Methods for reducing the total charge loss of batteries

Abstract

Method for reducing the total charge loss of battery cells (16, 18) by balancing charge states SoC, comprising the following process steps: a) checking boundary conditions, specifically whether a previous balancing step occurred at least a defined time interval t_wait, - the temperature of a balancing unit (64) is below an adjustable temperature limit, - the charge states SoC_i of all battery cells (16, 18) are > SoC_MIN, b) determining a balancing requirement by determining whether a maximum difference of all charge states SoC_i of all battery cells (16, 18) is > DELTA_SoC, c) if process steps a) and b) are affirmed, the battery cells (16, 18) are balanced by the balancing units (64), during which balancing resistors R_bal are connected to the respective battery cells (16, 18) for a time t_i.mitt_i=Q_i⋅R_balU_OCV(SoC_i)Q_i≙ charge to be dissipated from the battery cell iU_OCV ≙ battery cell voltage with open circuitSoC_i ≙ state of charge of the battery cell iR_bal ≙ resistance value of the balancing resistord) the selection of the time interval t_wait between two balancing steps such that this corresponds to the execution of a maximum number of cycles of the balancing units (64), characterized in that the maximum number of cycles of the balancing unit (64) and the coverage of any symmetrization requirement by the duration t_i of a balancing step are adapted to each other.,

Description

State of the art Hybrid and electric vehicles use lithium-ion battery packs, which consist of a large number of electrochemical battery cells connected in series. A battery management system monitors the battery and, in addition to safety monitoring, aims to ensure the longest possible lifespan. To achieve this, it is essential to ensure that the state of charge (SoC) of the individual battery cells is balanced despite differing self-discharge rates. This is accomplished through appropriate cell balancing. Cell balancing is typically performed resistively, i.e., using at least one ohmic resistor. Each battery cell is assigned a resistor and a switching element to allow individual cells to be discharged across this balancing resistor. In addition to differing self-discharge rates of the individual battery cells, the capacities of the battery cells also vary due to manufacturing tolerances. This effect is negligible at the beginning of the battery cells' lifespan, but can increase over time due to differences in cell aging and result in a capacity difference of several percent between individual battery cells. In battery systems where the capacity of the individual battery cells is unknown and resistive balancing is performed to achieve a common state of charge, the total charge to be balanced is very high, as charge is unnecessarily dissipated via the balancing resistors, far exceeding the pure compensation of the various self-discharges. This effect is illustrated in Fig. 1. Fig. 1 shows two battery cells 16 and 18 of different capacities, which are initially charged to 50% SoC (State of Charge) according to step 1. The first battery cell 16 has a lower capacity than the second battery cell 18, which is indicated in Fig. 1 by a shorter line length. In step 2, the cells are charged. In step 3, the SoC is equalized again, i.e., the first battery cell is partially discharged resistively. In the subsequent step 4, both battery cells 16 and 18 are discharged. Finally, the second battery cell 18 must be partially discharged resistively to achieve a uniform SoC with respect to the first battery cell 16.The balancing requirement during the transition from step 4 to step 5 is partially created by the balancing during the transition from step 2 to step 3. If the battery system is permanently balanced so that all battery cells have an identical state of charge (SoC), not only is charge unnecessarily dissipated via the balancing resistors required for balancing, but this also results in an unnecessarily high number of switching operations by the balancing unit, which can negatively impact its lifespan. DE 10 2009 045 519 A1 relates to a battery system and a method for balancing the battery cells of this battery system. The battery system comprises a first battery element. The positive terminal of the first battery element is conductively connected to the negative terminal of the second battery element. A discharge device is provided for the partial discharge of the first and second battery elements. A voltage divider is designed to generate, starting from the electrical potential of the negative terminal of the first battery element and the electrical potential of the positive terminal of the second battery element, a first electrical potential that corresponds to the target value of the electrical potential at the positive terminal of the first battery element and the negative terminal of the second battery element.Comparison devices are used to compare the first electrical potential with a second electrical potential applied to the positive terminal of the first battery cell