Intelligent matching method, device and equipment for multi-batch lead-acid storage batteries and medium

Abstract

The invention discloses an intelligent matching method, device and equipment for multi-batch lead-acid storage batteries and a medium, relates to the technical field of lead-acid storage batteries, and aims to carry out pretreatment and battery performance repair on the multi-batch lead-acid storage batteries. Performing battery capacity test on the repaired battery, determining the capacity grade of the battery qualified in the capacity test, and converting the capacity grade into a capacity value; discharging the battery qualified in the capacity test, and calculating the dynamic direct current internal resistance value of the battery to screen the pulse qualified battery; performing floating charging on the pulse qualified battery to obtain a floating charging steady-state voltage value so as to screen a target qualified battery; constructing a three-dimensional feature vector for the target qualified battery; performing standardization processing on the three-dimensional feature vector; and clustering and intelligent battery matching are carried out on the target qualified batteries by utilizing a clustering algorithm and based on the processed three-dimensional feature vectors, the number of each group of batteries and the clustering number, so that the problems that the dynamic consistency cannot be evaluated and the method depends on artificial experience are solved, and the reliability and the efficiency of battery re-matching are improved.

Description

Technical Field

[0001] This invention relates to the field of lead-acid battery technology, and in particular to a method, apparatus, equipment and medium for intelligent grouping of multiple batches of lead-acid batteries. Background Technology

[0002] High-power lead-acid batteries are the core of backup power systems for critical facilities such as data centers and UPS (Uninterruptible Power Systems). In these applications, battery banks need to withstand large current loads instantaneously, making their high-power discharge performance and long-term float charging reliability crucial. Furthermore, in these applications, batteries are typically used in series or series-parallel configurations, and their overall performance and lifespan depend heavily on the consistency of individual cells within the bank. Poor consistency not only reduces the overall capacity of the bank but also causes lagging cells to fail prematurely during charge-discharge cycles, leading to a chain reaction of failures throughout the entire battery bank.

[0003] In the production, storage and after-sales processes of high-power batteries, a large number of "non-grouped" surplus individual cells accumulate. These cells come from complex sources and have inconsistent performance. They include surplus cells from production line assembly and individual cells replaced from retired battery packs. They are produced from multiple different batches and even different production years, and their voltage, internal resistance, capacity and high-power performance indicators vary significantly.

[0004] Currently, the industry's methods for handling these types of batteries include: 1. Downgraded use: It is used in applications with lower requirements, but it cannot realize its original high power performance value, resulting in poor economic efficiency.

[0005] 2. Scrap Recycling: Directly treating a large number of batteries with remaining lifespan as waste lead not only causes huge waste of resources and economic losses, but also goes against the direction of green and environmentally friendly industrial development.

[0006] 3. Simple and crude matching: This matching method mainly relies on static parameters (such as open circuit voltage and static internal resistance) or a single capacity value for matching. This method cannot reflect the consistency of the dynamic performance of the battery under actual high-power discharge conditions.

[0007] 4. Some other advanced processes have introduced a "dual screening" mechanism based on capacity and float voltage, but they still suffer from problems such as being too simplistic, lacking intelligence, and having poor adaptability.

[0008] As can be seen from the above, how to solve the problems of being unable to assess dynamic consistency and relying on human experience, thereby maximizing the efficient recycling and utilization of battery resources and improving the reliability and efficiency of battery repackaging, is a problem that needs to be solved in this field. Summary of the Invention

[0009] In view of this, the purpose of this invention is to provide a method, apparatus, equipment, and medium for intelligent grouping of multiple batches of lead-acid batteries, which can solve the problems of inability to assess dynamic consistency and reliance on human experience, thereby maximizing the efficient recycling and utilization of battery resources and improving the reliability and efficiency of battery regrouping. The specific solution is as follows: In a first aspect, this application discloses a method for intelligently grouping multiple batches of lead-acid batteries, including: Pre-treatment and battery performance repair were performed on multiple batches of lead-acid batteries at preset power levels to obtain repaired batteries. The repaired battery is subjected to a battery capacity test to obtain a qualified battery and its corresponding discharge termination voltage. The capacity level of the qualified battery is determined based on the discharge termination voltage, and the capacity level is converted into a capacity value. The capacity-tested qualified batteries are discharged using a constant current pulse to calculate the dynamic DC internal resistance value of the batteries. Based on the dynamic DC internal resistance value of the batteries, pulse-qualified batteries are selected from the capacity-tested qualified batteries. The pulse-qualified batteries are float-charged to obtain a float-charge steady-state voltage value. Based on the float-charge steady-state voltage, target qualified batteries are selected from the pulse-qualified batteries. A three-dimensional feature vector is constructed for the target qualified battery; the three-dimensional feature vector includes the capacity value, dynamic DC internal resistance value, and float charge steady-state voltage value corresponding to the target qualified battery. The three-dimensional feature vector is standardized to obtain the processed three-dimensional feature vector. Based on the order requirements, the number of batteries in each group and the number of clusters are determined. Using a clustering algorithm and based on the processed three-dimensional feature vector, the number of batteries in each group, and the number of clusters, the target qualified batteries are clustered and intelligently matched to obtain each group of target qualified batteries.

[0010] Optionally, the pretreatment and battery performance repair of multiple batches of lead-acid batteries at preset power levels to obtain repaired batteries includes: A current-limited constant-voltage charging method was used to recharge multiple batches of lead-acid batteries at preset power levels. After the battery is recharged, acid is added, and a multi-stage stepped charge-discharge process is used to repair the battery performance, resulting in a repaired battery.

[0011] Optionally, determining the capacity level of a qualified battery based on the discharge termination voltage and converting the capacity level into a capacity value includes: Using standard capacity testing methods and based on preset voltage segmentation ranges and discharge termination voltages, capacity-test-qualified batteries are classified into capacity grades to determine their capacity grades. Add a unique identifier to the capacity level and convert the unique identifier into a capacity value.

[0012] Optionally, the step of discharging the capacity-tested battery using a constant current pulse to calculate the battery's dynamic DC internal resistance includes: The capacity-tested battery was discharged using a constant current pulse, and the instantaneous voltage before the pulse discharge and the ending voltage at the end of the pulse discharge were collected. Calculate the voltage difference between the instantaneous voltage and the final voltage; The ratio between the voltage difference and the pulse current is used as the dynamic DC internal resistance of the battery.

[0013] Optionally, the step of performing float charging on the pulse-qualified batteries to obtain a float charging steady-state voltage value, and selecting target qualified batteries from the pulse-qualified batteries based on the float charging steady-state voltage, includes: The qualified pulse battery was float-charged using a constant parameter current-limiting constant voltage mode, and the steady-state voltage value of the battery at the end of the float charge was recorded. The pulse-qualified batteries whose float charge steady-state voltage value is greater than the preset voltage threshold are regarded as abnormal pulse-qualified batteries. The qualified batteries with abnormal pulses are discarded to obtain the target qualified batteries.

[0014] Optionally, the step of using a clustering algorithm and based on the processed three-dimensional feature vector, the number of batteries in each group, and the number of clusters to cluster and intelligently group the target qualified batteries to obtain each group of target qualified batteries includes: The K-means clustering algorithm is used to cluster and divide the target qualified batteries based on the number of clusters, and the target qualified batteries in each cluster are obtained. Determine whether the number of qualified target batteries in each cluster is consistent with the number of batteries in each group; If they match, the three-dimensional feature vectors of the target qualified batteries in the same cluster will be matched for consistency. If the consistency matching passes, the target qualified batteries in the same cluster will be intelligently grouped to obtain each group of target qualified batteries. If there is a discrepancy, the principle of proximity to the centroid is adopted to supplement the qualified target cells within the inconsistent cluster, and then the consistency matching process is executed.

[0015] Optionally, after obtaining each group of target qualified batteries, the process further includes: The open-circuit voltage and the overall internal resistance of each group of target qualified batteries were measured to obtain the open-circuit voltage range and the internal resistance deviation. Determine whether the open-circuit voltage range and internal resistance deviation meet the requirements; If the open-circuit voltage range and internal resistance deviation meet the requirements, then the qualified target batteries that meet the requirements will be stored and shipped.

[0016] Secondly, this application discloses an intelligent matching device for multiple batches of lead-acid batteries, comprising: The pre-processing and battery performance repair module is used to pre-process and repair the battery performance of multiple batches of lead-acid batteries under preset power to obtain repaired batteries. The battery capacity testing module is used to perform battery capacity testing on the repaired battery, obtain a qualified battery and its corresponding discharge termination voltage, determine the capacity level of the qualified battery based on the discharge termination voltage, and convert the capacity level into a capacity value. The pulse qualified battery screening module is used to discharge the capacity-test qualified batteries using a constant current pulse to calculate the dynamic DC internal resistance value of the battery, and to screen out pulse qualified batteries from the capacity-test qualified batteries based on the dynamic DC internal resistance value of the battery. A float charging module is used to float charge the pulse qualified battery to obtain a float charging steady-state voltage value, and to select target qualified batteries from the pulse qualified batteries based on the float charging steady-state voltage. A three-dimensional feature vector construction module is used to construct a three-dimensional feature vector for the target qualified battery; the three-dimensional feature vector includes the capacity value, dynamic DC internal resistance value, and float charge steady-state voltage value corresponding to the target qualified battery. A standardization processing module is used to standardize the three-dimensional feature vector to obtain the processed three-dimensional feature vector. The clustering and intelligent battery matching module is used to determine the number of batteries in each group and the number of clusters according to the order requirements. It uses a clustering algorithm and based on the processed three-dimensional feature vector, the number of batteries in each group, and the number of clusters, to perform clustering and intelligent battery matching on the target qualified batteries to obtain each group of target qualified batteries.

[0017] Thirdly, this application discloses an electronic device, including: Memory, used to store computer programs; A processor is used to execute the computer program to implement the aforementioned intelligent grouping method for multiple batches of lead-acid batteries.

[0018] Fourthly, this application discloses a computer storage medium for storing a computer program; wherein, when the computer program is executed by a processor, it implements the steps of the aforementioned intelligent grouping method for multiple batches of lead-acid batteries.

[0019] As can be seen, this application pre-processes and repairs the performance of lead-acid batteries under preset power conditions in multiple batches. This addresses issues such as plate sulfation, electrolyte imbalance, and capacity decay caused by storage, use, and batch differences in lead-acid batteries. It improves the resource utilization rate of multiple batches of used / accumulated lead-acid batteries, reduces battery scrap rates and material loss costs for enterprises. After repair, the battery capacity is tested to obtain qualified batteries and their corresponding discharge termination voltages. Based on the discharge termination voltage, the capacity level of the qualified batteries is determined, and the capacity level is converted into a capacity value. Batteries with substandard core energy storage capacity are screened out, ensuring the basic performance lower limit of the battery pack. Qualitative capacity differences are transformed into calculable and comparable quantitative data. A constant current pulse is used to discharge the qualified batteries to calculate their dynamic DC internal resistance. Based on the dynamic DC internal resistance, pulse-qualified batteries are selected from the qualified batteries, accurately reflecting their performance under actual high-power conditions and improving the performance of the battery pack after pairing. The system achieves high-power operating condition reliability. It performs float charging on qualified pulse batteries to obtain their steady-state float charging voltage. Based on this voltage, target qualified batteries are selected from the pulse qualified batteries, eliminating batteries with large voltage drift and unstable performance under long-term float charging conditions. This addresses the problem of traditional battery grouping focusing only on a single operating condition and having poor adaptability. A three-dimensional feature vector is constructed for the target qualified batteries. This vector includes the corresponding capacity value, dynamic DC internal resistance value, and steady-state float charging voltage value, preventing a single indicator's large range from dominating the clustering results and ensuring fairness in multi-dimensional matching. The three-dimensional feature vector is standardized to obtain a processed three-dimensional feature vector. The number of batteries in each group and the number of clusters are determined according to order requirements. Using a clustering algorithm and based on the processed three-dimensional feature vector, the number of batteries in each group, and the number of clusters, the target qualified batteries are clustered and intelligently grouped. This avoids errors and omissions in manual grouping, reduces battery waste caused by rigid grouping rules, and improves the overall grouping success rate of multiple batches of batteries. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0021] Figure 1This is a flowchart of an intelligent grouping method for multiple batches of lead-acid batteries disclosed in this application; Figure 2 This is a flowchart illustrating the intelligent grouping of multiple batches of lead-acid batteries disclosed in this application. Figure 3 This is a schematic diagram of the structure of an intelligent matching device for multiple batches of lead-acid batteries disclosed in this application; Figure 4 This application provides a structural diagram of an electronic device. Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] High-power lead-acid batteries are the core of backup power systems for critical facilities such as data centers and UPS systems. In these applications, battery banks need to withstand high current loads instantaneously, making their high-power discharge performance and long-term float charging reliability crucial. Furthermore, in these applications, batteries are typically used in series or series-parallel configurations, and their overall performance and lifespan depend heavily on the consistency of individual cells within the bank. Poor consistency not only reduces the overall capacity of the bank but also causes lagging cells to fail prematurely during charge-discharge cycles, leading to a chain reaction of failures throughout the entire bank. During the production, storage, and after-sales processes of high-power batteries, a large number of "non-banked" surplus cells accumulate. These batteries have complex origins and inconsistent performance, including surplus cells from production line assembly and cells replaced from retired battery banks. They come from multiple different batches and even different production years, exhibiting significant differences in voltage, internal resistance, capacity, and high-power performance indicators. Currently, the industry handles these batteries using the following methods: 1. Downgrading: Using them in applications with lower requirements, but failing to realize their original high-power performance value, resulting in poor economic efficiency. 2. **Scrap Recycling:** Directly treating large quantities of batteries with remaining lifespan as scrap lead not only results in significant resource waste and economic losses but also contradicts the green and environmentally friendly industrial development direction. 3. **Simple and Inefficient Grouping:** This grouping method relies primarily on static parameters (such as open-circuit voltage and static internal resistance) or a single capacity value for matching. This method cannot reflect the dynamic performance consistency of batteries under actual high-power discharge conditions. 4. Some advanced processes have introduced a "dual screening" mechanism based on capacity and float charge voltage, but it still suffers from problems such as limited dimensionality, lack of intelligence, and poor adaptability. Therefore, solving the problems of failing to assess dynamic consistency and relying on human experience, maximizing the efficient recycling and utilization of battery resources, and improving the reliability and efficiency of battery regrouping are issues that need to be addressed in this field.

[0024] See Figure 1 As shown in the figure, this invention discloses a smart grouping method for multiple batches of lead-acid batteries, which specifically includes: Step S11: Pre-treat and repair the performance of lead-acid batteries under multiple batches of preset power to obtain repaired batteries.

[0025] In this embodiment, a current-limited constant-voltage charging method is used to replenish the lead-acid batteries under a preset power in multiple batches; after replenishing the batteries, acid is added, and a multi-stage stepped charge-discharge system is used to repair the battery performance of the batteries after acid addition, resulting in a repaired battery.

[0026] This application describes a pretreatment process involving charging and acid replenishment for multiple batches of lead-acid batteries at preset power levels. Acid replenishment restores the electrolyte level and density, and multiple staged charge-discharge activation cycles promote the dissolution and recrystallization of the lead sulfate passivation layer on the plates, thereby restoring battery capacity and dynamic performance.

[0027] Specifically, for lead-acid batteries that have exceeded their storage time limit, a current-limited constant-voltage charging method is used to restore them to a near-fully charged state, providing uniform initial conditions for subsequent acid replenishment and activation. The acid replenishment operation involves injecting a sulfuric acid solution with a density similar to the original electrolyte into the upper part of the busbar, supplemented by vacuum treatment, to achieve the following synergistic effects: first, restoring the normal electrolyte level and density, ensuring sufficient wetting of active materials; second, optimizing electrolyte conductivity and reducing battery ohmic internal resistance; and third, promoting the dissolution and transformation of the lead sulfate passivation layer on the plates through the coupling of acid concentration adjustment with subsequent charge and discharge processes. The acid replenishment density tolerance is controlled within... Within the range, to improve the consistency between batteries; the charge-discharge activation cycle adopts a multi-stage stepped charge-discharge regime, through the combination of 0.1C discharge, 0.1C charge, 0.05C charge and 0.03C charge, the electrochemical action is used to promote the dissolution and recrystallization of the lead sulfate passivation layer of the negative electrode plate into spongy lead, effectively restoring the reaction area and pore structure of the active material, thereby repairing the battery capacity and dynamic performance.

[0028] Step S12: Perform a battery capacity test on the repaired battery to obtain a qualified battery and its corresponding discharge termination voltage. Determine the capacity level of the qualified battery based on the discharge termination voltage and convert the capacity level into a capacity value.

[0029] In this embodiment, after obtaining the qualified capacity test battery and the corresponding discharge termination voltage, the qualified capacity test battery is classified into capacity grades based on the standard capacity test method and the preset voltage segmentation range and discharge termination voltage to determine the capacity grade of the qualified capacity test battery; a unique identifier is added to the capacity grade, and the unique identifier is converted into a capacity value.

[0030] This application employs a standard capacity testing method, using the discharge termination voltage as the quantitative basis for capacity grading. Capacity-test-qualified batteries are categorized into multiple fine grades and assigned unique identifiers. This grading method discretizes continuous voltage values ​​into ordered categorical variables, preserving the relative differences in battery energy fundamentals while transforming categorical characteristics into capacity values ​​(e.g., A=1, B=2…), laying the foundation for subsequent multi-dimensional data fusion. The ordered numerical representation of unique identifiers allows features of different dimensions to be uniformly incorporated into the data model.

