Sorting method, system and equipment for retired lithium ion batteries and medium

Abstract

The invention relates to the technical field of battery recycling, and discloses a sorting method, system and equipment for retired lithium ion batteries and a medium. The sorting method comprises the steps that use data of the retired lithium ion batteries in the service period are obtained; obtaining a capacity loss curve used for representing the capacity loss rate of the retired lithium ion battery in the service period and a capacity increment curve used for representing the internal aging mechanism of the retired lithium ion battery in the service period; inputting the capacity increment curve of the decommissioned lithium ion battery and a nonlinear loss area in the capacity loss curve into the trained LSTM model, predicting to obtain a future capacity attenuation path of the decommissioned lithium ion battery, and identifying a sudden change point of the capacity attenuation rate of the decommissioned lithium ion battery from the capacity attenuation path; and according to the residual capacity of the decommissioned lithium ion battery and the abrupt change point of the capacity fading rate of the decommissioned lithium ion battery, evaluating the secondary service potential of the decommissioned lithium ion battery so as to complete the sorting of the decommissioned lithium ion battery.

Description

Technical Field

[0001] The present invention relates to the technical field of battery recycling, and in particular to a method, system, equipment and medium for sorting retired lithium-ion batteries. Background Art

[0002] Transportation is a key area of ​​carbon emissions. Carbon emissions can be reduced by promoting the electrification of the transportation sector and developing zero-emission cars, buses and trucks. At the same time, carbon emissions can also be reduced by developing electric vehicles and replacing fuel transportation with electric transportation as soon as possible. The number of electric vehicles has been growing rapidly, and lithium-ion batteries are widely used in vehicle power batteries due to their high energy density and power density, as well as low self-discharge rate. As lithium-ion batteries age, the accumulation of side reaction products inside them will reduce their performance. When the capacity of the vehicle power battery is insufficient to support the mileage requirements of electric vehicle drivers, it will withdraw from the vehicle application scenario.

[0003] Power batteries are heavy assets. If retired lithium-ion batteries are directly discarded, the high cost will be borne directly by electric vehicle users or battery swap operators. The cascade utilization of retired lithium-ion batteries can significantly reduce the cost of battery applications on the vehicle side, or improve the asset turnover rate of battery swap companies, which has an important impact on both the daily use costs of electric vehicle users and the operations of battery swap companies. In addition, a large number of retired lithium-ion batteries will also put tremendous pressure on the environment. If the cascade utilization and recycling of retired lithium-ion batteries are ignored, the scrapped electric vehicles will bring a large number of retired lithium-ion batteries every year, and the discarded batteries will cause environmental pollution. Therefore, it is urgent to recycle and reuse retired lithium-ion batteries.

[0004] As the electrification of automobiles continues to deepen, the amount of retired lithium-ion batteries is huge. Traditional retired lithium-ion battery sorting measures require complex testing of battery packs on special benches, which is time-consuming and inefficient. It cannot cope with the sorting work of large-scale battery retirement, and the low sorting efficiency of retired lithium-ion batteries will hinder the batteries from further flowing into the next-level market for cascade utilization. Summary of the invention

[0005] The object of the present invention is to provide a method, system, device and medium for sorting retired lithium-ion batteries, which can solve the problem of low efficiency in sorting retired lithium-ion batteries.

[0006] In order to solve the above technical problems, an embodiment of the present invention provides a method for sorting retired lithium-ion batteries, comprising the following steps:

[0007] Obtain usage data of retired lithium-ion batteries during their service life;

[0008] According to the usage data of retired lithium-ion batteries during service, a capacity loss curve for characterizing the capacity loss rate of retired lithium-ion batteries during service and a capacity increment curve for characterizing the internal aging mechanism of retired lithium-ion batteries during service are obtained; wherein the capacity loss curve includes a linear loss region and a nonlinear loss region;

[0009] Input the capacity increment curve and the nonlinear loss area in the capacity loss curve of the retired lithium-ion battery into the trained LSTM model to predict the future capacity decay path of the retired lithium-ion battery, and identify the mutation point of the capacity decay rate of the retired lithium-ion battery from the capacity decay path; wherein the LSTM model is trained by the nonlinear loss area in the capacity increment curve and the capacity loss curve of several different retired lithium-ion batteries;

[0010] According to the remaining capacity of retired lithium-ion batteries and the mutation point of the capacity attenuation rate of retired lithium-ion batteries, the potential for secondary service of retired lithium-ion batteries is evaluated to complete the sorting of retired lithium-ion batteries.

[0011] Optionally, the step of inputting the capacity increment curve and the nonlinear loss area in the capacity loss curve of the retired lithium-ion battery into the trained LSTM model to predict the future capacity attenuation path of the retired lithium-ion battery includes:

[0012] Perform one-dimensional linear fitting on the capacity loss curve of retired lithium-ion batteries to obtain the linear component of the linear loss region in the capacity loss curve;

[0013] Obtaining a health state of the retired lithium-ion battery according to the initial capacity and current capacity of the retired lithium-ion battery, and using the health state of the retired lithium-ion battery to subtract a component in a linear loss region in a capacity loss curve to obtain a nonlinear component in a nonlinear loss region in the capacity loss curve;

[0014] After normalizing the component of the nonlinear loss region in the capacity loss curve, multiply it by the exponential cycle life to obtain the nonlinear characteristics of the nonlinear component;

[0015] The capacity increment curve of retired lithium-ion batteries and the nonlinear characteristics of the nonlinear components are input into the trained LSTM model to predict the future capacity attenuation path of retired lithium-ion batteries.