and the negative terminal of the second battery cell. Discharge devices are designed to discharge the first battery cell when the second electrical potential deviates positively from the first electrical potential, and to discharge the second battery cell when the second electrical potential deviates negatively from the first electrical potential. Description of the invention The present invention is based on the objective of enabling SoC-based balancing and / or balancing without knowledge of the current capacities of a number of series-connected battery cells in such a way that the loss of charge during balancing is as low as possible and the number of switching cycles of the balancing resistors is reduced and their service life is thereby increased. According to the invention, a method is proposed which aims at battery cell symmetry, i.e., cell balancing, and in which two successive balancing steps on a battery cell are only permitted after a minimum time interval has elapsed. This minimum time interval is, in particular, adjustable. An upper limit of this time interval results from the cell symmetry requirement, i.e., the cell balancing requirement, and the balancing charge per balancing step. Following the method proposed according to the invention, boundary conditions are first checked, specifically whether an electric or hybrid vehicle whose battery modules or battery cells are to be balanced is in park mode. The electric or hybrid vehicle must not be in either charging or discharging mode. Furthermore, it is checked whether the last balancing step occurred a defined period of time ago and that the temperature of a balancing unit, which compensates for charge differences between battery cells resistively (i.e., by means of a resistor), is below an adjustable temperature limit. In the first step of the process, it is also checked whether the state of charge (SoC) of all battery cells is above an adjustable limit. The next step of the process checks whether there is a need to perform a balancing operation.The system determines whether or not there is a need to balance the charge states of battery cells. To do this, it calculates the state-of-charge (SoC) difference of all battery cells that exceeds an adjustable threshold, DELTA_SoC. This involves determining the minimum SoC of all battery cells, SoC_MIN. If the individual state of charge of at least one battery cell, i, is greater than DELTA_SoC than SoC_MIN, balancing is required. If balancing is required according to the second process step and the conditions checked in the first process step are met, a balancing step is carried out according to the following boundary conditions: The battery cell balancing unit performs autonomous balancing by switching the relevant balancing units on the respective battery cells to be balanced for a specific period of time upon request. The balancing units are permitted to switch off if the temperature exceeds a certain threshold, but they must not switch back on automatically. In a subsequent fourth process step, the individual balancing requirement is determined for each battery cell. The required charge to be dissipated from battery cell i is calculated according to the equation Q_i = C_NOM · (SoC_i - SoC_MIN), where C_NOM represents the nominal capacity of the battery cells.Based on the voltage U_OCV at the given state of charge SoC of the cell and the value for the balancing resistor R_bal, the time during which the balancing steps should be carried out can now be determined using Ohm's law, according to the following relationship: with R_bal balancing resistor Q_i charge to be dissipated from cell i SoC_i current state of charge of battery cell i U_OCV battery cell voltage with open circuit. In a fifth process step, each balancing unit i is switched on for a given battery cell i for a time t_i, but for a maximum adjustable time. Adherence to an upper time limit serves to protect the balancing unit from overheating. Following the method proposed according to the invention, the balancing units, which resistively equalize charge states on the battery cells and connect the balancing resistors R_bal to the individual battery cells, switch off automatically when heated above a temperature threshold, without, however, switching on again automatically. The parameters Q_i, R_bal, U_OCV, SoC_i, and C_NOM can be adjusted according to the specific battery system and operating strategy. Ideally, the SoC_MIN parameter should be set above the desired charge reserve. It is also advisable that the state of charge (SoC) of all battery cells be above an adjustable limit, for example, 15%, to prevent accelerated deep discharge due to cell balancing. The parameter DELTA_SoC defines a permissible SoC variance below which balancing is not allowed. However, if balancing is performed towards this limit, the difference between the maximum and minimum state of charge (i.e., the SoC of the battery cells) is precisely DELTA_SoC. This value should be chosen to be large compared to the uncertainty interval in which the state of charge of the battery