Attribute Information Extracting Method for Air Quality Assessment of Buildings

Abstract

Extracting architectural elements from Industry Foundation Classes (IFC) files plays an important role on indoor air quality assessment. However, the traditional methods may extract useless instances and miss some necessary information, which results in poor air quality assessment. To address the above issues, this paper proposes an attribute extraction method for air quality assessment from IFC files, called as IFC-AAE. First the instances of the IFC file are preprocessed to remove the redundancies. Next the entity instances related to air quality assessment are extracted and then classified based on floors. Finally, the attribute information of these entities is extracted according to their reference relationship. The experimental results show that the IFC-AAE method is superior than the previous methods. Compared with the IFC file analyzer, the IFC-AEE method generates fewer invalid data. Compared with the Map-based extract method, the IFC-AEE method has an improvement by 4.78% on the precision rate on average.