Energy wasted in residential and commercial buildings from leaky windows and doors, open refrigerator doors, and unnecessarily switched on small to medium everyday appliances account for a large fraction of the energy budget of a country. Energy audit systems provide a remedy to this problem where certain types of energy wastage due to leakages and poor insulation can be analyzed and addressed. However, there is a critical need for low cost systems that can continuously monitor such wastage and provide recommendations to users on how to mitigate such wastage. To this end, this project developed a low-cost thermal imaging system, IRLeak that can detect open and drafty windows, open refrigerator and microwave doors, and computers left accidentally on. Moreover, proper thermal insulation yields optimum energy expenses in buildings by maintaining necessary heat gain or loss through the built envelope. However, improper thermal insulation causes significant energy wastage along with infusing various damages on indoor and outdoor walls of the buildings, for example, damp areas, black stains, cracks, paint bubbles etc. Therefore, it is important to inspect the temperature variations in different areas of the built environments in regular basis. The project designed a method for identifying temperature variance in building thermal images based on Symbolic Aggregated Approximation. In addition, air leakages pose a major problem in both residential and commercial buildings. They increase the utility bill and result in excessive usage of Heating Ventilation and Air Conditioning (HVAC) systems, which impacts the environment and causes discomfort to residents. Repairing air leakages in a building is an expensive and time intensive task. Even detecting the leakages can require extensive professional testing. The project designed a method to identify the leaky homes provided their energy consumption data is accessible from residential smart meters. It employed a Non-Intrusive Load Monitoring (NILM) technique to disaggregate the HVAC data from total power consumption for several homes. It developed a recurrent neural network and a denoising autoencoder based approach to identify the "ON" and "OFF" cycles of the HVACs and their overall usages. The project categorized the typical HVAC consumption of about 200 homes and any probable insulation and leakage problems using the Air Changes per Hour at 50 Pa (ACH50) metric. However, modeling buildings' heat dynamics is a complex process which depends on various factors including weather, building thermal capacity, insulation preservation, and residents' behavior. Gray-box models offer an explanation of those dynamics, as expressed in a few parameters specific to built environments that can provide compelling insights into the characteristics of building artifacts. The project presented a systematic study of Bayesian approaches to modeling buildings' parameters, and hence their thermal characteristics. It designed a Bayesian state-space model that can adapt and incorporate buildings' thermal equations and postulated a generalized solution that can easily adapt prior knowledge regarding the parameters. The project had educated undergraduate students in the classroom about low-cost infrared camera-based system and data analytics-based approaches to detect energy usage, air leakages and insulation problems in their residential environments.

Last Modified: 11/04/2020
Modified by: Nirmalya Roy