Parametric Building Performance Simulation in the Early Architectural Design Stage: Mid-Rise Apartment in Hot and Dry Climate

Ceaseless increase of energy demand in building sector has become a challenge for designers, which is often combines with some goals like indoor air quality, environmental impact, and building costs. To support designers, building performance simulation is a common technique in development-design stages, however, its implementation in early stages is limited, even though early decisions have higher impact upon final performance and costs of buildings. Architects have to design more energy-efficient buildings due to the requirements of energy efficiency regulations in various countries. There are several simulation tools, which may help designers during the design process, to estimate the thermal performance of the building under consideration. However, architects are reluctant to use these tools for several reasons: they are not user-friendly, need detailed information about the specifications of the building elements, which are not known at the initial phase of the architectural design; building simulation models are time-consuming and the interpretation of the simulation results is difficult for architects. In this paper, we present a method for energy efficiency optimization that can be applied in the initial architectural design process. This method will help architects to select the optimized floor plan regarding the functional, thermal and lighting parameters in the preliminary stage of building design. Here we implement sensitivity analysis and simulation-based optimization in order to optimize the thermal comfort, energy and daylight performance of residential buildings in Tehran. These objective functions were simulated using EnergyPlus and Radiance software programs for individual residential building configurations that were generated by parametric modeling techniques. Two thousand simulations for one hundred building floor plans were performed to create a comprehensive dataset covering full ranges of design parameters. The floor plans were created using an algorithm developed by Eugenio Rodrigues. The algorithm generates floor plans regarding the adjacency and dimensions of the rooms, location, and size of door and window, together with the entrance location. The main distinction of this study compared to the similar researches is including floor plan design as one of the parameters of optimization in the hot-dry climate of Tehran. A residential unit, which is situated on the middle floor of a mid-rise apartment, was selected as the base model. The present study considered building floor plan, building construction materials, glass type, insulation thickness, floor height, WWRs for kitchen, bedroom and living room and color of the floor finishing, walls, and ceiling as design variables to achieve the optimize Energy Use Intensity (EUI), useful daylight illumination, and occupants’ adaptive comfort. A simulation-optimization tool that couples a multi-objective genetic algorithm to a whole-building performance simulation engine was applied in order to find the optimal set of design variables, and finally, the results of the energy and daylight simulations were implemented into a set of regression and simple sensitive analysis equation to predict the most effective variable in each objective. Sensitivity analysis showed that the type of floor planning is most effective parameter for all objectives except that external wall material is an effective parameter for EUI, and occupant comfort and WWR are effective for daylight quality.