Intelligent Battery Management System Research Articles

Prediction of the Remaining Useful Life of Lithium-Ion Batteries Based on the 1D CNN-BLSTM Neural Network

Lithium-ion batteries are currently widely employed in a variety of applications. Precise estimation of the remaining useful life (RUL) of lithium-ion batteries holds significant function in intelligent battery management systems (BMS). Therefore, in order to increase the fidelity and stabilization of predicting the RUL of lithium-ion batteries, in this paper, an innovative strategy for RUL prediction is proposed by integrating a one-dimensional convolutional neural network (1D CNN) and a bilayer long short-term memory (BLSTM) neural network. Feature extraction is carried out through the input capacity data of the model using 1D CNN, and these deep features are used as the input of the BLSTM. The memory function of the BLSTM is applied to retain key information in the database and to better understand the coupling relationship among consecutive time series data along the time axis, thereby effectively predicting the RUL trends of lithium-ion batteries. Two different types of lithium-ion battery datasets from NASA and CALCE were used to verify the effectiveness of the proposed method. The results show that the proposed method achieves higher prediction accuracy, demonstrates stronger generalization capabilities, and effectively reduces prediction errors compared to other methods.

New Method for Designing Robust External Short-Circuit Protection Systems at Different Scales

External short circuit is one of the failure modes of Li-ion and other batteries that potentially leads to serious damage such as, overheating and/or thermal runaway. Therefore in this work we test and analyse the short circuit behaviour of different system sizes from coin cells, 10Ah pouch cells and upto large scale automotive high voltage battery packs. It was found that the complex short circuit behavior can be described by 3 regions independently of the size of the battery. In the first region 274C-rate is observed which is mainly governed by the cell's double and diffusion layer discharge. In the second region, the current drops significantly to 50–60C-rate where mass transport becomes the current limiting factor. The maximum temperature is reached and cell rupture, venting and electrolyte leakage may occur. Then the current stabilizes at the maximum mass transport region independently of the external short circuit. This complex time behavior need to be covered by different fuses, circuit breaker and intelligent battery management system. The protection strategy depends on several factors, which makes the design process complex. We introduce the normalized resistance ratio which allows the comparison of external shorts with different external resistance and cells and with different capacities. We demonstrate that the estimation of the maximum external short circuit current derived in this work is more accurate than other calculations at different levels. Furthermore we developed and present a generalized visualisation of the full time-current window which potentially help scientists and engineers to design more robust external short circuit protection systems. With this method we also demonstrate that the currently used external short circuit design methods may potentially lead to unsafe regions especially in the soft short region. We also propose a testing protocol which can be applied in standards, regulations without jeopardizing the safety of testing personnels and infrastructure.

System identification and state estimation of a reduced-order electrochemical model for lithium-ion batteries

Lithium-ion batteries commonly used in electric vehicles are an indispensable part of the development process of decarbonization, electrification, and intelligence in transportation. From intelligent designing, manufacturing to controlling, an intelligent battery management system plays a crucial role in the long life, high efficiency, and safe operation of lithium-ion batteries. As a first-principle model, the electrochemical parameters of the electrochemical model have physical meanings and reflect the internal state of the lithium-ion batteries. The application of electrochemical models in an advanced intelligent battery management system is a future trend that promises to mitigate battery life degradation and prevent safety incidents. The reduced-order electrochemical model is expected to alleviate the requirements of advanced battery management systems for high accuracy and fast computing of lithium-ion battery models. However, the existing model order reduction methods have the drawbacks of high computational complexity and small application scope, so that inconvenient to apply onboard. In order to solve the existing obstacles, this paper applies the pseudo-spectral method to solve the solid-phase diffusion equation, while the liquid-phase concentration equation is simplified by the Galerkin method. Subsequently, a particle swarm optimization algorithm is used to identify 11 parameters of the electrochemical model. To further improve the accuracy of the electrochemical model, the above system identification method is applied to segment identification, especially for high or low state-of-charge (SoC) conditions in this study. Finally, based upon the derived model, estimation of SoC is performed using a particle filter. The results show that the proposed reduced-order electrochemical model achieves a low Mean Absolute Error (MAE) of 8.4 mV and a MAE of 0.54 % on estimation of SoC based on the envisaged particle filter. This work is expected to provide the basis for the subsequent development of lithium-ion battery electrochemical models with smaller identification parameters and faster identification processes.

