Lithium-ion (Li-ion) batteries have become a crucial factor in the recent electro-mobility trend. People's increased interest in electric vehicles (EVs) has motivated several automotive manufacturers and research organizations to develop suitable drivetrain designs involving batteries. Especially the development of the 48V Li-ion battery has been of great importance to reduce CO2 emissions and meet emission standards. However, accurately modeling Li-ion batteries is a difficult task since multiple factors have to be considered. Conservative Methods are using pyhsico-chemical models or electrical circuits in order to mimic the battery behavior. This thesis deals with developing a Li-ion battery model using artificial neural network (ANN) algorithms to predict the state of charge (SOC) as one of the key battery management system states. Due to the rising power of GPUs and the amount of available data, ANNs became popular in recent years. ANNs are also applicable to different areas of battery technology. Using battery data like the battery voltage, temperature, and current as input features, a neural network is trained that predicts battery SOC. A novel approach based on ANNs and one of the most commonly used SOC estimation methods are presented in this thesis to model the battery behavior. Furthermore, an approach for dealing with the highly unbalanced data by creating multidimensional bins and compare different neural network architectures for time series forecasting is introduced. By creating the model, our main priority is to reduce the model's errors in extreme operating areas of the battery. According to our results, long short-term memory (LSTM) architectures appear to be the best fit for this task. Finally, the developed ANN model can successfully learn battery behavior, however the model's accuracy under harsh operating circumstances is highly dependent on the data quality gathered.

The functionality to both charge and discharge energy from and to the power grid to a Battery Electric Vehicle (BEV) is referred to as Vehicle-to-Grid (V2G). This allows the customer to buy energy when the spot price is low and sell energy when the price is high to make a profit, called energy arbitrage. However, when the battery is charging, discharging, or idling for storage, battery degradation occurs due to chemical properties and reactions. This thesis developed a mathematical optimisation model in Python, using the modelling language Pyomo. Mathematical equations are used to integrate energy arbitrage and degradation data to reduce the total cost in terms of degradation and energy by finding an optimised charge and discharge pattern. The model allows different user scenarios to be analysed by changing inputs such as charger power, battery cost or daily driving distance. When using V2G technology, the State-of-Charge (SoC) level of BEVs battery packs can be adjusted to find SoC levels which minimise the battery degradation, while allowing the user to make a profit from energy arbitrage. The result shows that the V2G charging protocol, compared to protocols without a bidirectional charger could be beneficial for the simulated time periods, by both reducing degradation and the total energy cost. The results also indicate that the degradation cost of the battery is often the determining factor in the decision of when to charge or discharge, i.e., the substantial cost-saving strategy is to control the storage and cycle degradation to reduce the total degradation, rather than controlling the energy arbitrage. The model and the result of this thesis can be used by car manufacturers to learn more about how battery cell types behave in V2G mode and influence further work on V2G control.

The electric vehicle industry is experiencing a boom infunding and public interest, and the formula student movementis following suit; an electric race car is currently being developedby the KTH Formula Student organisation (KTHFS) which is thecause of this work.Consumers desire increased speed and range, and are unwillingto compromise one quality for the other. This necessitates the useof lithium ion cells, which may explode and exhume toxic gasesif over-strained with respect to current, charge or temperature.A robust, reliable and provably safe battery management systemshould therefore be developed. There are numerous methods tofurther increase the mileage to get an edge on competitors, suchas cell balancing and live estimation of the State of Charge(SOC). It is also vital that old and/or deteriorated cells should beidentified and disposed off in due time, and State of health (SOH)estimation provides a means to do this. In this paper a completebattery management system software solution is developed andpresented, utilising methods like simulation and code generationto create a program that runs on a real time operating system(RTOS). Some real world test were conducted and some resultsare simulated. The finished BMS performed well in tests, meets allgoals and meets all timing constraints. The project can thereforebe considered as successful. / Intresset för elbilsindustrin har på sistone‌ vuxit något markant, och formula student-rörelsen har anpassat sig efter dessa trender; en elektriskt bil tillverkas just nu av KTH Formula Student organisationen (KTHFS) vilket ger upphov till detta arbete. Marknaden vill ha snabbare bilar som dessutom har förbättrad räckvidd, men vägrar offra den ena egenskapen för det andra. Lösningen är att använda litiumjonceller. Dessa har dock en säkerhetsrelaterad nackdel; om cellerna utsätts för alldeles för höga eller låga temperaturer, strömmar eller laddningsnivåer kan de explodera och utsöndra giftig gas i luften. Därför är det lämpligt att skapa ett batterimonitoreringssystem vars funktion och säkerhet kvalitativt kan utvärderas och bevisas. Det finns flera metoder för att få förbättrad prestanda ur sin ackumulator (batteriensemble); cellnivåbalansering och laddningsnivåestimering (SOC) implementeras i detta projekt. Föråldrade/utslitna celler bör identifieras och avskrivas i god tid. Celldeklineringsestimering (SOH) är ett sätt att lösa detta problem. I denna rapport presenteras en fullständig implementation av mjukvaran för ett batterimonitoreringssystem, där metoder som kodgenerering och simulering utnyttjas för att skapa ett program som kan köras på ett realtidsoperativsystem (RTOS). Vissa test gjordes i verkligheten och vissa resultat simulerades. Det färdiga batterimonitoreringssystemet presterade väl i test, alla mål samt mötte alla tidskrav. Projektet kan därför anses som lyckat. / Kandidatexjobb i elektroteknik 2022, KTH, Stockholm

