A Comprehensive Review of Available Battery Datasets, RUL Prediction Approaches, and Advanced Battery Management

Presently electric vehicles (EVs) are considered as most propitious solution for the replacement of internal combustion (IC) engine-based vehicle. The development of EV technologies is growing rapidly and the battery technology is an important concept for development of the electric vehicles. The EV performance mainly relies on the battery performance and battery management system (BMS). Recently, the Lithium-ion (Li-ion) battery is mainly used as a battery in EVs due to smaller weight, high energy density and capability of fast charging and discharging. Considering the dynamic performance, economy, safety friendliness to the environment of the EVs, the BMS is designed such a way to meet the challenges like the energy management of battery, reduction of heating-time at low temperature and enhancing remaining-useful life (RUL) with accuracy of prediction. The battery is managed and controlled by BMS and it is mainly focused to maintain the reliability and safety. The state estimation of the battery is essential for vehicle control and management of energy. The paper gives review on the strategies like battery modeling, state estimation and prediction. The state estimation for State of charge (SOC), State of power (SOP), State of health (SOH) and prediction of RUL are overviewed.

A comprehensive review of battery modeling and state estimation approaches for advanced battery management systems

Accurate prediction of battery degradation and remaining useful life (RUL) is critical for optimizing the performance and lifespan of battery-powered systems in electric vehicles and renewable energy storage applications. This paper introduces a novel machine learning approach utilizing Bidirectional Long Short-Term Memory (Bi-LSTM) networks to predict battery degradation and estimate RUL based on key parameters including voltage, current, temperature, and cycle count. Unlike conventional LSTM models that process data in a unidirectional manner, our Bi-LSTM architecture captures both past and future dependencies in battery behavior, significantly improving prediction accuracy. Through comprehensive evaluation on real-world battery datasets, we demonstrate that Bi-LSTM outperforms traditional LSTM systems by reducing root mean square error (RMSE) for state of health (SOH) prediction from 4.5–3.1% and improving R² values from 0.87 to 0.92. For RUL prediction, our model achieves an RMSE of 120 cycles compared to 150 cycles for standard LSTM. These improvements enable more reliable real-time battery health monitoring and proactive management strategies. The integration of Bi-LSTM into battery management systems (BMS) offers enhanced computational efficiency and superior convergence speed, making it particularly suitable for applications requiring precise battery management such as electric vehicles and grid-scale energy storage systems.

State of health and remaining useful life prediction of lithium-ion batteries based on a disturbance-free incremental capacity and differential voltage analysis method

The prediction of Remaining useful life (RUL) and the estimation of State of health (SOH) are extremely important issues for operating performance of Lithium-ion (Li-ion) batteries in the Battery management system (BMS). A multi-scale prediction approach of RUL and SOH is presented, which combines Wavelet neural network (WNN) with Unscented particle filter (UPF) model. The capacity degradation data of Li-ion batteries are decomposed into the low-frequency degradation trend and high-frequency fluctuation components by Discrete wavelet transform (DWT). Based on the WNN-UPF model, the long-term RUL of Li-ion batteries is predicted with the low-frequency degradation trend data. The high-frequency fluctuation data and RUL prediction results are integrated effectively to estimate the short-term SOH of Li-ion batteries. The experimental results show that the proposed method achieves high accuracy and strong robustness, even if the prediction starting point is set to the early stage of Li-ion batteries' lifespan.

The remaining useful life (RUL) prediction of lithium-ion batteries (LIBs) plays a crucial role in battery management, safety assurance, and the anticipation of maintenance needs for reliable electric vehicle (EV) operation. An efficient prediction of RUL can ensure its safe operation and prevent both internal and external failures, as well as avoid any unwanted catastrophic events. However, achieving precise RUL prediction for electric vehicles presents a challenging task due to several issues related to intricate operational characteristics and dynamic shifts in model parameters throughout the aging process, battery parameters data extraction, data preprocessing, and hyperparameters tuning of the prediction model. This phenomenon significantly impacts the advancement of electric vehicle technology. To address these challenges, this study offers a comprehensive overview of various RUL prediction methods, presenting a comparative analysis of their outcomes, advantages, drawbacks, and associated research constraints. Emphasis is placed on the necessity of a battery management system (BMS) to ensure the safe and reliable functioning of LIBs. The review delves into crucial implementation factors, including battery test bench considerations, data selection, feature extraction, data preprocessing, performance evaluation indicators, and hyperparameter tuning. Additionally, the issues and challenges related to RUL prediction approaches such as; thermal runaway, material selection, cell balancing, battery aging, relaxation impact, training algorithms, data acquisition, and hyperparameter tuning were outlined to provide an in-depth understanding of the recent situations. The outcome of this review comprehensively examines various methods for predicting the RUL of LIB in EV applications, offering insights into their advantages, limitations, and research challenges. Recommendations for future trends in LIBs technology comprise enhancing prognostic accuracy and developing robust approaches to guarantee sustainable operation and management.

