Prognostics Of Lithium-ion Batteries Research Articles

Voltage fault diagnosis and prognostic of lithium-ion batteries in electric scooters based on hybrid neural network and multiple thresholds

Voltage fault diagnosis and prognostics of lithium-ion batteries (LIBs) are critical in ensuring their safe and reliable operation, and timely voltage prediction can help detect the oncoming fault of LIBs effectively. To this end, this study develops an efficient voltage fault diagnosis and prognostic method for LIBs based on voltage prediction and multiple thresholds. Firstly, a hybrid neural network integrating convolutional neural network and gated recurrent unit is established to extract both temporal and spatial features from the current operation data and predict voltage. Then, a residual calculator is devised to generate the voltage residuals, and the maximum voltage residuals of the fault-free data during charge and discharge operations are leveraged to constitute the multiple thresholds of fault diagnosis and prognostics strategy. The robustness, reliability and feasibility of voltage prediction are validated under the conditions of different driving scenarios, different dynamic temperature and real-world operation data. Moreover, the validation results on the fault electric scooter reveals that the proposed method can reliably diagnose the battery over-voltage and under-voltage faults and their alarm levels. Additionally, the warning moment by this method is at least 630s earlier than the alarm time in the battery management system, highlighting its efficient prognostic functionality.

An Adaptive Modeling Method for the Prognostics of Lithium-Ion Batteries on Capacity Degradation and Regeneration

Accurate prediction of remaining useful life (RUL) is crucial to the safety and reliability of the lithium-ion battery management system (BMS). However, the performance of a lithium-ion battery deteriorates nonlinearly and is heavily affected by capacity-regeneration phenomena during practical usage, which makes battery RUL prediction challenging. In this paper, a rest-time-based regeneration-phenomena-detection module is proposed and incorporated into the Coulombic efficiency-based degradation model. The model is estimated with the particle filter method to deal with the nonlinear uncertainty during the degradation and regeneration process. The discrete regeneration-detection results should be reflected by the model state instead of the model parameters during the particle filter-estimation process. To decouple the model state and model parameters during the estimation process, a dual-particle filtering estimation framework is proposed to update the model parameters and model state, respectively. A kernel smoothing method is adopted to further smooth the evolution of the model parameters, and the regeneration effects are imposed on the model states during the updating. Our proposed model and the dual-estimation framework were verified with the NASA battery datasets. The experimental results demonstrate that our proposed method is capable of modeling capacity-regeneration phenomena and provides a good RUL-prediction performance for lithium-ion batteries.

A double broad learning approach based on variational modal decomposition for Lithium-Ion battery prognostics

Predicting the remaining life of lithium-ion battery equipment is becoming increasingly important as enterprises transition to smart manufacturing. Accurate prediction results can be used to effectively determine the battery's health status and improve operational safety. However, during the decline process, lithium-ion battery capacity regeneration occurs, resulting in significant fluctuations in the degradation data that can easily lead to insufficient prediction accuracies. At the same time, a factor influencing the prediction results is the unification of modal information and insufficient feature extraction of the battery capacity data in the prediction process. Therefore, in this paper, a novel model based on variational modal decomposition and double broad learning (VMD-DBL) is proposed. First, we use VMD to perform adaptive decomposition of the degraded data to form intrinsic mode function (IMF) components and residual components to solve the data noise problem. Second, these two modal data of the feature extraction and modal fusion are inputted into the trained DBL model. Finally, the two modes are connected to the output layer to obtain the predicted result. The NASA dataset is used for experimental validation in this paper, and the results show that our proposed method outperforms other methods in terms of accuracy and feasibility.