[pt] Por conta dos rápidos avanços tecnológicos, percebe-se uma mudança nos hábitos e das necessidades das pessoas. Há uma dependência cada vez maior de aparelhos eletrônicos como smartphones, notebooks etc. Construir baterias com grande capacidade energética é um dos desafios atuais para aumentar a autonomia dos aparelhos eletrônicos. Entretanto, uma alternativa que pode ajudar a manter aparelhos eletrônicos por mais tempo longe das tomadas é o compartilhamento de baterias. Existem na literatura muitos estudos envolvendo o compartilhamento de baterias no contexto de veículos elétricos, porém não são encontradas aplicações em smartphones. Um parâmetro importante a ser monitorado neste contexto é o estado de saúde (SoH). Até o momento, não há um consenso na literatura acerca do melhor modelo para estimar o SoH de baterias devido à falta de métodos bem estabelecidos. Assim, o objetivo geral desta dissertação foi construir um modelo para estimar a curva de estado de saúde, por meio do estado de carga, com vistas a estimar a saúde de baterias de íon-lítio. O modelo proposto foi baseado em redes neurais recorrentes. Para treinar e validar o modelo, foi construído um sistema para a realização de ensaios destrutivos, sendo possível estudar o comportamento de baterias de íon-lítio ao longo de toda vida útil. O modelo proposto foi capaz de estimar o SoH das baterias estudadas com boa exatidão, sob diferentes parâmetros de carga/descarga. O diferencial do modelo são baixa complexidade computacional, mesmo envolvendo modelos de redes neurais, e serem adotados parâmetros de entrada de fácil medição. / [en] Because of the fast technological advances, there is a change in people s habits and needs. There is an increasing dependence on electronic devices such as smartphones, notebooks etc. Building batteries with great energy capacity is one of the current challenges to increase the autonomy of electronic devices. However, an alternative that can help keep electronic devices longer away from sockets is battery swap. There are many studies in the literature involving the sharing of batteries in the context of electric vehicles, but no applications are found in smartphones. An important parameter to be monitored in this context is state of health (SoH). To date, there is no consensus in the literature about the best model for estimating battery SoH due to the lack of well-established methods. Thus, the objective of this dissertation is to build a model to estimate the state of health curve, with a view to classifying the health of lithium-ion batteries, through state of charge curve, for applications involving battery swap aiming to use in smartphones. The proposed model was based on recurrent neural networks. To train and validate the model, a system was built to perform destructive tests, being possible to study the behavior of lithium-ion batteries throughout its useful life. The proposed model was able to estimate the SoH of the batteries studied with good precision, under different charge / discharge parameters. The distinction of the model is low computational complexity, even involving neural network models, and easy-to-measure input parameters are adopted.

