Remaining Useful Life Assessment for Lithium-Ion Batteries Using CNN-LSTM-DNN Hybrid Method

Zraibi, Brahim; Okar, Chafik; Chaoui, Hicham; Mansouri, Mohamed

doi:10.1109/tvt.2021.3071622

Cited by 161 publications

(34 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, compared with the prediction results obtained by the prediction model established by its data (Section 5.2), the RMSE difference between these two models is tiny, generally within 0.1%, and the maximum is 0.3%. Thus, it shows that the SOH estimation method constructed by the CEEMDAN-GARG fusion prediction model [52] 0.013 0.008 CNN-LSTM [53] 0.0204 0.01536 CNN-LSTM-DNN [53] 0.0145 0.01897 CEEMDAN-GARG 0.0116 0.009 71 PA-LSTM [45] 0.0163 -RNN [45] 0.0331 -UPE [54] 0.0143 -CEEMD-OMKRVM [54] 0.014 -CEEMDAN-GARG 0.0102 0.0096 81 LSSVM [55] 0.0267 0.023 MC-RNN [26] 0.0255 0.0223 MC-GRU [26] 0.0282 0.025 BP [51] 0.0586 -NAR [51] 0.0493 -LSTM [51] 0.0352 -CEEMDAN-GARG 0.009 0.0077 91 TCN [52] 0.014 0.006 PA-LSTM [45] 0.0166 -AST-LSTM [56] 0.0186 -RNN [45] 0.0484 -BP [51] 0.0394 -NAR [51] 0.0373 -LSTM [51] 0.0302 -CEEMDAN-GARG 0.0081 0.007 101 RVM [45] 0.0141 -RNN [45] 0.0165 -PA-LSTM [45] 0.006 -MC-RNN [26] 0.0276 0.0219 MC-GRU [26] 0.0286 0.0258 BP [51] 0.0233 -NAR [51] 0.0282 -LSTM [51] 0.0118 -CEEMDAN-GARG 0.0043 0.0036 values than other prediction models. And with the increase of training data, the prediction is more accurate.…”

Section: Robustness Analysis Of the Fusion Modelmentioning

confidence: 97%

State‐of‐Health Prediction for Lithium‐Ion Batteries Based on Complete Ensemble Empirical Mode Decomposition with Adaptive Noise‐Gate Recurrent Unit Fusion Model

2022

View full text Add to dashboard Cite

State‐of‐health (SOH) estimation is one of the most critical battery management system (BMS) tasks. A challenge remains for the SOH prediction due to the complicated battery aging mechanism. The most common health indicator is the capacity of the lithium‐ion battery. The fluctuation of capacity caused by the capacity regeneration phenomenon can seriously affect the prediction performance. A new complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) and gate recurrent unit (GRU) based fusion prediction model for SOH estimation is proposed to solve the problem effectively. First, the CEEMDAN algorithm decomposes the original SOH into local fluctuations and global degradation trends. Then, the GRU network and autoregressive integrated moving average model are used to predict the above trends, respectively. Next, a sliding window is designed to calculate an average value of the global degradation trend prediction residuals. Then, the second GRU algorithm can be used to correct prediction residuals. Finally, the prediction results of the aforementioned parts are combined to obtain the final SOH estimation. The proposed method is verified by experimental battery data from NASA and CALCE datasets. The results show that the fusion method has both higher estimation accuracy and stronger robustness than other methods.

show abstract

Section: Robustness Analysis Of the Fusion Modelmentioning

confidence: 97%

State‐of‐Health Prediction for Lithium‐Ion Batteries Based on Complete Ensemble Empirical Mode Decomposition with Adaptive Noise‐Gate Recurrent Unit Fusion Model

2022

View full text Add to dashboard Cite

show abstract

“…Moreover, Zraibi et al proved that adding DNN layer after CNN-LSTM can further improve the performance of prediction. [99] Kara et al used PSO adaptively to adjust the hyperparameters of CNN-LSTM model to improve the prediction performance of the model. [100] Hong et al successively used dilated convolution to extract multi-scale features, 1D-convolution to generate compact features and fully connected layer output prediction, and constructed the first RUL prediction model from the original data to the prediction results, that is, "end-to-end" model.…”

Section: Convolutional Neural Network (Cnn)mentioning

confidence: 99%

A Review on the Prediction of Health State and Serving Life of Lithium‐Ion Batteries

Zhong

Wang

Yang

et al. 2022

The Chemical Record

View full text Add to dashboard Cite

The monitoring and prediction of the health status and the end of life of batteries during the actual operation plays a key role in the battery safety management. However, although many related studies have achieved exciting results, there are few systematic and comprehensive reviews on these prediction methods. In this paper, the current prediction models of remaining useful life of lithium‐ion batteries are divided into mechanism‐based models, semi‐empirical models and data‐driven models. Their advantages, technical obstacles, improvement methods and prediction performance are summarized, and the latest research results are shown by comparison. We highlight that the fusion models of convolution neural network, long short term memory network and so on, which have great practical application prospects because of their outstanding computing efficiency and strong modeling ability. Finally, we look forward to the future work in simplifying the model and improving its interpretability.

show abstract

“…Similarly, in order to develop CNN and LSTM in depth for deeper information mining in limited data, Ren et al proposed a new LIB RUL prediction method based on improved CNN and LSTM, namely Auto-CNN-LSTM [ 25 ]. At the same time, Zraibi et al used the hybrid network of CNN, LSTM and Deep Neural Networks (DNN) to accurately predict the RUL of lithium-ion batteries, which proved that the hybrid network was superior to the single network [ 26 ].…”

Section: Introductionmentioning

confidence: 99%

Real-Time Updating High-Order Extended Kalman Filtering Method Based on Fixed-Step Life Prediction for Vehicle Lithium-Ion Batteries

Wang

Wen

2022

Sensors

View full text Add to dashboard Cite

Lithium-ion batteries have become an important power source in low-carbon transportation energy, and the safe operation and remaining useful life prediction are of great significance. Aiming at the shortcomings of existing methods, such as low prediction accuracy and a short prediction period, this paper proposes a real-time update high-order extended Kalman filter method based on fixed-step life prediction for vehicle lithium batteries based on the principle of combining models and data. First, the state model describing the parameters in the dynamic energy attenuation model is established, and the energy attenuation model is regarded as the observation model of the system to meet the requirements of establishing the Kalman filter. Secondly, the multi-step prediction equation of the state model is established by iterative recursion. At the same time, the multi-step prediction equation between the existing energy output value and the future output value is established based on the multi-dimensional Taylor network (MTN). The multiplicative noise term introduced in the dynamic modeling process is regarded as the hidden variable of the system to meet the requirements of establishing the multi-step linear predictive Kalman filter. Finally, the effectiveness of the new method is verified by digital simulation examples.

show abstract

Remaining Useful Life Assessment for Lithium-Ion Batteries Using CNN-LSTM-DNN Hybrid Method

Cited by 161 publications

References 45 publications

State‐of‐Health Prediction for Lithium‐Ion Batteries Based on Complete Ensemble Empirical Mode Decomposition with Adaptive Noise‐Gate Recurrent Unit Fusion Model

State‐of‐Health Prediction for Lithium‐Ion Batteries Based on Complete Ensemble Empirical Mode Decomposition with Adaptive Noise‐Gate Recurrent Unit Fusion Model

A Review on the Prediction of Health State and Serving Life of Lithium‐Ion Batteries

Real-Time Updating High-Order Extended Kalman Filtering Method Based on Fixed-Step Life Prediction for Vehicle Lithium-Ion Batteries

Contact Info

Product

Resources

About