Thermal safety boundary of lithium-ion battery at different state of charge

Thermal runaway(TR)is a critical issue hindering the large-scale application of lithium-ion batteries(LIBs).Understanding the thermal safety behavior of LIBs at the cell and module level under different state of charges(SOCs)has significant implications for reinforcing the thermal safety design of the lithium-ion battery module.This study first investigates the thermal safety boundary(TSB)correspondence at the cells and modules level under the guidance of a newly proposed concept,safe electric quantity boundary(SEQB).A reasonable thermal runaway propagation(TRP)judgment indicator,peak heat transfer power(PHTP),is proposed to predict whether TRP occurs.Moreover,a validated 3D model is used to quantitatively clarify the TSB at different SOCs from the perspective of PHTP,TR trigger temperature,SOC,and the full cycle life.Besides,three different TRP transfer modes are discovered.The interconversion relationship of three different TRP modes is investigated from the perspective of PHTP.This paper explores the TSB of LIBs under different SOCs at both cell and module levels for the first time,which has great significance in guiding the thermal safety design of battery systems.

Incomplete charge transfer in CMOS image sensor caused by Si/SiO_(2)interface states in the TG channel

CMOS image sensors produced by the existing CMOS manufacturing process usually have difficulty achieving complete charge transfer owing to the introduction of potential barriers or Si/SiO_(2)interface state traps in the charge transfer path,which reduces the charge transfer efficiency and image quality.Until now,scholars have only considered mechanisms that limit charge transfer from the perspectives of potential barriers and spill back effect under high illumination condition.However,the existing models have thus far ignored the charge transfer limitation due to Si/SiO_(2)interface state traps in the transfer gate channel,particularly under low illumination.Therefore,this paper proposes,for the first time,an analytical model for quantifying the incomplete charge transfer caused by Si/SiO_(2)interface state traps in the transfer gate channel under low illumination.This model can predict the variation rules of the number of untransferred charges and charge transfer efficiency when the trap energy level follows Gaussian distribution,exponential distribution and measured distribution.The model was verified with technology computer-aided design simulations,and the results showed that the simulation results exhibit the consistency with the proposed model.

Deep learning-based battery state of charge estimation:Enhancing estimation performance with unlabelled training samples

The estimation of state of charge(SOC)using deep neural networks(DNN)generally requires a considerable number of labelled samples for training,which refer to the current and voltage pieces with knowing their corresponding SOCs.However,the collection of labelled samples is costly and time-consuming.In contrast,the unlabelled training samples,which consist of the current and voltage data with unknown SOCs,are easy to obtain.In view of this,this paper proposes an improved DNN for SOC estimation by effectively using both a pool of unlabelled samples and a limited number of labelled samples.Besides the traditional supervised network,the proposed method uses an input reconstruction network to reformulate the time dependency features of the voltage and current.In this way,the developed network can extract useful information from the unlabelled samples.The proposed method is validated under different drive cycles and temperature conditions.The results reveal that the SOC estimation accuracy of the DNN trained with both labelled and unlabelled samples outperforms that of only using a limited number of labelled samples.In addition,when the dataset with reduced number of labelled samples to some extent is used to test the developed network,it is found that the proposed method performs well and is robust in producing the model outputs with the required accuracy when the unlabelled samples are involved in the model training.Furthermore,the proposed method is evaluated with different recurrent neural networks(RNNs)applied to the input reconstruction module.The results indicate that the proposed method is feasible for various RNN algorithms,and it could be flexibly applied to other conditions as required.

Neural Network-Based State of Charge Estimation Method for Lithium-ion Batteries Based on Temperature

Lithium-ion batteries are commonly used in electric vehicles,mobile phones,and laptops.These batteries demonstrate several advantages,such as environmental friendliness,high energy density,and long life.However,battery overcharging and overdischarging may occur if the batteries are not monitored continuously.Overcharging causesfire and explosion casualties,and overdischar-ging causes a reduction in the battery capacity and life.In addition,the internal resistance of such batteries varies depending on their external temperature,elec-trolyte,cathode material,and other factors;the capacity of the batteries decreases with temperature.In this study,we develop a method for estimating the state of charge(SOC)using a neural network model that is best suited to the external tem-perature of such batteries based on their characteristics.During our simulation,we acquired data at temperatures of 25°C,30°C,35°C,and 40°C.Based on the tem-perature parameters,the voltage,current,and time parameters were obtained,and six cycles of the parameters based on the temperature were used for the experi-ment.Experimental data to verify the proposed method were obtained through a discharge experiment conducted using a vehicle driving simulator.The experi-mental data were provided as inputs to three types of neural network models:mul-tilayer neural network(MNN),long short-term memory(LSTM),and gated recurrent unit(GRU).The neural network models were trained and optimized for the specific temperatures measured during the experiment,and the SOC was estimated by selecting the most suitable model for each temperature.The experimental results revealed that the mean absolute errors of the MNN,LSTM,and GRU using the proposed method were 2.17%,2.19%,and 2.15%,respec-tively,which are better than those of the conventional method(4.47%,4.60%,and 4.40%).Finally,SOC estimation based on GRU using the proposed method was found to be 2.15%,which was the most accurate.