With the continuing transition to renewable inherently intermittent energy sources like solar- and wind power, electrical energy storage will become progressively more important to manage energy production and demand. A key technology in this area is Li-ion batteries. To operate these batteries efficiently, there is a need for monitoring of the current battery state, including parameters such as state of charge and state of health, to ensure that adequate safety and performance is maintained. Furthermore, such monitoring is a step towards the possibility of the optimization of battery usage such as to maximize battery lifetime and/or return on investment. Unfortunately, possible online measurements during actual operation of a lithium-ion battery are typically limited to current, voltage and possibly temperature, meaning that direct measurement of battery status is not feasible. To overcome this, battery modeling and various regression methods may be used. Several of the most common regression algorithms suggested for estimation of battery state of charge and state of health are based on Kalman filtering. While these methods have shown great promise, there currently exist no thorough analysis of the impact of so-called filter tuning on the effectiveness of these algorithms in Li-ion battery monitoring applications, particularly for state of health estimation. In addition, the effects of only adjusting the cell capacity model parameter for aging effects, a relatively common approach in the literature, on overall state of health estimation accuracy is also in need of investigation. In this work, two different Kalman filtering methods intended for state of charge estimation: the extended Kalman filter and the extended adaptive Kalman filter, as well as three intended for state of health estimation: the dual extended Kalman filer, the enhanced state vector extended Kalman filer, and the single weight dual extended Kalman filer, are compared from accuracy, performance, filter tuning and practical usability standpoints. All algorithms were used with the same simple one resistor-capacitor equivalent circuit battery model. The Li-ion battery data used for battery model development and simulations of filtering algorithm performance was the “Randomized Battery Usage Data Set” obtained from the NASA Prognostics Center of Excellence.  It is found that both state of charge estimators perform similarly in terms of accuracy of state of charge estimation with regards to reference values, easily outperforming the common Coulomb counting approach in terms of precision, robustness and flexibility. The adaptive filter, while computationally more demanding, required less tuning of filter parameters relative to the extended Kalman filter to achieve comparable performance and might therefore be advantageous from a robustness and usability perspective. Amongst the state of health estimators, the enhanced state vector approach was found to be most robust to initialization and was also least taxing computationally. The single weight filter could be made to achieve comparable results with careful, if time consuming, filter tuning. The full dual extended Kalman filter has the advantage of estimating not only the cell capacity but also the internal resistance parameters. This comes at the price of slow performance and time consuming filter tuning, involving 17 parameters. It is however shown that long-term state of health estimation is superior using this approach, likely due to the online adjustment of internal resistance parameters. This allows the dual extended Kalman filter to accurately estimate the SoH over a full test representing more than a full conventional battery lifetime. The viability of only adjusting the capacity in online monitoring approaches therefore appears questionable. Overall the importance of filter tuning is found to be substantial, especially for cases of very uncertain starting battery states and characteristics.

The energy crisis and environmental pollution have become increasingly prominent in recent years. Lithium batteries have attracted extensive attention due to their high energy density, safety, and low pollution. To further study how the battery works, it is necessary to establish an accurate model conforming to the battery characteristics. As the core function of a battery management system(BMS), accurate state of charge(SoC) estimation dramatically improves battery life and performance. This thesis selects a ternary lithium battery in the centre for advanced life cycle engineering(CALCE) dataset for a study of cell modeling and SoC estimation. The second-order Thevenin equivalent circuit model is selected as the cell model due to a trade-off between model complexity and accuracy. The parameters to identify include OCV, internal ohmic resistance, polarized internal resistance and capacitance. They were obtained with the MATLAB toolbox at various SoC state points under different temperatures. The ‘terminal voltage comparison’ method is utilized to verify the identification's accuracy. The simulation results turn out to be satisfactory. Then cell SoC can be estimated after cell modeling. First, the principles of the Coulomb counting method, OCV method and EKF method are analyzed. The state space equations required in SoC estimation are determined by discretizing the non-linear equivalent circuit model. The simulation results are compared with the experimental results in the HPPC discharge experiment. Furthermore, the robustness of the EKF algorithm is further investigated. The results prove that the EKF algorithm has high precision, fast convergence speed and strong anti-interference capability. Last but not least, the research on battery pack SoC estimation was continued. How to expand a single cell into a battery pack is analyzed, including aggregating cells into a pack and scaling a cell model to a pack. In addition, battery pack SoC is individually estimated by the 'Big cell' method and 'Short board effect' method. The result is not so good, indicating that further work can be done to improve the SoC estimation accuracy. / Energikrisen och miljöföroreningarna har blivit allt mer framträdande de senaste åren. Litiumbatteri har väckt stor uppmärksamhet på grund av sin höga energitäthet, säkerhet och låga föroreningar. För att ytterligare studera hur batteriet fungerar är det nödvändigt att etablera en exakt modell som överensstämmer med batteriets egenskaper. Som kärnfunktionen hos BMS förbättrar noggrann SoC-uppskattning dramatiskt batteriets livslängd och prestanda. Denna avhandling väljer ett ternärt litiumbatteri i CALCE-datauppsättningen för forskning. Dessutom slutförs cellmodellering och SoC-uppskattning baserat på det. Den andra ordningens Thevenins ekvivalenta kretsmodell väljs som cellmodell på grund av en avvägning mellan modellens komplexitet och noggrannhet. Parametrarna som måste identifieras inkluderar OCV, intern ohmsk resistans, polariserad intern resistans och kapacitans. De erhölls med MATLAB-verktygslådan vid olika SoC-tillståndspunkter under olika temperaturer. Metoden "terminalspänningsjämförelse" används för att verifiera identifieringens noggrannhet. Simuleringsresultaten visar sig vara tillfredsställande. Sedan kan cell SoC uppskattas efter cellmodellering. Först analyseras principerna för Coulomb-räknemetoden, OCV-metoden och EKF-metoden. Tillståndsrymdsekvationerna som krävs vid SoC-uppskattning bestäms genom att diskretisera den icke-linjära ekvivalenta kretsmodellen. Simuleringsresultaten jämförs med de experimentella resultaten i HPPC-utsläppsexperimentet. Dessutom, robustheten hos EKF-algoritmen undersöks ytterligare. Resultaten bevisar att EKF-algoritmen har hög precision, snabb konvergenshastighet och stark anti-interferensförmåga. Sist men inte minst fortsatte forskningen kring SoC-uppskattning av batteripaket. Hur man expanderar ett enskilt batteri till ett batteripaket analyseras, inklusive aggregering av celler till ett paket och skalning av en cellmodell till ett paket. Dessutom uppskattas batteripaketets SoC individuellt med "Big cell"-metoden och "Short board effect"-metoden. Resultatet är inte så bra, vilket indikerar att ytterligare arbete kan göras för att förbättra SoC-uppskattningens noggrannhet.

