A battery capacity self-learning calibration method based on SOC partitioning

Abstract

The application particularly relates to a battery capacity self-learning calibration method based on SOC partitioning; the method involves: combining specific working conditions in advance, proposing numerically quantized first, second and third disposal intervals; based on the collected power-off SOC value, the three disposal intervals are used to divide the faulty equipment; based on this, dynamic monitoring is performed on a type of faulty equipment, and recall disposal is performed on three types of faulty equipment; for the second type of faulty equipment: based on the series data of the battery power-off event, the actual SOC value of the second type of faulty equipment is calculated, based on this, the preset-calibration algorithm is combined to timely calculate the battery calibration capacity, so that the second type of faulty equipment-battery calibration is timely performed; and based on the credibility determination mechanism, the credibility involved in the second type of faulty equipment-battery calibration is systematically evaluated; based on the foregoing method, by analyzing the differentiated causes of different fault intervals and systematically evaluating the calibration result-credibility, the calibration accuracy and calibration efficiency can be improved.

Description

Technical Field

[0001] This invention relates to the field of battery management technology, and in particular to a battery capacity self-learning calibration method based on SOC partitioning. Background Technology

[0002] In the daily operation of electric vehicles (e.g., electric two-wheelers used by delivery riders), the battery management system (BMS) is usually used to dynamically estimate the actual available capacity of the battery through a capacity self-learning algorithm to calculate the SOC value and provide users with information on the remaining power. However, due to battery aging, temperature changes, and differences in charging and discharging habits, the capacity self-learning algorithm may deviate, leading to inaccurate SOC calculation, or even a situation where "the device suddenly shuts down when the remaining battery level is displayed" (i.e., power failure at high SOC).

[0003] Based on the foregoing, a battery capacity self-learning calibration method can be used to address the "high SOC power failure" phenomenon. However, existing calibration methods typically employ a "single full charge and discharge calibration strategy," requiring a complete charge and discharge cycle and failing to consider the differentiated causes of different fault ranges. This results in: long calibration time, low efficiency, and inability to achieve accurate calibration. Furthermore, existing calibration methods lack a "systematic evaluation of the reliability of calibration results," making it difficult to effectively distinguish faulty devices and perform timely calibration, recall, and monitoring.

[0004] To address the aforementioned technical issues, this application proposes a battery capacity self-learning calibration method based on SOC partitioning. Summary of the Invention

[0005] In view of the above-mentioned shortcomings and deficiencies of the prior art, the present invention provides a battery capacity self-learning calibration method based on SOC partitioning, which solves the current problem of "how to improve calibration accuracy and efficiency by analyzing the differentiated causes of different fault ranges and systematically evaluating the reliability of calibration results when performing self-learning calibration of electric vehicle battery capacity".

[0006] To achieve the above objectives, the main technical solutions adopted by the present invention include: This invention provides a battery capacity self-learning calibration method based on SOC partitioning; the battery capacity self-learning calibration method includes: S1. Synchronously collect a series of data on battery power failure events; Among them, the battery power failure event series data includes: power failure SOC value, charging amount, charging SOC difference, discharging amount, discharging SOC difference, jump SOC value, empty calibration status value, and rated capacity. S2. Based on the collected SOC value of the power outage, classify the faulty equipment according to the first handling interval [0, 15%), the second handling interval [15%, 40%], and the third handling interval (40%, 100%). Dynamic monitoring is implemented for Class I faulty equipment, and recall is implemented for Class III faulty equipment; Class I / Class III faulty equipment refers to faulty equipment whose power-off SOC value corresponds to the first / third handling interval; S3. Calculate the actual SOC value of the equipment with type II faults based on the empty calibration state value, power-off SOC value, and jump SOC value; calculate the battery calibration capacity of the equipment with type II faults in a timely manner based on the charging amount, charging SOC difference, discharging amount, discharging SOC difference, actual SOC value, and rated capacity through a preset calibration algorithm, so as to perform equipment battery calibration in a timely manner. Category II fault equipment refers to fault equipment whose power-off SOC value corresponds to the second handling interval; S4. If device battery calibration is performed, then: the calculated battery calibration capacity of the type II fault device is verified for reasonableness to obtain the reasonableness verification result; based on the reasonableness verification result, a calibration command is issued to correct the battery capacity self-learning parameters of the type II fault device.

