Prediction of the energy consumption of refrigerators during use

© 2018 Dr. Lloyd Wesley HarringtonRefrigerators are a common appliance in developed and emerging economies around the world. These appliances consume significant energy globally and improving their efficiency is an important aspect of future climate change mitigation. While refrigerating appliances are ubiquitous, the factors that influence their energy consumption have been poorly understood. Several important knowledge gaps were identified at the beginning of this research: information on typical temperatures in Australian homes, information on how room air temperature impacts on the energy consumption of refrigerating appliances, information on defrosting performance in the home and quantification of the energy impact of user interactions during use. A method was developed to disaggregate the energy consumption measured in homes into three important components: energy consumption driven by room air temperature, defrosting energy and energy induced by user interactions. Field measurements of energy consumption and room air temperature were then collected in around 273 households in a diverse range of climates from Melbourne and Gippsland in the south to Cairns in the north over the period 2011 to 2015. Measurements were mostly for longer periods, averaging seven months. Energy was separated into components using the method developed to give a better understanding the drivers of energy consumption. Indoor temperature data was further analysed to develop a model of indoor temperature based on outdoor temperature. Further analysis allowed the development of generalised curve of energy consumption in response to changes in room air temperature for the most common types of refrigerating appliances. This was based on extensive laboratory data and was confirmed with field data. Detailed analysis of the defrosting behaviour of monitored appliances allowed the development of a deeper understanding of defrost energy impacts during use. Measurements of defrost energy tend to be underestimated in a laboratory setting. Defrost intervals in the field can be accurately predicted for run-time defrost controllers but are more difficult to estimate from laboratory data for variable defrost controllers. The energy impact of user interactions has been an area of interest for researchers for many years. Most previous investigations have examined the energy impact of arbitrary user interactions in a test laboratory. Only a couple of studies attempted to quantify the impact of user interactions in the field with little success. Using the data on user interactions measured in homes, a statistical model of user interactions has been developed. This allows the energy impacts of user interactions to be estimated for different sizes of refrigerating appliances, different household sizes and different operating temperatures. This is the first model of its type. Finally, a generalised model to allow the energy consumption of household refrigerators to be predicted during use has been developed. This allows the energy consumption of different appliances to be estimated and compared in identical usage and operating conditions. These same appliances can also be compared across a wide range of operating conditions such as climate and demographics. This provides a more accurate tool for analysts and policy makers to better understand the energy consumption of refrigerating appliances during use