[0016] Optionally, before inputting the capacity increment curve of the retired lithium-ion battery and the nonlinear features of the nonlinear component into the trained LSTM model, the method further includes:

[0017] The initial capacity increment curve of the retired lithium-ion battery is obtained by using the constant current interval during the discharge process of the retired lithium-ion battery;

[0018] The SG filtering method is used to filter the initial capacity increment curve to obtain the capacity increment curve for input into the LSTM model.

[0019] Optionally, identifying a mutation point of the capacity decay rate of the retired lithium-ion battery from the capacity decay path includes:

[0020] Using the capacity of the retired lithium-ion battery, a first derivative of the number of cycles of the retired lithium-ion battery is obtained to obtain a curve showing that the capacity attenuation rate of the retired lithium-ion battery varies with the number of cycles;

[0021] Using the curve of the capacity attenuation rate of the retired lithium-ion battery versus the number of cycles, a first derivative of the number of cycles of the retired lithium-ion battery is obtained to obtain a curve of the capacity attenuation rate change rate of the retired lithium-ion battery versus the number of cycles;

[0022] The mutation point of the capacity decay rate of the retired lithium-ion battery is obtained through the curve of the capacity decay rate change rate of the retired lithium-ion battery as a inverse of the cycle number.

[0023] Optionally, the evaluating the potential of the retired lithium-ion battery for secondary service according to the remaining capacity of the retired lithium-ion battery and the mutation point of the capacity attenuation rate of the retired lithium-ion battery includes:

[0024] If the remaining capacity of the retired lithium-ion battery is between 100% and 80%, and there is no mutation point in the capacity attenuation rate of the retired lithium-ion battery, the retired lithium-ion battery can still be used in the vehicle power battery scenario;

[0025] If the remaining capacity of the retired lithium-ion battery is between 80% and 60%, and there is no mutation point in the capacity attenuation rate of the retired lithium-ion battery, the retired lithium-ion battery can be used in public energy storage scenarios with clear operating conditions and a good working environment;

[0026] If the remaining capacity of the retired lithium-ion battery is between 60% and 20%, and there is no mutation point in the capacity decay rate of the retired lithium-ion battery, the retired lithium-ion battery can be used for household energy storage scenarios with low usage frequency and low energy density requirements;

[0027] If the remaining capacity of retired lithium-ion batteries is less than 20%, or if the capacity attenuation rate of retired lithium-ion batteries has passed a mutation point, the retired lithium-ion batteries will enter the raw material recycling link and will have no further service value.

[0028] Optionally, the LSTM model is obtained by training mainly with capacity increment curves of several different retired lithium-ion batteries and supplemented with nonlinear loss regions in capacity loss curves.

[0029] Optionally, the usage data of the retired lithium-ion battery during its service period includes: the service time, charging current, charging voltage, charging power and charging temperature of the retired lithium-ion battery.

[0030] An embodiment of the present invention further provides a system for sorting retired lithium-ion batteries, comprising:

[0031] A battery data acquisition module is used to obtain usage data of retired lithium-ion batteries during service;

[0032] A data feature extraction module is used to obtain a capacity loss curve for characterizing the capacity loss rate of the retired lithium-ion battery during service, and a capacity increment curve for characterizing the internal aging mechanism of the retired lithium-ion battery during service, based on the usage data of the retired lithium-ion battery during service; wherein the capacity loss curve includes a linear loss area and a nonlinear loss area;

[0033] The capacity decay prediction module is used to input the capacity increment curve of the retired lithium-ion battery and the nonlinear loss area in the capacity loss curve into the trained LSTM model, predict the future capacity decay path of the retired lithium-ion battery, and identify the mutation point of the capacity decay rate of the retired lithium-ion battery from the capacity decay path; wherein the LSTM model is trained by the nonlinear loss area in the capacity increment curve and capacity loss curve of several different retired lithium-ion batteries;

[0034] The retired battery sorting module is used to evaluate the potential of retired lithium-ion batteries for secondary service based on the remaining capacity of the retired lithium-ion batteries and the mutation point of the capacity attenuation rate of the retired lithium-ion batteries, so as to complete the sorting of retired lithium-ion batteries.

[0035] An embodiment of the present invention also provides a computer device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the above-mentioned method for sorting retired lithium-ion batteries.

[0036] An embodiment of the present invention further provides a computer-readable storage medium storing a computer program, wherein the computer program implements the above-mentioned method for sorting retired lithium-ion batteries when executed by a processor.

[0037] The method for sorting retired lithium-ion batteries provided by the present invention has at least the following beneficial effects:

[0038] Through the use data of retired lithium-ion batteries during service, the capacity loss curve and capacity increment curve of retired lithium-ion batteries are obtained. The capacity loss curve reflects the capacity loss rate of retired lithium-ion batteries during service, and the capacity increment curve reflects the internal aging mechanism of retired lithium-ion batteries during service. Among them, the nonlinear loss area in the capacity loss curve will have a great interference with the estimation of the final capacity of the battery. Therefore, in the present invention, the capacity increment curves of several different retired lithium-ion batteries and the nonlinear loss area in the capacity loss curve are combined, and the LSTM model is trained from the mechanism level of retired lithium-ion batteries and the external signal rheology, so that the LSTM model can effectively predict the capacity decay path of retired lithium-ion batteries in the future. At this time, the mutation point of the capacity decay rate of retired lithium-ion batteries is identified from the capacity decay path of retired lithium-ion batteries, and the potential for secondary service of retired lithium-ion batteries is evaluated based on the remaining capacity of retired lithium-ion batteries and the mutation point, so as to realize the sorting of retired lithium-ion batteries, which not only has high sorting efficiency, but also can effectively improve the utilization rate of retired lithium-ion batteries. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] One or more embodiments are exemplarily described by the pictures in the corresponding drawings, and these exemplary descriptions do not constitute limitations on the embodiments.

[0040] Figure 1 is a flow chart of a method for sorting retired lithium-ion batteries provided according to an embodiment of the present invention;

[0041] Figure 2 (a) is a schematic diagram of a capacity loss curve provided according to an embodiment of the present invention;

[0042] Figure 2 (b) is a schematic diagram of a UC feature provided according to an embodiment of the present invention;

[0043] Figure 3 (a) is a schematic diagram of an IC curve before and after SG filtering according to an embodiment of the present invention;

[0044] Figure 3 (b) is a schematic diagram of an IC complete curve with different cycle numbers provided according to an embodiment of the present invention;

[0045] Figure 4 is a schematic diagram comparing capacity attenuation paths of multiple batteries provided according to an embodiment of the present invention;

[0046] Figure 5 is a schematic diagram of a battery retirement point and a diving point provided according to an embodiment of the present invention;

[0047] Figure 6 is a schematic diagram of a battery sorting scheme provided according to an embodiment of the present invention;

[0048] Figure 7 A schematic diagram of a method for sorting retired lithium-ion batteries according to an embodiment of the present invention Figure 1 ;

[0049] Figure 8 A schematic diagram of a method for sorting retired lithium-ion batteries according to an embodiment of the present invention Figure 2 ;

[0050] Fig. 9 It is a schematic diagram of a capacity attenuation path provided according to an embodiment of the present invention. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical scheme and advantages of the embodiments of the present invention clearer, the embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings. However, it will be appreciated by those skilled in the art that in the embodiments of the present invention, many technical details are proposed in order to enable the reader to better understand the present invention. However, even without these technical details and various changes and modifications based on the following embodiments, the technical scheme claimed in the present invention can be implemented. The division of the following embodiments is for the convenience of description and should not constitute any limitation on the specific implementation of the present invention. The various embodiments can be combined and referenced with each other without contradiction.

[0052] Due to the lack of effective means of sorting retired lithium-ion batteries, most battery packs can only be scrapped and recycled "across the board", losing their potential value for reuse. In addition, electric vehicles have a variety of models and their battery pack structures are different, making it difficult to establish an efficient battery pack disassembly line, and the labor cost is high. Disassembling and recycling all battery packs without testing and sorting will result in low profits for comprehensive recycling companies. In addition, due to the low efficiency of battery cell recycling, most battery recycling and reuse currently only reaches the battery module level, and most battery cells face the end of being discarded, and the risk of environmental pollution always exists. And because the battery cells cannot pass the recycling process and enter the life cycle of new batteries again in the form of raw materials, the carbon cost of the battery's entire life cycle is increased.

[0053] Therefore, the present invention proposes an efficient sorting solution for retired lithium-ion batteries that does not rely on bench testing and structural disassembly, and provides a feasible option for the cascade utilization of retired lithium-ion batteries, thereby reserving a more sufficient time window for subsequent recycling and utilization, significantly reducing the processing pressure caused by centralized battery retirement, and ensuring that each batch of retired lithium-ion batteries can be processed in a timely and effective manner.

[0054] An embodiment of the present invention relates to a method for sorting retired lithium-ion batteries. The implementation details of the method for sorting retired lithium-ion batteries of this embodiment are specifically described below. The following content is only the implementation details provided for the convenience of understanding and is not necessary for implementing this solution.

[0055] The specific process of the method for sorting retired lithium-ion batteries in this embodiment can be as follows: Figure 1 As shown, including:

[0056] Step 101, obtaining usage data of retired lithium-ion batteries during service.

[0057] Specifically, first obtain the battery usage data during its service period. The main application target is the on-board lithium-ion power battery. The data during its service period needs to be provided by the data center of the car manufacturer, and the usage data of the retired lithium-ion battery during its service period mainly includes: the service time, charging current, charging voltage, charging power and charging temperature of the retired lithium-ion battery.

[0058] Step 102, based on the usage data of the retired lithium-ion battery during service, obtain a capacity loss curve for characterizing the capacity loss rate of the retired lithium-ion battery during service, and a capacity increment curve for characterizing the internal aging mechanism of the retired lithium-ion battery during service; wherein the capacity loss curve includes a linear loss region and a nonlinear loss region.

[0059] Step 103, inputting the capacity increment curve and the nonlinear loss area in the capacity loss curve of the retired lithium-ion battery into the trained LSTM model, predicting the future capacity decay path of the retired lithium-ion battery, and identifying the mutation point of the capacity decay rate of the retired lithium-ion battery from the capacity decay path; wherein the LSTM model is trained by the nonlinear loss area in the capacity increment curve and the capacity loss curve of several different retired lithium-ion batteries.

[0060] Specifically, the training process of the LSTM model is first described in detail:

[0061] Since the capacity loss curve of lithium-ion batteries during the aging process will show certain nonlinear characteristics, the algorithm trained only with battery capacity and battery service time data has insufficient accuracy in predicting the final stage of the battery capacity loss curve. Therefore, in this embodiment, based on the original data, two artificial features are constructed using the aging mechanism and mathematical statistics mechanism inside the battery, namely, the battery capacity incremental feature (Incremental Capacity, IC) and the nonlinear correction feature (Nonlinear Correction, UC), and then these two features are combined for model training.

