A gradient descent technique coupled with a dynamic simulation to determine the near optimum orientation of floor plan designs

25 Jul 2013  ·  Eugénio Rodrigues, Adélio Rodrigues Gaspar, Álvaro Gomes ·

A prototype tool to assist architects during the early design stage of floor plans has been developed, consisting of an Evolutionary Program for the Space Allocation Problem (EPSAP), which generates sets of floor plan alternatives according to the architect's preferences; and a Floor Plan Performance Optimization Program (FPOP), which optimizes the selected solutions according to thermal performance criteria. The design variables subject to optimization are window position and size, overhangs, fins, wall positioning, and building orientation. A procedure using a transformation operator with gradient descent, such as behavior, coupled with a dynamic simulation engine was developed for the thermal evaluation and optimization process. However, the need to evaluate all possible alternatives regarding designing variables being used during the optimization process leads to an intensive use of thermal simulation, which dramatically increases the simulation time, rendering it unpractical. An alternative approach is a smart optimization approach, which utilizes an oriented and adaptive search technique to efficiently find the near optimum solution. This paper presents the search methodology for the building orientation of floor plan designs, and the corresponding efficiency and effectiveness indicators. The calculations are based on 100 floor plan designs generated by EPSAP. All floor plans have the same design program, location, and weather data, changing only their geometry. Dynamic simulation of buildings was effectively used together with the optimization procedure in this approach to significantly improve the designs. The use of the orientation variable has been included in the algorithm.

PDF Abstract
No code implementations yet. Submit your code now

Tasks


Datasets


  Add Datasets introduced or used in this paper

Results from the Paper


  Submit results from this paper to get state-of-the-art GitHub badges and help the community compare results to other papers.

Methods


No methods listed for this paper. Add relevant methods here