Open ADR– the State of the Standard

by Dave Hardin, EnerNOC Inc.

Utilities long have called upon demand-side resources to complement traditional generation. During the past several years, the sophistication, reliability and adoption of demand-side management programs have increased significantly. In particular, demand response (DR) programs, which offer energy users financial incentives to reduce their electric demand temporarily in response to system conditions, have become widely used.

According to the most recent Federal Energy Regulatory Commission (FERC) estimates in its “2010 Assessment of Demand Response and Advanced Metering,” total DR resources in the United States have increased more than 40 percent since 2007. The total potential contribution from existing DR programs exceeds 58,000 MW, representing some 7.6 percent of the nation’s peak demand, according to the assessment.

As DR has grown, more sophisticated data exchange and automation technologies also have proliferated. As a result, utilities are becoming more concerned that deployment of separate, proprietary solutions might create vendor lock in and so-called stranded assets that depend on specific hardware and software. They are pursuing the use of interoperable products based on standards that can help avoid these issues.

One standard being considered is related to automated demand response (AutoDR), which, despite its relatively limited current deployment, is considered widely to be a critical component of future DR programs. In particular, an emerging standard known as OpenADR recently has sparked discussion for its ability to provide a consistent framework for AutoDR. Several California utility programs have been using the standard to support AutoDR programs, and the OpenADR Alliance held its inaugural members’ meeting in early 2011. Though challenges remain, the continued refinement of the OpenADR standard will help ensure that AutoDR programs are cost-effective, scalable and accessible to many participants while flexible enough to address future utility challenges.

The Origin of the OpenADR Standard

OpenADR was conceived after the 2001 California energy crisis during which large-scale blackouts affected millions of ratepayers across the state. The California Energy Commission (CEC) proposed that safeguards and solutions be developed to help prevent another such event and envisioned AutoDR playing an important role. In 2002 the state’s Public Interest Energy Research Program and the Department of Energy began funding research at Lawrence Berkeley National Laboratory (LBNL) to develop what would become the OpenADR standard. LBNL’s Demand Response Research Center conducted the research with representatives from utilities, California Independent System Operator (CAISO), energy services firms and hardware manufacturers. Their efforts resulted in several successful pilot programs and the emergence of OpenADR version 1.0 as a de facto standard that presents several benefits, including:

  • Consistency. An open standard encourages uniformity across AutoDR programs, which avoids duplication of efforts and facilitates sharing of service and planning models among program designers.
  • Reduced costs. Open standards create markets for software tools and systems that reduce the costs of entry and enable current stakeholders, as well as new entities, to participate in AutoDR with less up-front investment.
  • Facilitated innovation. As a standard that is developed collaboratively, OpenADR can be fine-tuned to meet the changing needs of program designers and participants.
  • Increased choice. The ready availability of the OpenADR standard encourages hardware manufacturers, program managers and curtailment service providers (CSPs) to invest in solutions based on the common standard. In turn, utilities and end-use customers are able to choose among options to suit their needs.

OpenADR has been used in California pilot programs since 2005, and the technology has been tested successfully in facilities including commercial property, retail locations, data centers, university campuses and industrial facilities. When version 1.0 of the OpenADR standard was finalized in April 2009, the CEC reported that some 200 commercial and industrial facilities in California were providing an estimated 50 MW of AutoDR capacity through the OpenADR protocol. While this figure is significant, AutoDR represents a clear minority of overall available demand response resources, and OpenADR must grow to support broad-based, high-capacity AutoDR programs. The OpenADR Alliance is poised to help build an even stronger version of the standard.

