Research on the use of UAVs for detecting hidden defects in building envelopes and assessing their impact on heat loss in urban environments

Abstract

The study demonstrates the significant potential of unmanned aerial systems equipped with high-resolution thermal imaging modules and multispectral cameras for detecting hidden defects in the envelopes of residential and public buildings, as well as for quantifying their impact on increased heat loss in dense urban environments. The analysis includes characteristic thermal anomalies, including violations of thermal insulation in the interpanel joints of large-panel buildings, the processes of masonry moisture in post-war brick buildings, cold bridges in the areas of balcony slabs and window slopes of monolithic-brick buildings, and their distribution by type of structure, localization on facades, and intensity of manifestation. Special attention is paid to temperature gradients between abnormal and defect-free zones, the areas of defect distribution, the calculated values of heat flux density that exceed the standard values by 2-3 times, and the integral indicators of additional heat loss, which reach an average of 18,06 kWh/m² per year and account for up to 18,72% of the total loss in buildings of the first mass series. A strong correlation has been identified between the age of buildings, the orientation of their facades (with a predominance of defects on the northern and northeastern sides), and the amount of heat leakage. Regression models allow for the consideration of a range of factors, including the quality of major repairs. A comparative analysis confirms the superiority of aerial photography using quadcopters such as DJI Matrice 300 RTK and Mavic 3 Thermal over traditional ground-based and contact methods in terms of completeness of detection (94,7%), efficiency, cost, and safety of work in high-rise buildings, limited visibility, and the presence of thermal interference. The obtained data provide a basis for developing methodological approaches for converting temperature fields into energy indicators, accounting for meteorological influences and solar radiation, and forming integral criteria for the energy state of buildings. The results are highly practical for optimizing capital repair programs, prioritizing housing stock facilities, economic assessment of damage from defects, and scaling remote monitoring at the urban district level. The prospects of integration with digital information models of buildings, geoinformation systems, and machine learning algorithms for automatic classification of anomalies open up opportunities for transition to automated energy modeling of urban development, forecasting of energy resource needs, and formation of a regulatory framework for regulating such surveys in accordance with the requirements of air legislation and thermal protection standards.