Connecting consumers to the smart grid


By Henry Marcy

Energy generation, use, and conservation are at the root of many of the most pressing issues facing society today. Demand continues to rise fairly steadily while the ability to generate and deliver energy is increasing at a much slower rate. Furthermore, in the United States, more than half of the electricity produced is actually wasted due to power generation and distribution inefficiencies. [1] Very simply, using less energy in our daily lives, and making more efficient use of the energy we do produce, are fundamental to our continued collective prosperity and quality of life.

Residential energy use accounts for a full 38% of the total energy consumed in the United States. [2] Home appliances, heaters and air conditioners together represent more than half of this consumption, or 21% of the total energy used in the United States. Developing the Smart Grid and linking it to “smart” appliances with demand response capability, will reliably and predictably reduce appliance energy consumption in real-time, representing a significant energy efficiency and conservation opportunity.

Demand response will both complement and take us well beyond recent efforts of Google and Microsoft to make personal energy use data available to consumers [3]. Creation of a closed loop system, in which supply and demand are explicitly controlled, will not require continuous consumer intervention or even awareness. Controlling and more importantly, coordinating the energy demand of millions of networked smart appliances will allow us to make more effective use of the country’s current power generation capacity and enable us to take widespread advantage of renewable energy sources such as wind and solar energy.

While there are incremental opportunities for energy savings as consumers make behavioral changes in response to more readily available cost information and the consequences of choosing high consumption cycles on individual appliances, the hardwired, invisible and automatic peak demand reducing capabilities that smart appliances can deliver offer large-scale opportunities as whole communities and regions become connected.

Smart appliances will be capable of delivering targeted energy efficiency, through substantial curtailment of energy consumption, at the precise and varying times when the efficiency and environmental benefits are greatest. The benefits of these products include: 1) significantly reducing the need for power generation, transmission and distribution infrastructure to meet peak demand; 2) enhancing the value and effectiveness of episodic renewable energy sources such as wind and solar, whose variability might otherwise exacerbate peak demand issues on the grid; 3) providing consumers with real-time information on peak energy consumption modes and cycles of major appliances, leading to behavioral changes that would reduce both peak and total energy usage of products as high energy consuming modes are chosen less often; 4) lowering consumers’ cost of energy as smart appliances take advantage of time-of-use rates by modifying usage to lower-cost modes and times of operation; and increasing consumer energy education by providing information that encourages energy-wise practices throughout the home.

Despite well-documented benefits to society, utilities and consumers, demand response remains an underutilised resource. One of the key barriers to greater participation is the cost to utilities of installing equipment in homes to enable demand responsiveness. Massive investments are being made by many utilities installing smart meters [4], but very little is being done to enable the loads in the home to respond to price, peak use, and energy storage or fluctuation signals. Research also suggests that consumers are reluctant to have unknown controls installed in their homes [5]. Many have speculated that these barriers could be overcome if major energy consuming appliances came ready to participate in demand response programmes out-of-thebox (“DR-ready”) [6].

A fleet of DR-ready appliances would provide an installed base allowing utilities to better quantify the return on investment of demand response services and helping consumers to understand the benefits of participating in demand response programmes. Recent legislation moving through the Congress specifically contemplates creating incentives to seed the grid with home appliances that are not only highly efficient, but also capable of receiving and responding to a demand response signal.[7]

Whirlpool Corporation is committed to making the Smart Grid and smart appliances a world-wide reality. By 2015, all of the electronically controlled appliances that Whirlpool Corporation produces will be Smart Grid compatible and DR-ready. The architecture representing Whirlpool’s vision is shown in Fig. 1. This architecture is more completely described in the side bar “Demand Response System Architecture.”

Success requires demand response and coordination on three distinct levels to ensure that these capabilities deliver on the promise to increase the world’s energy efficiency. At the lowest level, individual smart appliances must have demand response profiles that are compatible with their operation and with consumers’ expectations. Simple on-off control of appliance loads is not an acceptable option for consumers. Furthermore, how a refrigerator responds to a load reduction request is different from how an air conditioner or a clothes dryer responds. Developing consumer-acceptable demand response profiles that also deliver significant energy efficiency benefits for each of the smart appliances in the home is a significant, but critical undertaking.

The basic premise is that a kilowatt curtailed or shifted is more valuable to the grid and the environment than a kilowatt saved without regard to when during the day it is saved. At maximum consumption, most major appliances consume significantly more energy than the blended annual rate measured by the Department of Energy energy-efficiency standards and the current ENERGY STAR criteria. By way of example, it is much more valuable for a dishwasher to shift or eliminate the 1200 watts of peak power consumption used during the sanitary cycle, than to lower the average annual energy consumption from 307 kWh per year (the equivalent of a continuously burning 35 watt light bulb) to 291 kWh per year (the equivalent of a continuously burning 33 watt light bulb).