SoC-Modified Core Temperature Estimation of Lithium-Ion Battery Based on Control-Oriented Electro-Thermal Model

Lithium-ion batteries and their control technologies are the key points to electrification and intelligence of transportation. Dynamic thermal management is one of the key technologies for intelligent battery management systems. Real-time monitoring of information about the temperature characteristics inside the battery is important for effective and safe thermal management. This paper firstly constructs a distributed control-oriented electro-thermal coupling model which contains multidimensional internal information about the cell. Based on the proposed model, improved parameter identification methods are used to construct offline database of model parameters. The electrical and thermal parameters are identified by applying recursive least squares with variable forgetting factor (VFFRLS) and particle swarm optimization (PSO) separately. Finally, a SoC-modified core temperature estimation method is proposed, which adopts discrete-time nonlinear observer (DNLO) to modify SoC and adapative Kalman filter (AKF) to estimate core temperature. The method takes into account the sensitivity of the output results to nonlinear, time-varying battery systems. The results show that the root-mean-square error (RMSE) of SoC estimation is 1.75 % and the mean absolute error (MAE) is 0.65 % for the proposed temperature method under wide temperature points (-5, 25, 45 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$^{\circ }$</tex-math></inline-formula> C). The proposed core temperature estimation method possesses better robustness and universality, with RMSE of 0.61 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$^{\circ }$</tex-math></inline-formula> C and MAE of 0.56 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$^ {\circ }$</tex-math></inline-formula> C. Compared with the open-loop prediction method, the accuracy is improved about 0.5 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$^ {\circ }$</tex-math></inline-formula> C under extreme loadfiles with uncertainty.

State of health estimation of lithium-ion battery using energy accumulation-based feature extraction and improved relevance vector regression

The state of health (SOH) of lithium-ion battery is an important component of intelligent battery management system. Precise SOH estimation provides feasible and safe guidance for the energy system driven by lithium-ion battery. A novel SOH estimation method using energy accumulation of equal discharge voltage difference (EAEDVD) features extraction and improved relevance vector regression (RVR) is developed. Firstly, the EAEDVD based health feature is extracted from the discharge process as the parameters to describe battery aging, which considers the limitation of incomplete discharge greatly hindering the possibility of extracting traditional aging feature from the whole cycle. Then, the health features with high correlation are selected via integrated correlation analysis from EAEDVD curve smoothed by filter method, which eliminate redundant information. Second, this paper constructs a RVR model to estimate SOH, aiming at the problem of RVR model parameter selection, sparrow search algorithm (SSA) is developed to optimize the model parameters for improving the performance via finding the optimal solution. Finally, the feasibility is verified based on two battery datasets from NASA and laboratory. For the four batteries in the NASA dataset, the error indicators are all within 1 %, and for the two batteries from the laboratory, the mean error indicators are 0.75 %, 1.05 % and 1.05 % respectively, which indicates that proposed method has high accuracy, strong robustness and applicability.

An imperative role of studying existing battery datasets and algorithms for battery management system

Numerous portable technologies, including electric vehicles, cell phones, and laptops, are powered by batteries. The use of batteries is increasing due to the widespread usage of battery energy storage in the generation of renewable energy. This has also resulted in an increase in the number of negative incidents related to batteries and had a significant negative economic impact on industries. The shortcomings in the traditional monitoring of lithium-ion batteries have been overcome through the implementation of advanced technologies. However, few studies have discussed the use of distinct datasets in the implementation of an intelligent battery management system (BMS). This paper presents a discussion on the choice of dataset variables and applied algorithms for the implementation of effective BMS with artificial intelligence (AI) and machine learning (ML). The study analyzed the use of different datasets, including the National Aeronautics and Space Administration (NASA) battery dataset, to improve BMS. It found that the dataset variables must include the terminal voltage, terminal current, charge current, charge voltage, internal resistance, temperature, and cycle to calculate the state of health (SoH). In future, BMS hardware will be developed to obtain more precise results using AI and ML-based prediction models, utilizing the selected dataset characteristics and variables to achieve longer battery life.

