一种基于深度学习和提拉法的导电型氧化镓的质量预测方法、制备方法及系统
By combining deep learning and the Czochralski method, and using a neural network model to predict and adjust the preparation parameters, the problem of parameter control in the preparation of conductive gallium oxide single crystals by the Czochralski method was solved, and stable preparation with a predetermined carrier concentration was achieved, thus improving the repeatability and consistency of the preparation.
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
In the existing Czochralski method for preparing conductive gallium oxide single crystals, the preparation parameters are difficult to control, the repeatability is poor, and it is impossible to stably obtain conductive gallium oxide single crystals with a predetermined carrier concentration.
A method based on deep learning and the Czochralski technique is adopted. By acquiring and preprocessing the preparation data, a trained neural network model is used to predict the quality of conductive gallium oxide single crystals, including seed crystal data, environmental data, and control data. Preparation parameters are then adjusted to achieve a predetermined carrier concentration.
This method enables accurate prediction and stable control of the quality of conductive gallium oxide single crystals, allowing the fabrication of conductive gallium oxide single crystals with predetermined carrier concentrations, thus improving the repeatability and consistency of the fabrication process.
Smart Images
Figure CN112863620B_ABST