[0031] Step S13: Discharge the qualified capacity test battery using a constant current pulse to calculate the dynamic DC internal resistance value of the battery, and select the qualified pulse battery from the qualified capacity test batteries based on the dynamic DC internal resistance value of the battery.

[0032] In this embodiment, the process for calculating the dynamic DC internal resistance of the battery is as follows: the battery that has passed the capacity test is discharged using a constant current pulse, and the instantaneous voltage before the pulse discharge and the ending voltage at the end of the pulse discharge are collected; the voltage difference between the instantaneous voltage and the ending voltage is calculated; the ratio between the voltage difference and the pulse current is used as the dynamic DC internal resistance of the battery.

[0033] This application applies a short-duration, high-current constant pulse to each qualified capacity test battery, collects the voltage before the pulse starts and during discharge, calculates the dynamic DC internal resistance value of the battery, and removes batteries with abnormal dynamic performance based on statistical thresholds.

[0034] In other words, a short-duration, high-current constant pulse is used to excite the battery's dynamic response, with a pulse discharge current ratio of 1.0C. 10 Up to 3.0C 10 The pulse duration ranges from 5 to 30 seconds. This pulse intensity effectively stimulates the battery's high-power discharge characteristics without causing irreversible damage. The battery's DCR (Dynamic DC Internal Resistance) value is calculated based on the ratio of the voltage difference between the voltage instant before and after the pulse application to the pulse current, and is considered the gold standard for characterizing a battery's high-current output capability. This method can accurately identify batteries with potential dynamic performance degradation that meet capacity standards (such as micro-short circuits on plates, aging of active materials, poor connections, etc.) and, based on statistical thresholds (such as the mean), further refine the analysis. Dynamic anomaly removal is performed to fundamentally prevent the risk of early failure.

[0035] Step S14: Perform float charging on the pulse qualified battery to obtain the float charging steady-state voltage value, and select the target qualified battery from the pulse qualified battery based on the float charging steady-state voltage.

[0036] In this embodiment, the qualified pulse battery is float-charged using a constant parameter current-limiting constant voltage mode, and the steady-state voltage value of the battery at the end of the float charge is recorded. Pulse qualified batteries whose steady-state voltage value is greater than a preset voltage threshold are identified as abnormal pulse qualified batteries. The abnormal pulse qualified batteries are then removed to obtain the target qualified battery.

[0037] In this embodiment, after pulse discharge in step S13, float charging (current-limited constant-voltage charging + resting) is performed to ensure that all batteries are in the same fully charged state when entering the float charging process. The float charging process adopts a fixed-parameter current-limited constant-voltage mode, and the steady-state voltage value of the battery at the end of the float charging stage is recorded. The float charging voltage is a core characteristic of the consistency of the battery under long-term standby conditions, and its dispersion directly reflects the differences in battery self-discharge rate, gas recombination efficiency, and grid corrosion state. Abnormal float charging batteries are removed based on statistical thresholds (such as mean ±250mV), which can effectively ensure the stability of the reconstituted battery pack during the float charging life.

[0038] Step S15: Construct a three-dimensional feature vector for the target qualified battery; the three-dimensional feature vector includes the capacity value, dynamic DC internal resistance value, and float charge steady-state voltage value corresponding to the target qualified battery.

[0039] In this embodiment, a three-dimensional feature vector (including capacity value, battery dynamic DC internal resistance value, and float charge steady-state voltage value) is constructed for each target qualified battery. This vector fully covers the three key performance dimensions of the battery: the capacity value represents the energy base of the battery, the battery dynamic DC internal resistance value represents the high power output capability, and the float charge steady-state voltage value represents the long-term float charge stability.

[0040] Step S16: Standardize the three-dimensional feature vector to obtain the processed three-dimensional feature vector.

[0041] In this embodiment, since the three features in the three-dimensional feature vector have different dimensions and numerical ranges, each feature in the three-dimensional feature vector needs to be Z-score standardized to transform it into a standard normal distribution with a mean of 0 and a standard deviation of 1, thereby eliminating the influence of dimensions.

[0042] Step S17: Determine the number of batteries in each group and the number of clusters according to the order requirements. Using a clustering algorithm and based on the processed three-dimensional feature vector, the number of batteries in each group, and the number of clusters, cluster the target qualified batteries and perform intelligent battery matching to obtain each group of target qualified batteries.

[0043] In this embodiment, the K-means clustering algorithm is used to cluster the target qualified batteries based on the number of clusters, resulting in target qualified batteries in each cluster. It is then determined whether the number of target qualified batteries in each cluster is consistent with the number of batteries in each group. If they are consistent, the three-dimensional feature vectors of the target qualified batteries in the same cluster are matched for consistency. If the consistency matching passes, the target qualified batteries in the same cluster are intelligently grouped to obtain target qualified batteries in each group. If they are inconsistent, the principle of closest distance to the centroid is used to supplement the target qualified batteries in the inconsistent clusters, and then the consistency matching process is executed.

[0044] This application employs the K-means clustering algorithm to cluster and group the processed three-dimensional feature vectors. The number of clusters K is dynamically determined based on the quantity N of each battery group required in the order and the total number M of batteries to be matched, as shown in the following formula: K = floor(M / N).

[0045] Then, determine whether the number of qualified target batteries in each cluster is consistent with the number of batteries in each group; if they are consistent, the qualified target batteries in the same cluster are determined to be the combination with the highest consistency in the three dimensions of capacity, dynamic DC internal resistance and float charge voltage, and are directly matched into a group.

[0046] To address situations where the number of qualified batteries within a cluster does not perfectly match order requirements, a nearest-centroid principle is used to supplement qualified batteries from the nearest cluster, achieving smooth adjustment of cluster boundaries. This strategy maximizes battery utilization and batch power while ensuring consistency within the cluster.

