Battery health state estimation method and device, medium and BMS

Abstract

The invention relates to the technical field of power batteries, in particular to a battery health state estimation method and device, a medium and a BMS. Comprising the following steps: estimating capacity variation and SOC variation of a battery; dividing the capacity variable quantity by the SOC variable quantity to obtain a self-learning capacity value of the single battery; weighting the self-learning capacity value of the battery monomer and the empirical capacity value of the battery monomer through the weight to obtain a final battery health state; the capacity variation of the battery is obtained by correcting the basic variation. According to the invention, the estimation precision of the health state can be improved, and system errors caused by insufficient capacity statistics are made up; especially for a small-capacity battery pack, self-discharge and AFE power consumption in the dormancy period cannot be ignored, and when the compensation capacity is not considered, the calculation deviation is too large, the method also avoids the great jump of the health state estimation result, indirectly avoids the great jump of the SOC estimation result, and finally improves the user experience.

Description

Technical Field

[0001] The present invention relates to the technical field of power batteries, and in particular to a battery health state estimation method, device, medium and BMS. Background Art

[0002] For new energy vehicles such as pure electric vehicles and hybrid vehicles, the state of health (SOH) of the on-board battery is an important indicator of battery aging. In current battery management systems, capacity is usually used as an indicator of health status. Accurate estimation of health status is a prerequisite for accurate estimation of remaining power, and will also greatly affect the end-user's car experience.

[0003] During use, the real capacity of the battery cannot be measured by the charger and discharge machine. The on-board capacity estimation methods can be divided into two categories, namely, empirical aging models and capacity self-learning methods. The first method establishes an empirical aging model based on the results of the aging test. However, the aging test is usually a bench test of a single package, which makes it difficult to consider the impact of the actual vehicle environment on aging, and the data volume is large. When the data volume is insufficient, the estimation effect is poor. The second method counts the capacity change and SOC change over a period of time, and the difference between the two is the real capacity value at this time. However, the accuracy of the algorithm depends on the accuracy of the calculation of the capacity change and SOC change. The capacity change is usually calculated by integrating the current in ampere hours, without considering the capacity change that cannot be obtained by integrating the entire package current due to self-discharge, balanced power consumption, and AFE power consumption, and causes a systematic deviation between the calculated capacity change and the real capacity change. Especially for small-capacity battery packs, the difference will have a more obvious impact. When the battery is in a charging condition, the estimated capacity value is too small, while during the battery discharge process, the obtained capacity value is too small, which ultimately causes deviations in SOC and allowable discharge energy, affecting the charging process, cruising range and other aspects of customer experience. Summary of the invention

[0004] The invention discloses a battery health state estimation method, device, medium and BMS, which can improve the battery health state estimation accuracy.

[0005] To achieve the above purpose, on the one hand, a battery health status estimation method is provided, and the specific method is as follows:

[0006] Estimate the capacity change and SOC change of the battery during the preset test period;

[0007] The capacity change is divided by the SOC change to calculate the self-learning capacity value of the battery cell;

[0008] The capacity variation of the battery is obtained by correcting the basic variation by the corrected variation, and the corrected variation includes any one or more of the following parameters:

[0009] Single cell balanced capacity, AFE consumption capacity and self-discharge consumption capacity.

[0010] Furthermore, before estimating the battery health status, trigger conditions are also included;

[0011] When the maximum and minimum module temperatures in the battery are within the preset temperature range and the single cell SOC is within the preset SOC range, the trigger condition is met, otherwise it is not met.

[0012] Furthermore, the basic variation is calculated by integrating the ampere-hour value collected by the current sensor.

[0013] Furthermore, if a dormancy process is experienced during the preset test period, the correction change amount of the dormancy process is calculated based on the dormancy time, temperature, and remaining balanced capacity;

[0014] If the sleep process is not experienced during the preset test period, the correction change of the power-on process is calculated based on the average module temperature and the balanced current.

[0015] Furthermore, the specific estimation method of the balanced capacity of the single cell is as follows:

[0016] The balancing current of each battery cell is obtained, and the balancing current of each battery cell is integrated by ampere-hour to calculate the balancing capacity of each battery cell in the preset test period.