cell is determined. The time interval between two balancing steps, t_wait, should be chosen to match the maximum number of cycles of the balancing unit. For example, if the balancing unit can withstand N temperature cycles, the time interval t_wait should be greater than the lifetime / N. On the other hand, the difference in the self-discharge rates of the individual battery cells must also be taken into account when choosing the time interval t_wait.The time interval t_wait should be chosen to be large enough to compensate for the cell balancing requirements that arise during this time within a single balancing step. To ensure that both boundary conditions are met—i.e., the number of balancing cycles and the complete coverage of the cell balancing requirements—the duration of each balancing step must be adjusted. As a possible variant of the method proposed according to the invention for reducing overall charge losses, the time management, i.e., the staggered operation of the individual balancing units, can be controlled via the battery management system in such a way that, instead of all battery cells, individual, individually selected battery cells or subsets of battery cells are subjected to cell balancing. In this case, all battery cells or any subset of battery cells are assigned a suitable timer. The prerequisite of whether balancing is required is checked independently for each timer. The resulting advantage is that the individual battery cells can be divided in such a way that all battery cells of a specific subset can be balanced simultaneously without the balancing unit performing the balancing heating up too quickly or excessively. Advantages of the invention The method proposed according to the invention enables the balancing of charge differences, i.e., cell balancing of battery cells in a battery module of a battery pack for a hybrid or electric vehicle. This charge balancing is possible even with capacity differences of up to 20% between individual battery cells, without explicit knowledge of the individual cell capacities. The amount of capacity difference that the system can ultimately tolerate without knowledge of the actual capacities depends significantly on the driving profile. Furthermore, the charge loss resulting from cell balancing can be considerably reduced. Additionally, the service life of the resistors R_bal used for cell balancing can be significantly increased.The method proposed according to the invention advantageously allows cell balancing to be individually adapted to the self-discharge characteristics of different individual battery cells by selecting a corresponding minimum time interval, thus taking into account the different aging cycles of individual battery cells. A further advantage of the solution proposed according to the invention is that the time between two balancing steps can be flexibly adjusted to the driver's driving pattern. A further significant advantage of the solution proposed according to the invention is that the service life of the balancing resistors can be extended through time management. The service life can be extended by limiting the number of switching cycles.Only as many switching cycles are permitted as are required for balancing on the one hand and allowed by the discharge means on the other. Brief description of the drawings The invention is described in more detail below with reference to the drawing. It shows: Fig. 1 the balancing of different charge states, i.e., battery cell balancing to maintain the same state of charge (SoC) of the two battery cells; Fig. 2 a first periodic charge / discharge profile plotted over 24 h; Fig. 3 a second charge / discharge profile with a different charge / discharge characteristic, also plotted over 24 h; Fig. 4.1 a setup of a balancing unit; Fig. 4.2 a flowchart for carrying out balancing measures (cell balancing steps) on battery cells with time managers; Fig. 5 a simulation of the charge / discharge profile according to the representation in Fig. 2 with 20 battery cells over 365 days with self-discharge components and balancing steps; and Fig. 6 a simulation result of the second charge / discharge profile shown in Fig. 3, also with 20 battery cells over 365 days, also with Self-discharge fractions and cell balancing fractions. Design variants Fig. 1 shows the continuous execution of cell balancing steps to maintain an identical state of charge (SoC) of two battery cells 16, 18. Starting from a state of charge 10, an SoC level of 50%, Fig. 1 shows in step 1 that a charging process 12 proceeds vertically upwards with respect to the SoC level 10, while a discharging process, see reference numeral 14, proceeds in the opposite direction. The first battery cell is designated by reference numeral 16, the second by reference numeral 18. In the first step according to Fig. 1, both battery cells 16 and 18 have an identical state of charge (SoC) level 10 of 50%. In the second step, it can be seen that both battery cells 16 and 18 are charged above the SoC level of 50%. With respect to the slope, the first battery cell 