[0047] After obtaining the target qualified batteries for each group, the process also includes: measuring the open-circuit voltage and the overall internal resistance of each group of target qualified batteries to obtain the open-circuit voltage range and internal resistance deviation; determining whether the open-circuit voltage range and internal resistance deviation meet the requirements; and storing and shipping the target qualified batteries that meet the requirements if the open-circuit voltage range and internal resistance deviation meet the requirements.

[0048] In other words, after obtaining the target qualified batteries for each group, the battery packs that have been matched are left to stand, and the open circuit voltage and internal resistance are tested. If they pass the test, they are stored separately and shipped. Batteries rejected during the process are downgraded or scrapped.

[0049] Taking 350 unassembled, high-power battery cells with inconsistent performance and numerous batches as an example, there are 350 unassembled batteries in the production and warehousing process. These batteries need performance repair and are required to be shipped in groups of 40. The battery's rated power is 630W, rated capacity C is 160Ah, and the finished electrolyte density is [not specified] at 15℃. The specific process for applying the technical solution of this application is as follows: Figure 2 As shown, the steps are as follows: Step 1: Pre-treatment and performance restoration of the battery: Recharge: Charge all stock batteries for 10 hours at 0.10C (16A) current limiting and constant voltage of 2.45V / cell to bring the batteries to near full charge. Adding acid: Remove the sealing cap of the valve-regulated lead-acid battery, pull out the safety valve, and use a syringe to inject acid into the battery at a density of 15°C. Pour sulfuric acid solution into the upper part of the manifold, evacuate the battery for 15 seconds, and let it stand for 2.5 hours; Charge-discharge activation cycle: 0.1C (16A) discharge for 6 hours; Charge at 0.1C (16A) for 5 hours; Charge at 0.05C (8A) for 4 hours; Charge at 0.03C (4.8A) for 2 hours; Let it stand for 2 hours, then use a syringe to extract the excess acid from the battery.

[0050] Step 2: Battery Capacity Testing and Capacity Grade Classification Battery capacity test: 0.1C (16A) discharge for 10 hours, 0.1C (16A) charge for 9 hours, 0.05C (8A) charge for 9 hours; Grading rules: Based on the discharge termination voltage, batteries that have passed the capacity test are divided into multiple grades, and each grade is assigned a unique identifier. The grading method is shown in Table 1. Table 1. Capacity Classification Standards

[0051] The unique identifier corresponding to the capacity level is converted into a capacity value A=1, B=2, ..., I=9; In this example, after battery capacity testing, a total of 330 batteries passed the capacity test, while 20 unqualified batteries were discarded.

[0052] Step 3: Dynamic Pulse Characteristics Evaluation Short-duration high-current pulse: Apply a constant current pulse discharge of 2.5C (400A) to each qualified capacity test battery for 10 seconds; Calculation of battery dynamic DC internal resistance: Acquisition of instantaneous voltage before pulse discharge. and the voltage at the end of the pulse discharge The dynamic DC internal resistance of each battery is calculated using the following formula: ; Abnormal Rejection: The dynamic DC internal resistance of this batch of batteries is approximately 2.0 mΩ. A tolerance of ±15% is set, meaning batteries with an internal resistance between 1.70 and 2.30 mΩ are retained, while those exceeding this range are rejected. This step rejects 20 batteries.

[0053] Step 4: Float charging: Post-pulse recharge: Charge the batteries that passed the pulse test at 0.10C (16A) current-limited constant voltage of 2.45V / cell for 2 hours, then let them rest for 2 hours to ensure that all batteries are in the same fully charged state.

[0054] Float charge test: Using 0.1C (16A) current limiting, 2.25V / cell constant voltage float charge for 8 hours, record the steady state voltage value at the end of float charge.

[0055] Abnormal rejection: The average float charge steady-state voltage of this batch is approximately 13.50V, with an allowable deviation of ±250mV. Batteries with voltages between 13.25-13.75V are retained, while those exceeding this range are rejected. This step rejects 10 batteries.

[0056] After steps three and four, a total of 300 qualified batteries remain.

[0057] Step 5: Construction of 3D Feature Vectors For each target qualified battery, construct a three-dimensional feature vector containing its capacity value, dynamic DC internal resistance value, and float charge steady-state voltage value. For example: Battery 1: [Capacity grade D (capacity value 4), DCR=2.1mΩ, float charge steady-state voltage=13.62V]; Battery 2: [Capacity grade E (capacity value 5), DCR=2.2mΩ, float charge steady-state voltage=13.58V].

[0058] Step Six: Standardization. Perform Z-score standardization on the three-dimensional feature data using the following formula: ; After Z-score standardization, the influence of dimensions can be eliminated: Battery 1: [Capacity characteristic = 0.97, DCR characteristic = 0.84, Float charge voltage characteristic = 0.72]; Battery 2: [Capacity characteristic = 0.15, DCR characteristic = 0.82, Float charge voltage characteristic = 0.58].

[0059] Step 7: Cluster Analysis Determine the number of clusters K: The order requires N=40 batteries per group, and the total number of battery groups to be matched is M=300, then K = floor(M / N) = floor(300 / 40) = 7; Perform K-means clustering: Input standardized 300 A 3D data matrix is ​​used, with K=7, to run the algorithm.

[0060] Step 8: Intelligent Grouping Batteries within the same cluster are considered to have the highest consistency in three dimensions: capacity, dynamic DC internal resistance, and float charge voltage, and are directly grouped together.

[0061] If the number of batteries in a cluster is not equal to 40, the batteries closest to the centroid of the cluster are added from the nearest neighboring cluster first, until each group has 40 batteries; if the number of batteries in a cluster exceeds 40, the excess batteries are adjusted to the neighboring clusters with insufficient numbers.

[0062] In this example, the 7 clusters each contain 38-42 cells. After fine-tuning, they are divided into 7 groups of 40 cells each. Cells that become outliers after clustering, i.e., cells that do not belong to any large cluster, can be processed separately or re-detected and reassigned to a similar group.