[0007] Optionally, in S3, the calculation of the actual SOC value of the type II faulty equipment involves: If the empty calibration state value is 1, then the formula for calculating the actual SOC value is: (1) If the empty calibration state value is 0 or 3, the formula for calculating the actual SOC value is: (2) Among them, the empty calibration status values ​​0 / 1 / 3 represent: no empty check / empty check present / protection triggered; Indicates the actual SOC value; Indicates the SOC value after power failure; This indicates a jump in SOC value.

[0008] Optionally, in S3, the preset-calibration algorithm involves: When condition A is met: the difference between the charging amount and the charging SOC is greater than 0, then: (3) When condition B is met: the difference between discharge quantity and discharge SOC is greater than 0, then: (4) When condition C is met: the actual SOC value corresponds to the range (0, 100), then: (5) If conditions A, B, and C are not met, then: mark it as insufficient calibration data and do not perform calibration; in, Indicates a Class II fault device - battery calibration capacity; Indicates the amount of charge; Indicates the difference in SOC during charging; Indicates the amount of discharge; Indicates the difference in discharge SOC; Indicates the rated capacity.

[0009] Optionally, the battery capacity self-learning calibration method also involves a reliability determination mechanism; The credibility determination mechanism includes: After calculating the actual SOC value, calculate the average of the actual SOC values; if the average of the actual SOC values ​​is not greater than 2%, it indicates that the battery is close to being fully discharged and no device battery calibration is required. If the calculated battery calibration capacity of a Class II faulty device is less than 65% of the rated capacity, it indicates that the battery has severely degraded and needs to be physically replaced.

[0010] Optionally, the credibility determination mechanism further includes: If the following conditions are met: the battery calibration capacity of a Class II fault device is greater than the rated capacity, or the device battery calibration trigger days are not less than 2 and the actual SOC value fluctuation is greater than 10%, then the reliability of the current device battery calibration is low. When condition D is met: the battery calibration capacity of a Class II fault device is 65% to 100% of the rated capacity, and if the following conditions are also met: the trigger days for device battery calibration are 1 and the calibration data is sufficient, then the reliability of the current device battery calibration is medium. If the following conditions are also met: the trigger days for device battery calibration are not less than 2 and the fluctuation of the actual SOC value is not greater than 10%, then the reliability of the current device battery calibration is high.

[0011] The beneficial effects of this application include: In this application, based on specific operating conditions, numerical quantification of the first, second, and third handling intervals is proposed in advance; based on the collected power outage SOC value, the faulty equipment is divided into the aforementioned three handling intervals; based on this, dynamic monitoring is performed on the first type of faulty equipment, and recall handling is performed on the third type of faulty equipment. For Category II fault devices: the actual SOC value of the Category II fault device is calculated based on the battery power failure event series data. Based on this, the battery calibration capacity is calculated in a timely manner in combination with the preset calibration algorithm, so as to perform Category II fault device-battery calibration in a timely manner; and the credibility of "Category II fault device-battery calibration" is systematically evaluated based on the credibility judgment mechanism. Based on the foregoing, by analyzing the differential causes of different fault ranges and systematically evaluating the reliability of calibration results, calibration accuracy and efficiency can be improved. Attached Figure Description

[0012] Figure 1 This is a schematic diagram of a battery capacity self-learning calibration process provided in one embodiment of the present invention; Figure 2 This is a schematic diagram of a self-learning calibration parameter structure provided in one embodiment of the present invention. Detailed Implementation

[0013] To better explain and facilitate understanding of the present invention, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0014] To better understand the above technical solutions, exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that the present invention can be understood more clearly and thoroughly, and that the scope of the present invention can be fully conveyed to those skilled in the art.

[0015] Example 1 This embodiment proposes a battery capacity self-learning calibration method based on SOC partitioning; Figure 1 This is a schematic diagram of the battery capacity self-learning calibration process. Figure 2 This is a schematic diagram of the self-learning calibration parameter structure.

[0016] In this embodiment, combined with Figure 1 , Figure 2 The aforementioned battery capacity self-learning calibration method involves the following steps S1-S4; S1. Synchronously collect a series of data on battery power failure events; In this embodiment, preferably, the battery-outage-calculate-stats data table is used to collect a series of battery outage events that occur within a continuous 30-day period. Invalid records with Calculate-soc-cap-source=3 in the aforementioned data table are removed to ensure the quality of the collected battery outage event data.

