SIP Forum's Smart Grid Group: Why SIP Is the Best Protocol and Architecture for Demand and Response

Toward the end of 2009, SIP Forum started a task force dedicated to evaluating the application of the Session Initiation Protocol in the Smart Grid ecosystem. The origins of this group were previously covered in an article I wrote for this portal here. Six months later, the group has over 100 members across a wide spectrum of companies (Utilities, Service Providers, OEMs, ISVs etc.) and has produced several key approach notes and white papers that further outline the relevance of the role SIP can play in the Smart Grid ecosystem.

Some of the documents created by members of this group will additional insight:

The bottom line, so to speak, is based on the contributions made, it is clear that the flexibility of SIP makes it an appropriate choice for a variety of functions across the Smart Grid space. SIP can significantly reduce costs and time to market for all the players, simply because SIP is already proven and widely deployed in both enterprise and residential networks today, handling VoIP, IM, presence, and event services, many of which are applicable to the Smart Grid network. 

One of the primary goals of this group is to educate people on the various functions of SIP beyond handling voice calls and to that goal, this group recently published a significant white paper on the applicability of SIP for implementing a scalable Demand and Response system. The paper, authored by Hughes Systique Corporation, can be downloaded here, and is posted on this portal as well.

Simply put, Demand and Response is a mechanism to manage consumption of electricity in response to supply conditions, price of electricity and other parameters. Defining it further, a Demand Response program is a critical component, typically developed and offered by Utility companies or ISOs that offer participants to contribute effectively for better energy load management/reduction. 

Typical participants include individuals as well as small and large corporations. A DR program typically ties in with a dynamic pricing scheme for electricity where participants, and depending on a variety of factors (such as time of day, price etc.) can actively participate in requesting a Utility for increase or reduction of electricity demand. While this helps in cost reduction for participants, it significantly helps Utility/ISOs reduce their own cost and manage distribution of electricity (which is a finite resource) better. Several Utilities/ISOs offer incentive programs to participants for taking part in DR programs. In addition to DR, automation of DR is a key concept which helps reduce human intervention and increases accuracy and responsiveness to the DR program.

Interestingly, DR is a surprisingly primitive operation today. Traditionally, many components of DR have been a manual process. Usually, an industrial administrator would monitor DR events and manually react to peak pricing indications. For enterprises which did not have dedicated personnel, there are several aggregators who liaise between the Utility and the enterprise customer and perform such monitoring; and when system peak load reaches close to critical peak, they inform the enterprise customers to start shedding load.

While there is a degree of automation here, this too involves significant manual intervention. Often, indications are delivered to customers over phone calls or even in-person visits. Furthermore, there is no standard in DR which is widely used. Therefore, any automation, however small is mostly proprietary.

Well, DR is being done now. Just not in an automated fashion. And because DR is not automated, it makes deployment of DR expensive for utilities. But why is DR important? Because unless you can do effective management of Demand and Response, Utilities will not be able to manage their peak loads, and will hit critical loads more often than required. The impact of this is significant. FERC has a white paper on the importance of DR which can be downloaded here.

Got it, so what are the Standard bodies doing about this and how is SIP better?

The OpenADR research has created a specification that lays out an architecture for an automated and interoperable end-to-end Demand and Response system. NIST has endorsed this architecture as well. Effectively, OpenADR specifies abstract template and procedures for an automated DR system. The Open ADR architecture is a perfect fit for SIP because SIP has a powerful Event system which is exactly what automated DR is all about - it's a mechanism to:

Interestingly, this is what forms the backbone of the SIP Events architecture and it is already proven - all enterprises and operators that deploy VoIP use this events architecture for presence!

The OpenADR has also implemented a version of the specifications via Web Services, but we believe that SIP offers a significantly more scalable and powerful solution while adhering to the OpenADR architecture at the same time (in other words, SIP is better than HTTP for this purpose).

For more detail, please read our paper here. Hughes Systique Corp. has also implemented a proof of concept of such a system by re-using existing open source SIP servers. It is our hope that more people evaluate the power of SIP for such applications and make their own decisions as a result.

For any questions related to this paper or approach, feel free to email [email protected].