WO2020155233A1 - 基于双级模型预测的锂离子电池组外部短路故障诊断方法 - Google Patents

基于双级模型预测的锂离子电池组外部短路故障诊断方法 Download PDF

Info

Publication number
WO2020155233A1
WO2020155233A1 PCT/CN2019/075795 CN2019075795W WO2020155233A1 WO 2020155233 A1 WO2020155233 A1 WO 2020155233A1 CN 2019075795 W CN2019075795 W CN 2019075795W WO 2020155233 A1 WO2020155233 A1 WO 2020155233A1
Authority
WO
WIPO (PCT)
Prior art keywords
model
battery
stage
voltage
external short
Prior art date
Application number
PCT/CN2019/075795
Other languages
English (en)
French (fr)
Inventor
陈泽宇
蔡雪
杨英
张�浩
张清
Original Assignee
东北大学
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 东北大学 filed Critical 东北大学
Publication of WO2020155233A1 publication Critical patent/WO2020155233A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
    • G01R31/367—Software therefor, e.g. for battery testing using modelling or look-up tables
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
    • G01R31/396—Acquisition or processing of data for testing or for monitoring individual cells or groups of cells within a battery

Definitions

  • the invention relates to the technical field of lithium-ion power battery safety, in particular to a method for diagnosing external short-circuit faults of a lithium-ion battery pack based on a two-stage model prediction.
  • the technical problem to be solved by the present invention is to provide a method for diagnosing external short-circuit faults of lithium-ion battery packs based on the two-stage model prediction in view of the above-mentioned shortcomings of the prior art; the method has simple steps, is easy to implement online, has high reliability, and is applicable For online fault diagnosis and safety management of electric vehicle power batteries.
  • the present invention provides a method for diagnosing external short circuit faults of lithium-ion battery packs based on two-stage model prediction, which includes the following steps:
  • Step 2 Establish a two-stage battery model for external short-circuit faults, and use the experimental data obtained in step 1 to separately identify the offline optimality parameters of the two-stage battery model;
  • the first-level battery model is an improved equivalent circuit model.
  • the method of improvement is: the battery state of charge SOC in the traditional equivalent circuit model is improved to the depth of discharge ⁇ E during the short circuit process, and the open circuit voltage is regarded as Polynomial function of discharge depth ⁇ E ;
  • k represents the current sampling time
  • ⁇ R p C p , U t , U p , and U oc respectively represent the terminal voltage, polarization voltage, and open circuit voltage of the battery pack
  • R p , and R 0 represent poles, respectively
  • C p represents the polarization capacitance
  • i L represents the current of the battery pack
  • i p represents the current flowing on R p
  • ⁇ t is the sampling step size
  • ⁇ E represents the depth of discharge in an external short-circuit fault.
  • the second-level battery model is a half-cell model, and the specific mathematical expression is:
  • Step 3 Use the battery management system to monitor the voltage of each cell in the battery pack in real time. When the voltage of some cells is lower than the critical threshold Vn, then go to step 4;
  • Step 4 Trigger the first-level battery model, treat the adjacent abnormal battery cell as an abnormal battery pack, use the battery pack current as the model input, and calculate the predicted voltage output by the model in real time;
  • Step 5 Calculate the degree of agreement ⁇ between the predicted voltage of the first-level battery model and the measured voltage for a duration of T 1 , if the degree of agreement ⁇ the critical threshold ⁇ 1 , eliminate the possibility of external short-circuit faults, and proceed to step 8 , Otherwise, it is initially defined as an external short-circuit fault, triggers the second-level battery model, and enters step 6;
  • Step 6 the battery current as the input to the second stage of the battery model and predict the real-time calculation model output voltage, calculating the goodness of fit between the second-stage model predictive battery voltage and the actual measurement voltage [sigma], the duration of time T 2, if the match If degree ⁇ >critical threshold ⁇ 2 , it is confirmed that the abnormality is caused by an external short-circuit fault, locate the position of the abnormal battery cell and go to step 8; otherwise, increase the diagnosis duration to T 3 and go to step 7;
  • Step 7 Use the second-level battery model to repeatedly judge the fit degree. If the fit degree ⁇ the critical threshold ⁇ 2 then eliminate the possibility of external short-circuit failure, if the fit ⁇ > the critical threshold ⁇ 2 , it is confirmed as an external short-circuit fault;
  • Step 8 Store and output the diagnosis result, return to step 3, and wait for the next operation.
  • the two-level battery circuit model in step 2 is divided into two levels, where the first-level battery model is an overall battery model, the second-level battery model is a half-battery model, and the second-level battery model is a half-battery model.
  • the second-level modeling method is to treat the battery as a two-part equivalent circuit model, including model 1 and model 2, that is, the sum of model 1 and model 2 is the overall battery model, and the second-level battery model refers specifically to model 2;
  • model 1 a variable voltage source Connected to the battery internal resistance R 0 and the short-circuit resistance R S as a loop; in model 2, there is a constant voltage source It is connected to the RC link and generates the terminal voltage Ut.
  • the RC link is composed of a capacitor C in parallel with the polarization internal resistance Rp; the open circuit voltage of the entire battery is a variable voltage source With constant voltage source Sum:
  • the model parameters in step 2 are identified for offline optimality, and the model identification process needs to identify the two-level model parameters separately, and the two-level model parameters are independent of each other.
  • the definition of the goodness ⁇ in step 5 is: the reciprocal of the root mean square error between the model prediction result and the actual test result within a certain duration, namely:
  • ⁇ is the number of samplings in the duration T
  • U t, m are the model prediction results of the terminal voltage
  • U t is the online measurement data of the terminal voltage
  • ⁇ n represents the model parameter matrix.
  • the critical threshold ⁇ 1 in step 6 is the critical threshold of goodness of fit of the first-level battery model, and the value of the critical threshold needs to be slightly lower than the calculation result of the goodness of model in the experiment;
  • the critical threshold ⁇ 2 in the step 7 is the critical threshold of goodness of fit of the second-level battery model, and the value of the critical threshold needs to be slightly lower than the calculation result of the goodness of model in the experiment.
  • the method for diagnosing external short circuit faults of lithium-ion battery packs based on the two-stage model prediction adopts a two-stage optimized equivalent circuit model, wherein the first-stage model is The overall battery model contains more parameters to be identified.
  • the model has good adaptability but slightly lower accuracy.
  • the second-level model is a half-cell model, which contains fewer parameters to be identified and the model accuracy is higher; the external short circuit of the battery pack is used
  • the data identifies the parameters of the two-stage model, and the on-line fault diagnosis of the external short circuit is performed based on the agreement between the measured data of the battery pack and the model prediction.
  • the method has simple steps, is easy to implement online, and has high reliability, and is suitable for online fault diagnosis and safety management of electric vehicle power batteries.
  • Fig. 1 is a two-stage equivalent circuit model of external short circuit fault diagnosis provided by an embodiment of the present invention, where a is the first-stage battery model; b is model 1 in the second-stage battery model; c is the second-stage battery model Model 2;
  • FIG. 2 is a flowchart of an external short circuit online diagnosis and estimation method based on two-stage model prediction according to an embodiment of the present invention
  • FIG. 3 is a diagram of the identification error analysis result of the external short-circuit two-stage model provided by an embodiment of the present invention, where a is the identification error analysis result diagram, and b is the error diagram of the identification error analysis result;
  • a-1 is a voltage diagnosis diagram
  • a-2 is a partial enlarged diagram at H in diagram a-1
  • b-1 is a two-stage model error diagram
  • B-2 is a partial enlarged view at Z in Figure b-1;
  • FIG. 5 is a diagram of the external short-circuit experiment results provided by an embodiment of the present invention, in which a-1 is a voltage experiment result diagram, a-2 is a partial enlarged diagram of AB in Figure a-1, and b is a current experiment result diagram.
  • 18650NMC cylindrical lithium-ion power battery is taken as an example. Its rated voltage is 3.6V, nominal capacity is 2.4Ah, and 6 battery cells with SOH value> 0.96 are used to form a battery pack; the experimental equipment adopts: NEU_ESCTEST02 test bench Cooperate with GD-2045D temperature control box of Haixiang instrument,
  • the method of this embodiment is as follows.
  • the present invention provides a method for diagnosing external short-circuit faults of lithium-ion battery packs based on two-stage model prediction, as shown in Fig. 2, including the following steps:
  • Step 2 Establish a two-stage battery model for external short-circuit faults, and use the experimental data obtained in step 1 to identify the offline optimality parameters of the two-stage battery model;
  • the parameters of the two-level model need to be identified separately, and the parameters of the two-level model are independent of each other.
  • the first-level battery model is an improved equivalent circuit model.
  • the improved method is: the battery state of charge SOC in the traditional equivalent circuit model is improved to the short circuit process
  • the depth of discharge ⁇ E , and the open circuit voltage is regarded as a polynomial function of the depth of discharge ⁇ E ;
  • k represents the current sampling time
  • ⁇ R p C p , U t , U p , and U oc respectively represent the terminal voltage, polarization voltage, and open circuit voltage of the battery pack
  • R p , and R 0 represent poles, respectively
  • C p represents the polarization capacitance
  • i L represents the current of the battery pack
  • i p represents the current flowing on R p
  • ⁇ t is the sampling step size
  • ⁇ E represents the depth of discharge in an external short-circuit fault.
  • the open circuit voltage is expressed by a polynomial, as shown in the following formula
  • N p is the degree of the polynomial
  • ⁇ i is the polynomial coefficient
  • ⁇ E is the depth of discharge in the external short-circuit fault.
  • Q R is the nominal capacity
  • the second-level battery model is a half-cell model.
  • the second-level modeling method is to treat the battery as a two-part equivalent circuit model, including model 1 and model 2, that is, the sum of model 1 and model 2 is the overall battery model.
