A method for diagnosing faults in an electrical energy storage device, particularly for electric vehicles.

Abstract

A method for diagnosing faults in an electrical energy storage device (10), particularly for electric vehicles, said device comprising a plurality of modules, each comprising a plurality of electrical energy storage cells, said method comprising a step (20) for detecting faults by comparing said cells, said method further comprising a step (30) for defining groups of cells (12) taking into account the lifespan of the electrical energy storage device, said fault detection step (20) being carried out independently for each of said groups. Figure for the abstract: Figure 4

Description

Title of the invention: Method for diagnosing faults in an electrical energy storage device, particularly for electric vehicles

[0001] The invention relates to a method for diagnosing faults in an electrical energy storage device, particularly for electric vehicles. It relates more specifically to electrical energy storage devices used for vehicle motorization, in particular high-voltage lithium-ion type electrical energy storage devices, for example 400 Volts.

[0002] In this field, electrical energy storage devices, also called battery packs, are known. These devices consist of a set of cells, each containing a portion of the electrical charge of the storage device. The cells are electrically connected in series to achieve the power levels required by the storage device. The cells are further grouped into modules that can be changed individually within the same storage device, for example, during maintenance operations.

[0003] In motor vehicles, a control system selectively allows a current supply to an electric machine used for moving the vehicle, from said storage device whose cells are then discharged, or a current supply to said storage device, for example from a terminal and / or, through said electric machine, by regenerative braking, the cells of said storage device then being recharged.

[0004] For optimized operation of storage devices, a conventional solution is to monitor the state of charge (SOC) difference between each cell. The state of charge of a cell refers to the amount of energy available in a cell at a given time compared to the maximum amount of energy available at that cell. A storage device is considered unbalanced when the cells do not have the same electrical state of charge. These differences can arise due to manufacturing processes, variations in the materials used, dimensional variations, or environmental conditions. However, maintaining all cells in the electrical energy storage device with the same electrical charge level is preferred to utilize the battery's total capacity.

[0005] Fig. 1 illustrates this in the form of a histogram in the case of a storage device comprising two cells Cl, C2 connected in series and exhibiting, at a given moment, an imbalance, here an electrical charge difference of 15% as represented by the first two bars 1a, 1b of the histogram.

[0006] If the battery is charged without first compensating for this imbalance, the charging process must stop as soon as one of the cells reaches its maximum electrical charge. Since the cells are connected in series, continuing the charging process would present numerous drawbacks. The other battery cell will then have reached only 85% of its full electrical charge storage capacity, as shown by the following two bars 2a and 2b of the histogram.

[0007] In the case of the discharge phase, the situation is identical. The electric vehicle can use the energy from the storage device until the first of the cells reaches its lowest electrical charge. Thus, there will be an unused charge of 15% stored in cell C1 of the storage device.

[0008] Consequently, for a 15% imbalance between charging and discharging losses, only 70% of the battery capacity is usable due to the cell imbalance. Ultimately, it is clear that, between charging and discharging, only a portion (70%) of the total charging range can be used as long as the cells exhibit such a 15% imbalance: the capacity loss is therefore rapid.

[0009] To overcome this drawback, it is known to balance the charge of the electrical cells with each other, using a selective charging and / or discharging circuit integrated into a control system for the storage devices. Such a balancing process is triggered periodically, for example when the vehicle is stationary.

[0010] Furthermore, a simple diagnostic currently used by the control system of storage devices, in order to check the good health of cells, consists of monitoring that the operation of the balancing is effective, and therefore of checking that the voltage difference between the highest cell and the lowest cell is less than a threshold, for example of 200mV for cells whose maximum voltage is on the order of 4 Volts.

[0011] This solution has drawbacks. In particular, when a module of the storage device is changed, the Ampere Hour (AH) capacities of the cells of the new module will be generally greater than for the other cells because they are new.

[0012] Figure 2 illustrates the problem encountered. It shows the evolution of the state of charge of the cells over time, with the cells subjected to a charge and then a discharge current H. To facilitate reading the figure, only an average state of charge of the cell modules is shown, for a battery with only three modules, each exhibiting a state of charge SOC1, SOC2, and SOC3. The SOC1 and SOC3 curves are essentially superimposed. The initial state of charge of all the modules is 50%.

