Improved Quantile Convolutional and Recurrent Neural Networks for Electric Vehicle Battery Temperature Prediction
The battery thermal management of electric vehicles can be improved using neural networks predicting quantile sequences of the battery temperature. This work extends a method for the development of Quantile Convolutional and Quantile Recurrent Neural Networks (namely Q*NN). Fleet data of 225 629 dri...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Tsinghua University Press
2024-06-01
|
| Series: | Big Data Mining and Analytics |
| Subjects: | |
| Online Access: | https://www.sciopen.com/article/10.26599/BDMA.2023.9020028 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832544871171227648 |
|---|---|
| author | Andreas M. Billert Runyao Yu Stefan Erschen Michael Frey Frank Gauterin |
| author_facet | Andreas M. Billert Runyao Yu Stefan Erschen Michael Frey Frank Gauterin |
| author_sort | Andreas M. Billert |
| collection | DOAJ |
| description | The battery thermal management of electric vehicles can be improved using neural networks predicting quantile sequences of the battery temperature. This work extends a method for the development of Quantile Convolutional and Quantile Recurrent Neural Networks (namely Q*NN). Fleet data of 225 629 drives are clustered and balanced, simulation data from 971 simulations are augmented before they are combined for training and testing. The Q*NN hyperparameters are optimized using an efficient Bayesian optimization, before the Q*NN models are compared with regression and quantile regression models for four horizons. The analysis of point-forecast and quantile-related metrics shows the superior performance of the novel Q*NN models. The median predictions of the best performing model achieve an average RMSE of 0.66°C and R2 of 0.84. The predicted 0.99 quantile covers 98.87% of the true values in the test data. In conclusion, this work proposes an extended development and comparison of Q*NN models for accurate battery temperature prediction. |
| format | Article |
| id | doaj-art-19e30b28d0ad4efb9b5d89f5f11b944d |
| institution | Kabale University |
| issn | 2096-0654 |
| language | English |
| publishDate | 2024-06-01 |
| publisher | Tsinghua University Press |
| record_format | Article |
| series | Big Data Mining and Analytics |
| spelling | doaj-art-19e30b28d0ad4efb9b5d89f5f11b944d2025-02-03T09:01:25ZengTsinghua University PressBig Data Mining and Analytics2096-06542024-06-017251253010.26599/BDMA.2023.9020028Improved Quantile Convolutional and Recurrent Neural Networks for Electric Vehicle Battery Temperature PredictionAndreas M. Billert0Runyao Yu1Stefan Erschen2Michael Frey3Frank Gauterin4Karlsruhe Institute of Technology (KIT), Institute of Vehicle System Technology, Karlsruhe 76131, Germany, and also with Bayerische Motoren Werke (BMW) AG, Munich 80788, GermanyTechnical University of Munich (TUM), Munich 80333, GermanyBMW AG, Munich 80788, GermanyKIT, Institute of Vehicle System Technology, Karlsruhe 76131, GermanyKIT, Institute of Vehicle System Technology, Karlsruhe 76131, GermanyThe battery thermal management of electric vehicles can be improved using neural networks predicting quantile sequences of the battery temperature. This work extends a method for the development of Quantile Convolutional and Quantile Recurrent Neural Networks (namely Q*NN). Fleet data of 225 629 drives are clustered and balanced, simulation data from 971 simulations are augmented before they are combined for training and testing. The Q*NN hyperparameters are optimized using an efficient Bayesian optimization, before the Q*NN models are compared with regression and quantile regression models for four horizons. The analysis of point-forecast and quantile-related metrics shows the superior performance of the novel Q*NN models. The median predictions of the best performing model achieve an average RMSE of 0.66°C and R2 of 0.84. The predicted 0.99 quantile covers 98.87% of the true values in the test data. In conclusion, this work proposes an extended development and comparison of Q*NN models for accurate battery temperature prediction.https://www.sciopen.com/article/10.26599/BDMA.2023.9020028battery temperaturedeep learningconvolutional and recurrent neural networkquantile forecasting |
| spellingShingle | Andreas M. Billert Runyao Yu Stefan Erschen Michael Frey Frank Gauterin Improved Quantile Convolutional and Recurrent Neural Networks for Electric Vehicle Battery Temperature Prediction Big Data Mining and Analytics battery temperature deep learning convolutional and recurrent neural network quantile forecasting |
| title | Improved Quantile Convolutional and Recurrent Neural Networks for Electric Vehicle Battery Temperature Prediction |
| title_full | Improved Quantile Convolutional and Recurrent Neural Networks for Electric Vehicle Battery Temperature Prediction |
| title_fullStr | Improved Quantile Convolutional and Recurrent Neural Networks for Electric Vehicle Battery Temperature Prediction |
| title_full_unstemmed | Improved Quantile Convolutional and Recurrent Neural Networks for Electric Vehicle Battery Temperature Prediction |
| title_short | Improved Quantile Convolutional and Recurrent Neural Networks for Electric Vehicle Battery Temperature Prediction |
| title_sort | improved quantile convolutional and recurrent neural networks for electric vehicle battery temperature prediction |
| topic | battery temperature deep learning convolutional and recurrent neural network quantile forecasting |
| url | https://www.sciopen.com/article/10.26599/BDMA.2023.9020028 |
| work_keys_str_mv | AT andreasmbillert improvedquantileconvolutionalandrecurrentneuralnetworksforelectricvehiclebatterytemperatureprediction AT runyaoyu improvedquantileconvolutionalandrecurrentneuralnetworksforelectricvehiclebatterytemperatureprediction AT stefanerschen improvedquantileconvolutionalandrecurrentneuralnetworksforelectricvehiclebatterytemperatureprediction AT michaelfrey improvedquantileconvolutionalandrecurrentneuralnetworksforelectricvehiclebatterytemperatureprediction AT frankgauterin improvedquantileconvolutionalandrecurrentneuralnetworksforelectricvehiclebatterytemperatureprediction |