THE MODELLING AND ANALYSIS OF TEMPERATURE-DEPENDENT SILICON AND SILICON CARBIDE-BASED POWER DEVICE PERFORMANCE FOR HIGH-VOLTAGE APPLICATIONS USING MACHINE-LEARNING APPROACHES
DOI:
https://doi.org/10.55766/sujst-2024-02-e02681Keywords:
Silicon carbide, On-resistance, Temperature, Regression, Machine-learniAbstract
This study has revealed that On-State-Drain to Source resistance RDS(on) is one of the most significant variables influencing the performance of silicon and SiC-based power semiconductor devices in high-voltage applications. The low RDS(on) of power devices helps to drastically cut energy usage, enabling customers to create greener systems and products that lower CO2 emissions. In SiC MOSFETs, low RDS(on) enables significant system size and power consumption reductions in a range of applications, including traction inverters for electric vehicles and switching power supply. This work analyses the characteristics of silicon (Si) and Silicon Carbide (SiC)-based power semiconductor devices’ performance in terms of total gate charge, drain current, power dissipation, thermal resistance and switching speed. This work affirms that SiC-based power devices provide superior performance as compared to silicon power devices for the abovementioned parameters at the same voltage rating of 650 V. The study highlights the effects of RDS (on) on Si and SiC-based power devices and provides insight into the predictive modeling of the analyzed parameters at elevated temperatures, which is a novel contribution to this field of research. The models and formulations generated from this work help the designers select the power devices with the required RDS(on) and correlate their effects by estimating the values of other important parameters, mainly, continuous drain current and zero voltage drain current at elevated temperature, with the exception of the real-time device testing procedure. The models and formulations generated using regression in this work are further validated using Mean Absolute Percentage Error (MAPE) and R2 techniques. The outcome of this work is expected to be implemented in the current machine-learning technology for device testing purposes.
References
Abhijit D.P., (2001). Mosfet/IGBT Drivers Theory and Applications. https://www.ix Abhijit D.P., (2024). Mosfet/IGBT Drivers Theory and Applications.
Germany: IXYS Corporation. Available from: https://shorturl.asia/1T29u. Accessed date: 15 July 2023.
Balen, J., Inokon, A., Johnima, A., and Browndi, I. (2022). Predictive analytics challenges based on cloud computing for urban planning. International Journal of Information System Management System, 23(19).
Bau, D., Zhu, J.Y., Strobelt, H., Lapedriza, A., Zhou, B., and Torralba, A. (2020). Understanding the role of individual units in a deep neural network. Proceedings of the National Academy of Sciences of the United States of America, 117(48):30,071-30,078. https://doi.org/10.1073/pnas.1907375117
Casady, J.B., Agarwal, A.K., Seshadri, S., Siergiej, R.R., Rowland, L.B., MacMillan, M.F., Sheridan, D.C., Sanger, P.A., and Brandt C.D. (1998). 4H-SiC power devices for use in power electronic motor control. Solid-State Electronics, 42(12):2,165-2,176. https://doi.org/10.1016/S0038-1101(98)00212-3
Charbuty, B., and Abdulazeez, A. (2021). Classification Based on Decision Tree Algorithm for Machine Learning. Journal of Applied Science and Technology Trends, 2(1):20-28. https://doi.org/10.38094/jastt20165
Cheng X. (2018a). Overview of Recent Progress of Semiconductor Power Devices Based on Widebandgap Material. IOP Conference Series: Materials Science and Engineering, 439(2):1-7. https://doi.org/10.1088/1757-899X/439/2/022033
Cheng, X., Khomtchouk, B., Matloff, N., and Mohanty. P. (2018b). Polynomial Regression As an Alternative to Neural Nets. Cornell Univeristy, Available from: https://doi.org/10.48550/arXiv.1806.06850 Accessed date: 5 Jun, 2023.
Chicco, D., Warrens, M.J. and Jurman, G. (2021). The coefficient of determination R-squared is more effective than SMAPE, MAE, MAPE, MSE and RMSE in regression analysis evaluation. PeerJ Computer Science, 7(26):1-24. https://doi.org/10.7717/peerj-cs.623
Chukhrova, N., and Johannssen, A. (2019). Fuzzy regression analysis: Systematic review and bibliography. Applied Soft Computing Journal, 84:105708. https://doi.org/10.1016/j.asoc.2019.105708
Ghaderpour, E., Pagiatakis, S.D., and Hassan, Q.K. (2021). A survey on change detection and time series analysis with applications. Applied Sciences (Switzerland), 11(13):2-29. https://doi.org/10.3390/app11136141
Glaser J.S., Nasadoski J.J., Losee P.A., Kashyap A.S., Matocha K.S., Garrett J.L., and Stevanovic L.D. (2011). Direct comparison of silicon and silicon carbide power transistors in high-frequency hard-switched applications. 2011 Twenty-Sixth Annual IEEE Applied Power Electronics Conference and Exposition (APEC), 1,049-1,056. https://doi.org/10.1109/APEC.2011.5744724
Hren, A., Truntič, M., and Mihalič, F. (2023). A Survey on the State-of-the-Art and Future Trends of Multilevel of Inverters in BEVs., Electronics, 12(13):3-25. https://doi.org/10.3390/electronics12132993
Junta, U., Newiduom, L., Opuiyo, A., and Browndi, I. (2022). Predictive Analysis of Urban Planning for Through The Operation Of Artificial Cloud Network. International Journal of Environmental Science and Technology, 62(42).
Kurama, V. (2023). Regression in Machine Learning: What It Is and Examples of Different Models. Builtin. Available from: https://builtin.com/data-science/regression-machine-learning. Accessed date: 5 May, 2023.
Li, X., Jiang, J., Huang, A.Q., Guo, S. Deng, X., Zhang, B. She, X. (2017). A SiC Power MOSFET loss model suitable for high-frequency applications. IEEE Transactions on Industrial Electronics, 64(10):8,268-8,276. https://doi.org/10.1109/TIE.2017.2703910
Maulud D. and Abdulazeez A.M. (2020). A Review on Linear Regression Comprehensive in Machine Learning. Journal of Applied Science and Technology Trends, 1(4):140-147. https://doi.org/10.38094/jastt1457
Myttenaere, A.D., Golden, B., Grand, B.L., and Rossi, F. (2015). Mean Absolute Percentage Error for regression models. Neurocomputing, 192:38-48. https://doi.org/10.1016/j.neucom.2015.12.114
Norman R.D. and Harry S. (1998). Applied Regression Analysis Wiley Series in Probability and Statistics, 716. https://doi.org/10.1002/9781118625590
Rohm Semiconductor (2024) What is ON Resistance?. Germany: Rohm Semiconductor. Available from: https://www.rohm.com/electronics-basics/transistors/trwhat9. Accessed date. 23 Apr, 2024.
