Comparing Hybrid Approaches of Deep Learning for Remaining Useful Life Prognostic of Lithium-Ion Batteries
Bibliographic record
Abstract
Many published journals used hybrid deep learning methods to predict batteries’ remaining useful life by adopting different rationales to select and combine deep learning methods aiming to propose the most accurate prediction model possible. The main contribution of this article consists of proposing, to the best of the authors’ knowledge, the most accurate hybrid deep learning prediction model, designed and configured by considering the theoretical strength of each of the selected deep learning models, combined with meticulous data preprocessing and feature engineering steps. A benchmark study is presented to confirm the theoretical design by comparing the prediction results of the selected hybrid model with other proposed hybrid deep learning algorithms. The selected prediction model is compared as well with previously published articles, specifically, the ones that have used hybrid deep learning methods, NASA datasets, and batteries #6, #7, and #18 selectively. The hybrid model refers to the combination of different types of deep learning architectures, such as Convolutional Neural Networks (CNNs), Deep Neural Networks (DNN), Long Short-Term Memory (LSTM), Bidirectional Long Short-Term Memory (bLSTM), recurrent neural network (RNN), Bidirectional recurrent neural network (bRNN), Gated recurrent units (GRU) and Bidirectional Gated recurrent units (bGRU). This combination includes CNN-LSTM-DNN, CNN-bLSTM-DNN, CNN-GRU-DNN, CNN-bGRU-DNN, CNN-RNN-DNN, and CNN-bRNN-DNN, and aims to leverage the strengths of each architecture in capturing spatial, temporal, and sequential patterns present in the battery dataset. The hybrid deep learning approaches are tested with multichannel inputs, encompassing parameters such as voltage, current, and temperature, as well as their respective time series averages. The objective is to predict the remaining useful life. Performance evaluation is conducted using error metrics, including Mean Absolute Error (MAE) and Root Mean Square Error (RMSE). The results revealed a remarkable 90.5% enhancement in RMSE, indicating substantial improvement.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".