Retired battery sorting method and device
By training the capacity prediction model, the residual capacity of the decommissioned battery is predicted using discharge characteristic data, which solves the problem of long capacity testing in the prior art and improves the efficiency and accuracy of the sorting of the decommissioned battery.
Patent Information
- Application Number
- CN202411741993.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-29
- Publication Date
- 2025-03-21
AI Technical Summary
In the existing retired battery sorting technology, capacity testing takes a long time, affecting the sorting efficiency.
By obtaining the discharge characteristic data of the sample battery, training and building a capacity prediction model, predicting the remaining capacity of the batteries to be sorted, and sorting the retired battery.
It significantly improves the sorting efficiency, shortens the sorting time, and avoids the time-consuming problem of capacity testing.
Smart Images
Figure CN119667479A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of cascade utilization of retired batteries, and in particular to a method and device for sorting retired batteries. Background Art
[0002] With the rapid development of the electric vehicle industry, the cascade utilization of retired power batteries has become a new research hotspot. When the power battery capacity decays to 80%, it cannot be used anymore, but it can be used in fields such as small electric bicycles. Due to the different aging degrees of retired power batteries, if they are not screened, the inconsistency between batteries will seriously affect the performance of the energy supply system, leading to safety accidents such as excessive battery aging, thermal runaway, and even battery explosion. Therefore, it is very necessary to find a fast and accurate method for sorting retired batteries.
[0003] Retired battery sorting is to classify batteries with similar performance and combine them into battery packs to improve performance. Existing retired battery sorting technology usually relies on capacity testing.
[0004] Capacity testing has the problem of taking a long time. Summary of the invention
[0005] In view of this, it is necessary to provide a method and device for sorting retired batteries to solve the problem of time-consuming testing in the prior art.
[0006] In order to solve the above problems, the present application provides a method for sorting retired batteries, comprising: Acquire remaining capacity data and discharge characteristic data of a plurality of sample batteries, wherein the discharge characteristic data includes peak characteristic data of an IC curve; Using the discharge characteristic data of the sample battery as input and the remaining capacity data as output, training the constructed initial capacity prediction model to obtain a fully trained capacity prediction model; Obtaining discharge characteristic data of the battery to be sorted; The discharge characteristic data of the battery to be sorted is input into the well-trained capacity prediction model to obtain the predicted remaining capacity, and the retired batteries are sorted based on the predicted remaining capacity.
[0007] In some possible implementations, the step of obtaining discharge characteristic data of a plurality of sample batteries includes: Acquire voltage variation data of several sample batteries, wherein the voltage variation data includes first half-width peak voltage range data of the IC curve during the complete discharge process and first rebound voltage data after the battery is discharged to a preset voltage; Acquiring discharge characteristic data of the sample battery when discharged in the first half-width peak voltage range; The step of obtaining discharge characteristic data of the battery to be sorted includes: Acquire second rebound voltage data of the battery to be sorted after it is discharged to a preset voltage; determining second half-width peak voltage range data based on the second rebound voltage data according to a relationship between the first half-width peak voltage range data and the first rebound voltage data; Obtain discharge characteristic data of the battery to be sorted when discharged in the second half-width peak voltage range.
[0008] In some possible implementations, the voltage variation data further includes first peak voltage data of the IC curve during a complete discharge process; The determining, according to the relationship between the first half-width peak voltage range data and the first rebound voltage data and based on the second rebound voltage data, second half-width peak voltage range data comprises: Determine a mapping relationship between a rebound voltage and a peak voltage based on the first rebound voltage data and the first peak voltage data, and determine second peak voltage data of the battery to be sorted based on the mapping relationship and the second rebound voltage data; Based on a preset clustering algorithm, the first peak voltage data is classified to obtain several clusters and their corresponding maximum half-width peak voltage range data, the maximum half-width peak voltage range data is determined by the first half-width peak voltage range data within the cluster, the second peak voltage data is assigned to a target cluster among the several clusters, and the maximum half-width peak voltage range data corresponding to the target cluster is used as the second half-width peak voltage range data.
[0009] In some possible implementations, the clustering algorithm adopts a K-means clustering algorithm.
[0010] In some possible implementations, the peak characteristic data includes a peak value and a peak voltage.
[0011] In some possible implementations, the discharge characteristic data further includes curve similarity between discharge characteristic curves.
[0012] In some possible implementations, the calculation formula for the curve similarity is:
[0013] In the formula, represents the curve similarity, Indicates the similarity between the two discharge characteristic curves, Indicates the difference between the two discharge characteristic curves, Represents the weight parameter used to balance the relationship between similarity and difference.
