Data Modeling for the Integrated and Intelligent Computer-Aided Design of Buildings and HVAC systems

Abstract

Building design includes design (namely selection and configuration) of HVAC (Heating, Ventilation, Air, and Conditioning) systems. Conventionally, building design goes first, followed by design of a HVAC system, which fits to the building designed and even constructed. This process of building construction not only likely leads to re-design and re-construction of a building, thus incurring extra cost and increased lead time, but also makes it impossible to simultaneously design a building and design a HVAC, thus losing the potentially improved design of a building along with its HVAC system otherwise. Therefore, the idea of integration of both design activities could be promising. To facilitate a computer system to enable the integration, the first step is to develop data models for the building along with its HVAC system. It is noted that data in this thesis include both information and knowledge; particularly, information carries the semantics of what and where and when it is, and knowledge carries the semantics of why and how to do it. Data models help not only to understand the semantics of a work domain (building along with its HVAC system) but also build a bridge between the human designers and computer. There are three specific objectives with this thesis study: (1) analyzing the work domain (building along with its HVAC system), including its integrated or concurrent design process; (2) developing an information model (i.e., data model for the information of building and HVAC in the context of building); (3) developing a knowledge model for design of HVAC systems to a building under design. It is noted that in this thesis design of HVAC is meant for selection and configuration of HVAC systems to a building under design.iv A tool called FCBPSS (F: function, C: context, B: behavior, P: principle, SS: state-structure) architecture of information and knowledge of a system was employed for the research defined in Objective 1 due to its promise of a comprehensive analysis of a work domain. For Objective (2), the E-R (Entity-Relationship) modeling approach was employed due to its simplicity and adequate capacity of expressing semantics of a work domain as opposed to various relational modeling approaches. For Objective (3), OWL (ontology web language) and SWRL (semantics web rule language) were employed due to their generality and wide popularity. As such, the ontology of buildings along with HVAC systems was built, the information model and knowledge model for buildings along with HVAC systems were built. The three objectives were achieved with three models, namely in short (i) the work domain model, (ii) the information model, and (iii) the knowledge model for a residential building along with its HVAC system. The implementation of a computer program system based on the three models and enables to support integrated design of a building and its HVAC system was also discussed. A case study was discussed to illustrate the effectiveness of the models as well as the computer program system. In summary, this thesis has demonstrated the feasibility of a computer-based integrated and intelligent design process for a building along with its HVAC system. It seems that nobody in the literature of building design has ever attempted the idea behind this thesis and a comprehensive work domain analysis along with data (information and knowledge) modeling. It is worthy of attention that all the models developed are open and extendable, and because of this feature, completeness of the work makes sense.