[0062] Regarding the UC characteristics, that is, the characteristics of the nonlinear loss area in the capacity loss curve:

[0063] For lithium-ion batteries, the capacity loss curve shows two obvious stages: Figure 2 As shown in (a), the smooth curve in the figure shows the capacity of the battery during each discharge cycle, and the curve with dots shows the capacity loss of each cycle. The capacity loss curve is divided into two stages. The first stage is the linear degradation zone, in which the capacity of the battery decays approximately linearly, and the capacity loss rate remains at a relatively gentle and stable level. The second stage shows an accelerated decrease in capacity, which is a nonlinear decay zone, and the capacity loss rate begins to rise. This type of nonlinear decay at the end of battery aging will greatly interfere with the estimation of the battery's final capacity. Therefore, it is particularly important to identify the nonlinear loss zone in the capacity loss curve and introduce it into the neural network as an independent feature after separation to identify the future capacity decay path of the battery capacity.

[0064] In a specific implementation, when the nonlinear loss region in the capacity loss curve is used for model training, the capacity loss curve of the retired lithium-ion battery is firstly subjected to a one-dimensional linear fitting to obtain the linear component of the linear loss region in the capacity loss curve, and then the health status of the retired lithium-ion battery is obtained according to the initial capacity and current capacity of the retired lithium-ion battery, as shown in the following formula:

[0065]

[0066] Where SOH(t) represents the health status of retired lithium-ion batteries at cycle number t, and the SOH curve of the entire life cycle (from 1 to 0.8) can be decomposed into linear and nonlinear components.

[0067] Therefore, by subtracting the component of the linear loss area in the capacity loss curve from the health status of the retired lithium-ion battery, the nonlinear component in the nonlinear loss area of ​​the capacity loss curve can be obtained, that is, the nonlinear component in the SOH curve of the entire life cycle can be obtained by calculating the residual, as shown in the following formula:

[0068]

[0069] Next, the component of the nonlinear loss region in the capacity loss curve is normalized and multiplied by the exponential cycle life to obtain the nonlinear characteristics of the nonlinear component, as shown in the following formula:

[0070]

[0071] Where R(t) represents the percentage of the remaining number of cycles to the end of life (EOL) of the current cycle, ranging from 1 to 0. In order to maintain the same range as the battery SOH to obtain accurate predictions, UCexp (t) is mapped to the interval [0.8, 1]. In geometric theory, UC exp (t) gets its maximum value at the inflection point, by changing UC exp (t) is set to 1 to obtain the UC(t) curve, such as Figure 2 (b) as shown.

[0072] but, Figure 2 (b) is the characteristic of the nonlinear loss region in the capacity loss curve described in this embodiment (ie, the nonlinear characteristic of the nonlinear component).

[0073] Regarding the IC characteristics, that is, the characteristics of the capacity increment curve of retired lithium-ion batteries:

[0074] From a statistical point of view, UC gives the probability of nonlinear attenuation through the change of battery aging rate, but ignores the internal aging mechanism of the battery. It is difficult to combine the mechanism level with the external signal rheology of the battery to predict the capacity change trend. Therefore, UC is not suitable as a single feature training network. IC is a method widely used to describe the battery aging process, which analyzes the battery aging mechanism from the electrode level. The IC curve represents the rate of change of capacity with voltage evolution during constant current charging / charging and discharging. In this embodiment, the IC curve is used to describe the attenuation process of internal materials, which is used to combine with the nonlinear characteristics of the nonlinear component for model training.

[0075] However, in a low temperature environment, the constant current process of fast charging is very short, and constant voltage charging will be quickly entered. Therefore, this embodiment first uses the constant current interval during the discharge process of the retired lithium-ion battery to obtain the initial capacity increment curve of the retired lithium-ion battery, as shown in the following formula:

[0076]

[0077] Where Q represents the battery capacity at time t, V represents the voltage, and t represents the sampling time.

[0078] In actual operation, it is difficult to obtain the peak value of IC because the parameter value is easily disturbed by measurement noise in the charging platform area. In order to solve this problem, the SG filtering method is used to filter the initial capacity increment curve to obtain the capacity increment IC curve for model training.

[0079] like Figure 3 As shown, Figure 3 (a) shows the IC curves before and after SG filtering. It can be seen that the IC curve after filtering is smooth and can be used to extract aging features. The complete IC curves with different cycle numbers are shown in Figure 3(b) shows that the IC curves of the battery at different cycles have similar shapes, and the peaks in the curves have unique heights and positions, reflecting the changes in the electrode materials during the battery charging and discharging process.

[0080] The characteristics of the nonlinear loss region in the capacity loss curves and the characteristics of the capacity increment curves of several different retired lithium-ion batteries are used to train a long short-term memory network (LSTM) to obtain the final LSTM model.

[0081] The following describes the training process of the LSTM model:

[0082] First, in the training process of the LSTM model, the capacity increment curves of several different retired lithium-ion batteries are used as the main training, and the nonlinear loss area in the capacity loss curve is used as the auxiliary training.

[0083] However, when the capacity performance of a vehicle-mounted power battery decays to 80% of its initial capacity, its driving range will be difficult to meet the driver's needs; when it decays to 70% of its initial capacity, it will exit the vehicle application scenario. In order to simulate the actual application scenario, this embodiment divides the battery aging data obtained in the experiment into a training set and a validation set in a ratio of 3:7, and only uses the first 30% of the capacity decay data to train the model, and verifies the prediction effect of the capacity decay path of the later model.