[0013] The modules exhibiting the state of charge SOC1 and SOC3 are original modules of the storage device, while the module exhibiting the state of charge SOC2 was exchanged during a previous step of repairing the storage device.

[0014] This illustrates a situation where modules with state of charge SOC1 and SOC3 have a residual capacity of 80%, while module SOC2, being new, has a capacity of 100%. When charging the pack to 100%, with all modules connected in series, module 2 will only be charged to 90% because modules 1 and 3 will have reached full charge and will terminate the charging process. A 10% difference in state of charge can thus be reached at the end of the charging cycle, solely due to battery wear. The same difference in state of charge will also be reached if the storage device is completely discharged.

[0015] Thus, the storage device's control system may conclude that a fault exists when it is merely a difference in wear between the modules. One solution is to choose fault signaling thresholds based on a 20% difference in state of charge, but this presents a risk because it could mask a real fault in one of the cells and not just wear.

[0016] The invention aims to overcome at least in part the previous drawbacks and proposes to this end a method for diagnosing faults in an electrical energy storage device, in particular for electric vehicles, said device comprising a plurality of modules each comprising a plurality of electrical energy storage cells, said method comprising a step of detecting faults by comparing said cells with each other, said method further comprising a step of defining groups of cells taking into account the life of the electrical energy storage device, said step of detecting faults being carried out independently for each of said groups.

[0017] Thus, according to the invention, all the cells are not systematically compared with each other, but this comparison is carried out within groups, moreover, not using predefined groups but groups that take into account the lifespan of the storage device and, consequently, the simultaneous presence in the storage device of cells that are more or less new. This avoids false positives that would be due not to an actual failure of one of the cells but to a difference in the wear of said cells.

[0018] In other words, the present invention proposes an implementation of the diagnosis of proper cell operation with a solution which makes it possible to manage, among other things, the case of changes of cell modules, for example from an after-sales repair.

[0019] According to various additional features of the invention, which may be taken together or separately and which constitute so many embodiments of the invention: - said group definition stage includes a step of detecting a change in the module(s) and a step of forming groups taking said changes into account, - said group formation stage is defined iteratively and includes, with each change of one or more of the modules, referred to as exchanged module(s): • - the formation of at least one group comprising the cells of the or of some of the aforementioned exchanged modules, • the formation of at least one other group comprising the cells of the modules (M1, M2, M3, ...) not exchanged and / or the cells (12) of the other exchanged modules, - This group formation stage includes the following steps: • the formation of a group comprising the cells of the said module(s) exchanged at a given time t, • the respective formation of a group comprising the cells of the aforementioned module(s) previously exchanged, for each of the other module changes that may have occurred, • the formation of a group comprising the non-exchanged cells, - said step of defining the cell groups (G1, G2, ...) is carried out using one or more initial pieces of information relating to each of said cells and / or modules, - This first piece of information is defined by the energetic health state of the cells. - said cell group definition step is carried out according to a nearest neighbor classification method, - there are N cells per module, N being an integer greater than or equal to two, - The fault detection step is carried out using one or more secondary pieces of information relating to each of said cells, - said process includes a step of measuring and / or evaluating said second piece of information, - said second piece of information is the state of charge of the cells, - said defect detection step includes a step of determining the greatest difference between the second pieces of information and / or a step of comparing said greatest difference with a threshold, - said step of determining the strongest difference includes a step of selecting said second strongest and / or weakest information, - said detection step generates fault information if said largest difference is greater than approximately 20%, or even approximately 10%, - said measurement and / or evaluation step takes place following a relaxation, that is to say a rest period of the cells of said device, - said measurement and / or evaluation step is carried out jointly for all cells, before the distribution of said cells into said groups, - said measurement and / or evaluation step is carried out separately for each of said groups, after distribution of said cells into said groups, - said method includes a step of starting a control system for said storage device, - said process includes a step of verifying the operating status of said cells.

[0020] The invention also relates to an embedded control system in a motor vehicle comprising hardware and / or software elements implementing the diagnostic process mentioned above.

[0021] The invention further relates to an electrical energy storage device and / or a motor vehicle comprising said control system.

[0022] The invention further relates to a computer program product comprising program code instructions recorded on a computer-readable medium to implement the steps of the diagnostic process mentioned above, when said program is running on a computer.

[0023] The invention also relates to a data recording medium, readable by a computer, on which is recorded a computer program comprising program code instructions for implementing the diagnostic process as mentioned above.

[0024] The invention will be better understood, and other objects, details, features and advantages thereof will become more apparent in the course of the detailed explanatory description that follows, of at least one embodiment of the invention given by way of purely illustrative and non-limiting example, with reference to the accompanying schematic drawings, among which:

[0025] [Fig-1] already mentioned schematically illustrates a charging / discharging process of cells of an electrical energy storage device to which the diagnostic method according to the invention is intended to be applied;

[0026] [Fig.2] already mentioned illustrates schematically a situation that risks leading to a false positive with state-of-the-art diagnostic procedures;

[0027] [Fig.3] illustrates schematically an example of an energy storage device to which the diagnostic method according to the invention is intended to be applied;

[0028] [Fig.4] schematically illustrates an example of implementation of the process according to the invention;

[0029] [Fig.5] schematically illustrates an example of grouping points according to a classification method using a nearest neighbors method;

[0030] [Fig.6] is a graph illustrating a first scenario allowing us to conclude that there is a defect according to an example of implementation of the process according to the invention;

[0031] [Fig.7] is a graph illustrating a second scenario allowing one to avoid concluding erroneously that there is a defect according to the same implementation example as in [Fig.6].

[0032] It should first be noted that the terms "first", "second", "third", ... are used only to distinguish the components concerned from each other and do not imply any order or possible importance of said components.

[0033] The invention relates to a method for diagnosing faults in an electrical energy storage device 10.

[0034] As illustrated in [Fig. 3], said device 10 comprises a plurality of electrical energy storage cells 12. Preferably, each of said cells 12 contains a portion of an electrical charge of the storage device.

[0035] The cells 12 are, for example, electrically connected in series to achieve the power levels required for the storage device. For example, each cell 12 has a nominal voltage of approximately 4 volts, and the storage device has a nominal voltage of approximately 400 volts, requiring at least one set of one hundred cells connected in series. Such sets can also be connected in parallel to increase the power of said electrical energy storage device.

[0036] Said cells 12 are grouped electrically and / or mechanically into modules M1, M2, M3, .... Said cells 12 are, for example, N in number per module, N being an integer greater than or equal to two (a typical value of N will be between 8 and 18). The modules are designed to be interchangeable independently of one another, particularly during maintenance or repair operations. By way of example, said modules are advantageously connected in series and / or in parallel.

[0037] Said storage device 10 forms, in particular, a battery of a motor vehicle, in particular a battery used for the motorization of the vehicle.

[0038] In such an application, a current supply is operated to an electric machine used to move the vehicle from said storage device 10 whose cells 12 are then discharged, or a current supply to said storage device, for example from a fixed terminal and / or, through said electric machine, by regenerative braking, the cells 6 of said storage device being then recharged.

[0039] In order to optimize the operation of said storage device, as explained in relation to [Fig. 1], it is advantageous to allow selective charging and / or discharging of each of the cells, particularly for the purpose of balancing their electrical charge. Such a balancing process is triggered periodically, for example when the vehicle is stationary.

[0040] It uses a control system, embedded in the motor vehicle, for example a control system for said storage device. Said control system includes, for example, hardware and / or software elements such as measuring sensors, digital processing units such as microprocessors, and / or data storage memories, said processing units using information received from the sensors and / or stored in said memories and delivering information causing the charging and / or discharging of said electrical energy storage device, in particular for such balancing. Said system further includes, in particular, an electrical charging and / or discharging circuit equipped with electrical components such as switches connecting said cells according to instructions received from said processing units.

[0041] The method according to the invention is intended to determine the possible occurrence of a defect in any of said cells 12. Said defect consists, in particular, of the occurrence of a micro short circuit. However, it may be any other type of defect, in particular defects having an influence on the state of charge of the defective cell(s), as will be understood below.