Pulse charging based intelligent battery management system for electric vehicle

Electric vehicles (EVs) are now an important part of the automotive industry for two main reasons: decreased reliance on oil and reduced air pollution, which helps us contribute to the development of an environmentally friendly environment. EV buyers examine overall vehicle mileage, recharge time, vehicle mileage after every charge, batteries charging/discharging security, lifespan, charged rate, capability, and temperature increase. A new improved pulse charging technique is proposed, in which the battery is charged using proportional integral derivative (PID) control action and a neural network. A PID controller is used to develop the charging unit in this design. The feed forward neural network was used to determine the values of the PID control parameters. The battery management system (BMS) ensures that this designed battery charging system takes less time to charge the battery efficiently. The system is built with MATLAB/Simulink.

Pulse charging based intelligent battery management system for electric vehicle

Electric vehicles (EVs) are now an important part of the automotive industry for two main reasons: decreased reliance on oil and reduced air pollution, which helps us contribute to the development of an environmentally friendly environment. EV buyers examine overall vehicle mileage, recharge time, vehicle mileage after every charge, batteries charging/discharging security, lifespan, charged rate, capability, and temperature increase. A new improved pulse charging technique is proposed, in which the battery is charged using proportional integral derivative (PID) control action and a neural network. A PID controller is used to develop the charging unit in this design. The feed forward neural network was used to determine the values of the PID control parameters. The battery management system (BMS) ensures that this designed battery charging system takes less time to charge the battery efficiently. The system is built with MATLAB/Simulink.

Battery State of Health Estimate Strategies: From Data Analysis to End-Cloud Collaborative Framework

Lithium-ion batteries have become the primary electrical energy storage device in commercial and industrial applications due to their high energy/power density, high reliability, and long service life. It is essential to estimate the state of health (SOH) of batteries to ensure safety, optimize better energy efficiency and enhance the battery life-cycle management. This paper presents a comprehensive review of SOH estimation methods, including experimental approaches, model-based methods, and machine learning algorithms. A critical and in-depth analysis of the advantages and limitations of each method is presented. The various techniques are systematically classified and compared for the purpose of facilitating understanding and further research. Furthermore, the paper emphasizes the prospect of using a knowledge graph-based framework for battery data management, multi-model fusion, and cooperative edge-cloud platform for intelligent battery management systems (BMS).

A novel pulse charger with intelligent battery management system for fast charging of electric vehicle

Electric vehicles contribute a major role in building an eco-friendly environment. Li-ion batteries are most widely used in electric vehicles. It is very important to maintain the operation of Li-ion batteries within their “safety operation area (SOA)”. Hence implementing a battery management system (BMS) becomes a necessity while using Li-ion batteries. This paper proposes an intelligent BMS for electric vehicles using proportional integral derivative (PID) control action along with artificial neural network (ANN). It prefers the improved pulse charging technique. The design consists of a battery pack containing four 12 V Li-ion batteries, MOSFETs, Arduino Uno, a transformer, a temperature sensor, a liquid-crystal displays (LCD), a cooling fan, and four relay circuit are used. Arduino Uno is used as a master controller for controlling the whole operation. Using this design approximately 38 minutes are required to fully charge the battery. Implementation results validate the system performance and efficiency of the design.

Distributed Intelligent Battery Management System Using a Real-World Cloud Computing System.

In this work, a decentralized but synchronized real-world system for smart battery management was designed by using a general controller with cloud computing capability, four charge regulators, and a set of sensorized battery monitors with networking and Bluetooth capabilities. Currently, for real-world applications, battery management systems (BMSs) can be used in the form of distributed control systems where general controllers, charge regulators, and smart monitors and sensors are integrated, such as those proposed in this work, which allow more precise estimations of a large set of important parameters, such as the state of charge (SOC), state of health (SOH), current, voltage, and temperature, seeking the safety and the extension of the useful life of energy storage systems based on battery banks. The system used is a paradigmatic real-world example of the so-called intelligent battery management systems. One of the contributions made in this work is the realization of a distributed design of a BMS, which adds the benefit of increased system security compared to a fully centralized BMS structure. Another research contribution made in this work is the development of a methodical modeling procedure based on Petri Nets, which establishes, in a visible, organized, and precise way, the set of conditions that will determine the operation of the BMS. If this modeling is not carried out, the threshold values and their conditions remain scattered, not very transparent, and difficult to deal with in an aggregate way.