[0017] In this embodiment, it should be noted that the battery power outage event series data includes: power outage SOC value, charging amount, charging SOC difference, discharging amount, discharging SOC difference, jump SOC value, empty calibration status value, rated capacity, device unique identifier, statistical date, and first power outage time.

[0018] In this embodiment, regarding the aforementioned "synchronous acquisition," it should be noted that: To achieve battery capacity self-learning calibration, it is necessary to ensure that the parameters such as "power-off SOC value, charging amount, charging SOC difference, discharging amount, discharging SOC difference, jump SOC value, empty calibration state value, and rated capacity" have corresponding timestamps. Based on "device unique identifier, statistical date, first power-off time" and "parameter values ​​with timestamps", it is possible to obtain parameter values ​​corresponding to the same batch of devices. Based on this, the effectiveness of battery capacity self-learning calibration can be ensured.

[0019] S2. Based on the collected SOC value of the power outage, classify the faulty equipment according to the first handling interval [0, 15%), the second handling interval [15%, 40%], and the third handling interval (40%, 100%). Dynamic monitoring is implemented for Class I faulty equipment, and recall is implemented for Class III faulty equipment; Class I / Class III faulty equipment refers to faulty equipment whose power-off SOC value corresponds to the first / third handling interval; In this embodiment, regarding the "first treatment interval [0, 15%), second treatment interval [15%, 40%], and third treatment interval (40%, 100%)", it should be noted that: The "high SOC power failure" fault can be covered in the range [0%, 100%], and different faults correspond to different handling ranges; specifically, the first handling range [0, 15%) corresponds to temporary temperature effects; the second handling range [15%, 40%] corresponds to capacity self-learning deviation; and the third handling range (40%, 100%) corresponds to excessive battery capacity decay.

[0020] In this embodiment, since the first treatment interval [0, 15%) corresponds to a temporary temperature effect, no immediate action is required; dynamic monitoring is sufficient until the temperature returns to normal operating conditions. Since the third treatment interval (40%, 100%) corresponds to excessive battery capacity decay, and excessive battery capacity decay will not automatically return to normal over time, manual intervention is required. Therefore, a recall treatment method is adopted.

[0021] S3. Calculate the actual SOC value of the equipment with type II faults based on the empty calibration state value, power-off SOC value, and jump SOC value; calculate the battery calibration capacity of the equipment with type II faults in a timely manner based on the charging amount, charging SOC difference, discharging amount, discharging SOC difference, actual SOC value, and rated capacity through a preset calibration algorithm, so as to perform equipment battery calibration in a timely manner. Category II fault equipment refers to fault equipment whose power-off SOC value corresponds to the second handling interval; In the aforementioned S3, the calculation of the actual SOC value of the second-class fault equipment involves the following calculation formulas (1) and (2): If the empty calibration state value is 1, then the formula for calculating the actual SOC value is: (1) If the empty calibration state value is 0 or 3, the formula for calculating the actual SOC value is: (2) Among them, the empty calibration status values ​​0 / 1 / 3 represent: no empty check / empty check present / protection triggered; Indicates the actual SOC value; Indicates the SOC value after power failure; This indicates a jump in SOC value.

[0022] In the aforementioned S3, the preset-calibration algorithm involves the following conditions A to C and calculation formulas (3), (4), and (5): When condition A is met: the difference between the charging amount and the charging SOC is greater than 0, then: (3) When condition B is met: the difference between discharge quantity and discharge SOC is greater than 0, then: (4) When condition C is met: the actual SOC value corresponds to the range (0, 100), then: (5) If conditions A, B, and C are not met, then: mark it as insufficient calibration data and do not perform calibration; in, Indicates a Class II fault device - battery calibration capacity; Indicates the amount of charge; Indicates the difference in SOC during charging; Indicates the amount of discharge; Indicates the difference in discharge SOC; Indicates the rated capacity.