温度自适应SMO算法估计锂离子电池的SOC

现有对锂离子电池荷电状态(SOC)的估计,没有考虑温度变化导致的SOC估计准确度降低。提出一种考虑温度的滑模观测(SMO)法进行SOC估计。基于混合脉冲功率测试(HPPC)实验的数据,得到18650型LiFePO4锂离子电池的SOC与温度、参数之间的拟合式,通过台风(Typhoon)系统进行半实物实验分析。温度自适应SMO算法在低温或常温工况下的平均误差较传统SMO算法降低0.3~0.5个百分点,直接通过拟合式所快速估计的SOC较温度自适应SMO算法平均误差在2%左右,常温25℃工况下误差低于1%,能够实现较高的估计精准度,为快速估计SOC提供了较好的算法参考。

引入PID反馈的SHAEKF算法估算电池SOC

电池荷电状态(SOC)的估算精度是电动汽车电池组的重要指标。为提升SOC估算精度,在融合Sage-Husa扩展卡尔曼滤波(SHEKF)算法与自适应扩展卡尔曼滤波(AEKF)算法的基础上,增加比例积分微分(PID)反馈环节,形成改进算法。采用粒子群优化(PSO)算法对二阶RC等效电路模型进行参数辨识;用开源电池数据集对模型和算法进行实验和分析。改进的SHAEKF算法在电池动态应力测试(DST)、北京动态应力测试(BJDST)和美国联邦城市驾驶(FUDS)等工况下的平均估计误差都在1%以内,与单纯的融合算法SHAEKF算法相比,最大误差可减小5%。

基于MMC的分布式储能系统及其快速SOC均衡控制策略

提高基于模块化多电平换流器(modular multilevel converter,MMC)的分布式储能系统(distributed energy storage systems,DESS)的能量利用率,解决储能子模块(energy sub-module,ESM)荷电状态(state of charge,SOC)均衡问题至关重要。针对现有的SOC均衡控制策略的不足,提出内外分层的快速SOC均衡控制策略。外层针对桥臂间或相间的SOC差异,通过改进MMC模型预测控制(model predictive predictive control,MPC),配合自适应均衡系数,快速调整功率差额。内层引入自适应虚拟电阻法,根据ESM的SOC情况确定主导ESM,自适应调节各单元的虚拟电阻,产生相应的电压梯度,结合MMC排序算法使ESM按照各自SOC进行功率分配,从而实现ESM的SOC快速均衡,提高DESS能量利用率。通过在Matlab/Simulink构建仿真模型,证明了所提控制策略的有效性和可行性。

基于多新息扩展卡尔曼滤波的锂离子电池SOC估计

锂电池具有高能量密度、循环寿命长等优点而被广泛应用于电动汽车动力装置,但车辆运行状况复杂多变,且电池内部呈现高度非线性的性质,导致电池荷电状态(state of charge, SOC)难以准确计算。为优化锂电池SOC估计精度,构建结合Warburg元件的分数阶二阶RC模型,采用自适应遗传算法进行参数辨识;融合多新息理论和扩展卡尔曼滤波算法,提出基于多新息扩展卡尔曼滤波(multi innovation extended Kalman filter, MIEKF)的锂离子电池SOC估计算法,并利用试验数据验证该方法的有效性,为提高SOC估计精度和车载锂电池的循环使用寿命提供了新的方法途径和实践支撑。

考虑SOC的混合储能功率分配与自适应虚拟惯性控制

电力电子化的直流配电网存在低惯性问题,不利于系统稳定运行。混合储能设备可向电网提供虚拟惯性,但不同类型的储能之间存在功率协调问题,并且储能的荷电状态(state of charge, SOC)对虚拟惯性的调节也有约束作用。针对上述问题,提出了一种自适应时间常数的分频控制策略,时间常数根据混合储能系统(hybridenergy storage system, HESS)的SOC而动态调整以改变功率分配。首先,通过分析储能SOC与虚拟惯性的关系,并考虑储能充放电极限问题,研究兼顾SOC、电压变化率以及电压幅值的自适应虚拟惯性控制策略,提高系统惯性。然后,建立控制系统的小信号模型,分析虚拟惯性系数对系统的影响。最后,基于Matlab/Simulink搭建直流配电网仿真模型,验证了所提控制策略能合理分配HESS功率,提高超级电容器利用率,改善直流电压与功率稳定性。

基于AR-ECM平均差异模型的串联电池组SOC、容量多尺度联合估计方法

考虑电池单体老化差异所致的电池组不一致性,针对串联电池组荷电状态(state of charge,SOC)、容量估计问题,提出一种基于自回归等效电路模型(autoregression equivalent circuit model,AR-ECM)的平均差异模型(mean-difference model,MDM)。基于此模型,提出串联电池组SOC、容量多尺度联合估计算法。该算法由2个部分组成,一是基于AR-ECM的MDM及差异化模型参数辨识策略:条件辨识策略和定频分组辨识策略;二是基于多时间尺度H无穷滤波(multi-timescale H infinity filter,Mts-HIF)的电池组SOC、容量联合估计算法。通过将所提出MDM中的自回归平均模型(autoregression mean model,AR-MM)与传统MDM中的n阶RC平均模型(nRC mean model,nRC-MM)比较,结果表明所提出的AR-MM在复杂运行工况下具有更优的动态跟随性能。依据最小化信息量准则(akaike information criterion,AIC),AR-MM具有更优的复杂度与精度的权衡。通过与基于多时间尺度扩展卡尔曼滤波(multi-timescale extended Kalman filter,Mts-EKF)联合状态估计算法比较,结果表明所提出的Mts-HIF状态估计算法具有更优的鲁棒性、精度和收敛速度。

基于LWOA-LSTM的大容量锂电池SOC估计

准确预测锂电池荷电状态(SOC)对电池安全运行至关重要,分析在电网不同模式下的SOC更是锂电池全面推广的基础。提出一种基于莱维飞行的鲸鱼优化算法(LWOA)优化长短时记忆神经网络(LSTM),对调频模式下的大容量锂离子电池SOC进行估计。首先,分析LSTM神经网络和LWOA算法,构建LWOA-LSTM模型,进行参数优化;然后,选取调频模式下大容量锂离子电池组实验数据,对数据进行预处理和模型训练;最后,实现调频模式下锂电池的SOC估计。试验结果表明:所构建模型能准确预测锂电池SOC,较WOA-LSTM模型,评估指标RMSE和MAE分别降低了25.55%、28.71%,R^(2)上升了0.76%。

基于改进AFFRLS-AUKF的锂电池SOC估计

准确估计锂电池荷电状态(SOC)是保障电池管理系统安全稳定运行的重要前提之一。为了提高锂离子电池SOC估计精度,提出一种改进自适应遗忘因子最小二乘法(AFFRLS)与自适应无迹卡尔曼滤波算法(AUKF)联合估计锂离子电池SOC的估计方法。利用改进AFFRLS对已建立的二阶RC等效电路模型进行参数辨识,再结合AUKF估计锂离子电池SOC。通过动态应力测试(DST)工况和城市道路循环(UDDS)工况验证得到联合估计方法的平均绝对误差为0.44%,均方根误差为0.61%,表明改进的AFFRLS-AUKF方法可提高参数辨识及电池SOC估计的准确性和鲁棒性。

基于分数阶模型多新息无迹卡尔曼滤波算法的超级电容SOC估计

对超级电容的SOC估计展开了研究。首先,搭建了超级电容测试平台,用于超级电容的参数辨识,并对超级电容进行了常规性能测试;其次,在不同的环境温度和动态工况下采用多种算法进行超级电容SOC估计。结果表明,采用分数阶模型多新息无迹卡尔曼滤波(FOMIUKF)算法对超级电容SOC的估计精度最高,对超级电容的路端电压跟随情况最好,估计结果的均方根误差和平均绝对误差的最大值分别约为1.8%和1.73%。

基于等效电路模型和数据驱动模型融合的SOC和SOH联合估计方法

针对电池SOC与SOH估计结果相互影响,单独估计准确度不高的问题,该文提出了一种基于等效电路模型和数据驱动模型融合的SOC和SOH联合估计方法。通过构建考虑老化和SOC的电池二阶RC等效电路模型,采用带遗忘因子的递推最小二乘法,在不同SOC和SOH的情况下,对电池的参数进行在线辨识,实现电池参数在线辨识与电池SOC和SOH估计的耦合。以锂离子电池自SOC=20%到恒流充电阶段结束所需时间为输入,电池SOH值为输出,训练GPR模型,实现电池SOH估计。将输出的SOH估计值与电池的额定容量相乘,得到电池的实际容量,更新二阶RC状态空间方程,采用扩展卡尔曼滤波算法对电池进行SOC估计,实现电池SOH估计和SOC估计之间的联合。采用牛津大学电池退化数据集和NASA随机使用电池数据集进行算法验证,结果表明,所提联合估计方法能够在电池的生命周期内较准确地跟随锂离子电池SOC和SOH的真实值。

Fuzzy Model for Estimation of the State-of-Charge of Lithium-Ion Batteries for Electric Vehicles

A fuzzy model was established to estimate the state of charge（SOC） of a lithium-ion battery for electric vehicles.The robust Gustafson-Kessel（GK） clustering algorithm based on clustering validity indices was applied to identify the structure and antecedent parameters of the model.The least squares algorithm was utilized to determine the consequent parameters.Validation results show that this model can provide accurate SOC estimation for the lithium-ion battery and satisfy the requirement for practical electric vehicle applications.