[0042] As illustrated in [Fig.4], said method includes a step 20 of detecting defects by comparing said cells 12 with each other.

[0043] According to the invention, said method further includes a step 30 of defining groups of cells 12 taking into account the life of the storage device.

[0044] By "life of the storage device", we mean, in particular, any event, external conditions and / or other, likely to cause a different state of the cells 12 depending on the time spent by each in said storage device.

[0045] Furthermore, said step 20 of fault detection is carried out separately for each of said groups Gl, G2, ..., as symbolized by the arrow marked 40. It is understood that said step 20 of fault detection is therefore carried out after said step 30 of definition of groups Gl, G2, ....

[0046] Thus, the entire set of cells 12 is not compared with each other, but this comparison is carried out within the defined groups G1, G2, ..., individually group by group. The comparison therefore takes place between cells that are more homogeneous. Moreover, by not using predefined groups but groups generated taking into account the lifespan of the storage device, in particular the simultaneous presence in the storage device of cells that are more or less new, this homogeneity is reinforced. This avoids false positives that would be due not to an actual failure of one of the cells but, for example, to a difference in the age of said cells.

[0047] According to a first example of implementation of the process according to the invention, said step of defining the groups Gl, G2, ... includes a step of detecting a change in the module(s) M1, M2, M3, ... and a step of forming the groups Gl, G2, ... taking into account said changes.

[0048] For this purpose, for example, module change information is generated during the said change(s). This module change information includes, in particular, the identification of the module changed, and even its date of change. This module change information is stored, for example, in the control system of said storage device and is available for use during the group formation step.

[0049] Said step of group formation Gl, G2, ... is defined iteratively and includes, at each change of one or more of the modules, called exchanged module(s):

[0050] - the formation of at least one group comprising cells 12 of the or certain of the aforementioned exchanged modules,

[0051] - the formation of at least one other group comprising the 12 cells of the modules not exchanged and / or cells 12 of the other exchanged modules.

[0052] Preferably, said step of forming groups Gl, G2, ... comprises the following steps:

[0053] - the formation of a group comprising the cells 12 of said module(s) exchanged at a given moment t,

[0054] - the respective formation of a group comprising cells 12 of said cell(s) modules previously exchanged, for each of the other module changes that may have occurred,

[0055] - the formation of a group comprising the 12 non-exchanged cells.

[0056] In other words, a new group is defined each time one or more modules are changed simultaneously. Thus, before the first change, there is only one group of cells. After the first change, there are two: a first group with the unchanged cells and a second group with the cells of the or modules changed. In other words, the number of groups is equal to one plus the number of module changes.

[0057] According to one embodiment, said step of defining the groups of cells (G1, G2, ...) is carried out using one or more first pieces of information relating to each of said cells and / or said modules.

[0058] Said first information is defined, for example, by a state of health, in particular energy, of the cells 12, also known by the Anglo-Saxon acronym SOHE (for "State Of Health Energy"). Alternatively, it consists, among other things, of parameters related to the modules M1, M2, ... such as the age of the modules, the mileage covered by the vehicle with the modules M1, M2, ..., the number of charge / discharge cycles of the modules, ..., the said parameter(s) being recorded and stored so that said control system knows their value at all times.

[0059] As illustrated in [Fig. 5], said step 30 of defining the groups Gl, G2, ... of cells 12 is carried out, for example, according to a nearest neighbor classification method. According to this method, the cells 12 are distributed as points 50 according to a value of the first piece of information associated with each of the cells 12 in order to distribute them into groups. It is then possible to define groups according to the proximity of the points to each other. Here, we see two groups, for example, a first group Gl corresponding to one of the recently changed modules and the others G2 corresponding to the rest of the modules, considering that they are all original modules or at least of the same age.

[0060] By way of example, in order to avoid excessive dispersion, at least one of the groups Gl, G2, ... generated by said step 30 of defining the cell groups comprises at least 2 x N cells. More generally, the number of groups Gl, G2, ... is strictly less than the number of modules M1, M2, M3, ....

[0061] Alternatively or cumulatively, the fault detection step is carried out using one or more second pieces of information relating to each of said cells.

[0062] In the case of group formation using one or more first pieces of information relating to each of the cells, the said second piece of information is advantageously uncorrelated as much as possible with the said first piece of information.

[0063] Referring again to [Fig. 4], it can be seen that, in the illustrated embodiment, the process includes a step 15 of measuring and / or evaluating the second piece of information for the cells 12. In the case where, as mentioned above, the physical quantity is a state of charge of the cells 12, this is estimated, in particular, from a voltage across the cells 12, especially in an open circuit. Tables may also be used for this purpose. giving said state of charge as a function of said open-circuit voltage. This step is advantageously carried out for all of said cells 12, for example module by module and, within each module, cell by cell.

[0064] Said voltage is measured and / or calculated, for example, by sensors of the control system.

[0065] By way of example, said defect detection step 20 includes a step of determining the largest difference between the second pieces of information recorded during said measurement and / or evaluation step and / or a step of comparing said largest difference with a threshold. Preferably, said step of determining the largest difference includes a step of selecting said second largest and / or smallest piece of information recorded during said measurement and / or evaluation step, within each group.

[0066] In other words, in the case of using the state of charge as the second characteristic information of the cells, we identify in each group the cell with the maximum value of the state of charge (SOCmax) and the cell with the minimum value of the state of charge (SOCmin), then we calculate the difference between these two values ​​and compare the difference with the chosen threshold.

[0067] By carrying out this or these steps, group of cells by group of cells, we avoid looking for a difference between extrema associated with cells 12 that are too different from each other, taking into account the life of the energy storage device.

[0068] Preferably, said detection step 20 generates a fault signal if said largest difference is greater than approximately 20%, or even approximately 15%, or even 10%. Indeed, thanks to the invention, it is possible to lower the fault signaling thresholds since the comparison is more relevant due to the grouping.

[0069] A fault detection is illustrated in the embodiment with definition of the groups according to the changes in module in relation to figures 6 and 7.

[0070] More specifically, the storage device here consists of eight modules of twelve cells each. The cells are numbered successively according to the order of the modules. Thus, the cells bearing the cell index 1 to 12 belong to module M1 and so on.

[0071] Module M2 was changed on a given date and its condition is probably different, normally better, than that of the other modules.

[0072] In this case, the number of groups therefore changes from one before the change of module M2 to two after the change of said module M2 and, using the information returned to said control system of the storage device, said control system is able to establish that:

[0073] - the first group comprises modules M1 and M3 to M8, that is, eighty- four cells,

[0074] - the second group comprises the M2 module, that is to say, twelve cells.

[0075] In a first case, corresponding to [Fig.5], cell 67 has an SOC of 70.5% whereas, in the first group, the difference between SOCmax and SOCmin is 21.3%.

[0076] Thus, with a defect threshold set at a deviation greater than 15%, the existence of a defect in the first group is highlighted and, in this case, it is determined that the defect corresponds to cell 67.

[0077] In a second case, corresponding to [Fig. 6], it is cell 17 of module M2 that has a SOC of 70.5%. It is not faulty because it still has an acceptable deviation from module 2. Indeed, in the second group, the difference between SOCmax and SOCmin is 10.5%, which is less than the threshold of 15%.

[0078] According to the embodiment illustrated in [Fig.4], said measurement and / or evaluation step 15 is carried out separately for each of said groups, after distribution of said cells 12 into said groups Gl, G2, ....

[0079] Alternatively, not illustrated, said measurement and / or evaluation step 15 is carried out in a common manner for all cells 12, before distribution of said cells into said groups Gl, G2, ....

[0080] Advantageously, said measurement and / or evaluation step 15 takes place after recharging and / or balancing of the cells of said storage device, for example when the vehicle is stopped.

[0081] By way of example, said method includes a step of starting up a control system for said storage device and / or a step of verifying the operating status of said cells, in particular a verification that none of said cells is in use, said storage device being neither charging, nor discharging, nor even balancing. Said verification step advantageously occurs before measurement and / or evaluation step 15.

[0082] The start-up step may occur after the charging and / or balancing steps which may have taken place in a previous phase after which said control system is stopped before being restarted to perform the diagnostic process according to the invention.