[0023] In this embodiment, it should be noted that the aforementioned battery capacity self-learning calibration method also involves a reliability determination mechanism; the reliability determination mechanism specifically includes the following four points: First, after calculating the actual SOC value, calculate the average of the actual SOC values; if the average of the actual SOC values ​​is not greater than 2%, it indicates that the battery is close to being fully discharged and there is no need to perform device battery calibration. Second, if the calculated battery calibration capacity of a Class II faulty device is less than 65% of the rated capacity, it indicates that the battery has severely degraded and needs to be physically replaced. Third, if the following conditions are met: the battery calibration capacity of a Class II fault device is greater than the rated capacity, or the trigger days for device battery calibration are not less than 2, and the fluctuation of the actual SOC value is greater than 10%, then it indicates that the reliability of the current device battery calibration is low. Fourth, when condition D is met: the battery calibration capacity of a Class II fault device is 65% to 100% of the rated capacity, and if the following conditions are also met: the trigger days for device battery calibration are 1, and the calibration data is sufficient, then the reliability of the current device battery calibration is medium; if the following conditions are also met: the trigger days for device battery calibration are not less than 2, and the fluctuation of the actual SOC value is not greater than 10%, then the reliability of the current device battery calibration is high.

[0024] S4. If device battery calibration is performed, then: the calculated battery calibration capacity of the type II fault device is verified for reasonableness to obtain the reasonableness verification result; based on the reasonableness verification result, a calibration command is issued to correct the battery capacity self-learning parameters of the type II fault device.

[0025] In this embodiment, it should be noted that the aforementioned rationality check range is 0.2 to 2 times the rated capacity. Based on this, parameter values ​​that exceed the check range are marked as abnormal, and rationality checks are not performed based on abnormal parameter values ​​to ensure the reliability of the rationality check.

[0026] Based on the battery capacity self-learning calibration method based on SOC partitioning described in the aforementioned Embodiment 1, the first, second, and third handling intervals of numerical quantification are proposed in advance in combination with specific operating conditions; based on the collected power-off SOC value, the faulty equipment is divided into the aforementioned three handling intervals; based on this, dynamic monitoring is performed on one type of faulty equipment, and recall handling is performed on three types of faulty equipment. For Category II fault devices: the actual SOC value of the Category II fault device is calculated based on the battery power failure event series data. Based on this, the battery calibration capacity is calculated in a timely manner in combination with the preset calibration algorithm, so as to perform Category II fault device-battery calibration in a timely manner; and the credibility of "Category II fault device-battery calibration" is systematically evaluated based on the credibility judgment mechanism. Based on the foregoing, by analyzing the differential causes of different fault ranges and systematically evaluating the reliability of calibration results, calibration accuracy and efficiency can be improved.

[0027] Example 2 This embodiment proposes a verification embodiment based on the battery capacity self-learning calibration method described in Embodiment 1 above; the verification embodiment specifically involves the following: (a) Initial validation (28 samples) This implementation selected 119 devices with high SOC power failure as the verification sample. The test period was from February 3 to February 20, 2026, for a total of 18 days (3 days for preparation, 4 days for calibration execution, and 14 days for effect observation). The verification results of the 28 core calibration samples corresponding to the interval [15%, 40%] are as follows: 1) Data extraction success rate: 100% (preset target ≥98%), data from the 2.8~3.6V charging range of all 28 devices were successfully extracted; 2) Calibration command issuance success rate: 100% (preset target ≥95%), calibration commands were successfully issued to all 28 devices; 3) Calibration execution success rate: 100% (preset target ≥90%), calibration process was completed for all 28 devices; 4) Fault recurrence rate: 7.14% (2 / 28), far lower than the preset target ≤8%; During the 14-day observation period, 26 out of the 28 devices did not trigger the high protection of the discharge cell.

[0028] (II) Large-scale verification (3028 units) Based on the initial verification described above, a large-scale follow-up verification was conducted from April 1st to 27th, 2026, involving the re-examination and classification of 3028 calibration devices. The results are as follows: 1) The calibration effectiveness rate reached 84.9% (no recurrence + significant improvement), fully verifying the effectiveness and reliability of this method; 2) Remote calibration accounted for 50.1% (1516 / 3028), indicating that a large number of devices could be repaired through remote calibration without physical intervention; 3) 141 devices recurred in the range [15%, 20%], with an actual average SOC of 16.6%, mainly concentrated in the range [15%, 18%], where calibration was partially effective; 4) 206 devices recurred in the range [20%, 30%], with an actual average SOC of 24.8%, indicating that calibration could not completely repair them; 108 devices had cell problems, with an actual average SOC of 52.2%, verifying the necessity of the recall strategy.

[0029] It should be noted that any reference numerals placed between parentheses in the claims should not be construed as limiting the claims. The word "comprising" does not exclude the presence of components or steps not listed in the claims. The words "a" or "an" preceding a component do not exclude the presence of a plurality of such components. The use of the terms first, second, third, etc., is for convenience only and does not indicate any order. These terms can be understood as part of the component names.