ESTIMATION METHOD ON THE BATTERY STATE OF CHARGE FOR HYBRID ELECTRIC VEHICLE

A combined algorithm for battery state of charge （SOC） estimation is proposed to solve the critical issue of hybrid electric vehicle （HEV）. To obtain a more accurate SOC, both coulomb-accumulation and battery resistance-capacitor （RC） model are weighted combined to compensate the deficiencies of individual methods. In order to solve the key issue of coulomb-accumulation, the battery thermal model is used. Based on the principle of energy conservation, the heat generated from battery charge and discharge process is converted into the equivalent electricity to calculate charge and discharge efficiency under variable current. The extended Kalman filter （EKF） as a closed loop algorithm is applied to estimate the parameters of resistance-capacitor model. The input variables do not increase much computing difficulty. The proposed combined algorithm is implemented by adjusting the weighting factor of coulomb- accumulation and resistance-capacitor model. In the end, four different methods including Ah-efficiency, Ah-Equip, RC-SOC and Combined-SOC are compared in federal testing procedure （FTP） drive cycle. The experiment results show that the proposed method has good robustness and high accuracy which is suitable for HEV application.

21700锂离子电池在不同SOC下的热失控实验研究

为提升电池热安全、减少新能源汽车热灾害,揭示不同荷电状态(SOC)下对电池热失控危害的影响机制。在SOC为100%~0%几个荷电状态下研究了21700锂电池的热失控特性,包括电池在热失控当中的表面温度、工作电压、质量损失、能量、TNT当量和破坏半径等。结果表明:电池的温升幅度随SOC的增大而升高,高电量电池热失控触发所需的时间更短,100%SOC电池在603 s触发热失控,相比于25%SOC缩短了59.1%,其危险系数更大;SOC越大,电池热失控后的质量损失也越大;电池热失控过程释放的能量、TNT当量与破坏半径均随SOC的增加而增大,电池的热失控危害性与SOC之间呈现出正相关关系。

State of charge estimation of Li-ion batteries in an electric vehicle based on a radial-basis-function neural network

The on-line estimation of the state of charge （SOC） of the batteries is important for the reliable running of the pure electric vehicle in practice. Because a nonlinear feature exists in the batteries and the radial-basis-function neural network （RBF NN） has good characteristics to solve the nonlinear problem, a practical method for the SOC estimation of batteries based on the RBF NN with a small number of input variables and a simplified structure is proposed. Firstly, in this paper, the model of on-line SOC estimation with the RBF NN is set. Secondly, four important factors for estimating the SOC are confirmed based on the contribution analysis method, which simplifies the input variables of the RBF NN and enhttnces the real-time performance of estimation. FiItally, the pure electric buses with LiFePO4 Li-ion batteries running during the period of the 2010 Shanghai World Expo are considered as the experimental object. The performance of the SOC estimation is validated and evaluated by the battery data from the electric vehicle.

基于SOC的串联连接锂电池能量均衡控制研究

串联锂电池的SOC均衡控制对提高电池寿命具有重要意义。针对锂电池单体SOC表现出离散性的不同情况,本文研究了一种主动均衡与被动均衡相结合的混合均衡方案,其中主动均衡器拓扑由多绕组反激变换器实现,被动均衡器由电阻与开关组成并联在单体电池两端,详细分析了混合均衡器的工作原理。在控制策略上讨论了锂电池SOC的离散性对均衡速度的影响,引入表征SOC离散度的标准差和表征离散原因的系数以实现SOC不同离散情况下的快速均衡。所提出的混合均衡器拓扑和控制方案能够使耗能与均衡速度获得优化,实验结果验证了文中理论的可行性。

A review of deep learning approach to predicting the state of health and state of charge of lithium-ion batteries

In the field of energy storage,it is very important to predict the state of charge and the state of health of lithium-ion batteries.In this paper,we review the current widely used equivalent circuit and electrochemical models for battery state predictions.The review demonstrates that machine learning and deep learning approaches can be used to construct fast and accurate data-driven models for the prediction of battery performance.The details,advantages,and limitations of these approaches are presented,compared,and summarized.Finally,future key challenges and opportunities are discussed.