Intelligent Building Agents

Achieving national goals of net zero energy buildings requires substantial reduction in the energy consumption of commercial building systems. Although significant progress has been made in the integration of building control systems through the development of standard communication protocols, such as BACnet and BACnet/IP, little progress has been made in making them “intelligent” or in optimizing the performance of building systems. The focus of this project is to develop and test distributed intelligent agents that can control and optimize the performance of building systems.

Objective:  To develop intelligent agent techniques for optimizing the control performance of interacting building systems and demonstrate their feasibility for reducing energy consumption in commercial buildings by 2014.

What is the new technical idea?  The new technical idea is to adapt advances in intelligent agent technology to solve the control optimization problem.  Intelligent agents have been successfully implemented in a variety of applications, including search engines and robotic systems, and a considerable amount of information already exists in the artificial intelligence community on different agent architectures (e.g., deliberating, reactive, and hybrid), agent design and implementation, and agent programming.   Intelligent agents know or can learn the performance and status of the systems and equipment they monitor and can communicate and collaborate with other agents to achieve a common goal, such as minimizing energy consumption and/or cost of operation, maximizing comfort, identifying and diagnosing problems, etc.  Intelligent agents make it possible to solve the problem of building system optimization in a “distributed manner” which greatly simplifies the computational methods required. An intelligent building agent simulation program (IBAS) has been developed and is currently being used to screen and evaluate prototype intelligent agents for their suitability to optimize the performance of building systems. Experiences gained from this simulation program will be used to further develop and rigorously test Intelligent Building Agents under more realistic condition for a wide range of equipment types.

What is the research plan?  The results to date demonstrate that intelligent agents have the potential to solve the optimization problem for building systems. A collaboration has been established with university researchers to explore technology options for a second generation IBAS platform with more flexibility and capability based on lessons learned from the first generation.  In FY2013 this collaboration will continue in parallel with further exploration of the potential of intelligent agents using the current generation IBAS. In addition, a new laboratory is under construction that will enable implementation of intelligent agent control of a mixed system of chillers, boilers, and air distribution components to conduct research on real equipment under controlled conditions. In FY 2013 that laboratory will be completed and a controls platform suited to intelligent agent research will be commissioned.

In FY2014 the new laboratory facility will be used for experimental testing of intelligent agent designs. Different approaches to “distributed optimization” in building HVAC applications will be explored including how agents can use negotiation techniques to optimize the performance of entire building HVAC systems.

After completion of this project it is expected that additional research will be needed that will involve CRADAs with control system manufactures to further developed and begin field testing of intelligent agent systems and speed commercialization of the technology. Lessons learned from this research will enable development of draft specifications for several intelligent building agents (e.g., an Air Handling Unit agent, a Chiller agent), including a list of functions to be included in each agent, technical design features, software architecture, and methods for system identification, communication, and negotiation among agents. NIST will work with standards writing organizations, such as ASHRAE, to develop guidelines and/or standards based on these specifications to facilitate the implementation of intelligent agents in building control applications.

Recent Results:

Outcomes:

Methods were developed for online performance identification and an intelligent building agent simulation program (IBAS) was developed using the intelligent agent framework (Fleeble3) and agent communication based upon the KWLM communication protocol.

IBAS prototype developed using quasi-steady-state models to simulate an HVAC system and containing fourteen agents that monitor and control eight VAV Boxes, two air handling units, two variable speed chillers, two cooling towers, and variable speed fans and pumps. The agents communicate and negotiate with each other using Publish/Subscribe “channels” within the Fleeble3 framework.

Outputs:

Detailed design including construction drawings and equipment specifications for the new laboratory facility completed.

“Optimization of Building HVAC Systems Using Intelligent Agents – A Proof of Concept Study.” Kelly, G.E. and Bushby, S.T., NIST TN 1707, 2011.

“Are Intelligent Agents The Key to Optimizing Building HVAC System Performance?”, Kelly, G.E. and Bushby, S.T., HVAC&R Journal (publication pending).

Standards and Codes: None at this time. As the technology matures it is anticipated that there will be a need for industry guidelines, standards and codes that include:

• best practice guidelines for applying and specifying intelligent agent technology;
• method of test standards for evaluating the performance of intelligent agents; and
• building energy code provisions regarding the use of intelligent agent technology, e.g.,  ASHARE Standard 189.1 and California Title 24.