September 10, 2025EnergiesOpen Access

Incorporating a Deep Neural Network into Moving Horizon Estimation for Embedded Thermal Torque Derating of an Electric Machine

View Full Paper

Ask AI

Bookmark

Authors

AWAlexander Winkler PSPranav Shah KBKatrin Baumgärtner

Discussion

Member takes

Implication

Novel approach integrates deep neural networks and moving horizon estimation for improved thermal control, indicating enhanced performance in automotive systems.

Key Points

The integration of deep neural networks into moving horizon estimation improves state estimation accuracy significantly.
A long short-term memory-based DNN trained on synthetic data enhances modeling efforts compared to traditional physics-based models.
Real-time implementation shows the DNN-based method exceeding real-time performance threefold in embedded systems.
Model-in-the-loop simulations confirm robust temperature estimation even under difficult sensor conditions.

Cite This Study

Winkler et al. (2025) studied this question.

synapsesocial.com/papers/68c2416eb210217d6479eeea https://doi.org/10.3390/en18143813

View Full Paper

Ask AI

Bookmark

Implication

Novel approach integrates deep neural networks and moving horizon estimation for improved thermal control, indicating enhanced performance in automotive systems.

Incorporating a Deep Neural Network into Moving Horizon Estimation for Embedded Thermal Torque Derating of an Electric Machine

Discussion

Member takes

Implication

Key Points

Cite This Study

Incorporating a Deep Neural Network into Moving Horizon Estimation for Embedded Thermal Torque Derating of an Electric Machine

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider