Adaptive Hybrid Approaches to Thermal Modeling of Building.

Thermal modeling is necessary if we want to control the thermal environment in a space within a building. Heating, Ventilation and Air-Conditioning (HVAC) equipment are usually installed in a building based on the empirical formula developed by the manufacturers considering the worst case scenarios of heat loss/gain through the walls, the windows and from other sources like body heat. In order to provide thermal comfort in building spaces and at the same time save energy, accurate modeling is required for adequate provisions and control of HVAC equipment. Better accuracy of a building's thermal model usually increases the complexity and thereby the demand for higher processing capability of the computer-based controller. This can increase the cost and the payback period and often become deterrent to implementation in large real-world buildings. This work is an attempt to study the existing modeling techniques and offer a low cost solution with an intuitive and simple modeling based on first principles validated by empirical data. The results of its implementation in a real-world building and the insights gained are also presented.