Piezoelectric modulus prediction using machine learning and graph neural networks

Abstract

Piezoelectric materials are widely used in many industries and our daily life. However, discovering high-performance piezoelectric materials is much more challenging than other material properties (formation energy, band gap). Here, we propose a comprehensive study on designing and evaluating advanced machine learning models for predicting piezoelectric modulus from materials’ composition/structures. We train prediction models based on extensive feature engineering combined with machine learning models and automated feature learning based on deep graph neural networks. We also use it to predict the piezoelectric coefficients for 12,680 materials and report the top 20 potential high-performance piezoelectric materials.