16 is slightly discharged during the transition from the second to the third step until, as shown in the third step, the two SoC levels 24 of the first battery cell 16 and the second battery cell 18 correspond to each other again. The partial discharge in the second step is designated by reference numeral 22. In the fourth step, as shown in Fig. 1, a discharge process 14 of both battery cells 16 and 18 takes place such that the current SoC level 24 falls below the SoC level 10 of 50%.To bring both battery cells 16 and 18 back to an identical SoC level, a resistive partial discharge of the second battery cell is required, as indicated by reference numeral 22 in step 4 according to Fig. 1, as indicated in step 4. In step 5, the two battery cells 16 and 18 are again "balanced", i.e., they have an identical SoC level 24, which is below the SoC level 10 of 50%. Fig. 2 shows a first charging / discharging profile 30, in which a charging rate is plotted over time, in this case over a 24 h day. The first charge / discharge profile 30 shown in Fig. 2 indicates that a state-of-charge (SoC) swing 34 occurs between 5 a.m. and 8 a.m. during the day 32, during which time a battery experiences a long SoC swing, i.e., goes through a discharge phase 38. The discharge phase 38 ends around 8 a.m. Around 9 p.m., a pronounced charge phase 40 follows, which lasts until midnight, after which the charge rate 42 is again zero. In contrast, Fig. 3 shows a second charging / discharging profile 44, in which, for example, a taxi vehicle experiences many short state-of-charge (SoC) swings 46 over the course of the day 32, i.e., charging and discharging phases 38, 40 alternate hourly, with four charging phases 40 and five discharging phases 38 occurring, except for late morning until 10 a.m. Unlike the first charging / discharging profile 30 shown in Fig. 2, the SoC swings 46, the changes in state of charge, are significantly more pronounced in amplitude in the second charging / discharging profile 44 shown in Fig. 3. The SoC swings 46 are counted along the arrow with the reference numeral 48. Starting from midday, five charging phases 40 will follow during the afternoon and early evening until 8 pm, each interrupted by short discharge phases 38, each lasting approximately 1 hour. Figure 4.1 shows a circuit setup for balancing charge states. Such a balancing unit 64 comprises balancing resistors R1, R2, ..., Rn, which are interconnected via a switching logic 66. The illustration according to Fig. 4.2 shows a flowchart 70, according to which the method proposed according to the invention for balancing charge differences (cell balancing) is carried out following the solution proposed according to the invention. A battery management system comprises a battery control unit (BCU) which determines the current state of charge (SoC) of all battery cells 16, 18 connected in series. A balancing unit for each of the battery cells 16 and 18 comprises a number of resistors R_bal that can be connected to one of the battery cells 16, 18, as well as switching logic 66 according to Fig. 4.1. Following the method proposed according to the invention, a first step checks whether an electric or hybrid vehicle is in park mode 72 or not. To carry out the method proposed according to the invention for balancing charge differences in battery cells 16, 18, an electric or hybrid vehicle must not be in charging or discharging mode. Furthermore, the first step checks whether the state of charge (SoC) of all battery cells is above an adjustable limit, for example, above a limit of 15%. This preliminary check prevents deep discharge of individual battery cells 16, 18 through cell balancing, i.e., through the discharge of charge, as shown in position 74 in the block diagram according to Fig. 4.2.Finally, the boundary condition is checked to determine whether a balancing unit 64 has a temperature below an adjustable temperature limit of, for example, 40°C or 45°C, so that the balancing unit 64, i.e., the switching logic 66 and the resistors via which charge equalization is carried out on the individual battery cells 16, 18 by means of resistive circuitry, can be used. It should be noted that the method proposed according to the invention is described here only with reference to two battery cells 16, 18. In practice, i.e., in the actual application, a battery system will be used that can contain up to 100 or more battery cells and in which the charge equalization method proposed according to the invention is implemented. In decision step 76, it is decided whether a balancing step to equalize charge differences, i.e., cell balancing, is even necessary. For this purpose, a maximum state-of-charge (SoC) difference of all battery cells is calculated, which must be above an adjustable threshold DELTA_SoC, for example 3%, such that the following condition is met: If the need for cell balancing is denied at decision point 76, a second branch 80 is referred and at 98 a continuation of normal operation is decided, in the present case not to carry out cell balancing. If, however, the decision