Remaining useful life prediction of lithium-ion batteries based on Monte Carlo Dropout and gated recurrent unit

Wei, Meng; Gu, Hairong; Ye, Min; Wang, Qiao; Xu, Xinxin; Wu, Changzhi

doi:10.1016/j.egyr.2021.05.019

Cited by 55 publications

(17 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To construct the unified neural network, the network consists of 12 layers: three 1D convolutional neural network (1D-CNN) layers, two Gated Recurrent Unit (GRU) [32] levels, two dropout layers, one maxpooling, one flatten layer, and three fully connected layers. The signals are initially routed via the first convolutional layer.…”

Section: D-cnn-gru Architecturementioning

confidence: 99%

Using Deep 1D Convolutional Grated Recurrent Unit Neural Network to Optimize Quantum Molecular Properties and Predict Intramolecular Coupling Constants of Molecules of Potential Health Medications and Other Generic Molecules

et al. 2022

View full text Add to dashboard Cite

A molecule is the smallest particle in a chemical element or compound that possesses the element or compound’s chemical characteristics. There are numerous challenges associated with the development of molecular simulations of fluid characteristics for industrial purposes. Fluid characteristics for industrial purposes find applications in the development of various liquid household products, such as liquid detergents, drinks, beverages, and liquid health medications, amongst others. Predicting the molecular properties of liquid pharmaceuticals or therapies to address health concerns is one of the greatest difficulties in drug development. Computational tools for precise prediction can help speed up and lower the cost of identifying new medications. A one-dimensional deep convolutional gated recurrent neural network (1D-CNN-GRU) was used in this study to offer a novel forecasting model for molecular property prediction of liquids or fluids. The signal data from molecular properties were pre-processed and normalized. A 1D convolutional neural network (1D-CNN) was then built to extract the characteristics of the normalized molecular property of the sequence data. Furthermore, gated recurrent unit (GRU) layers processed the extracted features to extract temporal features. The output features were then passed through several fully-connected layers for final prediction. For both training and validation, we used molecular properties obtained from the Kaggle database. The proposed method achieved a better prediction accuracy, with values of 0.0230, 0.1517, and 0.0693, respectively, in terms of the mean squared error (MSE), root mean square error (RMSE), and mean absolute error (MAE).

show abstract

Section: D-cnn-gru Architecturementioning

confidence: 99%

Using Deep 1D Convolutional Grated Recurrent Unit Neural Network to Optimize Quantum Molecular Properties and Predict Intramolecular Coupling Constants of Molecules of Potential Health Medications and Other Generic Molecules

et al. 2022

View full text Add to dashboard Cite

show abstract

“…To delineate the uncertainty of battery deterioration and avoid overfitting, Wei et al. establishes a model combining Monte Carlo dropout technique and GRU NN to forecast battery capacity with the input of constant voltage charge duration ( Wei et al., 2021 ).…”

Section: Machine-learning-based Soh Predictionmentioning

confidence: 99%

State of health prediction of lithium-ion batteries based on machine learning: Advances and perspectives

Shu

Shen

et al. 2021

iScience

View full text Add to dashboard Cite

Summary Accurate state of health (SOH) prediction is significant to guarantee operation safety and avoid latent failures of lithium-ion batteries. With the development of communication and artificial intelligence technologies, a body of researches have been performed toward precise and reliable SOH prediction method based on machine learning (ML) techniques. In this paper, the conception of SOH is defined, and the state-of-the-art prediction methods are classified based on their primary implementation procedure. As an essential step in ML-based SOH algorithms, the health feature extraction methods reported in the literature are comprehensively surveyed. Next, an exhausted comparison is conducted to elaborate the development of ML-based SOH prediction techniques. Not only their advantages and disadvantages of the application in SOH prediction are reviewed but also their accuracy and execution process are fully discussed. Finally, pivotal challenges and corresponding research directions are provided for more reliable and high-fidelity SOH prediction.

show abstract

“…34 Therefore, indirect aging characteristics based on aging data extraction is the most common data processing method today. 35 This method reflects the aging process of lithium-ion batteries by presenting regular variation characteristics through a large amount of current and voltage data, 36 such as equal voltage drop and charge/discharge time. However, all of them have some defects, so this paper tries to construct a new aging feature extraction and analysis method.…”

Section: Introductionmentioning

confidence: 99%

Research on the remaining useful life prediction method of lithium‐ion batteries based on aging feature extraction and multi‐kernel relevance vector machine optimization model

Qiu

Fan

Wang

et al. 2022

Intl J of Energy Research

View full text Add to dashboard Cite

Summary Lithium‐ion batteries are used in a wide range of applications due to their cleanliness and stability, and the health management of lithium‐ion batteries has become a necessity. The most important aspect of health management is the prediction of the remaining useful life (RUL) of the battery. Therefore, a RUL estimation model based on the aging factor of the charging process and an improved multi‐kernel relevance vector machine is proposed in order to achieve high accuracy estimation of the RUL of lithium‐ion batteries. First, eight aging features highly correlated with lithium‐ion batteries capacity degradation are extracted based on charging current, voltage, and temperature data, then, their correlation is proved using gray relation analysis. Secondly, the improved gray wolf constrained optimization algorithm is used to determine the kernel function combination coefficients of the multi‐kernel relevance vector machine, and the RUL prediction model of the improved multi‐kernel relevance vector machine is established. Finally, using the battery dataset from NASA, aging data of three datasets, 24°C, 43°C, and 4°C, with a total of 11 batteries, were selected for validation. The validation results show that the improved multi‐kernel relevance vector machine prediction model has higher prediction accuracy and more robust long‐term prediction capability, with RUL prediction error less than 10 cycles and MAE less than 0.05, both of which are better than that of the single‐kernel relevance vector machine model and other multi‐kernel relevance vector machine models.

show abstract

Remaining useful life prediction of lithium-ion batteries based on Monte Carlo Dropout and gated recurrent unit

Cited by 55 publications

References 29 publications

Using Deep 1D Convolutional Grated Recurrent Unit Neural Network to Optimize Quantum Molecular Properties and Predict Intramolecular Coupling Constants of Molecules of Potential Health Medications and Other Generic Molecules

Using Deep 1D Convolutional Grated Recurrent Unit Neural Network to Optimize Quantum Molecular Properties and Predict Intramolecular Coupling Constants of Molecules of Potential Health Medications and Other Generic Molecules

State of health prediction of lithium-ion batteries based on machine learning: Advances and perspectives

Research on the remaining useful life prediction method of lithium‐ion batteries based on aging feature extraction and multi‐kernel relevance vector machine optimization model

Contact Info

Product

Resources

About