OPTIMIZATION OF BUILDING MATERIAL SELECTION FOR ENERGY SAVING IN COMMERCIAL BUILDINGS IN DIFFERENT CLIMATIC CONDITIONS

Abstract

Most engineers predict future building energy consumption via simulation programs in the pre-design phase. In this process, many simulation steps have to be repeated to predict building energy consumption. The authors in this article proposed another way to select optimal building materials for saving commercial building energy in the U.S. using soft computing methods. To achieve the research goal, reliable public data that is provided by the U.S. Energy Information Administration was used. The data contain numerous energyrelated characteristics of buildings including gas, electricity, types of materials, and climate conditions of 6,700 commercial buildings located in the U.S. This study utilized two methods to find out optimal building materials for saving energy. First, the Principle Component Analysis was used to determine which building characteristics among over 400 characteristics have the greatest impact on gas and electricity consumption. Second, Association Rule Mining was used to extract combinations of optimal building materials. Since a building consists of a combination of various materials, energy simulation should predict for multiple factors rather than a single factor. The use of these methods would greatly reduce resources, such as limited budget and time, during the simulation process.