[0084] In addition, since the data of lithium-ion batteries are collected during the charging and discharging cycles, they belong to time series data. LSTM is widely used to process time series data, using the LSTM recursive unit structure to replace the state unit of the classic RNN. By introducing the input gate, forget gate and output gate mechanism, LSTM can effectively control the flow of information and retain important historical information, thereby improving the performance of the model.

[0085] At the same time, this embodiment introduces the attention mechanism, which is a machine learning technology that can help the neural network pay more attention to important features, thereby improving the performance and generalization ability of the model. In a neural network, each feature has a different impact on the result, but usually only one set of features plays a dominant role in the output. The attention mechanism learns according to the attention level of individual features in the sequence and integrates the features according to the attention level. Even if there is only one feature input to the LSTM network, when processing time series data, the attention mechanism can still capture the key information in the time series data better by weighting different time steps.

[0086] After the LSTM model is trained in the above manner, the capacity increment curve of the retired lithium-ion batteries to be sorted and the nonlinear loss area in the capacity loss curve are input into the trained LSTM model, and the future capacity attenuation path of the retired lithium-ion batteries can be predicted.

[0087] like Figure 4 As shown, a graph showing the change of the available capacity of a retired lithium-ion battery in a single cycle with the number of cycles obtained on the experimental bench (i.e., a capacity decay path graph) is shown. In the experiment, each battery was operated at minus 10°C, and constant current charging was performed using currents of 20A (#1 battery and #2 battery), 15A (#5 battery and #6 battery), and 10A (#9 battery and #10 battery) respectively, until the voltage reached 4.2V and switched to constant voltage charging. It can be observed from the figure that the capacity decay paths of #1 battery and #2 battery are close to a straight line, which is a typical linear capacity decay; while the capacity of the remaining 4 batteries all showed accelerated decay. For example, after the 27th cycle, the capacity decay rate of battery #1 increased significantly, resulting in a rapid decrease in the available capacity of the battery. In addition, the number of cycles that the #1 battery experienced to reach 80% of the initial capacity is also much smaller than that of the linearly decaying battery, and the subsequent capacity decay will still maintain a high rate, so it is not suitable for cascade utilization, but it is recommended to directly scrap and recycle key materials.

[0088] In other words, accurately predicting the battery capacity decay path and identifying the mutation point of the battery capacity decay rate, that is, the battery performance diving point, are the key to pre-selecting and sorting retired batteries for cascade utilization.

[0089] Among them, when identifying the mutation point of the capacity decay rate of the retired lithium-ion battery from the capacity decay path, firstly, the capacity of the retired lithium-ion battery is used to find a derivative of the number of cycles (i.e., the working cycle) of the retired lithium-ion battery to obtain a curve of the capacity decay rate of the retired lithium-ion battery varying with the number of cycles. Then, the curve of the capacity decay rate of the retired lithium-ion battery varying with the number of cycles is used to find a derivative of the number of cycles of the retired lithium-ion battery to obtain a curve of the capacity decay rate change rate (i.e., acceleration) of the retired lithium-ion battery varying with the number of cycles. Finally, the mutation point of the capacity decay rate of the retired lithium-ion battery is obtained through the curve of the capacity decay rate change rate of the retired lithium-ion battery varying with the number of cycles.

[0090] Because the rate of change of the capacity decay rate is near 0 before reaching the mutation point of the capacity decay rate, the rate of change of the capacity decay rate increases after reaching the mutation point of the capacity decay rate, and then returns to near 0. This process reflects that the battery capacity decay rate is: the decay rate is basically constant before the performance diving point, and the decay rate increases after the performance diving point, and gradually stabilizes at a higher rate. Therefore, this embodiment uses the second derivative of the capacity to the number of cycles (duty cycle), that is, the rate of change of the capacity decay rate as an indicator for finding the mutation point of the capacity decay rate of retired lithium-ion batteries.

[0091] Step 104, based on the remaining capacity of the retired lithium-ion batteries and the mutation point of the capacity decay rate of the retired lithium-ion batteries, the potential of the retired lithium-ion batteries for secondary service is evaluated to complete the sorting of the retired lithium-ion batteries.

[0092] Specifically, if the remaining capacity of retired lithium-ion batteries is between 100% and 80%, and there has not been a mutation point in the capacity decay rate of retired lithium-ion batteries, then the retired lithium-ion batteries can still be used in vehicle power battery scenarios; if the remaining capacity of retired lithium-ion batteries is between 80% and 60%, and there has not been a mutation point in the capacity decay rate of retired lithium-ion batteries, then the retired lithium-ion batteries can be used in public energy storage scenarios with clear operating conditions and good working environment; if the remaining capacity of retired lithium-ion batteries is between 60% and 20%, and there has not been a mutation point in the capacity decay rate of retired lithium-ion batteries, then the retired lithium-ion batteries can be used in household energy storage scenarios with low usage frequency and low energy density requirements; if the remaining capacity of retired lithium-ion batteries is less than 20%, or has passed the mutation point in the capacity decay rate of service lithium-ion batteries, then the retired lithium-ion batteries enter the raw material recycling link and have no value for re-service.