Shifting consumption from peak to non-peak periods is so valuable to the system that in Southern California, Edison’s service territory, consumers are charged a 14-cent per kilowatt premium during peak periods. Applying that differential to our dishwasher example would mean that a typical household that shifted its dishwasher usage from all on peak to all off peak would save $45 per year in utility costs. Additional peak energy use reduction capabilities for smart appliances are described in the side bar “Peak Demand Reduction – What are the Possibilities?

The overall response of various loads in a given consumer’s home must also be coordinated based on consumer preferences and current usage. Just as on-off control provides unacceptable outcomes for consumers, an all-or-nothing demand response from all smart appliances in the home is not a viable option. Personal priorities, family schedules, energy saving incentive options, and individual consumer’s desired levels of energy management must be taken into account. In effect, consumers, not utilities or other third parties, must be firmly in control of their home automation network. Providing this meaningful, coordinated demand response across all the energy loads in the home is the function of the intelligent network controller (INC) in Figure 1. As a simple example, delaying the delivery of hot water in the morning may not be as acceptable as reducing the run time of the air conditioner or heater. Similarly, in the evening, reducing the energy used by the clothes dryer may be preferred over reducing the energy used to heat the oven or cool the house. Creating a simple and unobtrusive demand response experience for consumers is a major, and absolutely essential, undertaking. Consumers must be able reduce their energy bills and provide benefits to society with essentially no effort.

For the big utility and societal benefits of demand response to be realised, the load characteristics across thousands and, ultimately, millions of homes must be understood, controlled and coordinated. This is the role of the Energy Management Server in Figure 1. The Energy Management Server utilises the Whirlpool SmartEnergy Algorithm to act as an energy arbiter for millions of appliances, reducing power used by the managed appliances in a way that shares the burden of energy efficiency across millions of consumers, minimising the performance impact on any individual consumer and appliance. This “network-based” energy management algorithm extracts the maximum energy-saving potential across the entire network, rather than providing just the summation of individual appliance savings.

Consumer energy control

Figure 1. Whirlpool’s demand response architecture is a secured, IP-based network providing bi-directional communication between demand response appliances (not only Whirlpool appliances) and the Whirlpool Integrated Service Environment (WISE) and/or the utilities’ smart meter system. Demand response is controlled and coordinated at the level of the individual appliance, the home, and across networks of thousands to millions of homes.

The SmartEnergy algorithm achieves aggregated results through coordination. To orchestrate this coordinated effort, information that is a natural part of a network-based structure is used. This information includes: the location of each managed appliance, real-time status of each managed appliance, the homeowner’s energy-saving preferences, and local energy pricing data. These input parameters enable the SmartEnergy algorithm to output demand response instructions that are then distributed to millions of appliances on a real-time basis. The SmartEnergy algorithm balances the interaction between a network of millions of smart appliances and the Smart Grid. The Smart Grid measures the energy loading in a specific geographical area or areas; loading which includes appliance use in that area. The SmartEnergy algorithm takes input data from the Smart Grid including, for example, a direct request from a utility or grid operator for a specific amount of energy use reduction and/or time-of-use or real-time pricing and transforms these inputs into energy-saving commands to be distributed to the smart appliances in the network. Individual smart appliances will reduce energy use based on these instructions and report their status and energy savings information back to the network. The Smart Grid will then generate new measurements to reflect the latest loading situation and repeat the process of issuing demand response instructions to the network. This process forms a realtime feedback loop of supply and demand.

An important function of the algorithm is maintaining a real-time database of the unrealised supply of energy-saving potential in various geographic areas such that operators always know exactly what level of demand response energy reduction is available. At any given moment, the Energy Management Server knows the status of each individual appliance and whether it is OK to impose an energy-saving cycle and/or the consumer is willing to run an energy-saving cycle. This provides utilities and grid operators with the quantitative information they need to provide the most reliable and cost-effective energy supply to their customers.

A very important policy paradigm shift that must accompany the systemic solutions and benefits provided by smart appliances connected to the Smart Grid, is the move from thinking about energy efficiency only in terms of the static contribution of an individual appliance, to a greater systemic efficiency realised by a whole home, neighborhood or region. As long as all smart appliances together can substantially reduce or shift consumption in response to a grid request, it is less important precisely which appliance delivers what portion of any particular household’s curtailment response. For example, one can imagine two consumers side by side in a neighborhood who both deliver the same amount of curtailment during a peak period, but through a very different combination of appliances and curtailments. The systemic benefit is the same and the system ought to celebrate and facilitate this sort of choice.

