Facility Smart Grid Information Model (FSGIM) Ready for Review and Feedback

Allen Jones, Independent Consultant allen@allencathern.onmicrosoft.com

Traditionally, facilities have been passive consumers of electricity.  The only data exchanges that occurred between a customer and the grid happened when a) the utility did a monthly reading of the customer’s electric meter, b) the utility sent the customer a bill, and c) the customer sent the utility a payment.  This is increasingly changing.

Facilities are becoming active consumers.  In some cases, this is due to the desire to limit demand charges.  In other cases, this is due to the desire of the customer to reap the benefits of participating in demand response programs.  In yet other cases it is due to the rapidly evolving activity surrounding Distributed Energy Resources and Transactive Energy.  Regardless of the driving force, facilities will need more energy-related information regarding their facilities in order to maximize the benefits of participating in the Smart Grid.

Where will this information come from?  In many cases it will come from the numerous systems that building owners, operators, and occupants already have within their facilities.  Building owners, operators, and occupants will not throw out their existing systems just to participate in the Smart Grid.  Rather, existing systems and the protocols that they use will adapt and contribute towards the facility’s Smart Grid-related goals.

But what happens if each of the protocol groups work on this adaptation totally independently? The result is likely to be confusion when it comes time to collect the data from the various systems and to provide information on how the facility as a whole will interact with the grid.  For example, suppose we ask several different systems in a facility “What is the electrical demand of your devices?”  We might get back any of the following answers:

• the “instantaneous” demand, i.e., the power that you would measure if you connected a power meter to your loads and read the value “right now”;
  

• the power being drawn by loads right now minus the power being supplied by onsite generation right now;
  

• the average power drawn by loads so far in the current five or fifteen minute meter billing interval;
  

• the average of the power drawn by loads minus the power being supplied by onsite generation so far in the current five or fifteen minute meter billing interval;
  

• the average power that was drawn by loads during the last five or fifteen minute meter billing interval;
  

• the average of the power drawn by loads minus the power supplied by onsite generation during the last five or fifteen minute meter billing interval;

Obviously, if we try to combine these various readings when there is not even a common agreement of what the term “demand” means, the results are likely to be meaningless.

What is needed is a common semantic model to which the various systems and their protocols can be mapped.  To address this issue, the Smart Grid Interoperability Panel created PAP 17 and tasked it with creating an abstract information model that was capable of supporting a wide range of energy management applications and grid interactions including but not limited to:

• on-site generation
  

• demand response
  

• electrical storage
  

• peak demand management
  

• forward power usage estimation
  

• load shedding capability estimation,
  

• end load monitoring (sub metering),
  

• power quality of service monitoring,
  

• utilization of historical energy consumption data, and,
  

• responsive energy control.

This abstract model could then be used by various protocol standards groups to adapt their protocols to provide the information needed to interact with the Smart Grid.

ASHRAE was assigned the lead role of coordinating the development of the standard and created a Standard Project Committee, SPC 201, to undertake the task. The SPC 201 committee consists of experts from residential, commercial, and industrial automation equipment manufacturers, protocol groups, utilities, Independent System Operators, governmental organizations, and academia.

The committee recently voted unanimously to release the result of its work, the Facility Smart Grid Information Model, for a Publication Public Review.  The Publication Public Review is the last major step before the draft standard will become an official ANSI/ASHRAE/NEMA standard.  The draft standard is also being considered for designation as an International Standard (ISO/WD 17800) through ISO/TC 205.  The Publication Public Review is scheduled to begin on August 7th and run through October 6th.  Beginning on August 7th, the public will be able to freely download a copy of the draft standard from the ASHRAE Public Review Draft Standards / Online Comment Database located at https://www.ashrae.org/standards-research--technology/public-review-drafts.  By clicking on the link in Annex A of the draft standard, reviewers will also be able to download an electronic version of the model itself.  The electronic version of the model is available in two formats:

• .EAP file – readable using a free EA Lite viewer from Sparx Systems or readable and modifiable using the Enterprise Architect Professional version also available from Sparx Systems.
  

• .xml files – XML Metadata Interchange formatted files that are importable into other modeling tools.

What is the FSGIM?

The FSGIM is an abstract information model of what the Smart Grid looks like from the perspective of a facility.  A facility can be a home, a commercial office building, a factory, or a campus.  It is basically anything on the customer side of the electric meter.  The FSGIM is not a protocol, but it can be used by other protocols to aid them in adapting to the Smart Grid.  The FSGIM models the information that would need to be exchanged within a facility to participate in the Smart Grid.  The FSGIM does not make assumptions regarding how much of this internal information the facility is willing to expose to the outside world.  The FSGIM also models certain information that a facility is likely need to exchange with the outside world such as demand, pricing, and weather information.

The SPC 201 committee began its work by reviewing a set of Use Cases developed by the Energy Information Standards Alliance.  These Use Cases describe, in general terms, the information that would need to be exchanged within a facility and between a facility and the grid in order for facilities to fully participate in the Smart Grid.  The SPC 201 committee then took this initial work, expanded on it, and began creating a model capable of providing the identified information.  This was done by creating four basic logical building blocks that can be used to model real devices within a facility.  These logical building blocks are

• The Load Component – used to model devices that consume energy
  

• The Generator Component – used to model devices that provide energy
  

• The Meter Component – used to model devices that measure energy, power, or pollutant emissions that result from generating power
  

• The Energy Manager Component – used to model devices that make decisions based on power, energy, emissions, costs, weather, etc.