The Road to OpenADR 2.0

During the alliance’s members’ meeting in January, stakeholders discussed how to improve and expand upon the current OpenADR standard before presenting it for consideration by the National Institute of Standards and Technology (NIST) and the Smart Grid Interoperability Panel (SGIP). The alliance continues to weigh the protocol’s relative strengths and weaknesses as it moves toward defining OpenADR 2.0, including:

Telemetry. Stakeholders must set expectations regarding two-way communications across the OpenADR standard. The current standard specification narrowly defines bi-directional communication. Utilities send DR signals through a Demand Response Automation Server (DRAS), and clients at end-use facilities confirm the receipt of those messages before enacting a predetermined energy-reduction plan. Currently C&I sites cannot communicate other messages during an AutoDR dispatch such as periodic updates on energy use, system state or other information relating to DR dispatch performance using OpenADR. Without this additional information provided through robust telemetry, utilities using OpenADR for AutoDR programs are unable to ensure that individual sites perform as expected during a dispatch and have limited opportunity to provide support or coaching to increase performance during a dispatch. Furthermore, full dispatch performance only can be calculated after the dispatch has concluded, meaning that any problems encountered during a DR dispatch will not be fully discovered until it is too late to address them.

End user participation in an AutoDR dispatch could vary for many reasons, including intentional opt out, changes in business conditions, shift in baseline consumption or loss of connectivity. As a result, the benefits of meaningful two-way communication between site and utility are potentially significant. Better information exchange would enable utilities and CSPs to measure and manage dispatches more effectively. If OpenADR is to be deployed on a larger scale, more advanced communications will be essential to ensure programwide performance. Better performance tracking would enable utilities to place greater confidence in AutoDR as a firm demand-side resource.

Cost-effectiveness. While more robust two-way communications may solve one problem for OpenADR, it also might introduce a second. Additional communication exchange comes at additional cost in the form of bandwidth, storage and server resources. It would be ironic if a low-cost standard such as OpenADR were to cost CSPs, utilities and grid operators more in the end through the increased cost of these exchanges. In the near term, OpenADR-based AutoDR programs are unlikely to reach sufficient size to create concerns over data transfer costs. If this issue is not addressed today, however, OpenADR risks facing it in the future. There are ways to circumvent this outcome by augmenting the standard Web services platform with another more scalable and cost-effective communications protocol. Though Web services have proven sufficient for current OpenADR-based programs, a change in communication architecture should be considered for future versions to support cost-effective growth.

Ancillary services support. In light of California’s pledge to reduce statewide greenhouse gas emissions and the ongoing transition to renewable energy sources, presenters at the OpenADR Alliance meeting discussed how AutoDR might be used to support systemic intermittencies from wind and solar power, as well as its potential use for cost-effective residential demand management. In its current state, OpenADR is not designed to provide these types of grid-balancing or ancillary services; however, the potential is there to do so.

Again, improved communications are the key. Ancillary services programs require relatively short response times and near real-time feedback on performance. Utilities and CSPs must be able to monitor energy-use information quickly, sending and receiving information in sub-minute intervals. While this capability exists in many manual and automated DR programs based on different communications architectures, the OpenADR standard does not yet provide this level of feedback in a scalable, cost-effective manner.

The Benefits of Building a Better Standard

The benefits of a better standard are clear. A carefully considered 2.0 release of OpenADR might be able to increase program accountability through improved two-way communications. It also could ensure cost-effective scalability while setting the stage for OpenADR-based AutoDR programs that could provide ancillary services for renewable generation.

There is widespread agreement that the smart grid transition will be expensive. Pike Research reports that smart grid investments exceeded $10 billion worldwide in 2009 and are expected to reach nearly $36 billion by 2013. With the proper investments of time, expertise and resources at this early stage, however, the industry might be able to save significant costs in the long term. By working within the OpenADR Alliance, stakeholders hope to create a mutually beneficial new version of the OpenADR standard while identifying areas for continued improvement. If the alliance succeeds in this mission, DR providers will be able to offer an attractive, low-cost form of AutoDR that is easy for their customers to implement and practice. The net effect would be strong growth for the DR market and a smarter, more resilient grid.


Dave Hardin is senior director of smart grid standards at EnerNOC. Reach him at


More Electric Light & Power Articles
Past EL&P Issues
Previous articleCS Week 35 Promises to be Dynamic
Next articleOptimize Maintenance Operations to Reduce Operating Costs

No posts to display