[0083] Preferably, said method further comprises a step of reporting the cell(s) for which the fault information was issued within one or more of the groups. This is, for example, a step subsequent to all those described above.

[0084] According to an advantageous embodiment of the invention, said process comprises the following steps, taken successively: - the control system startup stage, - the initial step of verifying the operating status of said cells 12, - said step 15 of measuring and / or evaluating said first information for each of the cells 12, in particular to obtain a reliable estimate of the state of charge of each cell from the voltage across the cell terminals, - said step 30 of defining the groups Gl, G2, ..., for example according to the change(s) in module that have occurred and / or according to their SOHE, - said step 20 of fault detection, by executing the following sub-steps, group by group: • We identify, within the group, the cells exhibiting the SOCmax and SOCmin values, • we calculate the difference (SOCmax - SOCmin), • This difference is compared to the chosen threshold, • if the difference is greater than the threshold, the said fault information is generated for the cell(s) involved in the group(s) concerned.

[0085] The invention also relates to the control system, mentioned above, which is configured to implement the diagnostic process described above. The invention further relates to the motor vehicle equipped with said system.

[0086] The invention further relates to a computer program product comprising program code instructions recorded on a computer-readable medium to implement the steps of the diagnostic process described above, when said program is running on a computer.

[0087] The invention further relates to a data recording medium, readable by a computer, on which is recorded a computer program comprising program code instructions for implementing the diagnostic process described above.

Claims

Demands

1. A method for diagnosing faults in an electrical energy storage device (10), particularly for electric vehicles, said device comprising a plurality of modules (M1, M2, M3, ...) each comprising a plurality of electrical energy storage cells (12), said method comprising a step (20) of fault detection by comparison of said cells, said method further comprising a step (30) of defining groups (G1, G2, ...) of cells (12) taking into account the life of the electrical energy storage device, said step (20) of fault detection being carried out independently for each of said groups (G1, G2, ...).

2. Method according to claim 1 wherein said group definition step (G1, G2, ...) comprises a step of detecting a change in the module(s) (M1, M2, M3, ...) and a group formation step (G1, G2, ...) taking into account said changes.

3. A method according to the preceding claim wherein said group formation step (G1, G2, ...) is defined iteratively and comprises, at each change of one or more of the modules (M1, M2, M3, ...), said exchanged module(s), - the formation of at least one group comprising the cells (12) of the or some of said exchanged modules, - the formation of at least one other group comprising the cells (12) of the non-exchanged modules and / or the cells (12) of the other exchanged modules.

4. A method according to the preceding claim in which said group formation step (G1, G2, ...) comprises the following steps: - the formation of a group comprising the cells (12) of said module(s) exchanged at a given time t, - the respective formation of a group comprising the cells (12) of said module(s) exchanged previously, for each of the other module changes that may have occurred, - the formation of a group comprising the non-exchanged cells (12).

5. A method according to claim 1 wherein said cell group definition step (G1, G2, ...) is carried out using of one or more initial pieces of information relating to each of said cells (12).

6. Method according to the preceding claim wherein said first information is defined by an energetic health state of the cells.

7. A method according to any one of claims 5 or 6 in which said step (30) of defining the groups (G1, G2, ...) of cells (12) is carried out according to a classification method using nearest neighbors.

8. A method according to any one of claims 5 to 7 wherein said cells (12) are of the number N per module, N being an integer greater than or equal to two, at least one of the groups generated by said step (30) of defining the groups (G1, G2, ... ) of cells (12) comprising at least 2 x N cells (12).

9. A method according to any one of the preceding claims wherein the defect detection step is carried out using one or more second pieces of information relating to each of said cells.

10. Method according to the preceding claim wherein said second information is a charge state of the cells.

11. On-board control system in a motor vehicle comprising hardware and / or software components implementing the diagnostic method according to any one of the preceding claims.

12. Motor vehicle comprising said control system according to the preceding claim.

13. Product computer program comprising program code instructions recorded on a computer-readable medium to implement the steps of the diagnostic process according to any one of claims 1 to 10, when said program is running on a computer.

14. A computer-readable data recording medium on which is recorded a computer program comprising program code instructions for implementing the diagnostic method according to any one of claims 1 to 10.

Images