point 76 determines that at least one of the battery cells 16, 18 requires cell balancing, and all the boundary conditions according to the first process step are met, the process proceeds to the first branch 78, and a cell balancing step is carried out according to the following boundary conditions: The balancing process, i.e., the cell balancing, takes place autonomously; that is, the BCU instructs the relevant balancing units 64 to switch the resistors R_bal to the respective battery cells 16, 18 for a specific period of time, so that resistive cell balancing occurs. The balancing units 64 themselves may switch off if they heat up above a certain temperature threshold; however, it must be ensured that they do not switch back on automatically. If cell balancing is permitted, i.e., the temperature of the balancing unit 64 is below a temperature of, for example, 40°C or 45°C, indicated by position 84 in Fig. 4.2 (second condition), and a third condition 88 is met, i.e., one balancing step has elapsed more than t_wait (e.g., 19 h), and a first condition 82 is met, according to which SoC_i > SoC_MIN + DELTA_SoC, e.g., 3%, the individual balancing requirement is determined for each of the battery cells i. The charge to be dissipated from a battery cell i is given by the following relationship: where C_NOM corresponds to a normal capacity of all battery cells 16, 18.Based on the voltage value U_OCV, which is established at a given state of charge of battery cell i, and the resistance value of the balancing resistor R_bal, the time during which cell balancing should be performed can be determined using Ohm's law according to: with R_bal balancing resistor Q_i charge amount to be discharged U_OCV battery cell voltage with open circuit SoC_i current state of charge of battery cell i. The charge equalization, i.e., the balancing, is now performed by the BCU, which connects a corresponding balancing resistor R_bal to the respective battery cell 16, 18 of battery cells i for a specifically defined time t_i. This connection is maintained for a maximum of t_i, but for an adjustable maximum time. This upper time limit, i.e., the maximum time, serves to protect the balancing unit 64 (BCU ≙ Balancing Unit) from overheating. This is indicated in the flow diagram 70 according to Fig. 4.2 by the first branch 90, which causes the battery control unit 94 to be controlled, so that the resistive cell balancing is carried out for a period of x minutes and a timer reset 96 is triggered at the timer 100 for the corresponding battery cell i of the battery cells on which the cell balancing is now initiated by resistive connection. If conditions 82, 84, 88 are not met, the second branch 92 leads to continuation 98; the same applies to a branch from Timer-Reset 96 to continuation 98. The parameters mentioned above, such as Q_i, SoC_i, SoC_MIN, the resistance values ​​for the balancing resistors R_bal, etc., can be adjusted according to the specific battery system to be balanced with regard to the charge differences of individual battery cells 16, 18, or according to the operating strategy of the batteries. For example, the parameter SoC_MIN should be above the desired charge reserve. DELTA_SoC defines the permissible variance with regard to the state of charge (SoC) of the individual battery cells 16, 18, below which cell balancing is omitted. If charge differences are to be balanced down to this limit, i.e., the permissible SoC variance, the difference between the maximum and minimum state of charge (SoC) of battery cells 16, 18 corresponds precisely to DELTA_SoC.This value should definitely be chosen so that it is large compared to an uncertainty interval in which the individual state of charge (SoC) of battery cells 16, 18 can be determined. The method proposed according to the invention increases the service life of the balancing units 64, and thus the service life of the balancing resistors R_bal, through time management by ensuring that a time interval t_wait of, for example, 12 hours or more elapses between two balancing steps performed by the balancing unit 64. This ensures that the battery cell balancing does not react to every difference in state of charge caused by respective capacity differences. By incorporating time management in conjunction with the balancing process, it can be made much more robust compared to the balancing shown in Fig. 1, thus achieving several advantages. Firstly, the service life of the balancing resistors R_bal is significantly extended. Furthermore, the balancing unit 64 used will perform considerably fewer switching cycles, since, by necessity,A predetermined time interval t_wait must elapse between balancing processes. Furthermore, as illustrated in connection with the description of Fig. 1, the number of switching operations for resistive cell balancing can be drastically reduced, so that even undesired charge loss, which is discharged to cell balancing according to the prior art procedure although not necessary, can remain in the battery cells 16, 18, thus making cell balancing significantly more