阅读说明：本技术 用于检测电池中的故障电池单元的方法 (Method for detecting a faulty battery cell in a battery ) 是由 B·德洛贝尔 M·奥里古奇 B·阿尔萨克 于 2019-01-15 设计创作，主要内容包括：本发明涉及一种用于检测电池组中的故障电池单元的方法,该方法包括以下步骤：-测量该电池组中的每个电池单元的端子处的空载电压的步骤；-根据在其期间不使用该电池组的天数来计算在所述电池单元的端子处测得的该空载电压与已知的平均空载电压之间的偏差的步骤；-计算所述计算出的偏差与变量n.σ之差的步骤,其中,n是不小于1的整数,并且σ是已知标准偏差,该已知标准偏差是在其期间不使用该电池组的天数的函数；以及-在所述差大于零的情况下检测故障电池单元的步骤。(The invention relates to a method for detecting a faulty battery cell in a battery pack, the method comprising the steps of: -a step of measuring the no-load voltage at the terminals of each battery cell in the battery pack; -a step of calculating the deviation between the no-load voltage measured at the terminals of said battery cells and a known average no-load voltage, according to the number of days during which the battery pack is not used; -a step of calculating the difference between said calculated deviation and a variable n.σ, where n is an integer no less than 1 and σ is a known standard deviation which is a function of the number of days during which the battery pack is not used; and-a step of detecting a faulty battery cell if said difference is greater than zero.)

1. A method for detecting a defective battery cell in a battery pack, the method comprising the steps of:

-a step of measuring the no-load voltage across the terminals of each battery cell in the battery pack,

-a step of calculating the deviation between the no-load voltage measured across the terminals of the battery cells and a known average no-load voltage according to the number of days in which the battery pack is not used,

-a step of calculating the difference between said calculated deviation and a variable n.σ, where n is an integer greater than or equal to one and σ is a known standard deviation depending on the number of days the battery is not used,

-a step of detecting a defective battery cell if the difference is greater than zero.

2. The detection method according to claim 1, wherein n is an integer between 2 and 10.

3. Method according to any of claims 1 and 2, characterized in that the method comprises a warning step when a defective battery cell is detected.

4. A method as claimed in claim 3, wherein the warning step uses a visual signal which is visible to a person carrying out the method.

5. A method as claimed in claim 3, characterized in that the warning step uses an audible audio signal for the person carrying out the method.

6. A method according to any one of claims 1 to 5, characterized in that it implements a step of preventing the battery pack from being recharged when at least one defective battery cell is detected.

7. The method of any one of claims 1 to 6, wherein if σ ≧ 5mV, then n ═ 6.

8. The method of any one of claims 1 to 6, wherein if σ ≧ 10mV, then n ═ 4.

9. The method of any one of claims 1 to 6, wherein if σ ≧ 20mV, then n-3.

10. Method according to any of claims 1 to 9, characterized in that it comprises a step of displaying a loss of autonomy in case a faulty battery cell is detected.

Technical Field

The present invention relates to a method for detecting a faulty battery cell in a battery.

Background

Currently, there are diagnostic procedures that make it possible to protect an unbalanced battery pack from faulty cells. However, this imbalance is not necessarily due to a faulty cell, but rather is very simply due to the deviation of the self-discharge.

Specifically, the diagnostic procedure typically consists in measuring the voltage in order to estimate the SOC (state of charge) difference between the most charged cell and the least charged cell. Thus, if the difference is greater than a predetermined threshold, vehicle recharging will be inhibited. Thus, once the vehicle is discharged, it will be held stationary and must be returned to the dealer for service. However, as we have been able to demonstrate, there may be "natural" self-discharge deviations inherent to the technology or production lot that may give a false impression (i.e., the battery pack contains at least one faulty cell). Thus, if the customer uses his vehicle only rarely, he will not benefit from balancing the cells for a long period of time. In this case, the customer may have a battery pack that is fully functional, but no longer able to be charged after a certain time, since the battery pack will be diagnosed as faulty, since the SOC difference between the battery cell with the minimum SOC and the battery cell with the maximum SOC will be greater than the predetermined threshold.

In order to detect battery cells that are unsuitable or end-of-life for a given use and thus do not continue to use them for that given use, patent US 8332342 discloses a method in which the internal electrochemical behavior of a battery cell as it draws a discharge current for a given use is mathematically modelled and from which the theoretical SOC and terminal voltage after that use are then derived. Thus, the battery cell continues or does not continue to be used for the given application, depending on the theoretical SOC and terminal voltage.

The method described in this document has the disadvantage of being very complex to implement, since it requires very accurate knowledge of the electrochemistry of the battery cells and very high computing power.

Thus, current methods either do not have sufficient performance to distinguish actual faults from natural discharges or are very complex to implement.

Disclosure of Invention

The detection method according to the invention makes it possible to safely, easily and reliably detect faulty cells in a battery pack, in particular by not creating confusion with natural self-discharge, while overcoming the drawbacks highlighted in the prior art.

One subject of the invention is a method for detecting defective battery cells in a battery pack.

The method according to the invention is primarily characterized in that it comprises the following steps:

-a step of measuring the no-load voltage across the terminals of each battery cell in the battery pack,

-a step of calculating the deviation between the no-load voltage measured across the terminals of the battery cells and a known average no-load voltage according to the number of days in which the battery pack is not used,

-a step of calculating the difference between said calculated deviation and a variable n.σ, where n is an integer greater than or equal to one and σ is a known standard deviation depending on the number of days the battery is not used,

-a step of detecting a defective battery cell if the difference is greater than zero.

Thus, the method makes it possible to individually test each battery cell forming the battery pack in order to detect an actual failure of at least one of the battery cells. The method is based on comparing a calculated deviation between the no-load voltage measurement of the battery cells and the average no-load voltage each battery cell should have with a variable proportional to the standard deviation, the calculated deviation depending on the number of days the battery is not used. Therefore, the criteria for establishing a diagnosis of an actual failure of the battery cells are more complete and more accurate than simply evaluating the deviation of the states of charge of the battery cells forming the battery pack. Such a method is preferably carried out in a motor vehicle by means of a computer. Advantageously, by means of the method according to the invention, the driver is informed of the presence of at least one faulty battery cell in the battery pack when starting his vehicle. The terms "failed" and "defective" assigned to a battery cell are equivalent.

Advantageously, n is an integer between 2 and 10.

Preferably, the method according to the invention comprises a warning step when a defective battery cell is detected. In particular, if a defective battery cell is detected, means for preventing the battery from being recharged are implemented. It is important to inform the driver of the vehicle about such a fault in order to take the required measures as early as possible.

The warning step preferably uses a visual signal visible to a person carrying out the method. Advantageously, the visual signal is provided by a specific indicator light activated on the screen of the dashboard of the vehicle.

Advantageously, the warning step uses an audible audio signal for the person carrying out the method.

Advantageously, the detection method according to the invention implements a step of preventing the battery pack from being recharged when at least one defective battery cell is detected. In this way, once a defective battery cell is detected, the battery can no longer be recharged, which means that the user must visit a professional to replace the defective battery cell.

Preferably, if σ ≧ 5mV, n ═ 6.

Preferably, if σ ≧ 10mV, n ═ 4.

Advantageously, if σ ≧ 20mV, n ═ 3.

Advantageously, the detection method according to the invention comprises a step of displaying a loss of autonomy in case a defective battery cell is detected. Therefore, when the battery unit fails, this is reflected in a decrease in autonomy of the vehicle, and the vehicle driver is immediately notified thereof.

The detection method according to the present invention is advantageous in that the actual failure of the battery cells in the battery pack is safely and reliably detected, in particular, by eliminating a configuration in which some battery cells naturally discharge faster than others over time.

Drawings

A detailed description of a preferred embodiment of the detection method according to the invention is given below with reference to the following figures:

figure 1 is a flow chart illustrating the main steps of a method for detecting defective battery cells according to the prior art,

figure 2 is a flow chart illustrating the main steps of a method for detecting defective battery cells according to the invention,

FIG. 3 is a graph illustrating a first example of a distribution of normally operating cells of a battery pack as a function of a measured no-load voltage,

FIG. 4 is a graph illustrating a second example of a distribution of normally operating cells of a battery pack as a function of the measured no-load voltage,

fig. 5 is a graph illustrating a third example of the distribution of normal operating cells of a battery pack as a function of the measured no-load voltage, showing defective cells.

Detailed Description

Referring to fig. 1, a method for detecting a faulty battery cell in a battery pack from the prior art includes:

-a step of measuring the voltage across the terminals of each battery cell in the battery pack,

-a step of estimating the SOC of each of the battery cells,

-a step of determining the SOC difference between the most charged and the least charged battery cell,

a step of detecting a faulty battery cell in case the difference is greater than or equal to a predetermined value (which may be equal to 20% or 30%, for example).

Detecting at least one faulty battery cell in a battery pack is typically accompanied by a step of inhibiting charging of said battery pack, since the battery pack is considered to be inoperable due to the presence of the at least one faulty battery cell. Therefore, before the battery pack can be recharged, it is necessary to replace each faulty cell with a new cell equipped with full capacity. Now, this method is relatively limited because, for example, in the case where the vehicle is not used for a long time, natural discharge of the battery cells over time is not taken into consideration. In other words, this method is not able to ultimately distinguish a faulty cell from a valid, but discharged, cell.

The detection method according to the invention has the following features: so that a faulty battery cell can be reliably and unambiguously distinguished from a healthy but discharged battery cell.

In this manner, referring to fig. 2, the method for detecting a faulty battery cell in a battery pack according to the present invention includes the steps of:

-a step of measuring the no-load voltage across the terminals of each battery cell in the battery pack,

-a step of calculating the deviation between the no-load voltage measured across the terminals of the battery unit and the known average no-load voltage according to the number of days in which the battery is not used,

-a step of calculating the difference between said calculated deviation and a variable n.σ, where n is an integer greater than or equal to one and σ is a known standard deviation depending on the number of days in which the battery is not used,

-a step of detecting a defective battery cell if the difference is greater than zero.

Compared to the above described method from the prior art, the method according to the invention proposes a more realistic method by combining the phenomenon of natural charge loss of the battery cells when the battery is not used for a certain number of days. In this way, such natural charge loss will no longer be prone to bias the conditions for detecting a faulty cell.

The variable n. sigma can be adjusted according to the value of the standard deviation sigma. In this way, a number of advantages are presented, by way of example,

if σ is>5mV, then n may be set to 6,

if σ is>10mV, then n may be set to 4,

if σ is>20mV, then n may be set to 3.

Fig. 3 illustrates a first exemplary distribution of battery cells of a battery pack as a function of a no-load voltage measured across terminals of each of the battery cells. In other words, each point represents the number of battery cells having the same measured no-load voltage. It can be observed that when none of the cells fail, all points are organized around a gaussian mean. The measured voltage changes and thus the state of charge changes between the cells are due to self-discharge phenomena that are more or less manifested between different cells.

Fig. 4 illustrates a second exemplary distribution of battery cells of a battery pack as a function of a no-load voltage measured across terminals of each of the battery cells. All points are coherent as they make it possible to define a continuous and regular distribution curve without having to emphasize a single point that will leave the curve. Thus, at the origin of all these voltage measurements, the battery pack is free of any faulty cells.

Fig. 5 illustrates a third exemplary distribution of battery cells of a battery pack as a function of a no-load voltage measured across terminals of each of the battery cells. The graph of fig. 5 shows a singularity 1 located slightly outside of the curve representing the cell distribution as a function of the no-load voltage measured across the terminals of the cells, compared to the regular curve illustrated in fig. 4. This point reflects the presence of a virtually faulty cell in the battery pack. In this case, the battery is kept stationary and can no longer be recharged.

The detection method according to the present invention is implemented in a vehicle equipped with a battery by: an on-board computer with appropriate software, and various devices for measuring the voltage across the terminals of each battery cell. In this way, when the driver turns on the ignition switch, he is immediately and directly informed of the state of his battery.

In this respect, once the presence of at least one faulty battery unit has been detected, the detection method according to the invention implements a warning step that allows the driver to be informed of the presence of a faulty battery unit. This warning step can be performed, for example, by an indicator light or a characteristic audio signal appearing on the dashboard.

The detection method according to the invention may also implement a step of automatically preventing the recharging of the battery if at least one of the battery cells proves to be faulty and has not been replaced.