[0030] Furthermore, it should be noted that in the description of this specification, the terms "one embodiment," "some embodiments," "embodiment," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Furthermore, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0031] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the claims should be interpreted to include both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0032] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, then this invention should also include these modifications and variations.

Claims

1. A battery capacity self-learning calibration method based on SOC partitioning, characterized in that: The battery capacity self-learning calibration method includes: S1. Synchronously collect a series of data on battery power failure events; Among them, the battery power failure event series data includes: power failure SOC value, charging amount, charging SOC difference, discharging amount, discharging SOC difference, jump SOC value, empty calibration status value, and rated capacity. S2. Based on the collected SOC value of the power outage, classify the faulty equipment according to the first handling interval [0, 15%), the second handling interval [15%, 40%], and the third handling interval (40%, 100%). Dynamic monitoring is implemented for Class I faulty equipment, and recall is implemented for Class III faulty equipment; Class I / Class III faulty equipment refers to faulty equipment whose power-off SOC value corresponds to the first / third handling interval; S3. Calculate the actual SOC value of the equipment with type II faults based on the empty calibration state value, power-off SOC value, and jump SOC value; calculate the battery calibration capacity of the equipment with type II faults in a timely manner based on the charging amount, charging SOC difference, discharging amount, discharging SOC difference, actual SOC value, and rated capacity through a preset calibration algorithm, so as to perform equipment battery calibration in a timely manner. Category II fault equipment refers to fault equipment whose power-off SOC value corresponds to the second handling interval; S4. If device battery calibration is performed, then: the calculated battery calibration capacity of the type II fault device is verified for reasonableness to obtain the reasonableness verification result; based on the reasonableness verification result, a calibration command is issued to correct the battery capacity self-learning parameters of the type II fault device.

2. The battery capacity self-learning calibration method according to claim 1, characterized in that: In S3, the calculation of the actual SOC value of the second type of faulty equipment involves: If the empty calibration state value is 1, then the formula for calculating the actual SOC value is: （1） If the empty calibration state value is 0 or 3, the formula for calculating the actual SOC value is: （2） Among them, the empty calibration status values ​​0 / 1 / 3 represent: no empty check / empty check present / protection triggered; Indicates the actual SOC value; Indicates the SOC value after power failure; This indicates a jump in SOC value.

3. The battery capacity self-learning calibration method according to claim 1, characterized in that: In S3, the preset-calibration algorithm involves: When condition A is met: the difference between the charging amount and the charging SOC is greater than 0, then: （3） When condition B is met: the difference between discharge quantity and discharge SOC is greater than 0, then: （4） When condition C is met: the actual SOC value corresponds to the range (0, 100), then: （5） If conditions A, B, and C are not met, then: mark it as insufficient calibration data and do not perform calibration; in, Indicates a Class II fault device - battery calibration capacity; Indicates the amount of charge; Indicates the difference in SOC during charging; Indicates the amount of discharge; Indicates the difference in discharge SOC; Indicates the rated capacity.

4. The battery capacity self-learning calibration method according to claim 1, characterized in that: The battery capacity self-learning calibration method also involves a reliability determination mechanism; The credibility determination mechanism includes: After calculating the actual SOC value, calculate the average of the actual SOC values; if the average of the actual SOC values ​​is not greater than 2%, it indicates that the battery is close to being fully discharged and no device battery calibration is required. If the calculated battery calibration capacity of a Class II faulty device is less than 65% of the rated capacity, it indicates that the battery has severely degraded and needs to be physically replaced.

5. The battery capacity self-learning calibration method according to claim 4, characterized in that: The credibility determination mechanism also includes: If the following conditions are met: the battery calibration capacity of a Class II fault device is greater than the rated capacity, or the device battery calibration trigger days are not less than 2 and the actual SOC value fluctuation is greater than 10%, then the reliability of the current device battery calibration is low. When condition D is met: the battery calibration capacity of a Class II fault device is 65% to 100% of the rated capacity, and if the following conditions are also met: the trigger days for device battery calibration are 1 and the calibration data is sufficient, then the reliability of the current device battery calibration is medium. If the following conditions are also met: the trigger days for device battery calibration are not less than 2 and the fluctuation of the actual SOC value is not greater than 10%, then the reliability of the current device battery calibration is high.

Images