No description available.

In the past decade, automobile manufacturers have gone through the initial adoption phase of electric mobility. The increasing momentum behind electric vehicles (EV) suggests that electrified storage systems will play an important role in electric mobility going forward. Lithium ion batteries have become one of the most common solutions for energy storage due to their light weight, high specific energy, low self-discharge rate, and non-memory effect. To fully benefit from a lithium-ion energy storage system and avoid its physical limitations, an accurate battery management system (BMS) is required. One of the key issues for successful BMS implementation is the battery model. A robust, accurate, and high fidelity battery model is required to mimic the battery dynamic behavior in a harsh environment. This dissertation introduces a robust and accurate model-based approach for lithium-ion battery management system. Many strategies for modeling the electrochemical processes in the battery have been proposed in the literature. The proposed models are often highly complex, requiring long computational time, large memory allocations, and real-time control. Thus, model-order reduction and minimization of the CPU run-time while maintaining the model accuracy are critical requirements for real-time implementation of lithium-ion electrochemical battery models. In this dissertation, different modeling techniques are developed. The proposed models reduce the model complexity while maintaining the accuracy. The thermal management of the lithium ion batteries is another important consideration for a successful BMS. Operating the battery pack outside the recommended operating conditions could result in unsafe operating conditions with undesirable consequences. In order to keep the battery within its safe operating range, the temperature of the cell core must be monitored and controlled. The dissertation implements a real-time electrochemical, thermal model for large prismatic cells used in electric vehicles' energy storage systems. The presented model accurately predicts the battery's core temperature and terminal voltage. / Thesis / Doctor of Philosophy (PhD)

