| Heating, ventilating and air conditioning (HVAC) systems accounted for about 40% of the total energy consumption in buildings in the United States. However, a variety of faults on the different types of HVAC components are responsible for about 10-40% of the amount of energy used. In order to save energy and improve the control of indoor environment, researchers have developed hundreds of computerized approaches and algorithms that can automatically and continuously analyze the conditions of HVAC systems. However, although previous research studies have validated the energy saving potential of these approaches, their complex information requirements are one of the major challenges for system operators to deploy them in real-world facilities.;In order to utilize the energy saving potential of these algorithms, I develop an integrated information support framework to facilitate the process of collecting, integrating and providing information required by the algorithms developed for the secondary HVAC systems. This research study consists of three main steps. First, in order to identify a general set of information requirements, I extended the Information Delivery Manual approach to formally identify and document information requirements of 104 algorithms. These information requirements were mapped to three data models (Industry Foundation Classes, Green Building XML and EnergyPlus) using an extended Model-View-Definition approach. Second, I formalized a semi-automated approach to integrate the data items that are stored in different data models. This approach enables the users to define the structures of the control loops in their HVAC systems. The structures are used to guide the integration of data models that contain different types of information at various levels of detail. Third, in order to automatically retrieve the information required by different algorithms, I formalized a domain-specific query language and the corresponding query mechanisms to enable the users to formulate query statements that describe their information needs. The query mechanisms are able to parse the query statements, reason about the contextual information that describe the information needs, and retrieve the information items. Finally, I evaluated the performance of the formalized approaches using a prototype framework and a real-world testbed. |