Combining Multi-Scale Convolutional Neural Network with Long Short-Term Memory Neural Network for State of Charge Estimation of Lithium-ion Batteries

To develop safe and intelligent battery management systems for electric vehicles, it is necessary to accurately estimate the state of charge (SOC) of lithium-ion batteries. At present, deep learning methods have been broadly applied in the field of SOC estimation of lithium-ion batteries. However, existing deep SOC estimators are difficult to capture global trends due to being too sensitive to the changes of continuous time data points. In addition, a single non-linear neural network tends to ignore the linear features of the data, which makes the robustness of the estimation not good enough. To address these two issues, this paper proposes a SOC estimation method by combining multi-scale convolutional neural network with long short-term memory neural network (called MCNN-LSTM). Specifically, on the one hand, multiple one-dimension convolutions with different dilation rates are used to extract features at different time scales, and LSTM neural networks are used to acquire the long-term dependencies of the data. On the other hand, an additional fully connected layer is used to extract the linear features, which reduces the volatility of the estimation and enhances the robustness of the estimation. The results suggest that MCNN-LSTM has higher estimation accuracies and better robustness than DNN, LSTM, and CNN-LSTM.

Lithium-ion cells for space applications: Aspects of durability

Lithium-ion cells (LIC) are emerging as the critical energy source in space applications for spacecraft. The analysis of the failure of small satellites reveals that 41.3% have either suffered partial or total failures in the space environment around Earth, resulting in the accumulation of debris. The defects in batteries caused more failures than thrusters. The present paper details a framework for ensuring the durability of LIC systems in space applications. It is discussed in three verticals viz. a) Durability through Design' that gives prominence to best practices in cell chemistry selection, processing and, design through parametric modelling; b) 'Durability through quality control and testing by the manufacturer and by identifying relevant reliability models; c)'On-orbit management' through the State of Health (SOH) assessment by Random Forest algorithm and a well-equipped intelligent Battery Management System (BMS). Lacuna in standards for cell development process is brought out. Importance of quantitative approach for design, reliability and asset management during space application is highlighted. Use of agile, contemporary techniques of artificial intelligence in this quantitative approach is also discussed. The framework tries to bring out a systematized protocol to be followed during the development and operational life cycle of LIC for ensuring highly reliable systems for space application.

A Comprehensive Review on Optimization and Artificial Intelligence Algorithms for Effective Battery Management in EVs

Globally, research on battery technology to be utilized in electric vehicle applications is rapidly expanding to solve the problems of greenhouse emissions and global warming. The efficiency of Electric Vehicles (EVs) are highly depends on the precise measurement of significant factors, as well as on the appropriate operation and analysis of the battery storage system. Unfortunately, inadequate battery storage system monitoring and safety measures can result in serious problems such battery over-charging, over-discharging, overloading, imbalanced cells, heat explosion, and combustion hazards. The quantity of a battery’s energy in respect to its capability is described to as the state of charge (SOC). SOC is measured in percentage points and is estimated as the distance between the battery’s maximum possible output and its average energy at a specific time under the same issues. State of health (SOH) is the evaluation of a battery’s maximum charge amount compared to its starting value when it is first discharged. SOH is calculated using percentage points as its variables. An efficient battery management system, which includes tailored to the content, charging-discharging control, thermal regulation, battery protection and security, is essential for addressing these issues. This paper’s objective is to provide a thorough analysis of various intelligent control strategies and battery management system methodologies used in the EV applications. Also, the review assesses the smart algorithms for estimating battery state in terms of their attributes, customization, arrangement, accuracy, benefits, and drawbacks. Finally, prospects and directions for developing a successful sophisticated algorithm and controller are presented in order to create an enhanced battery management system for applications in future, eco-friendly EV technology.