[0017] Furthermore, the AFE consumption capacity is specifically estimated as follows:

[0018] Obtain the operating current of each AFE sampling chip, perform ampere-hour integration on the operating current of each AFE sampling chip, and calculate the AFE consumption capacity;

[0019] When in sleep mode, the AFE consumption capacity is calculated by multiplying the low power mode current by the sleep time.

[0020] Furthermore, the self-discharge consumption capacity is specifically estimated by the following method:

[0021] The self-discharge current of each single cell during the dormant process is obtained, and the self-discharge current of each single cell is integrated in ampere-hours to calculate the self-discharge consumption capacity.

[0022] Furthermore, after the self-learning capacity value of the battery cell is calculated, the empirical capacity value of the battery cell is estimated by the empirical aging model, and the final battery health state is obtained by weighting the self-learning capacity value of the battery cell and the empirical capacity value of the battery cell.

[0023] Furthermore, the weight is determined by the proportion of the calculated capacity change estimation error and SOC change estimation error to the battery cell self-learning capacity value;

[0024] The greater the proportion of the capacity change estimation error and the SOC change estimation error to the battery cell self-learning capacity value, the smaller the weight corresponding to the battery cell self-learning capacity value.

[0025] Furthermore, the capacity change estimation error includes any one or more of the following parameters:

[0026] Single cell balanced capacity estimation error, AFE consumption capacity estimation error and self-discharge consumption capacity estimation error.

[0027] Furthermore, the estimation error of the balanced capacity of the single cell is calculated from the error range caused by the deviation between the extreme value and the normal value of Rdson.

[0028] Furthermore, the AFE consumption capacity estimation error is calculated by estimating a current sensor sampling error.

[0029] Furthermore, the self-discharge consumption capacity estimation error is calculated by estimating a current sensor sampling error.

[0030] To achieve the above object, on the other hand, a battery health status estimation device is provided, including a capacity change estimation module, an SOC change estimation module, a battery cell actual capacity value calculation module and a weight fusion module;

[0031] The capacity change estimation module estimates and corrects the capacity change of the battery within a preset test period;

[0032] The SOC change estimation module estimates the SOC change within a preset test period;

[0033] The actual capacity value calculation module calculates the battery cell self-learning capacity value by dividing the corrected capacity change by the SOC change;

[0034] The weight fusion module obtains the final battery health state by weighting the battery monomer self-learning capacity value and the battery monomer experience capacity value, and the battery monomer experience capacity value is estimated by the experience aging model.

[0035] To achieve the above-mentioned purpose, on the other hand, a storage medium is provided, which stores a plurality of instructions, and a processor loads the plurality of instructions to execute the above-mentioned battery health status estimation method.

[0036] In order to achieve the above-mentioned purpose, on the other hand, a BMS is provided, comprising the above-mentioned battery health status estimation device, and / or the above-mentioned storage medium.

[0037] Due to the adoption of the above technical solution, the present invention has the following beneficial effects:

[0038] 1. It can improve the estimation accuracy of the health status and compensate for the system error caused by insufficient capacity statistics. Especially for small-capacity battery packs, the self-discharge and AFE power consumption during sleep cannot be ignored. When the compensation capacity is not considered, the calculation deviation is too large. This method also avoids large jumps in the health status estimation results, indirectly avoids large jumps in the SOC estimation results, and ultimately improves the customer experience.

[0039] 2. According to the estimation error of the self-learning capacity, the self-learning capacity result and the estimated result of the empirical aging model are weighted averaged to obtain a relatively accurate actual capacity value.

[0040] It should be noted that the terms "first", "second" and similar terms used in this article are only for describing the various components in the technical solution, and do not constitute a limitation of the technical solution, nor can they be understood as an indication or suggestion of the importance of the corresponding elements; elements with "first", "second" and similar terms indicate that the corresponding technical solution contains at least one of the element. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solution of the present invention and facilitate further understanding of the technical effects, technical features and purposes of the present invention, the present invention is described in detail below in conjunction with the accompanying drawings. The accompanying drawings constitute an essential part of the specification and are used together with the embodiments of the present invention to illustrate the technical solution of the present invention, but do not constitute a limitation to the present invention.

