Session: 11-03-01: Artificial Intelligence, Machine Learning and Data Science for Thermal Processes, Heat Transfer and Energy Systems

Paper Number: 145806

145806 - Accelerating the Prediction of Thermal Conductivity and Radiative Properties Through Machine Learning 

Lattice dynamics is a fundamental mechanism that is important for many applications, including thermal conductivity and radiative property, but experimental measurements or first principles calculations including three-phonon and four-phonon scattering are expensive or even unaffordable, preventing the large-scale high-throughput screening of materials. Machine learning approaches that can achieve similar accuracy have been a long-standing open question. Despite recent progress, machine learning models using structural information as descriptors fall short of experimental or first principles accuracy.

To overcome the computational barriers, we first present a machine learning approach that can predict phonon scattering rates and thermal conductivity with experimental and first principles accuracy. The success of our approach is enabled by mitigating computational challenges associated with the high skewness of phonon scattering rates and their complex contributions to the total thermal resistance. Transfer learning between different orders of phonon scattering can further improve the model performance. Our surrogates offer up to two orders of magnitude acceleration compared to first principles calculations and would enable large-scale thermal transport informatics.

Inspired by the first machine learning model, we further developed a statistical sampling method to accelerate the phonon scattering. This is done by estimating scattering rates from a small sample of scattering processes using maximum likelihood estimation. The calculation of scattering rates and associated thermal conductivity and radiative properties are dramatically accelerated by three to four orders of magnitude. We also derive the confidence interval of our estimation, which is useful for choosing a proper sample size. The accuracy and efficiency of our approach make it ideal for the high-throughput screening of materials for thermal and optical applications. The work is incorporated as a new feature within the FourPhonon open-source package.

Besides acceleration, our model also enables us to use an unprecedented q-mesh (discretized grid in the reciprocal space) to study the lattice dynamics of materials, which was not possible before due to the high computational cost. By using a 32 × 32 × 32 q-mesh for calculating four-phonon scattering of silicon (Si), we achieve a converged thermal conductivity value that agrees much better with experiments. Besides, we did a sophisticated theoretical study of the lattice thermal conductivity in bulk hexagonal boron nitride (h-BN). We include three-phonon scattering, four-phonon scattering, and phonon renormalization.  Our predicted thermal conductivity is 363 W/(m·K) and 4.88 W/(m·K) for the in-plane and out-of-plane directions at room temperature, respectively. The results show good agreement with experimental results and reveal the contribution of both four-phonon scattering and phonon renormalization.

Presenting Author: Ziqi Guo Purdue University

Presenting Author Biography: Ziqi Guo, a third-year PhD candidate at Purdue University. His research interest lies at the intersection of physics simulation, energy transport and machine learning.

Authors:

Ziqi Guo Purdue University
Zherui Han Purdue University
Dudong Feng Purdue University
Prabudhya Roy Chowdhury Purdue University
Abdulaziz Alkandari Purdue University
Krutarth Khot Purdue University
Guang Lin Purdue University
Xiulin Ruan Purdue University