robust. The method proposed according to the invention, i.e., linking cell balancing with time management, is applied. Only as many switching cycles are performed by the balancing units 64 as are absolutely necessary for cell balancing and are currently permitted by the discharge means. In addition to the aforementioned time interval t_wait of 12 hours, time intervals of t_wait in the range of 19 hours, 20 hours, and more can also be defined, which must elapse between two balancing processes. The following achievable balancing profiles 60, 62 can be seen from the representations according to Figs. 5 and 6 by applying the method proposed according to the invention for reducing the total charge loss, the cell balancing proposed according to the invention. In Figs. 5 and 6, the charge losses 52 determined in ampere-hours during a simulation period are shown on the vertical axis. The charge losses 52 are divided into a part that is due to self-discharge 56 and a part that is due to balancing steps 58. If the [unclear] in Fig.2. If the first charging / discharging profile 30 shown is simulated on a number of battery cells 50 of twenty over 365 days, it can be shown that the total charge losses due to self-discharge 56 can be compensated for by the balancing steps 58 over each of the battery cells 1 to 20, so that overall a uniform state of charge is achieved for all twenty battery cells. With regard to the simulation result shown in Fig. 6, and a simulated charge / discharge cycle as shown in Fig. 3, a similar result is obtained. Here, as with the simulation result shown in Fig. 5, the self-discharges 56 are in some cases significantly higher compared to the proportions of the balancing steps 58.

Claims

Method for reducing the total charge loss of battery cells (16, 18) by balancing charge states SoC, comprising the following process steps: a) checking boundary conditions, specifically whether a previous balancing step occurred at least a defined time interval t_wait, - the temperature of a balancing unit (64) is below an adjustable temperature limit, - the charge states SoC_i of all battery cells (16, 18) are > SoC_MIN, b) determining a balancing requirement by determining whether a maximum difference of all charge states SoC_i of all battery cells (16, 18) is > DELTA_SoC, c) if process steps a) and b) are affirmed, the battery cells (16, 18) are balanced by the balancing units (64), during which balancing resistors R_bal are connected to the respective battery cells (16, 18) for a time t_i, with t _ i = Q _ i ⋅ R _ bal U _ OCV (S o C _ i) Q_ i ≙ charge to be dissipated from the battery cell i U_OCV ≙ Battery cell voltage in open circuit SoC_i ≙ State of charge of battery cell i R_bal ≙ resistance value of the balancing resistor d) the choice of the time interval t_wait between two balancing steps such that this corresponds to the execution of a maximum number of balancing cycles (64), characterized by , that the maximum number of cycles of the balancing unit (64) and the coverage of any symmetrization requirement by the duration t_i of a balancing step are adapted to each other. Method according to claim 1, characterized in that, according to method step c), the charge Q_i to be discharged from battery cell i is determined according to the following relationship: Q_i = C_NOM ⋅ ( SoC_i − SoC_MIN ) with C_NOM ≙ nominal capacity of the battery cells (16, 18). Method according to one of the preceding claims, characterized in that, according to method step b), the balancing requirement is affirmed if, for at least one battery cell i, the individual state of charge SoC_i is greater than DELTA_SoC than SoC_MIN. Method according to one of the preceding claims, characterized in that the balancing units (64) automatically switch off and remain switched off when heated above a temperature threshold TGrenz. Method according to one of the preceding claims, characterized in that the balancing units (64) remain switched on for the time t_i, but at most for an adjustable maximum time tmax. Method according to one of the preceding claims, characterized in that the parameter SoC_MIN, the minimum state of charge of an individual battery cell i, is above a desired charge reserve. Method according to one of the preceding claims, characterized in that the parameter DELTA_SoC defines a permissible SoC variance below which no balancing steps are performed. Method according to one of the preceding claims, characterized in that in a balancing process, balancing is performed on the parameter DELTA_SoC and the difference between the maximum and minimum state of charge SoC of the battery cells (16, 18) corresponds exactly to the value DELTA_SoC. Method according to one of the preceding claims, characterized in that the time interval t_wait is chosen to be so large that any need for symmetry arising during this time period is compensated during a balancing step. Method according to one of the preceding claims, characterized in that, according to the above method, individual battery cells i or subsets of battery cells i are individually balanced.

Images