[0042] The same reference numerals in the accompanying drawings represent the same components, specifically:

[0043] Figure 1 This is a schematic diagram of the battery health status estimation process in Example 1. DETAILED DESCRIPTION

[0044] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. Of course, the specific embodiments described below are only for explaining the technical solutions of the present invention, rather than for limiting the present invention. In addition, the parts described in the embodiments or the accompanying drawings are only examples of the relevant parts of the present invention, rather than the whole of the present invention.

[0045] Example:

[0046] A battery health status estimation method, such as Figure 1 As shown, the specific steps are as follows:

[0047] S1. Determine whether a trigger condition for battery health status estimation is currently met.

[0048] Specifically, when the maximum and minimum module temperatures in the battery are within a preset temperature range and the single cell SOC is within a preset SOC range, the trigger condition is met, otherwise it is not met.

[0049] In step S1, first determine whether the triggering condition of the capacity self-learning algorithm is met, that is, whether self-learning can be triggered. The triggering condition includes temperature condition and SOC condition. The triggering condition formula is as follows:

[0050]

[0051] When the maximum and minimum module temperatures collected in real time in the battery pack are within a reasonable range, the open circuit voltage is captured and the current single-cell SOC value is calculated based on the open circuit voltage. When the SOC estimation error is within a reasonable range, the capacity self-learning is triggered at time t0. After the self-learning is triggered, the SOC value SOC0 of all single cells at time t0 at the beginning of the self-learning sampling period is stored.

[0052] S2. Estimate the capacity change and SOC change of the battery.

[0053] S21. The capacity variation of the battery is obtained by correcting the basic variation by the corrected variation, where the corrected variation is the balanced capacity of the single cell, the AFE consumption capacity and the self-discharge consumption capacity.

[0054] S211. The basic change amount is calculated by integrating the ampere-hour value collected by the current sensor.

[0055] In step S211, the cumulative throughput ΔQ from the start of the sampling period to the current time is calculated in real time based on the ampere-hour integration of the entire packet current I Ah As the basic change, the calculation formula is as follows:

[0056]

[0057] S212, single cell balanced capacity, AFE consumption capacity and self-discharge consumption capacity, the calculation method is as follows:

[0058] S2121. Obtain the balancing current of each battery cell, perform ampere-hour integration on the balancing current of each battery cell, and calculate the balancing capacity of each battery cell in a preset test period.

[0059] S2122. Obtain the operating current of each AFE sampling chip, perform ampere-hour integration on the operating current of each AFE sampling chip, and calculate the AFE consumption capacity; when in sleep mode, calculate the AFE consumption capacity by multiplying the low power mode current by the sleep time.

[0060] S2123. Obtain the self-discharge current of each single cell, perform ampere-hour integration on the self-discharge current of each single cell, and calculate the self-discharge consumption capacity.

[0061] In step S212, the compensation capacity ΔQ needs to be calculated cmp As the correction change, ΔQcmp =ΔQ bal +ΔQ AFE +ΔQ selfdisc

[0062] ΔQ bal The balanced capacity of each single cell:

[0063] Among them, I(t) bal The balancing current of each cell is related to the current cell voltage. By integrating the ampere-hour of the balancing current of each cell, the balancing capacity of each cell from the start of the sampling period to the current moment can be obtained.

[0064] ΔQ AFE Capacity consumed by AFE:

[0065] Among them, I(t) AFE The operating current of the chip is sampled for each AFE. Since the operating current I(t) AFE It is measured offline and is temperature-dependent, so it changes in real time with temperature. By integrating the AFE operating current in ampere-hours, we can get the capacity consumed by all AFE operations from the start of the sampling period to the current moment. When the controller just wakes up, the current I(t) in the low power mode is obtained based on the temperature at the moment the controller just wakes up. AFEsleep , the capacity consumed by AFE is calculated as: ΔQ selfdisc =I(t) selfdisc ×Δt sleep .

[0066] ΔQ selfdisc The capacity consumed by the self-discharge of the battery during the dormant process: ΔQ selfdisc =I(t) selfdisc ×Δt sleep

[0067] Among them, I(t) selfdisc The self-discharge current of each cell. Since the self-discharge current is related to the temperature and SOC of the battery during storage, the self-discharge current during sleep is obtained by using the temperature and SOC value when the controller is just awakened.