[0063] Step Nine: Offline and Verification After the battery packs are assembled, let them stand for 24 hours, then measure the open-circuit voltage and the overall internal resistance. The open-circuit voltage range should be ≤50mV and the internal resistance deviation should be ≤10%. Qualified battery packs are stored separately, labeled, and then shipped. A total of 50 batteries are rejected throughout the process (including those with unqualified capacity, dynamic DC internal resistance exceeding the range, and float charge voltage exceeding the range), and are downgraded for use in low-power scenarios or disposed of as scrap.

[0064] This invention maximizes the recycling of battery resources and ensures that the recombined battery packs achieve commercially viable levels in terms of capacity, high-rate discharge performance, and float charging stability.

[0065] Addressing the shortcomings of existing technologies in handling multiple batches of non-grouped high-power lead-acid batteries—namely, their crude processing methods, poor dynamic consistency after reassembly, and inability to meet specific application requirements—this application proposes an intelligent grouping method for multiple batches of lead-acid batteries. The advantages are as follows: Precise dynamic performance diagnosis and significantly increased reliability: Through pulse characteristic evaluation, the dynamic DC internal resistance of the battery is accurately quantified, effectively identifying batteries with qualified static parameters but deteriorating dynamic performance. This fundamentally eliminates the risk of early failure of reassembled battery packs during high-power discharge, resulting in reliability far exceeding traditional methods; Multi-dimensional intelligent integration and a leap forward in scientific grouping: Employing clustering analysis algorithms, the method achieves comprehensive analysis of battery capacity, dynamic DC internal resistance, and float charging performance. The unified quantification and comprehensive optimal matching of the three key dimensions of voltage overcome the one-sidedness of human experience and ensure a high degree of consistency in the overall performance of the recombined battery pack; it has strong adaptability and maximizes resource utilization: the intelligent algorithm can adapt to mixed datasets of multiple batches of batteries from different sources and with different aging levels, automatically find the optimal grouping, and enable a large number of surplus batteries to be safely and reliably reused in high-power scenarios, resulting in significant economic benefits; the process system is complete and easy to promote industrially: this application proposes a complete process flow from preprocessing, repair, dynamic detection to intelligent decision-making, with clear logic, well-defined parameters, and the overall process is easy to implement in industrial batches on the basis of existing equipment.

[0066] This application pre-processes and repairs the performance of lead-acid batteries under preset power conditions in multiple batches. It addresses issues such as plate sulfation, electrolyte imbalance, and capacity decay caused by storage, use, and batch differences in lead-acid batteries. This improves the resource utilization rate of multiple batches of used / accumulated lead-acid batteries, reduces battery scrap rates and material loss costs for enterprises. The repaired batteries undergo capacity testing to obtain qualified batteries and their corresponding discharge termination voltages. Based on the discharge termination voltage, the capacity level of the qualified batteries is determined, and the capacity level is converted into a capacity value. Batteries with substandard core energy storage capacity are eliminated, ensuring the basic performance lower limit of the battery pack. Qualitative capacity differences are transformed into calculable and comparable quantitative data. A constant current pulse is used to discharge the qualified batteries to calculate their dynamic DC internal resistance. Based on this dynamic DC internal resistance, pulse-qualified batteries are selected from the qualified batteries, accurately reflecting their performance under actual high-power conditions and improving the performance of the assembled battery pack. High-power operating condition reliability: Pulse-qualified batteries are float-charged to obtain a steady-state float voltage. Based on this steady-state voltage, target qualified batteries are selected from the pulse-qualified batteries, eliminating batteries with large voltage drift and unstable performance under long-term float charging conditions. This addresses the problem of traditional battery grouping focusing only on a single operating condition and having poor adaptability. A three-dimensional feature vector is constructed for the target qualified batteries. This vector includes the capacity value, dynamic DC internal resistance value, and steady-state float voltage value corresponding to the target qualified battery, avoiding the dominance of a single indicator value in the clustering results and ensuring the fairness of multi-dimensional matching. The three-dimensional feature vector is standardized to obtain a processed three-dimensional feature vector. The number of batteries in each group and the number of clusters are determined according to order requirements. Using a clustering algorithm and based on the processed three-dimensional feature vector, the number of batteries in each group, and the number of clusters, the target qualified batteries are clustered and intelligently grouped. This avoids errors and omissions in manual grouping, reduces battery waste caused by rigid grouping rules, and improves the overall grouping success rate of multiple batches of batteries.

[0067] See Figure 3 As shown in the figure, an embodiment of the present invention discloses an intelligent pairing device for multiple batches of lead-acid batteries, which may specifically include: The pre-processing and battery performance repair module 11 is used to pre-process and repair the battery performance of multiple batches of lead-acid batteries under preset power to obtain repaired batteries. The battery capacity testing module 12 is used to perform battery capacity testing on the repaired battery, obtain a qualified battery and its corresponding discharge termination voltage, determine the capacity level of the qualified battery based on the discharge termination voltage, and convert the capacity level into a capacity value. The pulse qualified battery screening module 13 is used to discharge the capacity-test qualified battery with a constant current pulse to calculate the dynamic DC internal resistance value of the battery, and to screen out the pulse qualified battery from the capacity-test qualified battery based on the dynamic DC internal resistance value of the battery. The float charging module 14 is used to float charge the pulse qualified battery to obtain a float charging steady-state voltage value, and to select target qualified batteries from the pulse qualified batteries based on the float charging steady-state voltage. The three-dimensional feature vector construction module 15 is used to construct a three-dimensional feature vector for the target qualified battery; the three-dimensional feature vector includes the capacity value, dynamic DC internal resistance value, and float charge steady-state voltage value corresponding to the target qualified battery. Standardization processing module 16 is used to standardize the three-dimensional feature vector to obtain the processed three-dimensional feature vector; The clustering and intelligent battery matching module 17 is used to determine the number of batteries in each group and the number of clusters according to the order requirements. It uses a clustering algorithm and based on the processed three-dimensional feature vector, the number of batteries in each group, and the number of clusters, to perform clustering and intelligent battery matching on the target qualified batteries to obtain each group of target qualified batteries.

