Algorithm prediction of lifetime under high-power microwave based on T-type field plate HEMTs
Source abstract
Abstract
Abstract To predict the reliability of T-type field plate HEMTs under high-power microwave (HPM) stress, this study constructed a deep learning model and adopted TCAD technology to explore HPM’s influence on the devices. The model takes pulse width, power and duty cycle as input features, with its prediction relative error below 15% and most below 5%. Comparative experiments with five traditional machine learning (ML) algorithms (SVM, decision tree, KNN, ridge regression, linear regression) show the deep learning algorithm achieves the minimum average relative error of 6.654% and optimal performance. This data-driven model allows researchers without semiconductor expertise to quickly obtain device lifetime data under arbitrary conditions, and can reflect individual device HPM reliability for device design, boasting extremely broad application prospects in device lifetime prediction.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Advancements in Semiconductor Devices and Circuit Design · Radio Frequency Integrated Circuit Design · Advanced Data and IoT Technologies
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
References (26)
Show source-provided references
1. Challenges and opportunities for high-power and high-frequency AlGaN/GaN high-electron-mobility transistor (HEMT) applications: a review10.3390/mi13122133 ↗
2. AlGaN/GaN HEMTs-an overview of device operation and applications10.1109/JPROC.2002.1021567 ↗
3. A survey of wide bandgap power semiconductor devices10.1109/TPEL.2013.2268900 ↗
4. The modified Manson–Coffin curve method to estimate fatigue lifetime under complex constant and variable amplitude multiaxial fatigue loading10.1016/j.ijfatigue.2015.10.005 ↗
5. Probabilistic framework for strain-based fatigue life prediction and uncertainty quantification using interpretable machine learning10.1016/j.ijfatigue.2024.108647 ↗
6. Cumulative fatigue damage and life prediction theories: a survey of the state of the art for homogeneous materials10.1016/S0142-1123(97)00081-9 ↗
7. Deep learning for visual understanding: a review10.1016/j.neucom.2015.09.116 ↗
8. Deep learning in agriculture: a survey10.1016/j.compag.2018.02.016 ↗
9. Deep learning10.1038/nature14539 ↗
10. In-Situ early anomaly detection and remaining useful lifetime prediction for high-power white LEDs with distance and entropy-based long short-term memory recurrent neural networks10.1016/j.eswa.2023.121832 ↗
11. High throughput substrate screening for interfacial thermal management of β-Ga2O3 by deep convolutional neural network10.1063/5.0201699 ↗
12. Lifetime prediction of GaN power devices based on COMSOL simulations and long short-term memory (LSTM) networks10.3390/electronics14193883 ↗
13. Prediction of FinFET current-voltage and capacitance-voltage curves using machine learning with autoencoder10.1109/LED.2020.3045064 ↗
14. Review on machine learning methods for remaining useful lifetime prediction of lithium-ion batteries10.1109/GECOST55694.2022.10010569 ↗
15. Trap-assisted degradation mechanisms in E-mode p-GaN power HEMT: a review10.1016/j.microrel.2022.114800 ↗
16. Reliability issues and degradation mechanisms of p-GaN gated E-mode AlGaN/GaN power HEMTs: a critical review10.1109/ACCESS.2025.3569288 ↗
17. Degradation prediction of GaN HEMTs under hot-electron stress based on ML-TCAD approach10.3390/electronics11213582 ↗
18. Gate switching lifetime of P-Gate GaN HEMT: circuit characterization and generalized model10.1109/TPEL.2024.3443709 ↗
19. Physics-constrained neural network-based compact modeling of displacement damage for AlGaN/GaN HEMTs10.1109/TNS.2025.3611989 ↗
20. Lifetime prediction and analysis of AlGaN/GaN HEMT devices under temperature stress10.1016/j.mejo.2022.105370 ↗
21. Reliability of GaN high-electron-mobility transistors: state of the art and perspectives10.1109/TDMR.2008.923743 ↗
22. Mechanism analysis and protection design of GaN HEMTs induced by high-power microwave pulse10.1088/1361-6528/ae3c1b ↗
23. RF power degradation of GaN high electron mobility transistors10.1109/IEDM.2010.5703397 ↗
24. Investigation of trapping and hot-electron effects in GaN HEMTs by means of a combined electrooptical method10.1109/TED.2011.2160547 ↗
25. A study of GaN HEMTs current collapse impacts on Doherty multistage PA linearity10.1109/CSICS.2015.7314483 ↗
26. Degradation behavior and trap analysis based on low-frequency noise of AlGaN/GaN HEMTs subjected to radio frequency overdrive stress10.1109/TED.2020.3040698 ↗