Forward Selection Methodology for Phase Change Material Composite Optimization

Composite thermal energy storage (TES) materials containing a phase change material (PCM) component have been shown to act as highly efficient cooling components for the management of transient thermal loads. These material systems have the advantage of combining favorable capacitive properties of PCMs with thermally conductive elements allowing for simultaneous transportation and absorption of thermal energy. However, there is a lack of understanding of the role of component composition and distribution on the overall performance of these composite systems. This leads to systems that lack design intentionality and rely on post-manufacturing testing. Therefore, in this body of work we will lay the groundwork of developing a methodology for optimal system design.To provide efficient analysis of these systems, a finite difference analysis (FDA) model is developed. For the purposes of this study the total amount of thermal energy absorbed at a given time of interest is used as the performance metric calculated by the FDA for a given constant temperature thermal load. For the same thermal load, the FDA is then iterated with varying material composition fractions as a function of radial distance. Using the resulting dataset, a Gaussian Process Regressor is used to fit a response surface to the data. Different acquisition functions are then be tested to identify new points of the composite design space that correspond to high performing PCM thermal management systems. From there, different initial dataset sizes and numbers of additional points are tested to determine the most efficient methodology of obtaining a system of optimum performance.