Collaborative Prognostics of Lithium-Ion Batteries Using Federated Learning With Dynamic Weighting and Attention Mechanism

Collaborative Prognostics of Lithium-Ion Batteries Using Federated Learning With Dynamic Weighting and Attention Mechanism

Health prognostics of lithium-ion batteries based on universal voltage range features mining and adaptive multi-Gaussian process regression with Harris Hawks optimization algorithm

In the field of battery prognostics and health management (PHM), accurate State-of-Health (SOH) estimation is important for the safe and reliable operation of lithium-ion batteries (LIBs). To improve the applicability and accuracy of the SOH estimation model, a new SOH estimation framework is proposed. Firstly, considering the different usage habits of users, to meet the needs of more users, a new health indicator (HI) was put forward, which involves only a small universal voltage range of charging data. Secondly, to alleviate the negative impact of the inherent cell inconsistency on SOH estimation accuracy, an adaptive Multi-Gaussian Process Regression model based on Harris Hawks Optimizer (HHO-MGPR) was proposed. The core idea is to use the HHO algorithm to automatically generate the optimal training subset combination for multiple GPR base learners so as to improve the model generalization ability. Finally, four open-source datasets with different battery types, temperature and operating conditions are employed for performance validation. The experimental results show that in all cases, the proposed HI can effectively characterize the battery degradation and the developed HHO-MGPR can improve the model generalization ability, indicating the proposed framework has wide applicability.

Long short-term memory network with Bayesian optimization for health prognostics of lithium-ion batteries based on partial incremental capacity analysis

Long short-term memory network with Bayesian optimization for health prognostics of lithium-ion batteries based on partial incremental capacity analysis

Prognostics of Lithium-Ion batteries using knowledge-constrained machine learning and Kalman filtering

• Propose a knowledge-constrained machine learning method for batteries prognostics • Propose a data-driven stochastic model for battery degradation • Employ an ANN-based DEKF method to capture the dynamics of lithium-ion batteries • Validate the proposed approach using NASA dataset for battery prognostics Accurately predicting the remaining useful life (RUL) of lithium-ion rechargeable batteries remains challenging as the battery capacity degrades in a stochastic manner given the internal complex electrochemical reactions of the battery and the external operational/environmental conditions. In this work, a knowledge-constrained machine learning framework is developed to learn the stochastic degradation of battery performance over working cycles for health prognostics of Lithium-Ion batteries. An artificial neural network (ANN) model is first trained and synchronized using a Dual Extended Kalman Filter (DEKF) to obtain critical health information of Lithium-Ion batteries. With the obtained health information, a knowledge-constrained machine learning method (KcML) is then developed to predict the stochastic degradation of the battery capacity in operation. Specifically, prior knowledge in battery capacity fade can be formulated as extra constraints to facilitate the development of machine learning models with improved fidelity level for battery capacity predictions. A dataset published by NASA is utilized to demonstrate the effectiveness of the proposed knowledge-constrained machine learning approach.

Cloud-based in-situ battery life prediction and classification using machine learning

In-situ battery life prediction and classification can advance lithium-ion battery prognostics and health management. A novel physical features-driven moving-window battery life prognostics method is developed in this paper, which can be used to predict the battery remaining useful life (RUL) and knee-point, and for the first time to classify the battery life in real-time. The relationship between the physical features and battery life is captured by using machine learning. The proposed methodology is validated based on experimental data of more than 100 cell samples. The results show that the method predicts accurate RUL and knee-point, with the root mean squared error and mean absolute percentage error being, respectively, low to 55 cycles and 3.55%. The battery life is also classified accurately based on the data of only one single cycle, with the sorting accuracy up to 91.84%, facilitating fast and efficient sorting/screening of retired batteries in the future. Both the prediction and classification accuracies decrease as the moving-window moves forward, indicating accurate life prediction can still be obtained even when the battery has been put in operation for years.

Remaining useful life prognostics of lithium-ion batteries based on a coordinate reconfiguration of degradation trajectory and multiple linear regression

Lithium battery has been widely applied as new energy to cope with pressures in both form environment and energy. The remaining useful life (RUL) prognostics of lithium-ion batteries have become more critical. Convenient battery life prediction allows early detection of performance deficiencies to help maintain the battery system promptly. This paper proposes a RUL prognostics model of lithium-ion batteries based on a coordinate reconfiguration of degradation trajectory and multiple linear regression. First, a new sampling rule is used to reconfigure the coordinates of degradation data of new batteries and truncated similar batteries. Then, the relationship between similar and new lithium-ion batteries is established by using the reconfiguration data. Moreover, a new RUL prognostics model based on a coordinate reconfiguration of degradation trajectory and multiple linear regression is established by considering the influence of time-varying factors, which can improve prediction accuracy with small sample data and significantly reduce product development time and cost.