[0093] In the specific implementation, Figure 5 As shown in the figure, by analyzing the retirement point of the power battery (the standard is usually 70% of the initial capacity or 80% of the initial capacity) and the mutation point of the capacity decay rate of the retired lithium-ion battery, the relative position of the mutation point and the retirement point can be obtained to evaluate the potential of the battery for secondary service and realize pre-selection and sorting. The sorting scheme is as follows: Figure 6 shown.

[0094] In this embodiment, the capacity loss curve and capacity increment curve of the retired lithium-ion battery are obtained through the use data of the retired lithium-ion battery during service. The capacity loss curve reflects the capacity loss rate of the retired lithium-ion battery during service, and the capacity increment curve reflects the internal aging mechanism of the retired lithium-ion battery during service. Among them, the nonlinear loss area in the capacity loss curve will have a great interference with the estimation of the final capacity of the battery. Therefore, in the present invention, the capacity increment curves of several different retired lithium-ion batteries and the nonlinear loss area in the capacity loss curve are combined, and the LSTM model is trained from the mechanism level of the retired lithium-ion battery and the external signal rheology, so that the LSTM model can effectively predict the capacity decay path of the retired lithium-ion battery in the future. At this time, the mutation point of the capacity decay rate of the retired lithium-ion battery is identified from the capacity decay path of the retired lithium-ion battery, and the potential for the secondary service of the retired lithium-ion battery is evaluated based on the remaining capacity of the retired lithium-ion battery and the mutation point, so as to realize the sorting of the retired lithium-ion battery, which not only has high sorting efficiency, but also can effectively improve the utilization rate of the retired lithium-ion battery.

[0095] In one embodiment, the implementation process of the method for sorting retired lithium-ion batteries of the present invention is as follows: Figure 7 and Figure 8 As shown, the method includes the following steps: (1) obtaining usage data of the battery during its service life; (2) constructing artificial features; (3) processing the data set as model input; (4) neural network model prediction; (5) obtaining battery life attenuation path analysis and cascade utilization recommendations.

[0096] (1) Obtaining battery usage data during service

[0097] The data during the battery service period is used to process the input training set of the subsequent model, which should mainly include: battery service time, battery charging current, battery charging voltage, battery charging power, and battery charging temperature. The above data is collected by the on-board battery management system of new energy vehicles and transmitted to the car company or the new energy big data center. The data in this step needs to be collected by the user. The system proposed by the present invention does not contain hardware facilities for collecting data during the battery service period.

[0098] (2) Constructing artificial features

[0099] According to the formula in the detailed technical solution of the invention, the battery incremental capacity characteristics (IC) and nonlinear correction characteristics (UC) are processed. This step is implemented in the processor module of the system of the present invention, and the processing results are stored in the system storage unit.

[0100] (3) Processing the data set as model input

[0101] Based on the processed artificial features, the peak value and peak position of the IC feature of each battery working cycle, as well as the probability of rapid battery capacity decline given by the UC feature of each battery working cycle are extracted. The two are organized into a data set for inputting the model, that is, a data set of IC peak value, IC peak position, and UC probability of the battery working cycle. This step is completed by the interaction between the storage unit and the processor of the system of the present invention.

[0102] (4) Neural network model prediction

[0103] By transferring the input data set into the neural network model, the battery capacity decay curve can be predicted, such as Fig. 9 As shown. The circle in the figure is the actual capacity collected by the battery management system. Among them, EOL = 0.8 means that the battery is retired from the vehicle application scenario when it reaches 80% of the initial capacity. In this case, the service data of the first 20% of the battery life cycle is used to input the neural network model to obtain the attenuation curve from 80% to 60% capacity, as shown by the solid line in the figure below. It can be found that the neural network model accurately predicts the capacity attenuation trajectory of the battery (data with a battery capacity lower than 80% is not required in actual applications, and it is used here only to demonstrate the accuracy).

[0104] (5) Battery life attenuation path analysis and cascade utilization recommendations

[0105] From the analysis of the battery life capacity decay results given by the neural network model, it can be learned that when the battery in this case reaches 60% of the initial capacity, it is in linear decay and there is no performance drop point. According to the rapid sorting standard for cascade utilization in the detailed technical solution of the invention, the system proposed by the present invention recommends that the battery be used as a public energy storage battery in power stations and other places.

[0106] The method for sorting retired lithium-ion batteries of this embodiment has the following beneficial effects:

[0107] (1) Rapid sorting improves corporate profits

[0108] Currently, there is a lack of systems on the market that support fast and accurate sorting, which results in low efficiency of existing cascade utilization methods and inability to achieve effective profitability. The lithium-ion battery rapid sorting technology and system that considers performance drop points accurately analyzes the spatiotemporal relationship between battery performance drop points and retirement points through built-in model algorithms, and efficiently utilizes retired batteries in cascades. This significantly reduces the processing pressure caused by centralized battery retirement, ensures that each batch of batteries can be processed in a timely and effective manner, and improves profitability.

[0109] (2) Efficient and cascaded utilization helps environmental protection

[0110] The cascade utilization of retired batteries can extend the service life of batteries, reduce the demand for lithium-ion batteries, and reduce carbon emissions in the production process. At the same time, the cascade utilization of battery energy storage can smooth the peak and fill the valley for new energy power generation, reducing carbon emissions from fossil fuels. The recycling of scrapped battery metal materials can significantly reduce carbon emissions during the mining, smelting and transportation of raw materials, while avoiding environmental pollution by harmful materials such as heavy metals and promoting the reuse of precious metals. According to the inventor's estimate, the system can help comprehensive recycling companies save nearly 45% of material and labor costs, and help batteries reduce carbon emissions by about 40% over their entire life cycle.

