Method and system for rapidly detecting residual capacity of retired battery

Abstract

The method obtains the historical estimated value of the current capacity of the decommissioned battery by obtaining the historical operation data of the decommissioned battery and inputting the historical operation data into the neural network algorithm, obtains the existing estimated value of the current capacity of the decommissioned battery by combining the characteristic frequency impedance, and comprehensively considers the external parameter change of the battery in the actual working condition. According to the present invention, the characteristic screening is performed on the historical operation data and the characteristic frequency impedance of the battery according to the change of the internal structure and the material during the battery aging process, and the historical pre-estimated value and the existing pre-estimated value are fused to obtain the residual capacity of the retired battery, such that the accurate identification of the retired lithium battery residual energy detection can be improved; meanwhile, subsequent management of a new battery pack is facilitated, errors are reduced, and the accuracy of final model estimation is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of lithium battery detection, and in particular to a method and system for detecting residual energy of retired lithium batteries. Background Art

[0002] With the popularity of new energy vehicles, early automotive lithium batteries have begun to be eliminated in large numbers. The capacity of these eliminated lithium batteries is less than 80% of their initial capacity, but they can still be used in various energy storage fields. To achieve the cascade utilization of lithium batteries and regroup them, it is necessary to test the residual energy of lithium batteries to ensure that the reorganized new battery pack has a high consistency and prolong its service life.

[0003] A method for adaptively estimating the health state of a lithium-ion battery with publication number CN117129872A includes conducting a battery aging experiment, and performing capacity calibration at intervals of a certain number of charge and discharge cycles, while collecting electrochemical impedance spectra under different charge states; performing time scale identification on the electrochemical impedance spectrum and extracting electrochemical impedance time scale features; selecting electrochemical impedance time scale features that are highly correlated with the battery health state based on the Pearson correlation coefficient to form a decay feature data set; based on the decay feature data set, offline training a health state estimation model based on electrochemical impedance time scale features and integrated learning; in actual applications, obtaining the electrochemical impedance spectrum of the battery at the current charge state and extracting electrochemical impedance time scale features of the same type; inputting the electrochemical impedance time scale features into the trained health state estimation model to obtain the battery health state estimation result.

[0004] Publication No. CN110501651A discloses a method for detecting the capacity of retired batteries, including the following steps: obtaining the historical operating data of retired batteries and obtaining an estimated value of the current capacity of retired batteries based on the data and a neural network algorithm. At the same time, the present invention provides a device for detecting the capacity of retired batteries, including a module for obtaining the historical operating data of retired batteries, a module for calculating the estimated value of the current capacity of retired batteries, and a result output module. The present invention obtains an estimated value of the current capacity of retired batteries by obtaining the historical operating data of retired batteries and inputting the data into a neural network algorithm. It only requires the reading time of its recent historical data and the calculation time of the neural network algorithm, which greatly improves the detection efficiency of the method for detecting the capacity of retired batteries and reduces the detection cost.

[0005] Usually, when testing the battery capacity, a full charge and discharge test is generally used, that is, the battery is charged to the highest cut-off voltage, and then discharged to the lowest voltage in a constant current manner, and the total capacity of the current battery is determined based on the amount of electricity during the discharge process. However, in actual applications, retired batteries still have a large capacity, and it takes a lot of time to charge and discharge. Especially when more batteries need to be tested, it will take a long time to detect the battery capacity, affecting the efficiency of battery capacity detection. At the same time, the battery life is lost during the charging and discharging process, wasting manpower and material resources. In addition, when the existing detection method uses the above method for detection, a single parameter is usually used for detection, which is difficult to comprehensively reflect the performance of lithium batteries in actual working conditions, as well as changes in the internal structure and materials of the battery, reducing the accuracy of the residual energy detection of retired lithium batteries. Summary of the invention

[0006] In view of this, the present invention aims to propose a retired battery capacity detection method to improve the detection efficiency of retired batteries and reduce the battery life loss as much as possible.

[0007] To achieve the above object, the technical solution of the present invention is achieved as follows:

[0008] A method for quickly detecting the remaining capacity of a retired battery, comprising:

[0009] S1, obtaining historical operation data of retired batteries, wherein the historical operation data of retired batteries includes operation data of the retired batteries over their entire life cycle;

[0010] S2, identifying and correcting the historical operation data set of the retired battery to obtain an optimized historical operation data set of the retired battery;

[0011] S3, using the Spearman correlation coefficient method to perform feature screening on the optimized data set and establish a historical operation data feature database;

[0012] S4, performing an electrochemical impedance spectroscopy test on the retired battery to obtain an electrochemical impedance spectroscopy data set, wherein the electrochemical impedance spectroscopy test includes an electrochemical impedance spectroscopy test at different SOCs and an electrochemical impedance spectroscopy test at different open circuit voltages;

[0013] S5, identifying and correcting the obtained electrochemical impedance spectroscopy data set of retired batteries to obtain an optimized electrochemical impedance spectroscopy data set of retired batteries,

[0014] S6, using the Spearman correlation coefficient method to perform feature screening on the discharge data and electrochemical impedance spectroscopy data of retired batteries, and establish an impedance feature database;

[0015] S7, input the data of the historical operation data feature database into the neural network algorithm for training, input the impedance feature data into the multivariate nonlinear regression estimation model for training, use the neural network algorithm and the multivariate nonlinear regression estimation model to respectively extract the features and predict the remaining capacity of the battery, and then merge the prediction results of the two to obtain the final prediction result of the remaining capacity of the retired battery.

[0016] Furthermore, in S2 and S5, the data set is identified and corrected, including: using the DBSCAN algorithm to identify abnormal data.

[0017] Furthermore, the specific steps of using the DBSCAN algorithm to identify abnormal data are as follows: input a data set, randomly select an unvisited data point p from the data set to start clustering, use p as the access point, and start searching whether the Eps neighborhood of p contains Minpts data points. If so, p is marked as a core point, and a new cluster M is created for the core point p, and all points in the Eps neighborhood are added to the candidate set N; for all objects q that have not been selected in the candidate set N, check whether their Eps neighborhood contains at least Minpts data points. If so, these points are added to N. If q in the candidate set N is not added to any cluster, q is added to the new cluster M, and the unprocessed data points in the candidate set N are continued to be checked. When all points in the candidate set N have been processed, stop, and repeat the above steps until all points have been visited. At this time, any points that have not been added to any cluster are classified as abnormal data.

[0018] Furthermore, the operating data of the retired battery throughout its life cycle includes discharge curve data, discharge temperature, charge curve data, overcharge times, and overdischarge times.

[0019] Furthermore, the Spearman correlation coefficient calculation formula is:

[0020]

[0021] Where ρ is the Spearman correlation coefficient between sample X and sample Y, n is the total amount of correlation analysis data, X i is the i-th data of the sample, is the mean of sample X, Y i is the i-th data of the sample, is the mean of sample Y.

[0022] The steps of the further neural network algorithm are: a1, initializing connection weights and node thresholds; a2, inputting historical battery operation data; a3, taking a set of operation data as samples; a4, calculating hidden layer node outputs;

[0023] a5. Calculate the output of the output layer; a6. Update the connection weights and node thresholds; a7. Determine whether all data is included; if so, proceed to the next step; if not, return to step a2; a8. Determine whether the error is less than the lower limit of 3% or whether the number of learning times has reached the upper limit of 50,000 times; if so, proceed to the next step; if not, return to step a4; a9. Obtain the estimated value of the current capacity of the retired battery.

[0024] Furthermore, a multivariate non-linear regression is used to construct an estimation model. The real part impedance and imaginary part impedance of each characteristic frequency in the electrochemical impedance test are input, and the output is the remaining capacity of the retired lithium battery to be measured. The expression of the estimation model is: C = f(Z1, Z2); where Z1 is the real part impedance of each characteristic frequency in the electrochemical impedance test; Z1 is the imaginary part impedance of each characteristic frequency in the electrochemical impedance test; C is the remaining capacity of the retired lithium battery to be measured.

[0025] Furthermore, in S7, the prediction results are fused to obtain the final prediction result of the remaining capacity of the retired battery, including: using the weighted combination prediction method to calculate the weight coefficients of the two prediction results respectively, and according to the calculated weights, output the final prediction result of the remaining capacity of the retired battery; y3 = Ay1 + By2, where y1 and y2 respectively represent the results predicted based on historical operation data and the results predicted based on the impedance feature database, y3 is the final prediction result, and A and B respectively represent the weight coefficients of y1 and y2, and A + B = 1.

[0026] Furthermore, when y1 > y2, A is 0.4 and B is 0.6; when y1 < y2, A is 0.7 and B is 0.3; when y1 = y2, A is 0.5 and B is 0.5.

[0027] The present invention also provides a rapid detection system for the remaining capacity of retired batteries, characterized in that

[0028] including:

[0029] A data acquisition module for acquiring the historical operation data set and electrochemical impedance spectrum data set of the retired battery;

[0030] A data processing module for identifying and correcting the historical operation data set and electrochemical impedance spectrum data set to obtain an optimized data set;

[0031] A feature selection module for quantitatively analyzing the data using the Spearman correlation coefficient method and screening the data set with strong correlation;

[0032] A model construction module for constructing an estimation model based on the neural network algorithm and multivariate non-linear regression to construct a combined prediction model for the remaining capacity of the battery;

[0033] The prediction module is used to take a data set with strong correlation as the model input, extract and predict the model features, and then fuse the prediction results to obtain the battery remaining capacity prediction result.

[0034] Compared with the prior art, the method and system for detecting the remaining energy of retired lithium batteries of the present invention have the following beneficial effects:

[0035] (1) A combined prediction model based on historical operation data and electrochemical impedance spectroscopy data is used to predict the remaining capacity of the battery. The prediction results are then integrated through a weighted combination prediction method to complete the final remaining capacity prediction of the battery, which can fully improve the accuracy of battery capacity prediction.

[0036] (2) The present invention obtains the historical operating data of retired batteries and inputs them into the neural network algorithm to obtain the historical estimated value of the current capacity of the retired batteries. At the same time, by combining the characteristic frequency impedance, the existing estimated value of the current capacity of the retired batteries is obtained. The external parameter changes of the battery in actual working conditions and the changes of its internal structure and materials during the battery aging process are comprehensively considered, and the historical operating data and characteristic frequency impedance of the battery are feature screened. The historical estimated value and the existing estimated value are integrated to obtain the remaining capacity of the retired battery, which can improve the accurate identification of the residual energy detection of retired lithium batteries and facilitate the subsequent management of new battery packs.

[0037] (3) During the actual use of lithium batteries, it is usually necessary to maintain a SOC of 20%-80% to prevent irreversible loss of lithium batteries during actual use. When testing retired batteries at different SOCs, there will be a certain amount of false electricity during deep discharge (below 20% SOC) and overcharge (above 80% SOC). Therefore, only measuring the electrochemical impedance at different SOCs cannot fully and truly reflect the actual physical and chemical conditions inside the battery. Therefore, it is supplemented by the electrochemical impedance at different open circuit voltages. By combining the electrochemical impedance at different SOCs with the electrical impedance at different open circuit voltages, the accuracy of the results can be further improved.

[0038] Instruction manual drawings:

[0039] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0040] Figure 1 It is a schematic diagram of the process of the method for detecting the remaining capacity of retired batteries of the present invention; DETAILED DESCRIPTION

[0041] The following will be combined with the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] The present invention provides a method for quickly detecting the remaining capacity of a retired battery, which is characterized by comprising:

[0043] S1, obtain the historical operation data of the retired battery, the historical operation data of the retired battery includes the operation data of the retired battery in the whole life cycle and the voltage data and current data in several recent charge and discharge processes; the historical operation data of the retired battery also includes the voltage data and current data in several recent charge and discharge processes of the retired battery; specifically, the following retired battery nuclear capacity detection device can be directly connected to the BMS (BMS: Battery Management System) board in the retired battery pack to read the historical operation data of the retired battery pack. Among them, the charge-discharge process is counted as 1 complete process of obtaining voltage data and current data. The number of the recent charge and discharge processes of the retired battery is not less than 50, and the operation data of the retired battery in the whole life cycle includes discharge curve data, discharge temperature, charging curve data, overcharge times and over-discharge times.

[0044] S2, identifying and correcting the historical operation data set of the retired batteries to obtain an optimized historical operation data set of the retired batteries; and using a DBSCAN algorithm to identify abnormal data.

[0045] The specific steps of using DBSCAN algorithm to identify abnormal data are as follows: input the data set, randomly select an unvisited data point p from the data set to start clustering, take p as the access point, and start searching whether p contains Minpts data points in the Eps neighborhood. If so, p is marked as a core point, and at the same time, the core point p creates a new cluster M, and all the points in the Eps neighborhood are added to the candidate set N; for all the objects q that have not been selected in the candidate set N, check whether their Eps neighborhood contains at least Minpts data points. If so, these points are added to N. If q in the candidate set N is not added to any cluster, q is added to the new cluster M, and the unprocessed data points in the candidate set N are continued to be checked. When all the points in the candidate set N have been processed, stop, and repeat the above steps until all the points have been visited. At this time, any points that have not been added to any cluster are classified as abnormal data.

[0046] S3, using the Spearman correlation coefficient method to perform feature screening on the optimized data set and establish a historical operation data feature database;

[0047] The Spearman correlation coefficient calculation formula is:

[0048]

[0049] Where ρ is the Spearman correlation coefficient between sample X and sample Y, n is the total amount of correlation analysis data, X i is the i-th data of the sample, is the mean of sample X, Y i is the i-th data of the sample, is the mean of sample Y.

