一种基于深度学习和坩埚下降法的高阻型氧化镓制备方法
By combining deep learning and the crucible descent method with a neural network model, the problem of high-resistivity gallium oxide that cannot be prepared with a predetermined resistivity in existing technologies has been solved, and the precise preparation of high-resistivity gallium oxide single crystals has been achieved.
CN112863617BActive Publication Date: 2026-07-17HANGZHOU FUJIA GALLIUM TECH CO LTD
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- HANGZHOU FUJIA GALLIUM TECH CO LTD
- Filing Date
- 2020-12-31
- Publication Date
- 2026-07-17
AI Technical Summary
Technical Problem
Existing technologies cannot effectively prepare high-resistivity gallium oxide with a predetermined resistivity.
Method used
By employing deep learning and the crucible-lowering method, seed crystal, environmental, and control data are acquired and preprocessed. A trained neural network model is used to predict the resistivity of high-resistivity gallium oxide, and high-resistivity gallium oxide single crystals are prepared by combining this with the crucible-lowering method.
Benefits of technology
Accurate prediction and preparation of the resistivity of high-resistivity gallium oxide single crystals have been achieved, ensuring the acquisition of high-resistivity gallium oxide with a predetermined resistivity.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN112863617B_ABST
Abstract
本申请涉及基于深度学习和坩埚下降法的高阻型氧化镓制备方法，预测方法包括：获取高阻型氧化镓单晶的制备数据；制备数据包括：籽晶数据、环境数据、控制数据以及原料数据；原料数据包括：掺杂类型数据及掺杂浓度；对制备数据进行预处理；将预处理的制备数据输入训练好的神经网络模型，得到高阻型氧化镓单晶对应的预测性质数据；所述预测性质数据包括：预测电阻率。先将制备数据进行预处理，得到预处理的制备数据，将预处理的制备数据输入训练好的神经网络模型，得到高阻型氧化镓单晶对应的预测性质数据，通过调整制备数据，可以得到预定电阻率的高阻型氧化镓。
Need to check novelty before this filing date? Find Prior Art