Accelerated Stress Factors Based Nonlinear Wiener Process Model for Lithium-Ion Battery Prognostics

Accurate remaining useful life (RUL) of batteries plays an imperative role in ensuring safe operations and avoiding catastrophic accidents. However, in practice, complicated working conditions bring challenges to accurate battery prognostics. In this article, an accelerated stress factors-based nonlinear Wiener process model is proposed to enrich inadequate battery prognostic works at various operating conditions. To realize online individual battery prognostics, once a new measurement is available, the parameters of a state-space model constructed by the proposed model are posteriorly updated. Then, based on the Peukert law and the Arrhenius equation, two specific accelerated stress-relevant drift functions and their associated degradation models at different discharge rates and temperatures are respectively designed. Subsequently, RUL predictions are conducted using the proposed method. RUL predictions at different discharge rates and different temperatures demonstrate the accuracy and robustness of the proposed prognostic models. According to some general prognostic metrics, the proposed method is proved to be superior to four existing RUL prediction approaches.

Lithium-Ion Battery Prognostics through Reinforcement Learning Based on Entropy Measures

Lithium-ion is a progressive battery technology that has been used in vastly different electrical systems. Failure of the battery can lead to failure in the entire system where the battery is embedded and cause irreversible damage. To avoid probable damages, research is actively conducted, and data-driven methods are proposed, based on prognostics and health management (PHM) systems. PHM can use multiple time-scale data and stored information from battery capacities over several cycles to determine the battery state of health (SOH) and its remaining useful life (RUL). This results in battery safety, stability, reliability, and longer lifetime. In this paper, we propose different data-driven approaches to battery prognostics that rely on: Long Short-Term Memory (LSTM), Autoregressive Integrated Moving Average (ARIMA), and Reinforcement Learning (RL) based on the permutation entropy of battery voltage sequences at each cycle, since they take into account vital information from past data and result in high accuracy.

Multiple health indicators assisting data-driven prediction of the later service life for lithium-ion batteries

Data-driven method is an efficient tool for diagnostics and prognostics of lithium-ion batteries during their manufacturing and service period. Accurately predicting the later service life of batteries is a meaningful task. Still, it remains a challenge due to the nonlinear rapid capacity decay caused by the accumulation of inner electrochemical deterioration. Here, we use a classic deep neural network algorithm to study the degradation laws in later battery service life under the common role of multiple health indicators. A battery cyclic data pre-processing method is proposed and several characteristic parameters with a high correlation to battery life are carefully selected. Our models achieve an average test error within 5% using any continuous 30 cycles of data to predict the battery capacity curve in the next 200 cycles. This study highlights the promise of combining deliberate data processes with health indicators in data-driven modeling to predict the later service life of lithium-ion batteries.

Gaussian process-based prognostics of lithium-ion batteries and design optimization of cathode active materials

The increasing adoption of lithium-ion batteries (LIBs) in consumer electronics, electric vehicles, and smart grids poses two challenges: the accurate prediction of the battery health to prevent operational impairments and the development of new materials for high-performance LIBs. Characterized by their flexibility and mathematical tractability, Gaussian processes (GPs) provide a powerful framework for monitoring and optimization tasks. This study employs two GP-based techniques: co-kriging surrogate modelling and Bayesian optimization. The GP training data comes from the cycling performance test of five CR2032 cells with Ni contents of 0.0, 0.4, 0.5, 0.6, and 1.0 in their cathode active material LiNixMn2−xO4. The co-kriging surrogate predicts the capacity degradation profile of a cell by exploiting information from different cells. Bayesian optimization identifies new Ni compositions that can produce cells with high initial specific capacity and large cycle life. The study shows the predictive capabilities of the co-kriging surrogate when correlated data is available. Bayesian optimization predicts that a Ni content of 0.44 produces cells with an initial specific capacity of 103.4 ± 3.8 mAh g−1 and a cycle life of 595 ± 12 cycles. Furthermore, the Bayesian strategy identifies other Ni contents that can improve the overall quality of the current Pareto front.