In this research, the focus was on the development of a Battery Cycler system with the primary objective of accurately predicting both the State of Charge (SoC) and State of Health (SoH) for batteries. Machine learning techniques, specifically MLP Regression, K-Nearest Neighbor, and Decision Tree Regression, were harnessed for a comprehensive analysis. The data collection and processing phase spanned 44 days. The findings underscore the potential of employing relatively uncomplicated machine learning models for the prediction of SoC and SoH. Particularly noteworthy was the strong performance of K-Nearest Neighbor, especially after deliberate optimization efforts were applied. Despite the simplicity of these techniques, the results suggest significant promise for battery management and health assessment. Nevertheless, challenges linked to temperature fluctuations and current noise were identified as factors impacting predictive performance. Mitigating these challenges is imperative to enhance the robustness and precision of predictive models in future iterations. The implications of this work extend to broader applications in battery management systems, offering insights into potential avenues for optimizing battery usage, extending longevity, and enhancing overall performance. Leveraging machine learning methodologies in a straightforward manner, this research is anticipated to contribute to advancements in battery health monitoring and management, setting the stage for more intricate models in subsequent studies. / I denna forskning har vi tagit oss an utvecklingen av ett battericykelsystem med målet att noggrant förutsäga både laddningstillstånd (SoC) och hälsotillstånd (SoH) för batterier. Genom att utnyttja maskininlärningstekniker utförde vi en omfattande analys med hjälp av MLP-regression, K-Nearest Neighbor och Decision Tree regression. Över en tidsperiod av 44 dagar samlades batteridata in och bearbetades noggrant. Resultatet understryker möjligheten att använda relativt okomplicerade maskininlärningsmodeller för att förutsäga både laddningstillstånd och hälsotillstånd. Särskilt K-Nearest Neighbor framstår som en lovande kandidat med tanke på förutsägbarhetsnoggrannheten den visat, särskilt efter medvetna ansträngningar för optimering. Trots teknikernas enkelhet antyder våra resultat en betydande potential för batterihantering och hälsobedömning. Emellertid har utmaningar som temperaturvariationer och strömbuller identifierats som faktorer som påverkar förutsägelseprestanda. Att bemöta dessa utmaningar är avgörande för att öka robustheten och precisionen i våra förutsägelsemodeller i framtida utföranden.. Denna forsknings arbetsresultat sträcker sig till bredare användningsområden inom batterihanteringssystem och erbjuder insikter i möjliga riktningar för att optimera batterianvändning, livslängd och övergripande prestanda. Genom att utnyttja maskininlärningsmetoder förutser vi att denna forskning kommer att bidra till framsteg inom övervakning och hantering av batterihälsa, och lägga grunden för mer sofistikerade modeller i framtiden.

Les véhicules hybrides se démocratisent avec une utilisation croissante des éléments de stockage à base de lithium-ion. Dans ce contexte d'exploitation, le type d'usage est atypique et dépend fortement des stratégies de répartition des énergies au sein du véhicule. Parmi les hybridations, la catégorie "mild hybrid" conserve la motorisation thermique pour l'autonomie qu'elle apporte, et lui adjoint une machine électrique associée à un élément de stockage réversible, afin de permettre une récupération de l'énergie cinétique du véhicule. L'objet de ces travaux porte sur la mise en place d'algorithmes destinés à la détermination des états de charge (SoC), de santé (SoH) et de fonction (SoF) de chacune des cellules qui compose un pack batterie lithium-ion. Ces fonctionnalités sont implantées dans un système de gestion dénommé BMS pour Battery Management System. Dans un souci de réduction des coûts de production, nos travaux s'attachent à limiter la puissance de calcul et les moyens de mesure nécessaires à la détermination de ces états. À partir de mesures effectuées lors d'une utilisation de la batterie dans une application "mild hybrid", les méthodes développées permettent la détermination des états, ainsi que d'une partie des paramètres internes aux cellules. Cette utilisation est caractérisée par de forts courants et un maintien de l'état de charge autour de 50 %, ceci afin de maximiser la disponibilité de la batterie et d'en minimiser le vieillissement. L'utilisation d'observateurs et de méthodes en boucle ouverte, à partir d'une modélisation simplifiée de cellule, nous permet d'obtenir des résultats satisfaisants avec une puissance de calcul réduite / Hybrid vehicles are developing with increasing use of energy storage elements based on lithium-ion battery. In this context, the use of battery is atypical and highly dependent on energy allocation strategies within the vehicle. Among these vehicles, the mild hybrid category retains heat engine for the autonomy that offer and adds to it an electric machine associated with a reversible storage system, to allow the kinetic energy recovery of the vehicle. The object of this work involves the development of algorithms for determining the states of charge (SoC) and health (SoH) and function (SoF) of each cell that compose a lithium-ion battery pack. These features are implemented in a Battery Management System (BMS) for industrial production. In order to reduce production costs, our work attempts to limit the computing power and the measuring sensors necessary for these states determination. From battery measurements in a "mild hybrid" use, developed methods allow the states determination, as well as some of the internal parameters of cells. This application is characterized by high currents and maintaining a SoC of around 50%, in order to maximize the availability of the battery and to minimize aging. The use of observers and estimators, using a simplified model cell, allows us to achieve satisfactory results with a reduced computing power