A Composite State of Charge Estimation for Electric Vehicle Lithium-Ion Batteries Using Back-Propagation Neural Network and Extended Kalman Particle Filter

Accurate estimation of battery state of charge (SOC) plays a crucial role for facilitating intelligent battery management system development. Due to the high nonlinear relationship between the battery open-circuit voltage (OCV) and SOC, and the shortcomings of traditional polynomial fitting approach, it is an even more challenging task for predicting battery SOC. To address these challenges, this paper presents a composite SOC estimation approach for lithium-ion batteries using back-propagation neural network (BPNN) and extended Kalman particle filter (EKPF). First, a second order resistance capacitance model is established to make parameters identification of a lithium-ion battery cell using recursive least squares algorithm with forgetting factors (FFRLS) approach. Then, BPNN is used to fit the desired OCV-SOC relationship with relatively high precision. Next, by incorporating the extended Kalman filter (EKF) into the particle filter (PF), an expected EKPF approach is presented to realize the SOC estimation. Last, the performances of SOC estimation using different methods, namely the PF, EKF and the EKPF are compared and analyzed under constant current discharge and urban dynamometer driving schedule working conditions. The experimental results show that the proposed method has higher accuracy and robustness compared to the other two SOC estimation methods.

Flexible battery state of health and state of charge estimation using partial charging data and deep learning

Accurately monitoring battery states over battery life plays a central role in building intelligent battery management systems. This study proposes a flexible method using only short pieces of charging data to estimate both maximum and remaining capacities to simultaneously address the state of health and state of charge estimation problems. Different from conventional studies based on specific operating data to estimate one state, the proposed method is based on a convolutional neural network that only requires short-term charging data to estimate two states. The proposed method is first validated based on the degradation data of eight 0.74 Ah batteries. We show that the maximum and remaining capacities can be accurately estimated using arbitrary pieces of 1 C charging data collected within 400 s over battery life, and the root mean square error is lower than 12.68 mAh. The influence of the input data length and different loss weights of the two states is investigated to demonstrate the high flexibility of the proposed method. Interestingly, it is observed that the simultaneous estimation of two states achieves higher accuracy than individual state estimation. Further validations on other two types of batteries reveal that the proposed method can ensure reliable estimation in the cases of different battery chemistries and different working conditions. Our method offers a flexible and easy-to-implement approach to achieving an accurate estimation of multiple states over battery life.

Physics-Informed Recurrent Neural Networks with Fractional-Order Constraints for the State Estimation of Lithium-Ion Batteries

The state estimation of lithium-ion battery is the basis of an intelligent battery management system; therefore, both model-based and data-driven methods have been designed and developed for state estimation. Rather than using complex partial differential equations and the complicated parameter tuning of a model-based method, a machine learning algorithm provides a new paradigm and has been increasingly applied to cloud big-data platforms. Although promising, it is now recognized that big data for machine learning may not be consistent in terms of data quality with reliable labels. Moreover, many algorithms are still applied as a black box that may not learn battery inner information well. To enhance the algorithm generalization in realistic situations, this paper presents a fractional-order physics-informed recurrent neural network (PIRNN) for state estimation. The fractional-order characteristics from battery mechanism are embedded into the proposed algorithm by introducing fractional-order gradients in backpropagation process and fractional-order constraints into the convergence loss function. With encoded battery knowledge, the proposed fractional-order PIRNN would accelerate the convergence speed in training process and achieve improved prediction accuracies. Experiments of four cells under federal urban driving schedule operation conditions and different temperatures are conducted to illustrate the estimation effects of the proposed fractional-order PIRNN. Compared to the integer-order gradient descent method, the fractional-order gradient descent method proposed in this work can optimize network convergence and obtains regression coefficient larger than 0.995. Moreover, the experimental results indicate that the proposed algorithm can achieve 2.5% estimation accuracy with the encoding fractional-order knowledge of lithium-ion batteries.