Prognostics of Lithium-Ion Batteries Based on Capacity Regeneration Analysis and Long Short-Term Memory Network

The accurate prognostics of state of health (SOH) prediction of lithium-ion batteries is significant for manufacturers and consumers to determine the failure and optimize the usage in advance. This paper proposes a framework to decouple the capacity regeneration phenomena and the normal capacity degradation process to make predictions. The regeneration phenomena are automatically identified by clustering the time intervals of adjacent cycles and detecting the long time intervals. Then support vector regression is used to predict the capacity regeneration amplitude and the corresponding regeneration cycles. The complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) is utilized to decompose the normal degradation path into several intrinsic mode functions. The long short-term memory recurrent network is constructed to predict the components with the C-C method determining the appropriate model parameters. Ultimately, the prediction results are added to obtain the complete capacity degradation trajectory. The proposed framework is validated with five lithium-ion battery datasets from NASA Ames Prognostics Center of Excellence and Center for Advanced Life Cycle Engineering of the University of Maryland. The results demonstrate that the proposed method provides more accurate SOH prediction than the published methods.

A data-driven approach based on deep neural networks for lithium-ion battery prognostics

Remaining useful life estimation is gaining attention in many real-world applications to alleviate maintenance expenses and increase system reliability and efficiency. Deep learning approaches have recently provided a significant improvement in the estimation of remaining useful life (RUL) and degradation progression concerning machinery prognostics. This research presents a new data-driven approach for RUL estimation using a hybrid deep neural network that combines CNN, LSTM, and classical neural networks. The presented CNN–LSTM neural network aims to extract the spatio-temporal relations in multivariate time series data and capture nonlinear characteristics to achieve better RUL prediction accuracy. To improve the proposed model's performance, PSO is handled to simultaneously optimize the hyperparameters of the network consisting of the number of epochs, the number of convolutional and LSTM layers, the size of units (or filters) in each convolutional, and LSTM layers. Besides, the proposed model in this paper, called the CNN–LSTM–PSO, realizes the multi-step-ahead prediction. In the experimental studies, the popular lithium-ion battery dataset presented by NASA is selected to verify the CNN–LSTM–PSO approach. The experimental consequences revealed that the presented CNN–LSTM–PSO model gives better results than other state-of-the-art machine learning techniques and deep learning approaches considering various performance criteria.

A review on state of health estimations and remaining useful life prognostics of lithium-ion batteries

Lithium-ion batteries have been generally used in industrial applications. In order to ensure the safety of the power system and reduce the operation cost, it is particularly important to accurately and timely estimate the state of health (SOH) and predict the remaining useful life (RUL) of lithium-ion batteries. With the development of intelligent tools such as artificial intelligence, big data analysis and the Internet of Things, the methods of battery health assessment have been gradually diversified. Here, we have compiled four publicly available battery datasets. The SOH estimations and RUL prognostics of lithium-ion batteries are reviewed by analyzing the research status. To this end, after studying different scientific and technical literatures, the respective methods are divided into specific groups, and the advantages and limitations of the battery management system application are discussed. At the end, the future development trend and research challenges are analyzed. All key insights in this review will hopefully drive the development of battery health estimation and life prediction techniques.

