An integrated knowledge-based and procedural system for the design of energy conscious buildings

Abstract

The paper describes a methodology for supporting computationally all phases of an energy-conscious design and evaluation process, by combining procedural simulation and knowledge-based heuristic methods in one integrated system. The methodology is based on partitioning the design process into discrete phases and identifying the energy-conscious informational characteristics of each phase. These informational characteristics are expressed in the form of design variables (parameters) and the relationships between them. The methodology allows for testing the applicability of heuristics rules in non-standard cases, using context-sensitive application of default values and judicious selection of rules of thumb, thereby improving the overall predictive powers of the evaluation. The methodology is demonstrated through developing a KB-CAAD system for the schematic-conceptual design phase of solar and low energy buildings. The system is based on the integration of a knowledge-based model and a simulation model with a commercially available CAAD system for building representation. The knowledge base contains heuristic rules for the design of passive solar buildings. Whenever possible, the knowledge-base guides the designer through the decision making process. However, if the rules of thumb are not acceptable for the particular design problem, the system guides the architect by means of the procedural simulation model. A case study demonstrates how the system leads to the design of better energy conscious buildings, while taking less time and effort to introduce the geometrical data that is needed for the representation of a regular, non-solar building.