Normalization of Cooling Demand in Buildings, development, and evaluation of methods

Abstract: Because of the increasing demand for indoor thermal comfort and the better insulation performance of building envelopes, cooling demand is an increasingly important part of building energy consumption despite the cold climate in Stockholm. Finding a way to normalize cooling demand to climatic conditions will help better energy audits to improve the building performance to achieve energy-saving.  This thesis researched commercial building cooling demands in Stockholm from 2015 to 2020 and a normal year. It primarily includes work with the simulation software IDA-ICE where some building models are used to describe the variation of comfort cooling demands between different years and how the energy consumption distribute between the building systems. The following simple data processing and analysis works are carried out in Excel. The regression and correlation work for getting the correction factors are carried out in MATLAB with input independent variables and cooling demand variables. The independent variables include dry-bulb temperature, relative humidity, and solar irradiation. At the same time, building HVAC systems, including different types of air handling units, room coolers, and other systems also influence on cooling demands. Compared with the existing normalization methods, the method proposed in this thesis work, proved to be more accurate than the cooling degree days methods, which are only based on dry-bulb temperature, e.g., the Power Signature method. It matches the SMHI Kyl index well, especially in the summertime. It can also match the Solar Irradiation Factor well, instead of the factor taking both enthalpy and solar irradiation into account. In essence, most cooling demand is contributed by solar heat gains, although the dry-bulb temperature variation mainly reflects its result. The result of implementing the correction factor method in real buildings shows that it can well normalize the summer cooling demand, while in wintertime, there are many deviations when the based load dominates. It is difficult to quantitively measure and describe the weather-independent influences, which always bring large deviation on the base load. Only by communicating with the building operator in time can we know the specific pattern of the buildings to make better normalizations. The measurement equipment performance and accuracy are important factors that require more attention, sometimes bringing unpredictable significant errors that are always ignored.