This research project was in support of the 2nd phase of Ben Welle's PhD thesis. The project involved building into our whole building daylighting analysis tool (PIDO2Rad) software to automatically identify identical spaces in a large building model to reduce the overall building simulation time. A link to the published article is here.
Flexible problem formulation is required for product model-based thermal analysis using multidisciplinary design optimization (MDO) environments for speed, accuracy, scalability, and cost-effectiveness in the Architecture, Engineering, and Construction (AEC) industry. The integration of daylighting simulation into an MDO process, however, presents several implementation challenges. In current practice, the process of an architect, engineer, or daylighting consultant to determine how to analyze a given building design for daylighting performance is frequently subjective, time-consuming, and inconsistent. Furthermore, long simulation time requirements for daylighting significantly hinder the realization of many benefits from MDO. The determination of which spaces in a building are sufficiently different to warrant an independent daylighting analysis is based primarily on building physics, building design criteria, and operating schedules (e.g. occupancy schedules). This characteristic of daylighting analysis creates the opportunity to develop intelligent mechanisms to automate the identification of the building spaces for analysis using performance-based methods, simulation of spatial results, and the scaling of spatial simulation results to whole building performance metrics in a fraction of the time it takes in current practice. Such methods would result in improved speed, accuracy, scalability, and cost-effectiveness for MDO- and non-MDO based daylighting simulation. Currently, no such methods exist in literature or in practice. This paper fills these gaps by presenting a methodology for automated product model decomposition and recomposition for climate-based daylighting simulation using Radiance. The authors validate the research with the method’s application to several test cases and a large federal office building industry case study.