Reconsidering Uncertainty from Frequency Domain Thermoreflectance Measurement and Novel Data Analysis by Deep Learning
Wenqing Shen, Diego Vaca, Satish Kumar
Abstract
Frequency-domain thermoreflectance (FDTR) is a popular technique to investigate thermal properties of bulk and thin film materials. The FDTR data analysis involves fitting experimental data to a theoretical model whose accuracy may be affected by improper fitting approach and by convergence to local minima. This work proposes a novel data analysis approach using deep learning techniques. The developed deep learning model for FDTR (DL-FDTR) can accurately predict thermal conductivity, volumetric heat capacity and thermal boundary conductance with mean error below 5% for bulk samples coated with Au. DL-FDTR predictions can serve as an initial guess to the traditional fitting algorithms and can efficiently avoid local minima with regular fitting options, therefore improving the accuracy of data fitting and uncertainty evaluation.