Using U-Net convolutional neural network to model pixel-based electrostatic potential distributions in GaN power MIS-HEMTs

• Published in: Scientific reports Vol. 14; no. 1; p. 8151
• Main Authors: Chen, Bang-Ren, Hsiao, Yu-Sheng, Lin, Wei-Cheng, Lee, Wen-Jay, Chen, Nan-Yow, Wu, Tian-Li
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 08-04-2024; Nature Publishing Group; Nature Portfolio
• Subjects: 639/166; 639/925; Design; Electrostatic potential modeling; Electrostatic properties; GaN HEMT; High performance computing; Humanities and Social Sciences; Machine learning; multidisciplinary; Neural networks; Science; Science (multidisciplinary); Semiconductor research; Simulation; Transistors; U-Net; Electrostatic potential modeling; U-Net; GaN HEMT; Machine learning
• ISSN: 2045-2322; 2045-2322
• DOI: 10.1038/s41598-024-58112-9

This study demonstrates a novel use of the U-Net convolutional neural network (CNN) for modeling pixel-based electrostatic potential distributions in GaN metal–insulator-semiconductor high-electron mobility transistors (MIS-HEMTs) with various gate and source field plate designs and drain voltages. The pixel-based images of the potential distribution are successfully modeled from the developed U-Net CNN with an error of less than 1% error relative to a TCAD simulated reference of a 500-V electrostatic potential distribution in the AlGaN/GaN interface. Furthermore, the modeling time of potential distributions by U-Net takes about 80 ms. Therefore, the U-Net CNN is a promising approach to efficiently model the pixel-based distributions characteristics in GaN power devices.