Soft Computing for detecting thermal insulation failures in buildings

Improved detection of thermal insulation efficiency in buildings could substantially contribute to reductions in energy consumption and the carbon footprint of domestic heating systems. Thermal insulation standards are now contractual obligations at the construction stage, although they are not standardized in buildings that are already in operation: lighting, occupancy and temperature profiles, air conditioning, and ventilation, all add to the complexity of standardization processes. The identification of thermal insulation failure can help to reduce energy consumption in heating systems. Conventional methods can be greatly improved through the application of hybridized soft-computing techniques to detect thermal insulation failures when a building is in operation. The method proposed in this paper begins by considering local building and heating system regulations as well as the specific features of the climate zone as part of a three-step procedure. Firstly, the dynamic thermal performance of different variables is modelled, which relate to both the building and the climate zone. Secondly, Cooperative Maximum-Likelihood Hebbian Learning is used to extract their relevant features. Finally, neural projections and identification techniques are applied, in order to detect fluctuations in room temperatures and, in consequence, thermal insulation failures. Although a great deal of further research remains to be done in this field, the proposed system is expected to outperform conventional methods described in Spanish building codes that are used to calculate energetic profiles in domestic and residential buildings.