Artificial Intelligence-Driven Thermal Management in Electric Vehicle Traction Inverters: Cooling System Optimization and Real-Time Temperature Prediction
Open accessThe DOI matches its Crossref record; 35 of the 42 references it cites were found in a registry or an index.
Abstract
The accelerated adoption of electric vehicles (EVs) has intensified technical demands on traction inverters, particularly regarding power density, energy efficiency, and thermal management. Silicon Carbide (SiC) MOSFET modules, with their superior switching performance and thermal characteristics, offer significant advantages but introduce new thermal challenges due to their increased power density and compact packaging. The primary objective of this study is to enhance real-time thermal control and prediction in SiC-based dual traction inverters through the design and experimental validation of a high-performance cooling system integrated into a modular electric drive platform. Multiple cold plate configurations were evaluated using computational fluid dynamics (CFD) simulations, focusing on junction temperature, thermal uniformity, and pressure drop. To enable real-time thermal predictions, a reduced-order model (ROM) was derived from CFD data and combined with an artificial neural network (ANN) trained to estimate junction temperatures dynamically. The ANN was integrated into the inverter control system and supported by real-time sensor data, enabling continuous dynamic thermal monitoring during operation. The proposed approach offers a scalable and intelligent thermal management solution, supporting predictive maintenance and enhancing the reliability and efficiency of next-generation EV powertrains.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Citations
Cites
He, Z., Yan, Y., & Zhang, Z. (2021) · Energy · doi:10.1016/j.energy.2020.119223 · Cited by 869 · cited 1×
“These temperature-driven degradations are significant, with thermal factors accounting for approximately 55% of all power electronics failures [14], [15], [16], [17], [18].”
Semiconductor Power Devices: Physics, Characteristics, Reliability
Lutz, J., Schlangenotto, H., Scheuermann, U., & De Doncker, R. (2011) · Springer eBooks · doi:10.1007/978-3-642-11125-9 · Cited by 706
Inductors and Transformers for Power Electronics
Valchev, V. C., & Van den Bossche, A. (2018) · CRC Press eBooks · doi:10.1201/9781420027280 · Cited by 562 · cited 1×
“Assuming resistive conduction losses [38]: The total power loss of the MOSFET is the sum of the conduction and switching losses, given by The maximum allowable power dissipation is limited by the system’s thermal resistance, cooling capacity, and the acceptable temperature rise [39].”
Present Status and Future Trends in Electric Vehicle Propulsion Technologies
Rajashekara, K. (2013) · IEEE Journal of Emerging and Selected Topics in Power Electronics · doi:10.1109/jestpe.2013.2259614 · Cited by 446
State of the art of high temperature power electronics
Buttay, C., Planson, D., Allard, B., Bergogne, D., Bevilacqua, P., Joubert, C., Lazar, M., Martin, C., Morel, H., Tournier, D., et al. (2011) · Materials Science and Engineering: B · doi:10.1016/j.mseb.2010.10.003 · Cited by 432 · cited 1×
“Although the SiC devices are theoretically capable of operating at junction temperatures exceeding 400 C, practical constraints related to packaging reliability currently limit commercial operation to approximately 175 C [12].”
Silicon carbide power MOSFETs: Breakthrough performance from 900 V up to 15 kV
Palmour, J. W., Cheng, L., Pala, V., Brunt, E. V., Lichtenwalner, D. J., Wang, G.-Y., Richmond, J., O'Loughlin, M., Ryu, S., Allen, S. T., et al. (2014) · 2014 IEEE 26th International Symposium on Power Semiconductor Devices & IC's (ISPSD) · doi:10.1109/ispsd.2014.6855980 · Cited by 300 · cited 1×
“Their appeal lies in several advantageous characteristics, including reduced conduction losses, lower on-resistance, faster switching speeds, higher breakdown voltage, and superior high-temperature operation [7], [8], [9], [10].”
Air cooled heat sink geometries subjected to forced flow: A critical review
Khattak, Z., & Ali, H. M. (2018) · International Journal of Heat and Mass Transfer · doi:10.1016/j.ijheatmasstransfer.2018.08.048 · Cited by 289 · cited 1×
“These temperature-driven degradations are significant, with thermal factors accounting for approximately 55% of all power electronics failures [14], [15], [16], [17], [18].”
Reliability and performance limitations in SiC power devices
Singh, R. (2006) · Microelectronics Reliability · doi:10.1016/j.microrel.2005.10.013 · Cited by 284
Reliability of Power Electronic Systems for EV/HEV Applications
Blaabjerg, F., Wang, H., Vernica, I., Liu, B., & Davari, P. (2021) · Proceedings of the IEEE · doi:10.1109/jproc.2020.3031041 · Cited by 249
Li, Z., Luo, H., Jiang, Y., Liu, H., Xu, L., Cao, K., Wu, H., Gao, P., & Liu, H. (2024) · Applied Thermal Engineering · doi:10.1016/j.applthermaleng.2024.123612 · Cited by 243 · cited 1×
“These temperature-driven degradations are significant, with thermal factors accounting for approximately 55% of all power electronics failures [14], [15], [16], [17], [18].”
