A single-event effect prediction method for FDSOI devices based on deep learning

The single-particle effect prediction model of FDSOI devices is constructed based on deep learning, which solves the problems of large calculations, long time and non-convergence in traditional methods, and achieves fast and accurate single-particle effect prediction.

CN114925587BActive Publication Date: 2025-03-18XIDIAN UNIV
0 Cites 0 Cited by

Patent Information

Application Number
CN202210493522.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-05-07
Publication Date
2025-03-18
Estimated Expiration
2042-05-07

AI Technical Summary

Technical Problem

The single-particle effect prediction method of traditional FDSOI devices is large in calculation, time-consuming and does not converge, making it difficult to meet research needs.

Method used

Using a deep learning-based method, the influence parameters and single-particle effect characterization parameters of single-particle incident FDSOI devices are obtained through TCAD simulation software, and the deep learning network prediction model is constructed, and the loss function optimization model is used to achieve fast and accurate prediction.

Benefits of technology

The efficiency and accuracy of single-particle effect prediction of FDSOI devices are improved, the time-consuming and non-convergence problems in traditional methods are solved, and the rapid and accurate prediction results are achieved.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

A method for predicting single particle effects of FDSOI devices based on deep learning, using TCAD simulation software to obtain multiple groups of influencing parameters and corresponding single particle effect characterization parameters when single particles are incident on FDSOI devices as sample sets; the influencing parameters when single particles are incident on FDSOI devices include linear transmission energy value, particle incident position, particle incident angle and drain bias voltage, and the single particle effect characterization parameters include drain transient current pulse, drain transient current peak and total collected charge; the sample set is divided into training set, cross-validation set and test set, a deep learning network prediction model is constructed and trained and verified, and the model that meets the requirements can be used for single particle effect characterization parameters under different heavy ion incidence conditions. The present invention solves the problems of large amount of calculation, long time consumption and non-convergence in the prediction of single particle effects of FDSOI devices in traditional research.
Need to check novelty before this filing date? Find Prior Art