[0068] In step S21, if a sleep process is experienced during the preset test period, the corrected change amount of the sleep process is calculated based on the sleep time, temperature, and remaining balancing capacity; if a sleep process is not experienced during the preset test period, the corrected change amount of the power-on process is calculated based on the average module temperature and balancing current.

[0069] S22. Estimate SOC change.

[0070] In step S2, when the triggered capacity self-learning sampling period ends, the effectiveness of this self-learning is judged, and the capacity result and its error after self-learning are calculated. When the error is less than a certain value, this self-learning is considered effective. Since the compensation capacity is the battery discharging to the outside, the current direction of the discharge is defined as negative, so the capacity change is calculated as follows:

[0071] ΔQ=ΔQ Ah -ΔQ cmp

[0072] S3. Divide the capacity change by the SOC change to calculate the battery cell self-learning capacity value.

[0073] The battery cell self-learning capacity values ​​calculated based on self-learning are as follows:

[0074]

[0075] S4. Estimation of the battery cell empirical capacity value SOH based on the empirical aging model Model .

[0076] S5. Obtain the final battery health status by weighting the battery cell self-learning capacity value and the battery cell experience capacity value.

[0077] Specifically, the weight is determined by the proportion of the calculated capacity change estimation error and SOC change estimation error to the battery cell self-learning capacity value;

[0078] The capacity change estimation error ΔQ cmp Estimation error ΔQErr based on the balanced capacity of single cells bal , AFE consumption capacity estimation error ΔQErr AFE And self-discharge consumption capacity estimation error ΔQErr selfdisc Composition, ΔQ cmp =ΔQ bal +ΔQ AFE +ΔQ selfdisc .

[0079] There is an error in the calculation of the balancing capacity. Since the harness resistance and Rdson resistance are not taken into account, the balancing circuit resistance is too small, which leads to a large calculation of the balancing current, affecting the calculation accuracy. Since the harness resistance is small, it is temporarily ignored. The Rdson resistance is related to the temperature, but since the temperature is difficult to obtain accurately, only the error range caused by the deviation between the extreme value and the normal value of Rdson is considered as the error ΔQErr of the balancing capacity. bal The capacity consumed by AFE also has a calculation error ΔQErr due to the current sensor sampling error, etc. AFE Since the self-discharge current also has an error ΔQErr due to current sampling selfdisc .

[0080] The error of the SOC change amount ΔSOC is ΔSOCErr=SOCErr1+SOCErr1.

[0081] When the triggered capacity self-learning sampling period ends, the effectiveness of this self-learning is judged, and the capacity result and its error after self-learning are calculated. When the error is less than a certain value, this self-learning is considered valid.

[0082] The weight w is given by SOHErr Q When the error is large, the self-learning weight w decreases, SOHErr Q The calculation formula is as follows:

[0083]

[0084] Among them, SOHER Q The unit is %. When all three compensation capacities are valid, the SOHErr calculated at the same time is Q Within a certain range, this self-learning is considered effective.

[0085] The final weighted calculation formula for the battery health status is as follows:

[0086] SOH=w×SOH Q +(1-w)SOH Model

[0087] Repeat the above steps to achieve the final battery health status calculation.

[0088] Embodiment 2:

[0089] A battery health status estimation device, comprising a capacity change estimation module, a SOC change estimation module, a battery cell actual capacity value calculation module and a weight fusion module;

[0090] The capacity change estimation module estimates and corrects the capacity change of the battery within a preset test period;

[0091] The SOC change estimation module estimates the SOC change within a preset test period;

[0092] The actual capacity value calculation module calculates the battery cell self-learning capacity value by dividing the corrected capacity change by the SOC change;

[0093] The weight fusion module obtains the final battery health state by weighting the battery monomer self-learning capacity value and the battery monomer experience capacity value, and the battery monomer experience capacity value is estimated by the experience aging model.

[0094] It should be noted that the above embodiments are only for more clearly illustrating the technical solutions of the present invention. Those skilled in the art will understand that the implementation methods of the present invention are not limited to the above contents, and obvious changes, replacements or substitutions based on the above contents do not exceed the scope of the technical solutions of the present invention; other implementation methods will also fall within the scope of the present invention without departing from the concept of the present invention.