The Past, Present, and Future of Power Electronics Integration Technology in Motor Drives
University of Wisconsin-Madison, & Jahns, T. M. (2017) · CPSS Transactions on Power Electronics and Applications · doi:10.24295/cpsstpea.2017.00019 · Cited by 205
Shi, B., Ramones, A. I., Liu, Y., Wang, H., Li, Y., Pischinger, S., & Andert, J. (2023) · IET Power Electronics · doi:10.1049/pel2.12524 · Cited by 184 · cited 1×
“Their appeal lies in several advantageous characteristics, including reduced conduction losses, lower on-resistance, faster switching speeds, higher breakdown voltage, and superior high-temperature operation [7], [8], [9], [10].”
Rahman, S. M. I., Moghassemi, A., Arsalan, A., Timilsina, L., Chamarthi, P. K., Papari, B., Özkan, G., & Edrington, C. S. (2024) · IEEE Access · doi:10.1109/access.2024.3385429 · Cited by 168 · cited 1×
“Research into alternative thermal management technologies focuses on enhancing cooling efficiency with more affordable options, while also reducing production and operational / maintenance costs, all without compromising the performance of the power electronic system [28].”
Wide-Bandgap-Based Power Devices: Reshaping the power electronics landscape
Bindra, A. (2015) · IEEE Power Electronics Magazine · doi:10.1109/mpel.2014.2382195 · Cited by 159 · cited 1×
“…in power electronics, especially the integration of wide-bandgap (WBG) semiconductor materials, have demonstrated considerable potential in addressing this challenge by enabling higher switching frequencies, improved thermal characteristics, and greater overall system efficiency [4], [5], [6].”
Hosseinabadi, F., Chakraborty, S., Bhoi, S. K., Prochart, G., Hrvanovic, D., & Hegazy, O. (2024) · IEEE Open Journal of Power Electronics · doi:10.1109/ojpel.2024.3379294 · Cited by 139 · cited 2×
“Junction temperature is a critical factor affecting long-term reliability, as elevated temperatures accelerate failure mechanisms such as solder fatigue, material diffusion, and bond degradation [13].”
Self-citationImmersion liquid cooling for electronics: Materials, systems, applications and prospects
Zhang, C.-B., Wang, H., Huang, Y.-P., Zhang, L., & Chen, Y.-P. (2025) · Renewable & Sustainable Energy Reviews · doi:10.1016/j.rser.2024.114989 · Cited by 119 · cited 1×
“Each criterion was scored on a 1–5 scale, with 5 indicating the most favorable outcome, using supplementary data from relevant literatures [34] and [35].”
Łach, Ł., & Svyetlichnyy, D. S. (2025) · Energies · doi:10.3390/en18051302 · Cited by 115 · cited 1×
“AI-enhanced models can enable rapid estimation of junction temperatures and system-level thermal behavior, providing a powerful tool for both design and control [20], [21].”
Asim, M., Baig, T., Siddiqui, F. R., Khan, S., Khan, S. A., Babar, H., Said, Z., Zhao, J., & Abidi, I. H. (2025) · Journal of Energy Storage · doi:10.1016/j.est.2025.115344 · Cited by 112 · cited 1×
“High Cooling Efficiency: An efficient cooling sys-tem minimizes energy losses, prevents overheating, and enhances system performance by maintaining optimal operational conditions, thereby maximizing overall sys-tem efficiency [27].”
Artificial Intelligence-Aided Thermal Model Considering Cross-Coupling Effects
Zhang, Y., Wang, Z., Wang, H., & Blaabjerg, F. (2020) · IEEE Transactions on Power Electronics · doi:10.1109/tpel.2020.2980240 · Cited by 98 · cited 1×
“Zhang et al. [24] proposed an AI-aided thermal model for power electronic systems that explicitly accounts for thermal cross-coupling (TCC) effects.”
Li, A., Yuen, A. C. Y., Wang, W., Chen, T. B. Y., Lai, C. S., Yang, W., Wu, W., Chan, Q. N., Kook, S., & Yeoh, G. H. (2022) · Batteries · doi:10.3390/batteries8070069 · Cited by 88 · cited 1×
“Li et al. [22] proposed a novel approach to optimize battery thermal management systems (BTMS) by integrating a three-dimensional electro-thermal model with CFD and ANN.”