[0068] In some specific embodiments, the preprocessing and battery performance repair module 11 may specifically include: The charging module is used to charge multiple batches of lead-acid batteries at preset power using a current-limited constant-voltage charging method. The battery performance repair module is used to replenish acid in the battery after it has been recharged, and to repair the battery performance by using a multi-stage stepped charge and discharge system, resulting in a repaired battery.

[0069] In some specific embodiments, the battery capacity testing module 12 may specifically include: The capacity class classification module is used to classify the capacity of qualified batteries using standard capacity testing methods and based on preset voltage segmentation ranges and discharge termination voltages, so as to determine the capacity class of qualified batteries. The unique identifier conversion module is used to add a unique identifier to the capacity level and convert the unique identifier into a capacity value.

[0070] In some specific embodiments, the pulse qualified battery screening module 13 may specifically include: The discharge module is used to discharge the qualified capacity battery using a constant current pulse, and to collect the instantaneous voltage before the pulse discharge and the end voltage when the pulse discharge ends. The differential pressure calculation module is used to calculate the differential pressure between the instantaneous voltage and the final voltage; The battery dynamic DC internal resistance determination module is used to determine the ratio between the voltage difference and the pulse current as the battery dynamic DC internal resistance value.

[0071] In some specific embodiments, the float charging module 14 may specifically include: The voltage value recording module is used to perform float charging on the pulse qualified battery and record the steady-state voltage value of the battery at the end of float charging using a fixed parameter current limiting constant voltage mode. The abnormal pulse qualified battery determination module is used to identify pulse qualified batteries whose float charge steady-state voltage value is greater than a preset voltage threshold as abnormal pulse qualified batteries. The rejection module is used to reject the qualified batteries with abnormal pulses to obtain the target qualified batteries.

[0072] In some specific embodiments, the clustering and intelligent battery matching module 17 may specifically include: The clustering and grouping module is used to use the K-means clustering algorithm to cluster and group the target qualified batteries based on the number of clusters, and obtain the target qualified batteries in each cluster. The first judgment module is used to determine whether the number of qualified target batteries in each cluster is consistent with the number of batteries in each group. The battery intelligent grouping module is used to perform consistency matching on the three-dimensional feature vectors of the target qualified batteries in the same cluster if they are consistent. If the consistency matching is successful, the target qualified batteries in the same cluster are intelligently grouped to obtain each group of target qualified batteries. The battery replenishment module is used to replenish qualified target batteries within inconsistent clusters by using the principle of closest proximity to the centroid, and then execute the consistency matching process if there is any inconsistency.

[0073] In some specific embodiments, the clustering and intelligent battery matching module 17 may specifically include: The open-circuit voltage and overall internal resistance measurement module is used to measure the open-circuit voltage and overall internal resistance of each group of target qualified batteries, and to obtain the open-circuit voltage range and internal resistance deviation. The second judgment module is used to determine whether the open-circuit voltage range and internal resistance deviation meet the requirements. The storage and shipping module is used to store and ship the target qualified batteries that meet the requirements if the open-circuit voltage difference and internal resistance deviation are within acceptable limits.

[0074] Figure 4This is a schematic diagram of an electronic device provided in an embodiment of this application. The electronic device 20 may specifically include: at least one processor 21, at least one memory 22, a power supply 23, a communication interface 24, an input / output interface 25, and a communication bus 26. The memory 22 stores a computer program, which is loaded and executed by the processor 21 to implement the relevant steps in the intelligent grouping method for multiple batches of lead-acid batteries disclosed in any of the foregoing embodiments.

[0075] In this embodiment, the power supply 23 is used to provide operating voltage for each hardware device on the electronic device 20; the communication interface 24 can create a data transmission channel between the electronic device 20 and external devices, and the communication protocol it follows can be any communication protocol applicable to the technical solution of this application, and is not specifically limited here; the input / output interface 25 is used to acquire external input data or output data to the outside world, and its specific interface type can be selected according to specific application needs, and is not specifically limited here.

[0076] In addition, the memory 22, as a carrier for resource storage, can be a read-only memory, random access memory, disk or optical disk, etc. The resources stored on it include operating system 221, computer program 222 and data 223, etc., and the storage method can be temporary storage or permanent storage.

[0077] The operating system 221 manages and controls the various hardware devices on the electronic device 20 and the computer program 222 to enable the processor 21 to perform calculations and processing on the data 223 in the memory 22. The operating system 221 can be Windows, Unix, Linux, etc. The computer program 222, in addition to including a computer program capable of performing the intelligent grouping method for multiple batches of lead-acid batteries disclosed in any of the foregoing embodiments, may further include computer programs capable of performing other specific tasks. The data 223 may include data received by the intelligent grouping device for multiple batches of lead-acid batteries from external devices, and may also include data collected by its own input / output interface 25.

[0078] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0079] Furthermore, this application also discloses a computer-readable storage medium storing a computer program. When the computer program is loaded and executed by a processor, it implements the steps of the intelligent grouping method for multiple batches of lead-acid batteries disclosed in any of the foregoing embodiments.

[0080] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0081] The present invention provides a detailed description of an intelligent grouping method, apparatus, equipment, and storage medium for multiple batches of lead-acid batteries. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for intelligent grouping of multiple batches of lead-acid batteries, characterized in that, include: Pre-treatment and battery performance repair were performed on multiple batches of lead-acid batteries at preset power levels to obtain repaired batteries. The repaired battery is subjected to a battery capacity test to obtain a qualified battery and its corresponding discharge termination voltage. The capacity level of the qualified battery is determined based on the discharge termination voltage, and the capacity level is converted into a capacity value. The capacity-tested qualified batteries are discharged using a constant current pulse to calculate the dynamic DC internal resistance value of the batteries. Based on the dynamic DC internal resistance value of the batteries, pulse-qualified batteries are selected from the capacity-tested qualified batteries. The pulse-qualified batteries are float-charged to obtain a float-charge steady-state voltage value. Based on the float-charge steady-state voltage, target qualified batteries are selected from the pulse-qualified batteries. A three-dimensional feature vector is constructed for the target qualified battery; the three-dimensional feature vector includes the capacity value, dynamic DC internal resistance value, and float charge steady-state voltage value corresponding to the target qualified battery. The three-dimensional feature vector is standardized to obtain the processed three-dimensional feature vector. Based on the order requirements, the number of batteries in each group and the number of clusters are determined. Using a clustering algorithm and based on the processed three-dimensional feature vector, the number of batteries in each group, and the number of clusters, the target qualified batteries are clustered and intelligently matched to obtain each group of target qualified batteries.