[0111] The step division of the various methods above is only for clear description. When implemented, they can be combined into one step or some steps can be split and decomposed into multiple steps. As long as they include the same logical relationship, they are all within the protection scope of the present invention. Adding insignificant modifications or introducing insignificant designs to the algorithm or process without changing the core design of the algorithm and process are all within the protection scope of the invention.

[0112] Another embodiment of the present invention relates to a sorting system for retired lithium-ion batteries. The implementation details of the sorting system for retired lithium-ion batteries of this embodiment are described in detail below. The following content is only for the convenience of understanding the implementation details provided, and is not necessary for the implementation of this solution. The sorting system for retired lithium-ion batteries of this embodiment includes:

[0113] A battery data acquisition module is used to obtain usage data of retired lithium-ion batteries during service;

[0114] A data feature extraction module is used to obtain a capacity loss curve for characterizing the capacity loss rate of the retired lithium-ion battery during service, and a capacity increment curve for characterizing the internal aging mechanism of the retired lithium-ion battery during service, based on the usage data of the retired lithium-ion battery during service; wherein the capacity loss curve includes a linear loss area and a nonlinear loss area;

[0115] The capacity decay prediction module is used to input the capacity increment curve of the retired lithium-ion battery and the nonlinear loss area in the capacity loss curve into the trained LSTM model, predict the future capacity decay path of the retired lithium-ion battery, and identify the mutation point of the capacity decay rate of the retired lithium-ion battery from the capacity decay path; wherein the LSTM model is trained by the nonlinear loss area in the capacity increment curve and capacity loss curve of several different retired lithium-ion batteries;

[0116] The retired battery sorting module is used to evaluate the potential of retired lithium-ion batteries for secondary service based on the remaining capacity of the retired lithium-ion batteries and the mutation point of the capacity attenuation rate of the retired lithium-ion batteries, so as to complete the sorting of retired lithium-ion batteries.

[0117] It is not difficult to find that this embodiment is a system embodiment corresponding to the above method embodiment, and this embodiment can be implemented in conjunction with the above method embodiment. The relevant technical details and technical effects mentioned in the above embodiment are still valid in this embodiment, and in order to reduce repetition, they are not repeated here. Accordingly, the relevant technical details mentioned in this embodiment can also be applied in the above embodiment.

[0118] It is worth mentioning that all modules involved in this embodiment are logic modules. In practical applications, a logic unit can be a physical unit, a part of a physical unit, or a combination of multiple physical units. In addition, in order to highlight the innovative part of the present invention, this embodiment does not introduce units that are not closely related to solving the technical problem proposed by the present invention, but this does not mean that there are no other units in this embodiment.

[0119] Another embodiment of the present invention relates to a computer device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method for sorting retired lithium-ion batteries in the above-mentioned embodiments.

[0120] Among them, the memory and the processor are connected in a bus manner, and the bus may include any number of interconnected buses and bridges, and the bus connects various circuits of one or more processors and memories together. The bus can also connect various other circuits such as peripherals, voltage regulators, and power management circuits, which are well known in the art and are therefore not further described herein. The bus interface provides an interface between the bus and the transceiver. The transceiver can be one element or multiple elements, such as multiple receivers and transmitters, providing a unit for communicating with various other devices on a transmission medium. The data processed by the processor is transmitted on a wireless medium via an antenna, and further, the antenna also receives data and transmits the data to the processor.

[0121] The processor is responsible for managing the bus and general processing, and can also provide various functions, including timing, peripheral interfaces, voltage regulation, power management, and other control functions. Memory can be used to store data used by the processor when performing operations.

[0122] Another embodiment of the present invention relates to a computer-readable storage medium storing a computer program, which implements the above method embodiment when executed by a processor.

[0123] That is, those skilled in the art can understand that all or part of the steps in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a program, and the program is stored in a storage medium, including a number of instructions for a device (which can be a single-chip microcomputer, chip, etc.) or a processor (processor) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as: ROM), random access memory (Random Access Memory, referred to as: RAM), disk or optical disk and other media that can store program codes.

[0124] Those skilled in the art will appreciate that the above embodiments are specific embodiments for implementing the present invention, and in actual applications, various changes may be made thereto in form and detail without departing from the spirit and scope of the present invention.

Claims

1. A method for sorting retired lithium-ion batteries, characterized in that: include: Obtain usage data of retired lithium-ion batteries during their service life; According to the usage data of retired lithium-ion batteries during service, a capacity loss curve for characterizing the capacity loss rate of retired lithium-ion batteries during service and a capacity increment curve for characterizing the internal aging mechanism of retired lithium-ion batteries during service are obtained; wherein the capacity loss curve includes a linear loss region and a nonlinear loss region; Input the capacity increment curve and the nonlinear loss area in the capacity loss curve of the retired lithium-ion battery into the trained LSTM model to predict the future capacity decay path of the retired lithium-ion battery, and identify the mutation point of the capacity decay rate of the retired lithium-ion battery from the capacity decay path; wherein the LSTM model is trained by the nonlinear loss area in the capacity increment curve and the capacity loss curve of several different retired lithium-ion batteries; According to the remaining capacity of retired lithium-ion batteries and the mutation point of the capacity attenuation rate of retired lithium-ion batteries, the potential for secondary service of retired lithium-ion batteries is evaluated to complete the sorting of retired lithium-ion batteries.