Online remaining useful life prediction of lithium-ion batteries using bidirectional long short-term memory with attention mechanism

Energy storage systems (ESSs) and electric vehicle (EV) batteries depend on battery management systems (BMSs) for their longevity, safety, and effectiveness. Battery modeling is crucial to the operation of BMSs, as it enhances temperature control, fault detection, and state estimation, thereby maximizing efficiency and preventing malfunctions. This paper thoroughly examines the most recent advancements in battery and BMS modeling, including data-driven, thermal, and electrochemical methods. Advanced modeling approaches are explored, including physics-based models that incorporate mechanical stress and aging effects, as well as artificial intelligence (AI)-driven state estimation. New technologies that facilitate data-driven decision-making, real-time monitoring, and simplified systems include digital twins (DTs), cloud computing, and wireless BMSs. Nonetheless, there are still issues with cost optimization, cybersecurity, and computing efficiency. This study presents key advancements in battery modeling and BMS applications, including defect diagnostics, temperature management, and state-of-health (SOH) prediction. A comparison of machine learning (ML) methods for SOH prediction is given, emphasizing how well neural networks (NNs) and transfer learning function with real-world datasets. Additionally, future research objectives are described, with an emphasis on next-generation sensor technologies, cloud-based BMSs, and hybrid algorithms. Distinct from existing reviews, this paper integrates academic modeling with industrial benchmarking and highlights the convergence of hybrid physics-informed and data-driven techniques, multi-physics simulations, and intelligent architecture. For high-performance EV applications, this analysis offers insight into creating more intelligent, adaptable, and secure BMSs by addressing current constraints and utilizing state-of-the-art technologies.

To safeguard the security and dependability of battery management systems (BMS), it is essential to provide reliable forecasts of battery capacity and remaining useful life (RUL). However, most of the current prediction methods use the measurement data directly to carry out prediction work, which ignores the objective measurement noise and capacity increase during the aging process of batteries. In this study, an integrated prediction method is introduced to highlight the prediction of lithium-ion battery capacity and RUL. This approach incorporates several techniques, including variational modal decomposition (VMD) with entropy detection, a double Gaussian model, and a gated recurrent unit neural network (GRU NN). Specifically, the PE−VMD algorithm is first utilized to perform a noise reduction process on the capacity data obtained from the measurements, and this results in a global degradation trend sequence and local fluctuation sequences. Afterward, the global degradation prediction model is established by employing the double Gaussian aging model proposed in this paper, and the local prediction models are built for each local fluctuation sequence by GRU NN. Lastly, the proposed hybrid prediction methodology is validated through battery capacity and RUL prediction studies on experimental data from three sources, and its accuracy is also compared with prediction algorithms from the recent related literature. Experimental results demonstrate that the proposed hybrid prediction method exhibits high precision in the predicting future capacity and RUL of lithium-ion batteries, along with strong robustness and predictive stability.

Edge computing for vehicle battery management: Cloud-based online state estimation

This study highlights the increasing demand for battery-operated applications, particularly electric vehicles (EVs), necessitating the development of more efficient Battery Management Systems (BMS), particularly lithium-ion (Li-ion) batteries used in energy storage systems (ESS). This research addresses some of the key limitations of current BMS technologies, with a focus on accurately predicting the remaining useful life (RUL) of batteries, which is a critical factor for ensuring operational efficiency and sustainability. Real-time data are collected from sensors via an Internet of Things (IoT) device and processed using Arduino Nano, which extracts values for input into a Long Short-Term Memory (LSTM) model. This model employs the National Aeronautics and Space Administration (NASA) Li-battery dataset and current, voltage temperature, and cycle values to predict the battery RUL. The proposed model demonstrates significant forecasting precision, attaining a root mean square error (RMSE) of 0.01173, outperforming all comparative models. This improvement facilitates more effective decision-making in BMS, particularly in resource allocation and adaptability to transient conditions. However, the practical implementation of real-time data acquisition systems at a scale and across diverse environments remains challenging. Future research will focus on enhancing the generalizability of the model, expanding its applicability to broader datasets, and automating data ingestion to minimize integration challenges. These advancements are aimed at improving energy efficiency in both industrial and residential applications in accordance with the Sustainable Development Goals (SDGs) of the UN.