The logical building blocks are further broken down into classes that are modeled using Unified Modeling Language (UML).  For those who are not familiar with UML, the draft standard includes an informative annex that provides the reader with a tutorial sufficient to allow the reader to be able to understand the model.

Real-life devices are modeled by assembling combinations of these basic logical building blocks.  For example, an energy storage system could be modeled as a combination of a Load Component and a Generator Component.  The Load Component would model the characteristics of the energy storage system when it was charging; the Generator Component would model the characteristics of the energy storage system when it was discharging.  If the energy storage system made its own decisions regarding when to charge and discharge based on factors such as the cost of electricity or the anticipated weather conditions, then it would also include the Energy Manager Component.  The model can be applied to systems ranging from simple direct load control to transactive energy-based systems.

Reusing Existing Standards

The FSGIM makes use of a number of existing standards and specifications.  The OASIS EMIX and Energy Interoperation specifications were used as the basis for modeling market interactions between facilities and the grid.  The OASIS WS-Calendar PIM specification was used as the basis for modeling schedules.  The IEC CIM (Common Information Model) guided the representation of meters in the model while the IEC 61850 standard guided the representation of generators in the model.  Finally, the WXXM weather model served as the basis for representing the weather data that might influence a facility’s behavior relative to the grid.

The committee did extensive work harmonizing these existing standards.  For example, there were initially three different, incompatible models for representing a power measurement.  There is now one representation that can be readily transformed into any of the other forms as needed.  The committee also worked extensively with several of the standards groups, providing feedback on portions of their models that had missing or inconsistent information or that had modeling errors.

Conforming to the FSGIM

The classes in the FSGIM can be used in many combinations, but to help with interoperability when the FSGIM is applied to real-life scenarios, the FSGIM is also grouped into Conformance Blocks.  These Conformance Blocks encapsulate a group of classes and associations that are necessary in order to achieve a certain functionality.  For example, there is a Conformance Block for curtailable loads that includes all of the classes that apply to a load whose demand can be adjusted in response to a signal from the grid.  Protocol groups that wish to claim that portions of their protocols conform to the FSGIM are free to decide, within certain bounds, which Conformance Blocks they will implement and how their protocol will map to those Conformance Blocks.  Protocol groups self-certify their conformance to the FSGIM.  If a protocol group claims conformance, they must publicly state how their protocol maps to the Conformance Blocks that they claim to implement.  End products achieve conformance to the FSGIM by implementing a conforming protocol.

Both grid-level and facility level protocols can be mapped to the FSGIM.  For example, work is currently being done within the BACnet Smart Grid Working Group to map both BACnet Web Services and OpenADR 2.0b to the FSGIM.  This will allow BACnet Web Services to support OpenADR 2.0b.

Making the FSGIM Easier to Use

[an error occurred while processing this directive] Since the FSGIM is very large, several things have been done to make it easier to use.

• The FSGIM is very modular.  This allows users to ignore parts that do not apply to their domain.  For example, if the user does not do anything with generation or energy storage, they can ignore the Generator Component. 
  

• Within each logical component, there are one or more class diagrams that explain the component.  A listing of the classes in each diagram, along with a description of each class, appears immediately after each drawing.  Many users will not need a more detailed understanding than what they will learn from these diagrams and class descriptions.
  

• If the user needs a detailed understanding of a given class in the diagram, they can click on a hyperlink that will take them to explanations of all of the attributes and associations for that particular class.
  

• The printed and pdf versions of the draft standard also have information that ties each section of the written standard to the appropriate part of the electronic model.
  

• The draft standard contains an informative annex that provides implementation examples for some of the concepts that may be difficult to understand without a concrete example.
  

• The Use Cases that the committee used when creating the draft standard are included in an informative annex along with a cross-reference explaining how the information in those Use Cases maps to the information in the FSGIM.
  

Once the public review period opens, you are encouraged to download the draft standard and the electronic model.  Explore the model and provide feedback using the ASHRAE Online Comment Database.  Let’s get one step closer to allowing automated buildings to be full participants in the Smart Grid!

About the Author

Allen Jones is an independent consultant specializing in the interaction between facilities and the Smart Grid.  He has almost ten years of experience working on Smart Grid related topics, over ten years of engineering management experience, and over twenty years of project management and embedded systems development experience.  He serves on several national and international Smart Grid-related committees and also serves as an adviser to the BACnet Smart Grid Working Group.

Previously, Allen was with Schneider Electric where he was responsible for managing standards activity and analyzing energy-related trends for the Buildings Line of Business.  In that role he helped determine the requirements for a new energy reporting product, evaluated the gaps in a newly acquired energy reporting company, and served as a Smart Grid domain expert for a global proof-of-concept rapid development project.

He can be reached at allen@allencathern.onmicrosoft.com.

[an error occurred while processing this directive]
[Click Banner To Learn More]

[Home Page]  [The Automator]  [About]  [Subscribe ]  [Contact Us]