Remaining Useful Life Estimation for Prognostics of Lithium-Ion Batteries Based on Recurrent Neural Network

Prognostic and condition-based maintenance of lithium-ion batteries is a fundamental topic, which is rapidly expanding since a long battery lifetime is required to ensure economic viability and minimize the life cycle cost. Remaining useful life (RUL) estimation is an essential tool for prognostic and health management of batteries. In this article, a hybrid approach based on both condition monitoring and physic model is presented to improve the accuracy and precision of RUL estimation for lithium-ion battery. An artificial intelligence estimation method based on recurrent neural network (RNN) is integrated with a state-space estimation technique, which is typical of filtering-based approach. The state-space estimation is used to generate a big dataset for the training of the RNN. Some additional deep layers are used to improve the prediction of nonlinear trends (typical of batteries), while the performance optimization of the RNN is ensured using a genetic algorithm. The performances of the proposed method have been tested using a battery degradation dataset from the data repository of Prognostics Center of Excellence at NASA. Two different degradation models are compared, the widely known empirical double exponential model and an innovative single exponential model that allows to ensure optimal performance with fewer parameters required to be estimated.

Experimental verification of lithium-ion battery prognostics based on an interacting multiple model particle filter

A new data-driven prognostic method based on an interacting multiple model particle filter (IMMPF) is proposed for use in the determination of the remaining useful life (RUL) of lithium-ion (Li-ion) batteries and the probability distribution function (PDF) of the uncertainty associated with the RUL. An IMMPF is applied to different state equations. The battery capacity degradation model is very important in the prediction of the RUL of Li-ion batteries. The IMMPF method is applied to the estimation of the RUL of Li-ion batteries using the three improved models. Three case studies are provided to validate the proposed method. The experimental results show that the one-dimensional state equation particle filter (PF) is more suitable for estimating the trend of battery capacity in the long term. The proposed method involving interacting multiple models demonstrated a stable and high prediction accuracy, as well as the capability to narrow the uncertainty in the PDF of the RUL prediction for Li-ion batteries.

An HI Extraction Framework for Lithium-Ion Battery Prognostics Based on SAE-VMD

The signals of lithium-ion battery degradation are non-stationary and nonlinear. To adaptively extract the health indicator(HI) that can accurately represent the battery degradation characters and improve the prediction precision of battery remaining useful life (RUL), a stacked auto encoder-variational mode decomposition(SAE-VMD) based HI construction framework is proposed. Firstly, the stacked auto encoder(SAE) is used to reduce the noises of battery parameters and lower the data dimensionality and construct a syncretic HI that contains the battery degradation characters. Then the variational mode decomposition(VMD) is employed for effectively separating the syncretic HI into three modalities: the global attenuation, the local regeneration and the noises. The three modalities are selected as HIs to eliminate the HI noises and improve the RUL prediction precision. The RUL prediction results of lithium-ion battery indicate that the HI extracted by using the present method can obtain a better RUL prediction precision and verify the high quality of the extracted HI.

Prognostics Comparison of Lithium-Ion Battery Based on the Shallow and Deep Neural Networks Model

Prognostics of the remaining useful life (RUL) of lithium-ion batteries is a crucial role in the battery management systems (BMS). An artificial neural network (ANN) does not require much knowledge from the lithium-ion battery systems, thus it is a prospective data-driven prognostic method of lithium-ion batteries. Though the ANN has been applied in prognostics of lithium-ion batteries in some references, no one has compared the prognostics of the lithium-ion batteries based on different ANN. The ANN generally can be classified to two categories: the shallow ANN, such as the back propagation (BP) ANN and the nonlinear autoregressive (NAR) ANN, and the deep ANN, such as the long short-term memory (LSTM) NN. An improved LSTM NN is proposed in order to achieve higher prediction accuracy and make the construction of the model simpler. According to the lithium-ion data from the NASA Ames, the prognostics comparison of lithium-ion battery based on the BP ANN, the NAR ANN, and the LSTM ANN was studied in detail. The experimental results show: (1) The improved LSTM ANN has the best prognostic accuracy and is more suitable for the prediction of the RUL of lithium-ion batteries compared to the BP ANN and the NAR ANN; (2) the NAR ANN has better prognostic accuracy compared to the BP ANN.