Claims

1. A method for estimating a battery health state, characterized in that: The specific method is as follows: Estimate the capacity change and SOC change of the battery during the preset test period; The capacity change is divided by the SOC change to calculate the self-learning capacity value of the battery cell; The capacity variation of the battery is obtained by correcting the basic variation by the corrected variation, and the corrected variation includes any one or more of the following parameters: Single cell balanced capacity, AFE consumption capacity and self-discharge consumption capacity.

2. The battery health status estimation method according to claim 1, characterized in that: Before estimating the battery health status, trigger conditions are also included; When the maximum and minimum module temperatures in the battery are within the preset temperature range and the single cell SOC is within the preset SOC range, the trigger condition is met, otherwise it is not met.

3. The battery health status estimation method according to claim 1, characterized in that: The basic variation is calculated by integrating the ampere-hour value collected by the current sensor.

4. The battery health status estimation method according to claim 1, characterized in that: If a dormancy process occurs during the preset test period, the correction change amount of the dormancy process is calculated based on the dormancy time, temperature, and remaining balanced capacity; If the sleep process is not experienced during the preset test period, the correction change of the power-on process is calculated based on the average module temperature and the balanced current.

5. The battery health status estimation method according to claim 4, characterized in that: The specific estimation method of the balanced capacity of the single cell is as follows: The balancing current of each battery cell is obtained, and the balancing current of each battery cell is integrated by ampere-hour to calculate the balancing capacity of each battery cell in the preset test period.

6. The battery health status estimation method according to claim 4, characterized in that: The specific estimation method of the AFE consumption capacity is as follows: Obtain the operating current of each AFE sampling chip, perform ampere-hour integration on the operating current of each AFE sampling chip, and calculate the AFE consumption capacity; When in sleep mode, the AFE consumption capacity is calculated by multiplying the low power mode current by the sleep time.

7. The battery health status estimation method according to claim 4, characterized in that: The self-discharge consumption capacity is specifically estimated by the following method: The self-discharge current of each single cell during the dormant process is obtained, and the self-discharge current of each single cell is integrated in ampere-hours to calculate the self-discharge consumption capacity.

8. The battery health status estimation method according to claim 1, characterized in that: After the self-learning capacity value of the battery cell is calculated, the empirical capacity value of the battery cell is estimated by the empirical aging model, and the final battery health status is obtained by weighting the self-learning capacity value of the battery cell and the empirical capacity value of the battery cell.

9. The battery health status estimation method according to claim 8, characterized in that: The weight is determined by the proportion of the calculated capacity change estimation error and SOC change estimation error to the battery cell self-learning capacity value; The greater the proportion of the capacity change estimation error and the SOC change estimation error to the battery cell self-learning capacity value, the smaller the weight corresponding to the battery cell self-learning capacity value.

10. The battery health status estimation method according to claim 9, characterized in that: The capacity change estimation error includes any one or more of the following parameters: Single cell balanced capacity estimation error, AFE consumption capacity estimation error and self-discharge consumption capacity estimation error.

11. The battery health status estimation method according to claim 10, characterized in that: The estimated error of the single cell balanced capacity is calculated from the error range caused by the deviation between the extreme value and the normal value of Rdson.

12. The battery health status estimation method according to claim 10, characterized in that: The AFE consumption capacity estimation error is calculated by estimating the current sensor sampling error.

13. The battery health status estimation method according to claim 10, characterized in that: The self-discharge consumption capacity estimation error is calculated by estimating the current sensor sampling error.

14. A battery health status estimation device, characterized in that: It includes a capacity change estimation module, an SOC change estimation module, a battery cell actual capacity value calculation module and a weight fusion module; The capacity change estimation module estimates and corrects the capacity change of the battery within a preset test period; The SOC change estimation module estimates the SOC change within a preset test period; The actual capacity value calculation module calculates the battery cell self-learning capacity value by dividing the corrected capacity change by the SOC change; The weight fusion module obtains the final battery health state by weighting the battery monomer self-learning capacity value and the battery monomer experience capacity value, and the battery monomer experience capacity value is estimated by the experience aging model.

15. A storage medium, characterized in that: A plurality of instructions are stored, and a processor loads the plurality of instructions to execute the battery health state estimation method described in any one of claims 1 to 13.

16. A BMS, characterized in that: Includes the battery health status estimation device as described in claim 14, and / or the storage medium as described in claim 15.

Images