  • the second-level battery model refers specifically to model 2; in model 1, as shown in Figure 1, as shown in Figure b, a variable voltage source Connected to the battery internal resistance R 0 and the short-circuit resistance R S as a loop; in model 2, as shown in Figure 1, as shown in Figure c, there is a constant voltage source It is connected to the RC link and generates the terminal voltage Ut.
  • the RC link is composed of a capacitor C in parallel with the polarization internal resistance Rp; the open circuit voltage of the entire battery is a variable voltage source With constant voltage source Sum:
  • the second-level battery model is a half-cell model, and the specific mathematical expression is:
  • the experimental data is used to separately identify the two-stage battery model for offline optimality parameters.
  • the identification method can adopt a global optimization method. In this embodiment, a genetic algorithm is used for parameter identification. The selection method does not limit the present invention.
  • the identification error is shown in Figure 3. It can be seen that the model constructed in this way has a very high prediction accuracy of the second-level battery model. After the identification is completed, the record identification results are shown in Table 1 to Table 2:
  • Step 3 Use the battery management system to monitor the voltage of each cell in the battery pack in real time. If the voltage of some cells is lower than the critical threshold Vn, go to step 4;
  • the battery management system is a new energy vehicle battery management system, which mainly has current, voltage, and temperature collection functions , Battery state estimation, overvoltage protection and safety management system;
  • Step 4 Trigger the first-level battery model, treat the adjacent abnormal battery cell as an abnormal battery pack, use the battery pack current as the model input, and calculate the predicted voltage output by the model in real time;
  • Step 5 Calculate the degree of agreement ⁇ between the predicted voltage of the first-level battery model and the measured voltage for a duration of T 1 , if the degree of agreement ⁇ the critical threshold ⁇ 1 , eliminate the possibility of external short-circuit faults, and proceed to step 8 , Otherwise, it is initially defined as an external short-circuit fault, triggers the second-level battery model, and enters step 6;
  • goodness ⁇ is: the reciprocal of the root mean square error between the model prediction result and the actual test result within a certain duration, namely:
  • ⁇ is the number of samplings in the duration T
  • U t, m are the model prediction results of the terminal voltage
  • U t is the online measurement data of the terminal voltage
  • ⁇ n represents the model parameter matrix.
  • the critical threshold ⁇ 1 is the critical threshold of the fit of the first-level battery model, and the critical threshold must be slightly lower than the calculated results of the model fit in the experiment;
  • the critical threshold ⁇ 2 is the critical threshold of the fit of the second-level battery model, and the critical threshold needs to be slightly lower than the calculated results of the model fit in the experiment.
  • the critical threshold of model fit is determined based on the experimental results.
  • the critical threshold must be slightly lower than the model fit calculated in the experiment, so as to ensure that there will be no missed judgments in the diagnosis process; according to the experimental results of external short circuits, the model
  • the calculation results of good fit are shown in Table 3:
  • Step 6 the battery current as the input to the second stage of the battery model and predict the real-time calculation model output voltage, calculating the goodness of fit between the second-stage model predictive battery voltage and the actual measurement voltage [sigma], the duration of time T 2, if the match If degree ⁇ >critical threshold ⁇ 2 , it is confirmed that the abnormality is caused by an external short-circuit fault, locate the position of the abnormal battery cell and go to step 8; otherwise, increase the diagnosis duration to T 3 and go to step 7;
  • Step 7 Use the second-level battery model to repeatedly judge the degree of fit. If the degree of fit ⁇ the critical threshold ⁇ 2 then eliminate the possibility of external short-circuit failure, if the degree of fit ⁇ > the critical threshold ⁇ 2 , it is confirmed as an external short-circuit fault;
  • Step 8 Store and output the diagnosis result, return to step 3, and wait for the next operation.
  • the battery management system is used to monitor the voltage of each cell of the battery pack in real time.
  • the battery pack composed of 6 batteries is short-circuited, and the battery pack voltage quickly drops below 0.5V, which is below the critical threshold. Therefore, the first-level battery model is triggered, the abnormal battery cells are formed into abnormal battery packs according to adjacent individuals, the battery pack current is used as the model input, and the predicted voltage output by the model is calculated in real time, as shown in the solid line of a-1 in Figure 4 As shown, the measurement results of the terminal voltage of the battery pack are obtained online at the same time, as shown by the dotted line in a-1 in Fig. 4, and b-1 in Fig. 4 is a 3-7 second two-stage model error diagram.
  • the first-level battery model The goodness ⁇ 6.34 triggers the second-level battery model.
  • the error between the model prediction result and the measured data is less than 20mV.
  • the second-level battery model goodness ⁇ 93.7> critical threshold ⁇ 2 according to the judgment criteria, it is confirmed as an external short-circuit fault, the online diagnosis process is completed, and the diagnosis result is stored and output, as shown in Figure 5.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Secondary Cells (AREA)

Abstract

本发明提出一种基于双级模型预测的锂离子电池组外部短路故障诊断方法，涉及锂离子动力电池安全技术领域。首先,对锂离子电池组进行外部短路实验，构建电池组外部短路双级等效电路模型，利用被测实验数据对电池模型参数进行离线最优性辨识；然后，运行时根据电池测量数据判断电池组中电池状态，发现部分电池电压出现异常时，对产生异常的相邻电池单元标记为整体，记作异常电池组，启动第一级电池模型，若第一级电池模型误差小于临界阈值，则触发第二级电池模型，计算获得模型误差；最后，通过实测数据与双级模型吻合度，对异常电池进行故障诊断。该方法步骤简单，易于在线实现，且可靠性高，适用于电动汽车动力电池在线故障诊断与安全管理。

Description

基于双级模型预测的锂离子电池组外部短路故障诊断方法 技术领域
本发明涉及锂离子动力电池安全技术领域，尤其涉及一种基于双级模型预测的锂离子电池组外部短路故障诊断方法。
背景技术
近年来电动汽车发展迅速，发展电动汽车被视为是解决环境污染、降低燃油消耗、建设绿色、环保的城市交通的一种有效途径，然而在电池汽车的应用过程中时常出现起火爆炸的安全事故，其根源往往是由于电池故障引发的热失控。外部短路，是电池故障中十分常见且较为严重的故障之一，外部短路故障时电池组产生大电流，容易引发电池高温高热，外部短路故障的持续时间往往只有几十秒，因此如何对外部短路故障进行有效、准确且迅速的在线故障诊断，是一个十分重要的技术问题。
目前现有的电池管理系统多数是针对电池的状态估计、寿命预测等，而对于电池安全问题及故障诊断的方法尚不成熟，尤其是大功率电池组的外部短路故障诊断技术较为匮缺。
发明内容
本发明要解决的技术问题是针对上述现有技术的不足，提供一种基于双级模型预测的锂离子电池组外部短路故障诊断方法；该方法步骤简单，易于在线实现，且可靠性高，适用于电动汽车动力电池在线故障诊断与安全管理。
为解决上述技术问题，本发明所采取的技术方案是：
本发明提供一种基于双级模型预测的锂离子电池组外部短路故障诊断方法，包括以下步骤：
步骤1：进行电池组外部短路实验，记录实验数据，包括电流的测量数据I c＝[I c1,I c2,…,I cN] T、端电压的测量数据U c＝[U c1,U c2,…,U cN] T，其中N为数据采样数量，N的取值取决于外部短路试验中电流持续时间和采样步长,T表示矩阵的转置；
步骤2：建立外部短路故障的双级电池模型，并通过步骤1中所得到的实验数据分别对双级电池模型进行离线最优性参数辨识；
第一级电池模型为一种改进的等效电路模型，改进的方法为：将传统等效电路模型中的电池荷电状态SOC，改进为短路过程中放电深度ξ E，并将开路电压视为放电深度ξ E的多项式函数；
第一级电池模型具体数学表达形式为：
Figure PCTCN2019075795-appb-000001
其中,k表示当前采样时刻，τ＝R pC p，U t,U p,和U oc分别表示电池组的端电压、极化电压、和开路电压；R p，和R 0则分别表示极化内阻和欧姆内阻，C p代表极化电容，i L表示电池组电流，i p表示R p上流过的电流，Δt为采样步长，ξ E表示在外部短路故障中的放电深度。
第二级电池模型为半电池模型，具体数学表达形式为：
Figure PCTCN2019075795-appb-000002
其中
Figure PCTCN2019075795-appb-000003
代表恒定电压源；
步骤3：利用电池管理系统实时监测电池组每个单体电压，当部分电池单体电压低于临界阈值Vn，则进入步骤4；
步骤4：触发第一级电池模型，将相邻的异常电池单体视为一个异常电池组，将电池组电流作为模型输入，实时计算模型输出的预测电压；
步骤5：计算第一级电池模型预测电压与实测电压之间的吻合度σ，持续时间T 1时刻，如果吻合度σ＜临界阈值χ 1，则排除外部短路故障的可能性，并进入步骤8，否则，初步界定为外部短路故障，触发第二级电池模型，并进入步骤6；
步骤6：将电池组电流作为第二级电池模型的输入并实时计算模型输出的预测电压，计算第二级电池模型预测电压与实测电压之间的吻合度σ，持续时间T 2时刻，如果吻合度σ＞临界阈值χ 2，则确认该异常是由外部短路故障引起，定位发生异常电池单体的位置并进入步骤8；否则，将诊断持续时间增加到T 3，并进入步骤7；
步骤7：采用第二级电池模型重复判断吻合度，如果吻合度σ＜临界阈值χ 2则排除外部短路故障的可能性，如果吻合度σ＞临界阈值χ 2，则确认为外部短路故障；
步骤8：储存并输出诊断结果，返回步骤3，等待进行下一次操作。
所述步骤2中的双级电池电路模型一共分为两级，其中第一级电池模型为一个电池整体模型，第二级电池模型为半电池模型；第二级电池模型为半电池模型，第二级的建模方法是将电池视为两部分等效电路模型，包括模型1和模型2，即模型1与模型2之和为电池整体模型，第二级电池模型特指其中模型2；在模型1中，由一个可变电压源
Figure PCTCN2019075795-appb-000004
与电池内阻R 0、 短路电阻R S连为回路；在模型2中，有一个恒定电压源
Figure PCTCN2019075795-appb-000005
与RC环节连并产生端电压Ut，RC环节由一个电容C与极化内阻Rp并联组成；整个电池的开路电压为可变电压源
Figure PCTCN2019075795-appb-000006
与恒定电压源
Figure PCTCN2019075795-appb-000007
之和：
Figure PCTCN2019075795-appb-000008
所述步骤2中的离线最优性参数辨识为将实验电流测量值I c作为模型输入，端电压输出U＝[U 1,U 2,…,U N] T作为模型输出，使用全局优化算法对步骤2中的模型参数进行离线最优性辨识，且模型的辨识过程需对两级模型参数分别辨识，两级模型的参数相互独立。
所述步骤5中吻合度σ的定义为：在一定持续时间内模型预测结果与实际测试结果的均方根误差的倒数，即：
Figure PCTCN2019075795-appb-000009
其中ρ为持续时间T内的采样次数，U t,m为端电压的模型预测结果，U t为端电压的在线测量数据，θ n代表模型参数矩阵。
所述步骤6中的临界阀值χ 1为第一级电池模型的吻合度临界阀值，临界阈值的取值需略低于实验时的模型吻合度计算结果；
所述步骤7中的临界阀值χ 2为第二级电池模型的吻合度临界阀值，临界阈值的取值需略低于实验时的模型吻合度计算结果。
采用上述技术方案所产生的有益效果在于：本发明提供的基于双级模型预测的锂离子电池组外部短路故障诊断方法，该方法采用一种双级优化等效电路模型，其中第一级模型为电池整体模型，包含有较多的待辨识参数，模型适应性好但精度略低，第二级模型为半电池模型，包含有较少的待辨识参数且模型精度较高；使用电池组外部短路数据对双级模型参数进行辨识，通过电池组实测数据与模型预测的吻合度，来进行外部短路的在线故障诊断。该方法步骤简单，易于在线实现，且可靠性高，适用于电动汽车动力电池在线故障诊断与安全管理。
附图说明
图1为本发明实施例提供的外部短路故障诊断双级等效电路模型，其中a为第一级电池模型；b为第二级电池模型中的模型1；c为第二级电池模型中的模型2；
图2为本发明实施例提供的基于双级模型预测的外部短路在线诊断估计方法流程图；
图3为本发明实施例提供的外部短路双级模型辨识误差分析结果图，其中a为辨识误差分析结果图，b为辨识误差分析结果的误差示意图；
图4为本发明实施例提供的电池组外部短路诊断图，其中a-1为电压诊断图，a-2为图a-1中H处的局部放大图，b-1为双级模型误差图，b-2为图b-1中Z处的局部放大图；
图5为本发明实施例提供的外部短路实验结果图，其中a-1为电压实验结果图，a-2为图a-1中AB出的局部放大图，b为电流实验结果图。
具体实施方式
下面结合附图和实施例，对本发明的具体实施方式作进一步详细描述。以下实施例用于说明本发明，但不用来限制本发明的范围。
实施例中以18650NMC圆柱形锂离子动力电池为例，其额定电压为3.6V，标称容量为2.4Ah，采用SOH值＞0.96的6块电池单体组成电池组；实验设备采用：NEU_ESCTEST02试验台配合海向仪器GD-2045D温控箱，
本实施例的方法如下所述。
本发明提供一种基于双级模型预测的锂离子电池组外部短路故障诊断方法，如图2所示，包括以下步骤：
步骤1：进行电池组外部短路实验，记录实验数据，包括电流的测量数据I c＝[I c1,I c2,…,I cN] T、端电压的测量数据U c＝[U c1,U c2,…,U cN] T，其中N为数据采样数量，N的取值取决于外部短路试验中电流持续时间和采样步长,T表示矩阵的转置；
步骤2：建立外部短路故障的双级电池模型，并通过步骤1中所得到的实验数据分别对双级电池模型进行离线最优性参数辨识；离线最优性参数辨识为将实验电流测量值I c作为模型输入，端电压输出U＝[U 1,U 2,…,U N] T作为模型输出，使用全局优化算法对步骤2中的模型参数进行离线最优性辨识，且模型的辨识过程需对两级模型参数分别辨识，两级模型的参数相互独立。
第一级电池模型为一种改进的等效电路模型，如图1所示中a图所示，改进的方法为：将传统等效电路模型中的电池荷电状态SOC，改进为短路过程中放电深度ξ E，并将开路电压视为放电深度ξ E的多项式函数；
第一级电池模型具体数学表达形式为：
Figure PCTCN2019075795-appb-000010
其中,k表示当前采样时刻，τ＝R pC p，U t,U p,和U oc分别表示电池组的端电压、极化电压、和开路电压；R p，和R 0则分别表示极化内阻和欧姆内阻，C p代表极化电容，i L表示电池组电流，i p表示R p上流过的电流，Δt为采样步长，ξ E表示在外部短路故障中的放电深度。
开路电压用多项式来表示，如下式所示
Figure PCTCN2019075795-appb-000011
式中N p是多项式的次数，α i表示多项式系数，ξ E表示在外部短路故障中的放电深度，具体计算方法如下式所示：
Figure PCTCN2019075795-appb-000012
式中Q R为标称容量。
第二级电池模型为半电池模型，第二级的建模方法是将电池视为两部分等效电路模型，包括模型1和模型2，即模型1与模型2之和为电池整体模型，第二级电池模型特指其中模型2；在模型1中，如图1所示中b图所示，由一个可变电压源
Figure PCTCN2019075795-appb-000013
与电池内阻R 0、短路电阻R S连为回路；在模型2中，如图1所示中c图所示，有一个恒定电压源
Figure PCTCN2019075795-appb-000014
与RC环节连并产生端电压Ut，RC环节由一个电容C与极化内阻Rp并联组成；整个电池的开路电压为可变电压源
Figure PCTCN2019075795-appb-000015
与恒定电压源
Figure PCTCN2019075795-appb-000016
之和：
Figure PCTCN2019075795-appb-000017
第二级电池模型为半电池模型，具体数学表达形式为：
Figure PCTCN2019075795-appb-000018
其中
Figure PCTCN2019075795-appb-000019
代表恒定电压源；
对于第一级电池模型，待辨识参数θ 1＝[α 1,α 2,…,α 10,τ,R p,R 0]共计13个参数，对于第二级电池模型，带辨识参数θ 2＝[U 0,τ,R p]共计3个参数。使用实验数据分别对双级电池模型进行离线最优性参数辨识，辨识方法可以采用全局最优化方法，在本实施例中采用遗传算法进行参数辨识，选用方法对本发明不构成限定。辨识误差如图3所示，可以看出，这样所构建的模型，第二级电池模型预测精度会非常高。辨识完成后，记录辨识结果如表1-表2所示：
表1 第一级电池模型参数辨识结果
参数 辨识结果 参数 辨识结果
α1 16.5 α8 -3.3
α2 13.0 α9 2.4
α3 -78.9 α10 10.7
α4 -101.7 τ(s) 205.5
α5 240.5 R p(mΩ) 501.2
α6 -116.0 R 0(mΩ) 145.5
α7 9.6    
表2 第二级电池模型参数辨识结果
参数 辨识结果
U 0(mV) 589.7
τ(s) 6.9
R p(mΩ)) 6.5
步骤3：利用电池管理系统实时监测电池组每个单体电压，如果部分电池单体电压低于临界阈值Vn，则进入步骤4；
本实施例中设置单体电压临界阈值Vn＝2.0V，临界阈值的设定略微低于电池正常放电截至电压；所述电池管理系统是新能源汽车的电池管理系统，主要有电流电压温度采集功能、电池状态估计、过压保护以及安全管理系统；
步骤4：触发第一级电池模型，将相邻的异常电池单体视为一个异常电池组，将电池组电流作为模型输入，实时计算模型输出的预测电压；
步骤5：计算第一级电池模型预测电压与实测电压之间的吻合度σ，持续时间T 1时刻，如果吻合度σ＜临界阈值χ 1，则排除外部短路故障的可能性，并进入步骤8，否则，初步界定为外部短路故障，触发第二级电池模型，并进入步骤6；
吻合度σ的定义为：在一定持续时间内模型预测结果与实际测试结果的均方根误差的倒数，即：
Figure PCTCN2019075795-appb-000020
其中ρ为持续时间T内的采样次数，U t,m为端电压的模型预测结果，U t为端电压的在线测量数据，θ n代表模型参数矩阵。
本实施例中设置持续时间T 1＝1.0s，T 2＝3.0s，T 3＝10.0s；设置临界阈值χ 1＝3.5，临界阈值χ 2＝30。
临界阀值χ 1为第一级电池模型的吻合度临界阀值，临界阈值的取值需略低于实验时的模型吻合度计算结果；
临界阀值χ 2为第二级电池模型的吻合度临界阀值，临界阈值的取值需略低于实验时的模 型吻合度计算结果。
模型吻合度的临界阈值是根据实验结果决定的，临界阈值的取值需略低于实验时的模型吻合度计算结果，这样可以保证诊断过程不会出现漏判；根据外部短路的实验结果，模型吻合度计算结果如表3所示：
表3 模型吻合度
实验次数 第一级模型吻合度 第二级模型吻合度
1 3.9 33.8
2 4.1 32.4
3 4.4 31.7
4 3.7 35.5
5 4.6 37.1
因此，设置临界阈值χ 1＝3.5，临界阈值χ 2＝30。
步骤6：将电池组电流作为第二级电池模型的输入并实时计算模型输出的预测电压，计算第二级电池模型预测电压与实测电压之间的吻合度σ，持续时间T 2时刻，如果吻合度σ＞临界阈值χ 2，则确认该异常是由外部短路故障引起，定位发生异常电池单体的位置并进入步骤8；否则，将诊断持续时间增加到T 3，并进入步骤7；
步骤7：采用第二级电池模型重复判断吻合度，如果吻合度σ＜临界阈值χ 2则排除外部短路故障的可能性，如果吻合度σ＞临界阈值χ 2，则确认为外部短路故障；
步骤8：储存并输出诊断结果，返回步骤3，等待进行下一次操作。
在线运行，利用电池管理系统实时监测电池组每个单体电压，在本实施例中，对6块电池组成的电池组进行短路，电池组电压迅速下降到0.5V以下，低于了临界阈值，因此触发了第一级电池模型，将异常电池单体按相邻个体组成异常电池组，将电池组电流作为模型输入，实时计算模型输出的预测电压，如图4中a-1图的实线所示，同时在线获取电池组端电压测量结果，如图4中a-1图的虚线所示，如图4中b-1为3-7秒的双级模型误差图。
计算模型预测电压与实测电压之间的吻合度σ，并将吻合度与临界阈值进行比较，根据发明内容中所述的逻辑判断流程，进行故障诊断，在本实施例中，第一级电池模型吻合度σ≈6.34，触发第二级电池模型，在第二级电池模型运行过程中，模型预测结果与实测数据误差低于20mV，经计算，第二级电池模型吻合度σ≈93.7＞临界阈值χ 2，根据判断准则，确认为外部短路故障，在线诊断过程完成，储存并输出诊断结果，如图5所示。
最后应说明的是：以上实施例仅用以说明本发明的技术方案，而非对其限制；尽管参照前述实施例对本发明进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述实施例所记载的技术方案进行修改，或者对其中部分或者全部技术特征进行等同替换；而 这些修改或者替换，并不使相应技术方案的本质脱离本发明权利要求所限定的范围。

Claims (5)

  1. 一种基于双级模型预测的锂离子电池组外部短路故障诊断方法，其特征在于：包括以下步骤：
    步骤1：进行电池组外部短路实验，记录实验数据，包括电流的测量数据I c＝[I c1,I c2,…,I cN] T、端电压的测量数据U c＝[U c1,U c2,…,U cN] T，其中N为数据采样数量，N的取值取决于外部短路试验中电流持续时间和采样步长,T表示矩阵的转置；
    步骤2：建立外部短路故障的双级电池模型，并通过步骤1中所得到的实验数据分别对双级电池模型进行离线最优性参数辨识；
    第一级电池模型为一种改进的等效电路模型，改进的方法为：将传统等效电路模型中的电池荷电状态SOC，改进为短路过程中放电深度ξ E，并将开路电压视为放电深度ξ E的多项式函数；
    第一级电池模型具体数学表达形式为：
    Figure PCTCN2019075795-appb-100001
    其中,k表示当前采样时刻，τ＝R pC p，U t,U p,和U oc分别表示电池组的端电压、极化电压、和开路电压；R p，和R 0则分别表示极化内阻和欧姆内阻，C p代表极化电容，i L表示电池组电流，i p表示R p上流过的电流，Δt为采样步长，ξ E表示在外部短路故障中的放电深度；
    第二级电池模型为半电池模型，具体数学表达形式为：
    Figure PCTCN2019075795-appb-100002
    其中
    Figure PCTCN2019075795-appb-100003
    代表恒定电压源；
    步骤3：利用电池管理系统实时监测电池组每个单体电压，当部分电池单体电压低于临界阈值Vn，则进入步骤4；
    步骤4：触发第一级电池模型，将相邻的异常电池单体视为一个异常电池组，将电池组电流作为模型输入，实时计算模型输出的预测电压；
    步骤5：计算第一级电池模型预测电压与实测电压之间的吻合度σ，持续时间T 1时刻，如果吻合度σ＜临界阈值χ 1，则排除外部短路故障的可能性，并进入步骤8，否则，初步界定为外部短路故障，触发第二级电池模型，并进入步骤6；
    步骤6：将电池组电流作为第二级电池模型的输入并实时计算模型输出的预测电压，计 算第二级电池模型预测电压与实测电压之间的吻合度σ，持续时间T 2时刻，如果吻合度σ＞临界阈值χ 2，则确认该异常是由外部短路故障引起，定位发生异常电池单体的位置并进入步骤8；否则，将诊断持续时间增加到T 3，并进入步骤7；
    步骤7：采用第二级电池模型重复判断吻合度，如果吻合度σ＜临界阈值χ 2则排除外部短路故障的可能性，如果吻合度σ＞临界阈值χ 2，则确认为外部短路故障；
    步骤8：储存并输出诊断结果，返回步骤3，等待进行下一次操作。
  2. 根据权利要求1所述的一种基于双级模型预测的锂离子电池组外部短路故障诊断方法，其特征在于：所述步骤2中的双级电池电路模型一共分为两级，其中第一级电池模型为一个电池整体模型，第二级电池模型为半电池模型；第二级电池模型为半电池模型，第二级的建模方法是将电池视为两部分等效电路模型，包括模型1和模型2，即模型1与模型2之和为电池整体模型，第二级电池模型特指其中模型2；在模型1中，由一个可变电压源
    Figure PCTCN2019075795-appb-100004
    与电池内阻R 0、短路电阻R S连为回路；在模型2中，有一个恒定电压源
    Figure PCTCN2019075795-appb-100005
    与RC环节连并产生端电压Ut，RC环节由一个电容C与极化内阻Rp并联组成；整个电池的开路电压为可变电压源
    Figure PCTCN2019075795-appb-100006
    与恒定电压源
    Figure PCTCN2019075795-appb-100007
    之和：
    Figure PCTCN2019075795-appb-100008
  3. 根据权利要求1所述的一种基于双级模型预测的锂离子电池组外部短路故障诊断方法，其特征在于：所述步骤2中的离线最优性参数辨识为将实验电流测量值I c作为模型输入，端电压输出U＝[U 1,U 2,…,U N] T作为模型输出，使用全局优化算法对步骤2中的模型参数进行离线最优性辨识，且模型的辨识过程需对两级模型参数分别辨识，两级模型的参数相互独立。
  4. 根据权利要求1所述的一种基于双级模型预测的锂离子电池组外部短路故障诊断方法，其特征在于：所述步骤5中吻合度σ的定义为：在一定持续时间内模型预测结果与实际测试结果的均方根误差的倒数，即：
    Figure PCTCN2019075795-appb-100009
    其中ρ为持续时间T内的采样次数，U t,m为端电压的模型预测结果，U t为端电压的在线测量数据，θ n代表模型参数矩阵。
  5. 根据权利要求1所述的一种基于双级模型预测的锂离子电池组外部短路故障诊断方 法，其特征在于：所述步骤6中的临界阀值χ 1为第一级电池模型的吻合度临界阀值，临界阈值的取值需略低于实验时的模型吻合度计算结果；
    所述步骤7中的临界阀值χ 2为第二级电池模型的吻合度临界阀值，临界阈值的取值需略低于实验时的模型吻合度计算结果。
PCT/CN2019/075795 2019-01-28 2019-02-22 基于双级模型预测的锂离子电池组外部短路故障诊断方法 WO2020155233A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910078743.3 2019-01-28
CN201910078743.3A CN109738811B (zh) 2019-01-28 2019-01-28 基于双级模型预测的锂离子电池组外部短路故障诊断方法

Publications (1)

Publication Number Publication Date
WO2020155233A1 true WO2020155233A1 (zh) 2020-08-06

Family

ID=66366282

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/075795 WO2020155233A1 (zh) 2019-01-28 2019-02-22 基于双级模型预测的锂离子电池组外部短路故障诊断方法

Country Status (2)

Country Link
CN (1) CN109738811B (zh)
WO (1) WO2020155233A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114994543A (zh) * 2022-08-01 2022-09-02 湖南华大电工高科技有限公司 储能电站电池故障诊断方法、装置及存储介质

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111257753B (zh) * 2020-03-10 2022-07-26 合肥工业大学 一种电池系统故障诊断方法
CN111470067B (zh) * 2020-06-23 2020-10-09 中航金城无人系统有限公司 基于模型预测的串联式混合动力系统故障诊断系统和方法
CN112098850B (zh) * 2020-09-21 2024-03-08 山东工商学院 基于sdo算法的锂离子电池电压故障诊断方法及系统
CN112632850A (zh) * 2020-12-14 2021-04-09 华中科技大学 一种锂电池组中异常电池的检测方法及系统
CN113711070A (zh) * 2020-12-15 2021-11-26 东莞新能德科技有限公司 电池内短路侦测方法、电子装置和存储介质

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2016091872A (ja) * 2014-11-07 2016-05-23 トヨタ自動車株式会社 二次電池の異常検出方法及び異常検出装置
CN106526493A (zh) * 2016-11-01 2017-03-22 北京理工大学 基于bp神经网络的动力电池外部短路故障诊断及温升预测方法和系统
CN108693478A (zh) * 2018-04-17 2018-10-23 北京理工大学 一种锂离子动力电池的漏液检测方法

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105827200B (zh) * 2016-03-01 2019-05-03 华为技术有限公司 光电系统中电池组串故障的识别方法、装置和设备
KR20180043571A (ko) * 2016-10-20 2018-04-30 주식회사 엘지화학 2차 전지
CN107576917A (zh) * 2017-09-21 2018-01-12 中国检验检疫科学研究院 一种通用型单体锂电池外部短路试验方法
CN108363016B (zh) * 2018-02-22 2021-02-26 上海理工大学 基于人工神经网络的电池微短路定量诊断方法
CN108318775B (zh) * 2018-05-11 2020-06-09 北京市亿微科技有限公司 在线诊断电池短路故障的方法及装置

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2016091872A (ja) * 2014-11-07 2016-05-23 トヨタ自動車株式会社 二次電池の異常検出方法及び異常検出装置
CN106526493A (zh) * 2016-11-01 2017-03-22 北京理工大学 基于bp神经网络的动力电池外部短路故障诊断及温升预测方法和系统
CN108693478A (zh) * 2018-04-17 2018-10-23 北京理工大学 一种锂离子动力电池的漏液检测方法

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114994543A (zh) * 2022-08-01 2022-09-02 湖南华大电工高科技有限公司 储能电站电池故障诊断方法、装置及存储介质

Also Published As

Publication number Publication date
CN109738811A (zh) 2019-05-10
CN109738811B (zh) 2020-12-01

Similar Documents

Publication Publication Date Title
WO2020155233A1 (zh) 基于双级模型预测的锂离子电池组外部短路故障诊断方法
Yang et al. On-board diagnosis of soft short circuit fault in lithium-ion battery packs for electric vehicles using an extended Kalman filter
Zheng et al. Micro-short-circuit cell fault identification method for lithium-ion battery packs based on mutual information
Seo et al. Online detection of soft internal short circuit in lithium-ion batteries at various standard charging ranges
Lu et al. A method of cell-to-cell variation evaluation for battery packs in electric vehicles with charging cloud data
Li et al. A weighted Pearson correlation coefficient based multi-fault comprehensive diagnosis for battery circuits
Xu et al. A vehicle-cloud collaborative method for multi-type fault diagnosis of lithium-ion batteries
CN116401585B (zh) 一种基于大数据的储能电池失效风险评估方法
Sun et al. A multi-fault advanced diagnosis method based on sparse data observers for lithium-ion batteries
CN115494400B (zh) 一种基于集成学习的锂电池析锂状态在线监控方法
CN116381541B (zh) 一种储能锂电池系统的健康评估方法及系统
Xu et al. Online Soft Short Circuit Diagnosis of Electric Vehicle Li-ion Batteries Based on Constant Voltage Charging Current
Binelo et al. Mathematical modeling and parameter estimation of battery lifetime using a combined electrical model and a genetic algorithm
Sun et al. Relative entropy based lithium-ion battery pack short circuit detection for electric vehicle
CN113687251A (zh) 一种基于双模型的锂离子电池组电压异常故障诊断方法
Li et al. Multi-dimension statistical analysis and selection of safety-representing features for battery pack in real-world electric vehicles
Diao et al. Research on Electric Vehicle Charging Safety Warning based on A-LSTM Algorithm
CN105759217A (zh) 一种基于可测数据的铅酸蓄电池组在线故障诊断方法
Wu et al. Research on short-circuit fault-diagnosis strategy of lithium-ion battery in an energy-storage system based on voltage cosine similarity
Tang et al. An aging-and load-insensitive method for quantitatively detecting the battery internal-short-circuit resistance
Xiao et al. Lithium-ion batteries fault diagnosis based on multi-dimensional indicator
Xu et al. The Effect of Charge Behavior on Lithium Battery SOH
CN111190116A (zh) 一种锂离子电池安全性管理方法及系统
Haiying et al. Research on the consistency of the power battery based on multi-points impedance spectrum
Meng et al. Research on fault diagnosis of electric vehicle power battery based on attribute recognition

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19912476

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19912476

Country of ref document: EP

Kind code of ref document: A1