Dynamic modeling and simulation of frost and defrost conditions for refrigeration systems

Heating and cooling systems are widely used throughout the world in an array of applications as a means of providing thermal energy transfer. Recent decades have seen economic and environmental awareness come to a forefront and innovation resultingly being pushed toward a standard of greater performance and efficiency. With their nearly universal presence, it therefore comes as no surprise that heating and cooling systems have also been subject to such guidelines. One of the biggest contributors to performance degradation and inefficiency in air conditioning and refrigeration applications today is the accumulation of frost on heat exchanger surfaces. The presence of frost often incurs a decrease in the amount of cooling capacity achievable, resulting in the need for defrost cycles. Much work has been done to date to establish strategies aimed at improving defrost decision making and allow for improved operational efficiency. Nonetheless, some of the most common methods still permit defrost to run when unnecessary. Modeling has therefore been used to provide greater insight and prediction capability of these processes. However, current methods lack the qualities necessary to allow for the implementation of a model-based defrost strategy aimed at further improving upon existing defrost techniques. This thesis makes contributions toward improving refrigeration operational efficiency through the development of a model for dynamically simulating refrigeration systems operating under both frost and defrost conditions. Modeling provides the tools necessary to form more accurate defrost decisions through a greater understanding of frosting behavior and its effect on system performance. Such insight is achieved by means of developed frost growth and melt models discussed in the first part of this thesis. Validation results demonstrate good prediction capability of both frost thickness as a function of time and estimated time to defrost. The second portion of this thesis addresses implementation of frost growth/melt modeling into an existing air conditioning and refrigeration simulation library. Coupling of frost and evaporator dynamics under operation in cooling and defrost could therefore be represented; something existing models do not allow for. Simulation results indicate that identified trends in system degradation are consistent with those commonly seen in application. Agreement demonstrates the promise shown by developed models and the potential utility of such a tool in aiding defrost decision making and further mitigating key inefficiencies commonly plaguing the refrigeration industry today