[0014] In some possible implementations, the similarity is calculated and obtained by using any one of a cosine similarity measurement method and a Pearson correlation coefficient measurement method, and the difference is calculated and obtained by using any one of a Euclidean distance measurement method and a Manhattan distance measurement method.
[0015] In some possible implementations, the capacity prediction model uses a support vector machine.
[0016] The present application also provides a retired battery sorting device, comprising: A data acquisition unit, used to acquire the remaining capacity data and discharge characteristic data of a number of sample batteries, wherein the discharge characteristic data includes the peak characteristic data of the IC curve, and is also used to acquire the discharge characteristic data of the battery to be sorted; A model training unit, taking the discharge characteristic data of the sample battery as input and the remaining capacity data as output, training the constructed initial capacity prediction model to obtain a fully trained capacity prediction model; The model prediction unit is used to input the discharge characteristic data of the battery to be sorted into the trained capacity prediction model to obtain the predicted remaining capacity, and sort the retired batteries based on the predicted remaining capacity.
[0017] The beneficial effect of the present application is that the retired battery sorting method provided by the present application predicts the remaining capacity of the battery by analyzing the voltage changes during the battery charging and discharging process, extracting the dynamic peak characteristic data of the IC curve, and ensuring the accuracy of the capacity prediction. At the same time, the model prediction does not need to rely on the time-consuming capacity test, which significantly improves the sorting efficiency and shortens the sorting time. BRIEF DESCRIPTION OF THE DRAWINGS
[0018] Figure 1 A schematic diagram of an embodiment of the retired battery sorting method provided in this application; Figure 2 For this application Figure 1 A schematic flow chart of an embodiment of step S101; Figure 3 For this application Figure 1 A schematic flow chart of an embodiment of step S103; Figure 4 For this application Figure 3 A schematic flow chart of an embodiment of step S302; Figure 5 A schematic structural diagram of an embodiment of a retired battery sorting device provided in the present application. DETAILED DESCRIPTION
[0019] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.
[0020] It should be understood that the schematic drawings are not drawn to scale. The flowcharts used in this application illustrate operations implemented according to some embodiments of the present application. It should be understood that the operations of the flowcharts may be implemented out of order, and steps without logical contextual relationships may be reversed in order or implemented simultaneously. In addition, those skilled in the art may add one or more other operations to the flowcharts, or may remove one or more operations from the flowcharts, under the guidance of the content of this application. Some of the block diagrams shown in the accompanying drawings are functional entities, and do not necessarily correspond to physically or logically independent entities. These functional entities may be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor systems and / or microcontroller systems.
[0021] The descriptions of "first", "second", etc. involved in the embodiments of the present application are only used to describe the implicit purpose, and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the technical features defined as "first" and "second" may explicitly or implicitly include at least one of the features. "And / or" describes the association relationship of the associated objects, indicating that three relationships may exist, for example: A and / or B, which can represent: A exists alone, A and B exist at the same time, and B exists alone.
[0022] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.
[0023] The present application provides a method for sorting retired batteries, which is described below.
[0024] Figure 1 A schematic diagram of an embodiment of the retired battery sorting method provided in this application is as follows: Figure 1 As shown, the retired battery sorting method includes: S101, obtaining remaining capacity data and discharge characteristic data of a number of sample batteries, where the discharge characteristic data includes peak characteristic data of an IC curve; S102, using the discharge characteristic data of the sample battery as input and the remaining capacity data as output, training the constructed initial capacity prediction model to obtain a fully trained capacity prediction model; It should be noted that the capacity prediction model uses a support vector machine.
[0025] S103, obtaining discharge characteristic data of the battery to be sorted; S104, inputting the discharge characteristic data of the battery to be sorted into a well-trained capacity prediction model to obtain the predicted remaining capacity, and sorting the retired batteries based on the predicted remaining capacity.
[0026] Compared with the prior art, the present application predicts the remaining capacity of the battery by analyzing the voltage changes during the battery charging and discharging process, extracting the dynamic peak characteristic data of the IC curve, and ensuring the accuracy of the capacity prediction. At the same time, the model prediction does not rely on the time-consuming capacity test, which significantly improves the sorting efficiency and shortens the sorting time.
[0027] In order to further improve the sorting efficiency, in some embodiments, such as Figure 2 As shown, the step S101 obtains discharge characteristic data of several sample batteries, specifically including: S201, obtaining voltage change data of several sample batteries, the voltage change data including first half-width peak voltage range data of the IC curve during the complete discharge process and first rebound voltage data after the battery is discharged to a preset voltage; S202, obtaining discharge characteristic data of the sample battery when discharged in the first half-width peak voltage range; Correspondingly, such as Figure 3 As shown, step S103 specifically includes: S301, obtaining second rebound voltage data of the battery to be sorted after being discharged to a preset voltage; S302, determining second half-width peak voltage range data based on the second rebound voltage data according to the relationship between the first half-width peak voltage range data and the first rebound voltage data; S303, obtaining discharge characteristic data of the battery to be sorted when it is discharged within the second half-width peak voltage range.
[0028] In this way, according to the correlation between the rebound voltage and the half-width peak voltage range, it is only necessary to perform a short-time discharge test on the battery to be sorted to obtain the discharge characteristic data and improve the battery sorting efficiency.
[0029] In order to better determine the relationship between the half-width peak voltage range data and the rebound voltage data, in some embodiments, the voltage variation data also includes the first peak voltage data of the IC curve during the complete discharge process; It should be noted that the complete discharge process is to perform charge and discharge tests at 0.3C~1C after standing at room temperature for a sufficiently long time. The remaining capacity of the sample battery and the discharge characteristic curve during the complete discharge process can be obtained. The discharge characteristic curve is processed by dQ / dV (where Q represents capacity and V represents voltage) to obtain a large current IC curve. The obtained IC curve is then peak-searched to obtain the peak value of the strongest peak of the sample battery, the voltage value corresponding to the peak value, i.e. the peak voltage, and the voltage range at the half-width peak.
[0030] Correspondingly, such as Figure 4 As shown, step S302 specifically includes: S401, determining a mapping relationship between a rebound voltage and a peak voltage based on first rebound voltage data and first peak voltage data, and determining second peak voltage data of a battery to be sorted based on the mapping relationship and the second rebound voltage data; S402. Based on a preset clustering algorithm, the first peak voltage data is classified to obtain several clusters and their corresponding maximum half-width peak voltage range data, the maximum half-width peak voltage range data is determined by the first half-width peak voltage range data within the cluster, the second peak voltage data is assigned to a target cluster among the several clusters, and the maximum half-width peak voltage range data corresponding to the target cluster is used as the second half-width peak voltage range data.
[0031] It should be noted that the clustering algorithm adopts the K-means clustering algorithm. The calculation steps of the K-means clustering algorithm specifically include: first, select the number of clusters K (≤5) to be divided, and randomly select K sample points as the initial cluster center points; for each sample point, calculate the distance (Euclidean distance) between it and the center point of each cluster, and assign the sample point to the cluster with the closest distance; for each cluster, calculate the mean of all sample points in the cluster and use it as the new cluster center point; finally, K clusters are obtained, and the sample points in each cluster have similar characteristics, and the half-width peak maximum voltage range corresponding to the peak of each cluster is calibrated.
[0032] In order to improve the prediction accuracy of the model, in some embodiments, the discharge characteristic data includes the curve similarity between the discharge characteristic curves and the peak characteristic data of the IC curve, the peak characteristic data includes the peak value and the peak voltage, and the curve similarity is calculated and obtained by the elastic matching principle. The elastic matching is a matching method based on similarity and difference. The principle is to evaluate the matching degree of two objects by comprehensively considering the similarity and difference between them. The calculation formula is:
[0033] In the formula, represents the curve similarity, To express the similarity between two discharge characteristic curves, various similarity measurement methods can be used, such as cosine similarity, Pearson correlation coefficient, etc. Indicates the difference between two discharge characteristic curves. Various difference measurement methods can be used, such as Euclidean distance, Manhattan distance, etc. Represents the weight parameter used to balance the relationship between similarity and difference, which is a value between 0 and 1.
[0034] In order to better implement a retired battery sorting method in an embodiment of the present application, based on a retired battery sorting method, correspondingly, as Figure 5 As shown, the embodiment of the present application also provides a retired battery sorting device 500, comprising: The data acquisition unit 501 is used to acquire the remaining capacity data and discharge characteristic data of a number of sample batteries, the discharge characteristic data including the peak characteristic data of the IC curve, and is also used to acquire the discharge characteristic data of the battery to be sorted; The model training unit 502 takes the discharge characteristic data of the sample battery as input and the remaining capacity data as output, trains the constructed initial capacity prediction model, and obtains a fully trained capacity prediction model; The model prediction unit 503 is used to input the discharge characteristic data of the battery to be sorted into a well-trained capacity prediction model to obtain the predicted remaining capacity, and sort the retired batteries based on the predicted remaining capacity.