Developing the Smart Grid, and the smart appliances and other systems that will interact with it, requires a massive market transformation involving manufacturers, utilities, regulators, and government bodies. A major step toward realisation of the Smart Grid and smart appliances is being driven through the American Recovery and Reinvestment Act which is putting nearly US$4 billion of government funding and a corresponding amount of public and private investment in place to enable large scale Smart Grid demonstration projects by multiple utilities and a host of technology and service provider partners. In the words of the Department of Energy [8] these projects are aimed at developing a smart, strong and secure electrical grid that will help integrate renewable resources onto the grid, deliver power more reliably and effectively with less environmental impact, and create new jobs across the country.

There is also significant regulatory and legislative effort under way to incentivise utilities and product manufacturers to provide demand response services and smart appliances. For example, Section 144 of the Waxman-Markey bill [9], if enacted, would require that states establish peak-load reduction goals and promote the development and transmission of energy from renewable sources. Achieving these goals and maximising the return on the nation’s investment will be much greater and happen much more quickly if, as portions of the grid become “smart,” millions of households having DR-ready products are already in place to take advantage of these national investments. Smart appliances can immediately begin to deliver cost savings to consumers and reduced greenhouse gas emissions. A rapid market uptake, which will be stimulated by the manufacturer and retailer incentives provided for in Section 214 of the bill, even without a fully implemented national Smart Grid, will promote these benefits in the utility markets that have already adopted time-of-use electricity pricing.

Finally, there is a lot of focus on interoperability and rapidly establishing standards that will allow products from all manufacturers to “plug-and-play” together on the Smart Grid. [10]

These developments are critical for ensuring that the investments being made are effective over the long term.

There is a lot to be done, but there has never been a more urgent need for energy efficiency, energy conservation and the reduction of green house gas and carbon emissions. This need, coupled with the public investment environment created by the recent economic crisis, is providing the means for unparalleled transformation. It’s a truly exciting time and massive opportunity for us all.

Peak Demand Reduction – What are the Possibilities?
A typical dishwasher consumes up to 1.2 kilowatts at peak consumption. We believe the consumption can be delayed and reduced to essentially 0 watts during critical peak hours if consumers shift their use to off peak times at night and in the early morning. A typical washing machine consumes roughly 220 watts on average and up to 700 watts at peak consumption. Peak consumption can be delayed for periods of time ranging from 5 minutes to 10 minutes reducing energy use to ~120 watts, a 45% reduction without consumers observing any change in performance. A typical dryer consumes roughly 3800 watts on average and up to 6500 watts at peak consumption. Peak consumption can be delayed for periods of time ranging from 20 minutes up to 1 hour, with energy use being reduced to – 200 watts, a 97% reduction, without the consumer observing any change in performance. A typical water heater consumes 2.5 kilowatts and those with quick recovery consume up to 4.5 kilowatts. When hot water is not needed in the home, one or both heating elements can be turned off and quick-recovery can be deactivated for up to 8 hours without consumers observing any change in performance. A typical refrigerator consumes roughly 170 watts on average and up to 700 watts at peak consumption. Peak consumption can be delayed for periods of time ranging from 30 minutes up to two hours and the total energy use reduced to close to 20 watts, a 97% reduction, without consumers observing any change in performance. If all the refrigerators in service in the US were smart, the amount of curtailable energy from refrigerators alone for an average 3 peak demand hours per day is roughly 435 GW/year.

[1] Us Energy Flow Trends 2002 – (
[2] ERPI Assessment of Achievable Potential from Energy Efficiency and Demand Response Programmes in the US (2010-2030)
[3] Google Smart Meter ( Microsoft Hohm press release – (
[4] Edison Foundation Publication on Utility-Scale Smart Meter Deployment Plans and Proposals – April 2009 (
[5] Continental Automated Buildings Association State of the Connected Home Market Survey 2008 – (, “Developing Appliances to Meet the Smart Grid” Custon Home Online 6/3/2009
[6] Enabling DR-Ready Appliances – EPRI Project 170.006 (
[7] Waxman-Markey bill, American Clean Energy and Security Act of 2009,” section 214 (
[8] Obama Administration Announces Availability of $3.9 Billion to Invest in Smart Grid Technologies and Electric Transmission Infrastructure Recovery Act Funding Will Create Jobs, Help Modernise Nation’s Electric Grid – DoE Press Release, June 25, 2009
[9] Waxman-Markey bill, American Clean Energy and Security Act of 2009,”
[10] NIST Smart Grid Interoperability Standards Project (