| Battery energy storage system is the energy buffer of smart micro-grids(MGs)and the power source of electric vehicles(EVs).In many energy storage media,lithium-ion battery(LIB)has been widely used in MGs and pure EVs due to its advantages of high power density and energy density,long cycle life,low self-discharge rate,and moderate price.However,the LIB based energy storage system is a very complex system which is composed of hundreds of LIB cells.It shows high nonlinearity,temperature sensi-tivity,obvious aging characteristics as well as inevitable inconsistency which make the monitoring and control of the LIB-based energy storage system face many challenges.Improper monitoring and control will lead to the decline of battery performance,rapid degradation,and even fire or explosion.Therefore,how to monitor the internal state parameters of batteries real-time,accurately and robustly,evaluate the aging degree quantitatively,predict the remaining life,diagnose and repair battery faults are of great theoretical and practical value to ensure and improve the safety,reliability,durability as well as energy efficiency of LIB energy storage system.Moreover,they can also promote the efficient operation and market promotion of MGs and pure EVs.The grand main problem can thus be formulated as to how can LIBs be safely,reliably and efficiently operated in practical applications.This problem can be broken down into several smaller problems:1)How to establish mathematical descriptions of the dynamic behavior of the LIBs energy storage system accurately under variable am-bient temperature and current conditions?How can we estimate the battery internal states and parameters which cannot be directly measured according to currents,tem-peratures,and terminal voltages,and the mathematical model?2)How to quantify the aging level and predict the remaining lifetime of LIBs?3)How to realize the joint esti-mation of real-time state of charge and battery aging index,and how to obtain the aging index of battery while estimating the state of charge,thus providing a quantitative basis for health assessment and life prediction?4)How to quickly detect and identify the internal faults of batteries from the battery operation data based on the battery dynamic behavior models and state estimation methods?5)On the basis of the state estimation results of cells,how to estimate the operation state of large-scale series battery packs in real time and design a scheme to avoid the performance degradation of energy storage system caused by inconsistent failure based on the state estimation results of battery packs?Motivated by the above five problems,the groundwork and main contributions of this work can be summarized as:1)Aiming at the dynamic behavior description,state estimation and on-line pa-rameter correction of LIBs under variable ambient temperature and complex current operating environment,a mathematical model for describing the dynamic characteris-tics of LIBs is established based on the first-order equivalent circuit model.And then the initial values of the model parameters are identified based on the recursive least square method.Finally,a method for model parameter self-adaptive correction and state pa-rameter estimation based on the extended Kalman filter and the proportional integral observer is proposed.This method improves the adaptability of the LIB mathematical model and the accuracy of state estimation under complex temperature conditions.2)Aiming at the aging modeling,health state quantification estimation and re-maining life prediction of LIBs,the characteristics of the current curve during constant voltage charging process are summarized according to the aging data analysis results,which are suitable for character:izing the aging degree of batteries.Secondly,a non-linear mapping model between aging characteristics and quantification indexes of bat-teries is established based on support vector regression.Furthermore,on the basis of this model and particle filter,a method of health state estimation and lifetime prediction is proposed.Finally,the validity of the model and method is verified on several aging battery data.3)Aiming at the simultaneous estimation of the state of charge and aging index of LIBs,a novel state space equation model without prior knowledge of battery parameters is proposed based on the Lyapunov observer.Then,an observer gain selection method is determined to ensure the stability and convergence of the observer based on the Lya-punov direct method.The experimental results show that this method can estimate the state of charge,as well as the internal resistance and capacity which are commonly used to quantify the aging degree based on the only measured terminal voltages and currents.4)Aiming at the thermal fault diagnosis of LIB energy storage system,first of all,an electro-thermal coupling model of battery is established based on the coupling effect of electrical and thermal behavior.Secondly,the battery internal resistance and core temperature estimation method is proposed on the basis of the Lyapunov state observer and extended Kalman filter.Finally,based on the estimation results of internal resis-tances and core temperatures,a battery thermal fault diagnosis method is proposed.The simulation results show that the proposed model and fault diagnosis method are effec-tive.5)Aiming at the performance degradation caused by series connection of cells,the core control variables affecting the inconsistency and balance control strategy of batter,ies are analyzed firstly,and then a dual-time-scale extended Kalman filter is designed to monitor the maximum and minimum state of charge in batteries in real time and reduce the computational burden.On the basis of cell state estimation and active balance topol-ogy of the bi-directional flyback transformer,a dual-granularity balance control strategy by switching current is proposed to repair the performance degradation caused by cell inconsistency.The experimental results show that the parameter estimation method can estimate the extreme value of the state of charge in battery pack effectively.Moreover,the proposed active balance control strategy can improve the overall performance of the battery pack and balance accuracy effectively,and shorten the balance time. |
PDF Full Text Request