Zhou Hao, Wang Xiang, Wang Shulong, Liu Chenyu, Chen Dongliang, Li Jiarui, Ma Lan, Zhang Guohao
School of Microelectronics, Xidian University, Xi'an 710071, China.
School of Information Engineering, Guangdong University of Technology, Guangzhou 510006, China.
Micromachines (Basel). 2025 May 15;16(5):583. doi: 10.3390/mi16050583.
This work employs a deep learning method to develop a high-precision model for predicting the breakdown voltage () and forward characteristics of silicon carbide Schottky barrier diodes (SiC SBDs). The model significantly reduces the testing costs associated with destructive experiments, such as breakdown voltage testing. Although the model requires a certain amount of time to establish itself, it supports linear variations in related variables once developed. A predicted model for with an accuracy of up to 99% was successfully developed using 600 sets of input data after 200 epochs of training. After training for 1000 epochs, the deep learning-based model could predict not only point values like but also curves, such as forward characteristics, with a mean squared error (MSE) of less than 10. Our research shows the applicability and high efficiency of introducing deep learning into device characteristic prediction.
这项工作采用深度学习方法来开发一个高精度模型，用于预测碳化硅肖特基势垒二极管（SiC SBDs）的击穿电压（）和正向特性。该模型显著降低了与破坏性实验相关的测试成本，如击穿电压测试。虽然该模型需要一定时间来建立，但一旦开发完成，它支持相关变量的线性变化。经过200个训练轮次后，使用600组输入数据成功开发出了一个预测准确率高达99%的模型。经过1000个训练轮次后，基于深度学习的模型不仅可以预测像这样的点值，还可以预测曲线，如正向特性，其均方误差（MSE）小于10。我们的研究表明了将深度学习引入器件特性预测的适用性和高效率。
