users who participate also find indirect benefits from raised awareness for better energy management overall. Demand response is an enormous gateway to the benefits of overall energy efficiency and conservation. Because real-time, internet-based metering and communication is required, end-users get a real-time view on total energy consumption. Information is power and the vast majority of end-users did not previously have this decision-making support. Once end-users feel the benefits of programme participation, they often want to experience more and more, and this often comes in the form of what EnerNOC calls ‘Total Energy Management’.


Although this Negawatt Network has many unique features that make its performance transparent and accountable, there are dozens of other Negawatt Networks being developed throughout the world. Many states and load control areas in US have active programmes or are actively developing programmes, including New York, California, Texas, Pennsylvania, New Jersey and Maryland. Canada, Australia, Italy and China all are actively pursuing the deployment of Negawatt Networks that meet very specific needs, from explosive demand growth to congested control areas where traditional infrastructure cannot be sited. Although Negawatt Networks can never meet the baseload needs that traditional infrastructure will always be required to meet, the Negawatt Network is extremely flexible and can be deployed in a multitude of fashions to meet very specific requirements.

Perhaps most importantly, from an end-user perspective, participating in demand response and being part of the Negawatt Network raises the awareness of energy management and the savings opportunities that lie fallow. Once the technology is in place to participate in the Negawatt Network and once end-users become accustomed to the measures taken to participate in demand-response events, end-users become much more receptive to changing the ways they purchase and use energy. Demand response acts as the training wheels in the process of tapping the potential of total energy management. EnerNOC’s customers have found that the potential for total energy savings is often 25% or more of their total energy spending. This all becomes possible through technology-enabled demand-response participation.


The total value of this Negawatt Network has not yet been achieved. More value propositions are being developed as we enable assets. In fact, as we learn more about how each end-user type can participate and we learn about their actual response during demand-response events, we find more ways to enable more megawatts with more customers. While academics are debating how the ‘smart grid’ will evolve, we’re busy building it and showcasing its actual value.

Gregg Dixon is Vice President of Marketing and Sales with EnerNOC, Inc., Boston, US.
Fax: +1 617 224 9910

To learn more about this Negawatt Network go to

To comment on this article or to see related features from our archive, go to