Battery energy storage and management systems constitute an enabling technology for more sustainable transportation and power grid systems. On the one hand, emerging materials and chemistries of batteries are being actively synthesized to continually improve their energy density, power density, cycle life, charging rate, etc. On the other hand, advanced battery management systems (BMSs) are being intensively developed to guarantee the safety, reliability, efficiency, and cost-effectiveness of batteries in realistic operations, as well as their integration with mechatronics. Owing to their multi-physics nature, designing high-performance batteries and their management systems requires multidisciplinary approaches, with an ever-increasing synergy of electrochemi- cal, material, mechatronics, computer, and control disciplines.

We discuss the estimation of lithium-ion battery states using an extended Kalman filter based on the Doyle, Fuller and Newman pseudo two-dimensional (P2D) electrochemical model [1]. The P2D model is solved using our computationally efficient Matlab implementation [2] that uses Chebyshev orthogonal collocation. In order to predict the cell capacity and power fade, the P2D model is coupled to a bulk thermal model and a degradation model describing the formation of the solid-electrolyte interphase (SEI) [3]. This state estimation algorithm could be used online for advanced battery management systems (BMS) for electric vehicles and other applications. BMSs are embedded systems that monitor and control individual cells of a battery pack to ensure safe and optimal energy use. The estimation of battery states such as state-of-charge (SOC), state-of-health (SOH) and instantaneous power capability is an important function fulfilled by the BMS, requiring measurements of current, voltage and temperature combined with an accurate battery model. Current approaches usually employ low-order equivalent circuit models (ECMs) due to their low computational requirements. However, these are only accurate within a narrow operating range, and accounting for changes in dynamics and performance caused by degradation is difficult due to the lack of physical meaning of model parameters. Battery models such as the P2D model [1] are based on equations of conservation and are accurate over a wider operating range accounting for various effects, including temperature and degradation. However, these are usually too computationally intense for embedded applications. Instead, we focus on efficient numerical methods called spectral methods that have been recently applied by us and others to the P2D model to reduce the computational burden while maintaining accuracy [2, 4, 5]. We have also recently demonstrated this method applied to super-capacitor modeling [6]. We have implemented an efficient solution of the P2D model coupled to a bulk thermal model, with temperature- and concentration-dependent parameters, using the orthogonal collocation on Chebyshev polynomials spectral method to spatially discretize the partial differential equations of the model. This reduces the order of the discretized model by a factor of 100-1000 compared to finite difference or finite element methods whilst maintaining a similar accuracy. Figure 1 shows the voltage predicted using the spectral method (markers) compared to a finite element solution using Comsol (solid-lines), both are in very good agreement to high C-rates, and the solution time is 50-100 times faster with spectral methods. Modelling degradation is of utmost importance for automotive applications in order to estimate and predict the capacity and power fade of cells. The loss of lithium and the increase of battery impedance due to the formation and growth of the solid-electrolyte interphase (SEI) are believed to be the major degradation mechanisms in commercial graphite anode-based cells. We coupled our efficient implementation of the thermal-P2D model to a degradation model describing the evolution of the SEI layer thickness [3, 7]. The coupled thermal-P2D and degradation model is efficiently solved in Matlab using a unified control-engineering state space approach. Several techniques are available for the state estimation of such a non-linear state-space model, including the extended Kalman filter (EKF), the unscented Kalman filter (UKF) or moving horizon estimation (MHE). Although the UKF and MHE may provide a more accurate estimation, these are computationally intensive. Instead, we focus on a modified EKF that shows satisfactory performance with lower computational resources requirements. The state estimation algorithm is able to estimate the concentration profiles within the cell and the SEI growth during the cell operation. Due to its relatively low computational requirements, our state estimation algorithm could be solved online in a BMS to accurately monitor and predict SOC, SOH and instantaneous power capability. Acknowledgements This work is funded by Samsung Electronics Co. Ltd. through a Global Research Outreach award.

Electric Vehicles (EVs) have transformed the automotive industry and are becoming a more reliable and consistent mode of public transportation. The development of a pollutionfree environment and improved ecological surroundings is being significantly assisted by battery-powered vehicles. Lithium-ion (Li-ion) batteries are the most widely used type of batteries in EVs because of their superior performance as compared to their counterparts. The core of EVs is their battery management systems (BMS), which can unarguably improve a battery’s performance, operation, safety, and lifespan. Li-ion battery state estimation is one of the most important parts of the implementation of BMS, as it serves an important role in safe and reliable battery operation. Recently, researchers are working on the development of digital twin models to automate and optimize the BMS state estimation process by utilizing machine learning (ML) algorithms and cloud computing. The objective of this study is to review, characterize, and compare various ML-based approaches for the state estimation of different Li-ion battery states. Firstly, this study describes and characterizes several Li-ion battery state estimation approaches proposed in recent years. Secondly, the battery state estimation of electric vehicles is discussed. In addition, the challenges and prospects of Li-ion battery state estimation are put forward.

A naive Bayes model for robust remaining useful life prediction of lithium-ion battery