2. The method for sorting retired lithium-ion batteries according to claim 1, characterized in that: The capacity increment curve of the retired lithium-ion battery and the nonlinear loss area in the capacity loss curve are input into the trained LSTM model to predict the future capacity attenuation path of the retired lithium-ion battery, including: Perform one-dimensional linear fitting on the capacity loss curve of retired lithium-ion batteries to obtain the linear component of the linear loss region in the capacity loss curve; Obtaining a health state of the retired lithium-ion battery according to the initial capacity and current capacity of the retired lithium-ion battery, and using the health state of the retired lithium-ion battery to subtract a component in a linear loss region in a capacity loss curve to obtain a nonlinear component in a nonlinear loss region in the capacity loss curve; After normalizing the component of the nonlinear loss region in the capacity loss curve, multiply it by the exponential cycle life to obtain the nonlinear characteristics of the nonlinear component; The capacity increment curve of retired lithium-ion batteries and the nonlinear characteristics of the nonlinear components are input into the trained LSTM model to predict the future capacity attenuation path of retired lithium-ion batteries.

3. The method for sorting retired lithium-ion batteries according to claim 2, characterized in that: Before inputting the capacity increment curve of the retired lithium-ion battery and the nonlinear features of the nonlinear component into the trained LSTM model, the method further includes: The initial capacity increment curve of the retired lithium-ion battery is obtained by using the constant current interval during the discharge process of the retired lithium-ion battery; The SG filtering method is used to filter the initial capacity increment curve to obtain the capacity increment curve for input into the LSTM model.

4. The method for sorting retired lithium-ion batteries according to claim 1, characterized in that: The method of identifying a mutation point of the capacity decay rate of a retired lithium-ion battery from a capacity decay path includes: Using the capacity of the retired lithium-ion battery, a first derivative of the number of cycles of the retired lithium-ion battery is obtained to obtain a curve showing that the capacity attenuation rate of the retired lithium-ion battery varies with the number of cycles; Using the curve of the capacity attenuation rate of the retired lithium-ion battery versus the number of cycles, a first derivative of the number of cycles of the retired lithium-ion battery is obtained to obtain a curve of the capacity attenuation rate change rate of the retired lithium-ion battery versus the number of cycles; The mutation point of the capacity decay rate of the retired lithium-ion battery is obtained through the curve of the capacity decay rate change rate of the retired lithium-ion battery as a inverse of the cycle number.

5. The method for sorting retired lithium-ion batteries according to claim 1, characterized in that: The method of evaluating the potential of the retired lithium-ion battery for secondary service based on the remaining capacity of the retired lithium-ion battery and the mutation point of the capacity attenuation rate of the retired lithium-ion battery includes: If the remaining capacity of the retired lithium-ion battery is between 100% and 80%, and there is no mutation point in the capacity attenuation rate of the retired lithium-ion battery, the retired lithium-ion battery can still be used in the vehicle power battery scenario; If the remaining capacity of the retired lithium-ion battery is between 80% and 60%, and there is no mutation point in the capacity attenuation rate of the retired lithium-ion battery, the retired lithium-ion battery can be used in public energy storage scenarios with clear operating conditions and a good working environment; If the remaining capacity of the retired lithium-ion battery is between 60% and 20%, and there is no mutation point in the capacity decay rate of the retired lithium-ion battery, the retired lithium-ion battery can be used for household energy storage scenarios with low usage frequency and low energy density requirements; If the remaining capacity of retired lithium-ion batteries is less than 20%, or if the capacity attenuation rate of retired lithium-ion batteries has passed a mutation point, the retired lithium-ion batteries will enter the raw material recycling link and will have no further service value.

6. The method for sorting retired lithium-ion batteries according to claim 1, characterized in that: The LSTM model is obtained by training mainly with the capacity increment curves of several different retired lithium-ion batteries and supplemented with the nonlinear loss area in the capacity loss curve.

7. The method for sorting retired lithium-ion batteries according to any one of claims 1 to 6, characterized in that: The usage data of the retired lithium-ion battery during its service period includes: the service time, charging current, charging voltage, charging capacity and charging temperature of the retired lithium-ion battery.

8. A system for sorting retired lithium-ion batteries, characterized in that: include: A battery data acquisition module is used to obtain usage data of retired lithium-ion batteries during service; A data feature extraction module is used to obtain a capacity loss curve for characterizing the capacity loss rate of the retired lithium-ion battery during service, and a capacity increment curve for characterizing the internal aging mechanism of the retired lithium-ion battery during service, based on the usage data of the retired lithium-ion battery during service; wherein the capacity loss curve includes a linear loss area and a nonlinear loss area; The capacity decay prediction module is used to input the capacity increment curve of the retired lithium-ion battery and the nonlinear loss area in the capacity loss curve into the trained LSTM model, predict the future capacity decay path of the retired lithium-ion battery, and identify the mutation point of the capacity decay rate of the retired lithium-ion battery from the capacity decay path; wherein the LSTM model is trained by the nonlinear loss area in the capacity increment curve and capacity loss curve of several different retired lithium-ion batteries; The retired battery sorting module is used to evaluate the potential of retired lithium-ion batteries for secondary service based on the remaining capacity of the retired lithium-ion batteries and the mutation point of the capacity attenuation rate of the retired lithium-ion batteries, so as to complete the sorting of retired lithium-ion batteries.

9. A computer device, characterized in that: include: at least one processor; And, a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method for sorting retired lithium-ion batteries as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method for sorting retired lithium-ion batteries according to any one of claims 1 to 7 is implemented.

Images