2. The intelligent grouping method for multiple batches of lead-acid batteries according to claim 1, characterized in that, The process of pre-treating and repairing multiple batches of lead-acid batteries at preset power levels to obtain repaired batteries includes: A current-limited constant-voltage charging method was used to recharge multiple batches of lead-acid batteries at preset power levels. After the battery is recharged, acid is added, and a multi-stage stepped charge-discharge process is used to repair the battery performance, resulting in a repaired battery.

3. The intelligent grouping method for multiple batches of lead-acid batteries according to claim 1, characterized in that, The process of determining the capacity level of a battery that has passed the capacity test based on the discharge termination voltage, and converting the capacity level into a capacity value, includes: Using standard capacity testing methods and based on preset voltage segmentation ranges and discharge termination voltages, capacity-test-qualified batteries are classified into capacity grades to determine their capacity grades. Add a unique identifier to the capacity level and convert the unique identifier into a capacity value.

4. The intelligent grouping method for multiple batches of lead-acid batteries according to claim 1, characterized in that, The step of discharging the capacity-tested battery using a constant current pulse to calculate the battery's dynamic DC internal resistance includes: The capacity-tested battery was discharged using a constant current pulse, and the instantaneous voltage before the pulse discharge and the ending voltage at the end of the pulse discharge were collected. Calculate the voltage difference between the instantaneous voltage and the final voltage; The ratio between the voltage difference and the pulse current is used as the dynamic DC internal resistance of the battery.

5. The intelligent grouping method for multiple batches of lead-acid batteries according to claim 1, characterized in that, The step of performing float charging on the qualified pulse batteries to obtain a steady-state float charging voltage value, and selecting target qualified batteries from the qualified pulse batteries based on the steady-state float charging voltage, includes: The qualified pulse battery was float-charged using a constant parameter current-limiting constant voltage mode, and the steady-state voltage value of the battery at the end of the float charge was recorded. The pulse-qualified batteries whose float charge steady-state voltage value is greater than the preset voltage threshold are regarded as abnormal pulse-qualified batteries. The qualified batteries with abnormal pulses are discarded to obtain the target qualified batteries.

6. The intelligent grouping method for multiple batches of lead-acid batteries according to claim 1, characterized in that, The method utilizes a clustering algorithm and, based on the processed three-dimensional feature vector, the number of batteries in each group, and the number of clusters, performs clustering and intelligent battery grouping on the target qualified batteries to obtain each group of target qualified batteries, including: The K-means clustering algorithm is used to cluster and divide the target qualified batteries based on the number of clusters, and the target qualified batteries in each cluster are obtained. Determine whether the number of qualified target batteries in each cluster is consistent with the number of batteries in each group; If they match, the three-dimensional feature vectors of the target qualified batteries in the same cluster will be matched for consistency. If the consistency matching passes, the target qualified batteries in the same cluster will be intelligently grouped to obtain each group of target qualified batteries. If there is a discrepancy, the principle of proximity to the centroid is adopted to supplement the qualified target cells within the inconsistent cluster, and then the consistency matching process is executed.

7. The intelligent grouping method for multiple batches of lead-acid batteries according to any one of claims 1 to 6, characterized in that, After obtaining the target qualified batteries for each group, the process also includes: The open-circuit voltage and the overall internal resistance of each group of target qualified batteries were measured to obtain the open-circuit voltage range and the internal resistance deviation. Determine whether the open-circuit voltage range and internal resistance deviation meet the requirements; If the open-circuit voltage range and internal resistance deviation meet the requirements, then the qualified target batteries that meet the requirements will be stored and shipped.

8. An intelligent grouping device for multiple batches of lead-acid batteries, characterized in that, include: The pre-processing and battery performance repair module is used to pre-process and repair the battery performance of multiple batches of lead-acid batteries under preset power to obtain repaired batteries. The battery capacity testing module is used to perform battery capacity testing on the repaired battery, obtain a qualified battery and its corresponding discharge termination voltage, determine the capacity level of the qualified battery based on the discharge termination voltage, and convert the capacity level into a capacity value. The pulse qualified battery screening module is used to discharge the capacity-test qualified batteries using a constant current pulse to calculate the dynamic DC internal resistance value of the battery, and to screen out pulse qualified batteries from the capacity-test qualified batteries based on the dynamic DC internal resistance value of the battery. A float charging module is used to float charge the pulse qualified battery to obtain a float charging steady-state voltage value, and to select target qualified batteries from the pulse qualified batteries based on the float charging steady-state voltage. A three-dimensional feature vector construction module is used to construct a three-dimensional feature vector for the target qualified battery; the three-dimensional feature vector includes the capacity value, dynamic DC internal resistance value, and float charge steady-state voltage value corresponding to the target qualified battery. A standardization processing module is used to standardize the three-dimensional feature vector to obtain the processed three-dimensional feature vector. The clustering and intelligent battery matching module is used to determine the number of batteries in each group and the number of clusters according to the order requirements. It uses a clustering algorithm and based on the processed three-dimensional feature vector, the number of batteries in each group, and the number of clusters, to perform clustering and intelligent battery matching on the target qualified batteries to obtain each group of target qualified batteries.

9. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor for executing the computer program to implement the intelligent grouping method for multiple batches of lead-acid batteries as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, Used to store computer programs; wherein, when the computer programs are executed by a processor, they implement the intelligent grouping method for multiple batches of lead-acid batteries as described in any one of claims 1 to 7.

Images