SOH prediction of lithium battery based on IC curve feature and BP neural network

Precise battery SOH (state of health) prediction and monitoring are of extreme importance for the future intelligent battery management system (BMS). In this paper, battery discharge experiments at different temperatures were carried out. A battery SOH prediction model based on incremental capacity analysis and BP neural network is proposed to predict battery SOH at different ambient temperatures. By analyzing the correlation between the characteristics of IC curve and SOH, the mapping relationship between temperature and IC curve characteristics is established by using the least square method, and the SOH prediction model at different temperatures is obtained. At the same time, combined with ICA, an online real-time correction prediction model is established, and the characteristic data is continuously updated to ensure the SOH prediction accuracy under different aging states. Finally, the feasibility of the prediction method proposed in this paper is verified by comparing the model test results and experimental results, the average error of the model prediction results is 1.16%. Thus, by establishing the relationship between temperature and IC curve characteristics, the battery SOH at different temperatures can be predicted.

Optimizing battery charging efficiency and longevity through smart charging techniques in microcontroller-based systems

This paper explores the topic of optimizing battery charging efficiency and longevity through smart charging techniques in microcontroller-based systems. With the increasing demand for portable and wireless devices, efficient battery charging and longer battery life have become essential. The use of microcontroller-based systems allows for more precise and effective control of battery charging, resulting in improved efficiency and longevity. The current research is on intelligent battery charging systems and battery management systems and discusses various smart charging techniques that can be employed in microcontroller-based systems, including fuzzy logic control, model predictive control, and hybrid algorithms. The paper also highlights the importance of battery management and the need for a comprehensive battery management system to optimize charging efficiency and battery longevity. Finally, the paper presents several case studies and experimental results to demonstrate the effectiveness of these smart charging techniques in improving battery charging efficiency and longevity in microcontroller-based systems.

An Intelligent BMS With Probabilistic MO-GSA Based CDMAS Integrating Edge Controller Analytics

One of the most challenging facets of any Battery Management System is scheduling the charging and discharging cycles of each battery without compromising the uninterruptible power supplies to meet demand. A battery-powered system may include several batteries of varying sorts, models, makes, sizes, and lifespans, etc which employs a diversity of charging techniques. Subsequently, each battery would have its own charging-discharging path. Even when a battery of the same capacity and make is used with the same load profile, the charging curves differ. This is because batteries have differential electrochemical characteristics and deteriorate to some extent with each use. Accordingly, the Battery Management System should schedule the batteries so that degradation and usage are kept to a minimum. This paper developed an adaptive Intelligent Battery Management System that can schedule batteries with minimal power loss, increased battery life, and higher financial benefit, even when batteries of various sizes, capacities, production types, lifespans, charge cycle models etc are incorporated. Multi-zonal approaches are used, combining the benefits of edge analytics and Centralized Data Management and Analytics System. Every millisecond, the battery parameters will be monitored using an energy monitoring circuit integrated with an edge controller. However, the edge controller alone will not be able to process such a vast volume of data on its own, the data will be divided into two categories and analyzed in two phases. All of the big data is delivered to a Centralized Data Management and Analytics System using Low-power wide-area network, and this data is labeled as primary data. A second set of data is extracted from big data within the edge controller via metadata processing before it is transmitted to Centralized Data Management and Analytics System. This secondary data is processed against safety standards stored in read only memory and rapid judgments are performed using edge analytics if necessary. The Centralized Data Management and Analytics System employs a number of analytics techniques. An Auto-Regressive Integrated Moving Average method will be used to forecast the State of Charge of batteries. With the help of this forecasted data, the Multi-Objective Gravitational Search Algorithm is then used to schedule the best battery allocation based on a number of objectives such as battery temperature runaway, unit cost of consumption including span of service, last used time period, State of Charge (%), State of Charge (WH), and so on. Between Auto-Regressive Integrated Moving Average and Gravitational Search Algorithm, a Naive Bayes probabilistic estimator is encased to identify the best general population for Gravitational Search Algorithm, avoiding repeated battery swapping and improving power efficiency. The whole device is evaluated in Hardware in a Loop model. When comparing the performance of the developed model to that of other optimization models, it is evident that Gravitational Search Algorithm outperforms other methods when population is constrained.