Bailey, C. (2008) · IEEE Electronics Packaging Technology Conference (EPTC) · doi:10.1109/eptc.2008.4763487 · Cited by 84 · cited 1×
“These temperature-driven degradations are significant, with thermal factors accounting for approximately 55% of all power electronics failures [14], [15], [16], [17], [18].”
Mohanty, A. K., Vivekanandhan, S., Tripathi, N., Roy, P., Snowdon, M. R., Drzal, L. T., & Misra, M. (2023) · Composites Part C Open Access · doi:10.1016/j.jcomc.2023.100380 · Cited by 82 · cited 1×
“Increased vehicle weight can lead to several negative outcomes, including reduced range, heightened crash impact, longer braking distances, accelerated road wear, lower efficiency, and increased passenger fatality risk [29].”
Enabling Technologies for Compact Integrated Electric Drives for Automotive Traction Applications
Chowdhury, S., Gurpinar, E., Su, G.-J., et al. (2019) · IEEE Transportation Electrification Conference and Expo · doi:10.1109/itec.2019.8790594 · Cited by 75 · cited 1×
“Integrated thermal management systems that combine power electronics and electric motors can address these spatial limitations [31], [32].”
Cooling of high heat flux miniaturized electronic devices using thermal ground plane: An overview
Rakshith, B. L., Asirvatham, L. G., Angeline, A. A., et al. (2022) · Renewable and Sustainable Energy Reviews · doi:10.1016/j.rser.2022.112956 · Cited by 73 · cited 1×
“These temperature-driven degradations are significant, with thermal factors accounting for approximately 55% of all power electronics failures [14], [15], [16], [17], [18].”
Shuai, Z., He, S. T., Xue, Y., Zheng, Y., Gai, J., Li, Y., Li, J., & Li, J. (2023) · eTransportation · doi:10.1016/j.etran.2023.100241 · Cited by 71
Review of Recent Trends in Design of Traction Inverters for Electric Vehicle Applications
Goli, C. S., Essakiappan, S., Sahu, P. K., Manjrekar, M. D., & Shah, N. (2021) · IEEE International Symposium on Power Electronics for Distributed Generation Systems (PEDG) · doi:10.1109/pedg51384.2021.9494164 · Cited by 64 · cited 1×
“These demands are centered on achieving high power density, energy efficiency, effective thermal management, and seamless integration within compact system architectures [2], [3].”
Si, SiC and GaN power devices: An unbiased view on key performance indicators
Deboy, G., Treu, M., Häberlen, O. D., et al. (2016) · Technical digest - International Electron Devices Meeting/Technical digest · doi:10.1109/iedm.2016.7838458 · Cited by 50 · cited 1×
“…in power electronics, especially the integration of wide-bandgap (WBG) semiconductor materials, have demonstrated considerable potential in addressing this challenge by enabling higher switching frequencies, improved thermal characteristics, and greater overall system efficiency [4], [5], [6].”
Thermal management of harsh-environment electronics
Ohadi, M. M., & Qi, J. (2004) · doi:10.1109/stherm.2004.1291329 · Cited by 40
Direct liquid cooling of high performance Silicon Carbide (SiC) power modules
McPherson, B., McGee, B., Simco, D., Olejniczak, K., & Passmore, B. (2017) · 2017 IEEE International Workshop On Integrated Power Packaging (IWIPP) · doi:10.1109/iwipp.2017.7936760 · Cited by 29 · cited 1×
“This architecture significantly reduces thermal resistance, enhancing cooling efficiency and eliminating additional thermal barriers [41].”
Martin, G. E., Baghdadi, M. E., & Hegazy, O. (2025) · IEEE Access · doi:10.1109/access.2025.3599627 · Cited by 27 · cited 2×
“Therefore, effective thermal management is essential not only for safeguarding the performance and reliability of power electronic components but also for optimizing the overall energy efficiency of the vehicle [11].”
Self-citationThe prediction of maximum temperature for single chips’ cooling using artificial neural networks
Özsunar, A., Arcaklıoğlu, E., & Dur, F. N. (2008) · Heat and Mass Transfer · doi:10.1007/s00231-008-0445-x · Cited by 16 · cited 1×
“Ozsunar et al. [25] developed an ANN model to predict the maximum temperature in a chip subjected to constant heat power, aiming to reduce the computational burden of traditional CFD simulations.”
