Physics-based Modeling Of Li-Ion Batteries And Inconsistency Analysis Of Parallel Modules

On 9 October 2019,the Royal Swedish Academy of Sciences awarded this year’s Nobel Prize in Chemistry to John Goodenough,Stanley Wittingham,and Akira Yoshino for their contribution to the "development of lithium-ion batteries".There is no doubt that the technological innovation of lithium-ion batteries has given the automotive industry a significant boost into the new era of electrification.As the "heart" of the electric vehicles,the onboard battery system is usually made up of hundreds of lithium-ion cells connected in parallel and then in series to meet the power and energy requirements of the vehicles.The Battery Management System(BMS)acts as the "brain",managing the enormous onboard battery system in real-time to ensure safe and efficient operation.The multi-spatial scale information of the battery system is the critical basis for the BMS to perform its functions.However,due to cost and technical limitations,the current BMSs can only measure each parallel module’s total current and voltage and the temperatures at critical locations in the battery system.The internal micro-states of the battery cells and the inconsistencies among parallel modules cannot be obtained directly by the BMSs,which undoubtedly poses a severe challenge to the intelligent battery management oriented toward health and safety.On the one hand,changes in micro-states such as lithium-ion concentration,solid/liquid phase potential,and kinetic reaction rate within the battery cells are a symptom of charging/discharging,aging,and failure.Based on this information,control algorithms that sense the micro-state of the battery can be designed in BMSs.On the other hand,cell-to-cell inconsistencies in current,temperature,and aging in parallel modules can easily lead to localized cell abuse(overheating,overcharging,over-discharging,etc.),thus making the lifespan of parallel module much lower than that of the single-cell level.To address these two limitations,this thesis takes the single battery cell and parallel battery module as objects.It focuses on the physics-based modeling of lithium-ion battery cells and the inconsistency analysis of parallel battery modules.In terms of the physics-based modeling of battery cell:(1)To address the problems of low model accuracy and significant estimation errors caused by the nonlinear solid-phase diffusion,an equivalent circuit model considering the effect of nonlinear solid-phase diffusion is proposed for lithium-ion batteries.The model is developed based on the first principles of the solid-phase diffusion process.By introducing the concepts of the surface state of charge(SOC)and average SOC,the solid-phase diffusion overvoltage is expressed as the open circuit voltage difference between the surface SOC and the average SOC.The solid-phase diffusion equivalent circuit parameters are regarded as functions of the surface SOC.The model parameters are identified based on the voltage recovery period of the pulse discharge test.The experimental results verify that the model has the advantages of high accuracy,good robustness,and fast calculation speed.(2)To address the computational cost and online application problems,a simplified electrochemical model with prediction accuracy and computational speed is developed.The model is represented as a fifth-order diagonal system,in which the idea of computational domain decomposition is combined with existing polynomial approximation and single particle assumption for the first time in the simplification of electrolyte diffusion,and only two independent state variables represent the electrolyte diffusion process.The simulation and experimental results show that the model can guarantee the prediction accuracy of macro and micro variables and improve the computational speed at least 600 times compared with the traditional model.(3)To address the problem of the identification difficulties caused by the numerous parameters of the electrochemical model,a parameter identification method over a wide temperature range is established.The model is generalized to eliminate redundant parameters,and the effect of lithium-ion concentration on solid-phase diffusion is deliberately considered to accurately describe the lithium-ion diffusion mechanism in electrode particles.Subsequently,the dynamics of the battery cell are analyzed,and the unknown parameters are grouped,and then the specific tests are designed for stepwise parameter identification.The discharge tests over a wide temperature range(-20℃～45℃)show the effectiveness of the parameter identification method.In terms of the inconsistency analysis of parallel module:(1)To address the problems of low efficiency and short lifespan caused by cell-to-cell inconsistencies within the parallel battery module,the influencing factors of inconsistency in parallel modules were explored,and the parallel battery modules and their discharge test schemes were designed respectively taking into account the influence of parameter variation,environmental factor and electrical connection;parameters such as open-circuit voltage and internal resistance of cells were extracted during the discharge process,and the relationship between the different parameters and the current distribution was analysed to reveal the evolution law of the current distribution.(2)Considering that the design of battery modules in practical applications is rather complex and the cell-to-cell inconsistency is influenced by the coupling factors,the inconsistency analysis and comprehensive evaluation of air-cooled parallel modules with different parallel topologies are carried out.A multi-physics model is introduced to describe the electrochemical,aging,electrical and thermal characteristics of the modules,and the model is validated comprehensively at the single-cell and multi-cell levels.Furthermore,constant current discharge and cycling simulations are carried out for different connected modules to analyze the cell-to-cell inconsistencies in current,temperature and aging,and parametric studies are carried out regarding the inlet air velocity,cell spacing and interconnection resistance.Six performance indicators are extracted from the simulation results to represent different aspects of module performance,and the comprehensive evaluation results show the superiority of the cross-configured parallel topology.The research on the mechanism modeling of li-ion battery cells and the inconsistency analysis of parallel battery modules emphasizes the importance of the internal micro-states of battery cells and the inconsistent information of parallel modules for the safe and efficient operation of battery systems,laying a solid foundation for the continuous innovation of theories,methods and technologies for on-board battery control,and advancing the intelligent process of battery management systems.