[0035] The retired battery sorting device 500 provided in the above embodiment can implement the technical solution described in the above retired battery sorting method embodiment. The specific implementation principles of the above units can refer to the corresponding contents in the above retired battery sorting method embodiment, which will not be repeated here.
[0036] The above is a detailed introduction to a retired battery sorting method provided by the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for technical personnel in this field, according to the idea of the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.
[0037] The above is only a preferred specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the technical field within the technical scope disclosed in the present application should be covered within the protection scope of the present application.
Claims
1. A method for sorting retired batteries, characterized in that: include: Acquire remaining capacity data and discharge characteristic data of a plurality of sample batteries, wherein the discharge characteristic data includes peak characteristic data of an IC curve; Using the discharge characteristic data of the sample battery as input and the remaining capacity data as output, training the constructed initial capacity prediction model to obtain a fully trained capacity prediction model; Obtaining discharge characteristic data of the battery to be sorted; The discharge characteristic data of the battery to be sorted is input into the well-trained capacity prediction model to obtain the predicted remaining capacity, and the retired batteries are sorted based on the predicted remaining capacity.
2. The method for sorting retired batteries according to claim 1, characterized in that: The step of obtaining discharge characteristic data of a plurality of sample batteries includes: Acquire voltage variation data of several sample batteries, wherein the voltage variation data includes first half-width peak voltage range data of the IC curve during the complete discharge process and first rebound voltage data after the battery is discharged to a preset voltage; Acquiring discharge characteristic data of the sample battery when discharged in the first half-width peak voltage range; The step of obtaining discharge characteristic data of the battery to be sorted includes: Acquire second rebound voltage data of the battery to be sorted after it is discharged to a preset voltage; determining second half-width peak voltage range data based on the second rebound voltage data according to a relationship between the first half-width peak voltage range data and the first rebound voltage data; Obtain discharge characteristic data of the battery to be sorted when discharged in the second half-width peak voltage range.
3. The method for sorting retired batteries according to claim 2, characterized in that: The voltage variation data also includes the first peak voltage data of the IC curve during the complete discharge process; The determining, according to the relationship between the first half-width peak voltage range data and the first rebound voltage data and based on the second rebound voltage data, second half-width peak voltage range data comprises: Determine a mapping relationship between a rebound voltage and a peak voltage based on the first rebound voltage data and the first peak voltage data, and determine second peak voltage data of the battery to be sorted based on the mapping relationship and the second rebound voltage data; Based on a preset clustering algorithm, the first peak voltage data is classified to obtain several clusters and their corresponding maximum half-width peak voltage range data, the maximum half-width peak voltage range data is determined by the first half-width peak voltage range data within the cluster, the second peak voltage data is assigned to a target cluster among the several clusters, and the maximum half-width peak voltage range data corresponding to the target cluster is used as the second half-width peak voltage range data.
4. The method for sorting retired batteries according to claim 3, characterized in that: The clustering algorithm adopts K-means clustering algorithm.
5. The method for sorting retired batteries according to claim 1, characterized in that: The peak characteristic data includes a peak value and a peak voltage.
6. The method for sorting retired batteries according to claim 1, characterized in that: The discharge characteristic data also includes curve similarity between discharge characteristic curves.
7. The method for sorting retired batteries according to claim 6, characterized in that: The calculation formula of the curve similarity is: In the formula, represents the curve similarity, Indicates the similarity between the two discharge characteristic curves, Indicates the difference between the two discharge characteristic curves, Represents the weight parameter used to balance the relationship between similarity and difference.
8. The method for sorting retired batteries according to claim 7, characterized in that: The similarity is calculated and obtained by using any one of a cosine similarity measurement method and a Pearson correlation coefficient measurement method, and the difference is calculated and obtained by using any one of a Euclidean distance measurement method and a Manhattan distance measurement method.
9. The method for sorting retired batteries according to claim 1, characterized in that: The capacity prediction model adopts support vector machine.
10. A retired battery sorting device, characterized in that: include: A data acquisition unit, used to acquire the remaining capacity data and discharge characteristic data of a number of sample batteries, wherein the discharge characteristic data includes the peak characteristic data of the IC curve, and is also used to acquire the discharge characteristic data of the battery to be sorted; A model training unit, taking the discharge characteristic data of the sample battery as input and the remaining capacity data as output, training the constructed initial capacity prediction model to obtain a fully trained capacity prediction model; The model prediction unit is used to input the discharge characteristic data of the battery to be sorted into the trained capacity prediction model to obtain the predicted remaining capacity, and sort the retired batteries based on the predicted remaining capacity.