Pounds that Kill: The External Costs of Vehicle Weight
Anderson, M., & Auffhammer, M. (2011) · National Bureau of Economic Research · doi:10.3386/w17170 · Cited by 16
Comparative Thermal Analysis of Cooling Methods for Dual Inverter Applications in Electric Vehicles
Martin, G. E., Hosseinabadi, F., Chakraborty, S., Baghdadi, M. E., Romano, C., Tranchero, M., & Hegazy, O. (2023) · European Conference on Power Electronics and Applications · doi:10.23919/epe23ecceeurope58414.2023.10264584 · Cited by 5
Self-citationKulkarni, V., & Kospach, A. (2024) · doi:10.1109/ecceeurope62508.2024.10751869 · Cited by 5 · cited 1×
“Kulkarni and Kospach [23] presented a machine learning (ML)-driven reduced order model (ROM) framework for accurately and efficiently predicting the full temperature field in power electronics cooling systems within EV applications.”
Narsina, D., Kamisetty, A., Thompson, C. R., et al. (2024) · ABC Journal of Advanced Research · doi:10.18034/abcjar.v13i1.786 · Cited by 1
Global EV Outlook
(2024)
UnresolvedStatus and development of power semiconductor devices and its applications
Qian (2014) · Zhongguo Dianji Gongcheng Xuebao/Proc. Chin. Soc. Electr. Eng.
UnresolvedIs an antiparallel SiC-Schottky diode necessary? Calorimetric analysis of SiC-MOSFETs switching behavior
Kreutzer (n.d.)
UnresolvedMilitary Handbook. Reliability Prediction of Electronic Equipment AMSC N/A Distribution Statement A: Approved for Public Release; Distribution Unlimited
(1990)
UnresolvedSemiconductor Device Fundamentals
Pierret (1996)
UnresolvedEvaluation of thermal management solutions for power semiconductors
Balakrishnan (2016)
UnresolvedHeat Transfer: A Practical Approach
Çengel (2003)
Unresolved
Cited by
Radhakrishnan, A., Prasad, R. V., Railis, D. J., et al. (2026) · Journal of Thermal Analysis and Calorimetry · doi:10.1007/s10973-026-15314-9 · Cited by 5
“Martin et al. [34] introduced AI and machine learning (ML) models to predict and optimize thermal management strategies in EVs, offering improvements in energy efficiency and battery lifespan.”
Design Optimization of EV Drive Systems: Building the Next Generation of Automatic AI Platforms
Jiang, H., Wang, Y., Lei, G., et al. (2026) · World Electric Vehicle Journal · doi:10.3390/wevj17010035 · Cited by 1
“Inverter optimization was extended for thermal management by integrating high-fidelity CFD modeling with artificial neural networks in [122].”
Zhang, C., & Negri, R. (2026) · World Electric Vehicle Journal · doi:10.3390/wevj17010043
AI in the Adoption of Electric Vehicles: Advantages, Challenges and Future Outlook: A Review
AlZohbi, G., & Alradhi, N. (2026) · Premier journal of science. · doi:10.70389/pjs.100260
Prognostix-XAI: Interpretable Modeling of Mechanical Degradation Dynamics
Makesh, N. S., Naveenkumar, M., Kavitha, J., et al. (2026) · International Conference on Intelligent Data Communication Technologies and Internet of Things (IDCIoT) · doi:10.1109/idciot67589.2026.11455807
Metadata
- Authors
- Egin Martin, G., Verbrugge, B., Mahedi Hasan, M., El Baghdadi, M., Romano, C., & Hegazy, O.
- Affiliations
- MOBI-EPOWERS Research Group, ETEC Department, Vrije Universiteit Brussel (VUB), Brussels, Belgium
- Vrije Universiteit Brussel (BE)
- Ideas and Motion S.r.l., Cherasco, Italy
- Venue
- IEEE Access
- Published
- 2025
- Volume
- 13
- Pages
- 167632-167648
- ISSN
- 2169-3536 (ISSN-L)
- OpenAlex
- W4414348261
- Semantic Scholar
- e59416dc325c5c8cda909300a53b7cd22e031661
- Cited by
- 8
- References
- 42
- Citations by year
- 2026: 8
- FWCI
- 1.80
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- IEEE Access (publishedVersion, cc-by-nc-nd)
- VUBIR (Vrije Universiteit Brussel) (submittedVersion, other-oa)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Funders
- HiEFFICIENT Project from the Electronic Components and Systems for European Leadership (ECSEL) Joint Undertaking (101007281)
- Flanders Make
- Topics
- Physical Sciences › Engineering › Electrical and Electronic Engineering › Silicon Carbide Semiconductor Technologies
- Physical Sciences › Engineering › Mechanical Engineering › Heat Transfer and Optimization
- Physical Sciences › Engineering › Electrical and Electronic Engineering › Electric Motor Design and Analysis
- Keywords
- electric vehicle traction inverter, thermal management, cold plate design, junction temperature prediction, model reduction, artificial neural networks, real-time thermal monitoring, cooling system optimization, pressure drop, temperature uniformity, predictive maintenance
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.