[0050] S4, performing an electrochemical impedance spectroscopy test on the retired battery to obtain an electrochemical impedance spectroscopy data set, wherein the electrochemical impedance spectroscopy test includes an electrochemical impedance spectroscopy test at different SOCs and an electrochemical impedance spectroscopy test at different open circuit voltages;

[0051] The electrochemical impedance spectroscopy test process under different SOCs is as follows: when the ambient temperature is room temperature, constant current charging is performed at a rate of 0.5C, and constant voltage charging is performed after reaching the charging cut-off voltage until the current decreases to the charging cut-off current of 0.05C, and then the battery is left standing for 1 hour; first, the battery is subjected to an electrochemical impedance spectroscopy test at 100% SOC, and then 5% SOC is discharged at a constant current rate of 0.2C, and an electrochemical impedance spectroscopy test is performed at 95% SOC after standing for 30 minutes; 5% SOC constant current discharge and electrochemical impedance spectroscopy test are performed in a cycle, and the electrochemical impedance of a group of batteries is measured at every 5% SOC interval, until the battery is subjected to an electrochemical impedance spectroscopy test at 5% SOC, and the electrochemical impedance spectroscopy test process at different SOCs is completed.

[0052] The electrochemical impedance spectroscopy test process under different open circuit voltages is as follows: when the ambient temperature is room temperature, constant current charging is performed at a rate of 0.5C, and constant voltage charging is performed after reaching the charging cut-off voltage until the current decreases to the charging cut-off current of 0.05C, and then standing for 1 hour; first, the battery is tested for electrochemical impedance spectroscopy at the cut-off voltage, and then constant current discharge is performed at a rate of 0.05C to make the voltage drop by 0.2V, and then the electrochemical impedance spectroscopy test is performed after standing for 30 minutes; constant current discharge and electrochemical impedance spectroscopy test are performed in a cycle, and the electrochemical impedance of a group of batteries is measured every time the voltage drops by 0.2V, until the battery voltage drops to the discharge cut-off voltage, and the electrochemical impedance spectroscopy test process of different open circuits is completed.

[0053] We know that during the actual use of lithium batteries, it is usually necessary to maintain a SOC of 20%-80% to prevent irreversible loss of lithium batteries during actual use. For retired batteries, when testing different SOCs, there will be certain virtual electricity during deep discharge (below 20% SOC) and overcharge (above 80% SOC). Therefore, only measuring the electrochemical impedance at different SOCs cannot fully and truly reflect the true physical and chemical conditions inside the battery. Therefore, it is supplemented by the electrochemical impedance at different open circuit voltages. By combining the electrochemical impedance at different SOCs with the electrical impedance at different open circuit voltages, the accuracy of the results can be further improved.

[0054] S5, identifying and correcting the obtained electrochemical impedance spectroscopy data set of retired batteries to obtain optimized electrochemical impedance spectroscopy data of retired batteries,

[0055] The DBSCAN algorithm is used to identify abnormal data. The specific steps of using the DBSCAN algorithm to identify abnormal data are as follows: input the data set, randomly select an unvisited data point p from the data set to start clustering, use p as the access point, and start searching whether the Eps neighborhood of p contains Minpts data points. If so, p is marked as a core point. At the same time, the core point p creates a new cluster M, and all points in the Eps neighborhood are added to the candidate set N; for all objects q that have not been selected in the candidate set N, check whether their Eps neighborhood contains at least Minpts data points. If so, these points are added to N. If q in the candidate set N is not added to any cluster, q is added to the new cluster M, and the unprocessed data points in the candidate set N are continued to be checked. When all points in the candidate set N have been processed, stop, and repeat the above steps until all points have been visited. At this time, any points that are not added to any cluster are classified as abnormal data.

[0056] S6, using the Spearman correlation coefficient method to perform feature screening on the discharge data and electrochemical impedance spectroscopy data of retired batteries, and establish an impedance feature database;

[0057] The Spearman correlation coefficient calculation formula is:

[0058]

[0059] Where ρ is the Spearman correlation coefficient between sample X and sample Y, n is the total amount of correlation analysis data, X i is the i-th data of the sample, is the mean of sample X, Y i is the i-th data of the sample, is the mean of sample Y.

[0060] S7. Input the data in the historical operation data feature database into the neural network algorithm for training, and input the impedance feature data into the multivariate nonlinear regression to construct an estimation model for training. Use the neural network algorithm and the multivariate nonlinear regression to construct an estimation model to extract features and predict the remaining capacity of the battery respectively, and then fuse the two prediction results to obtain the final prediction result of the remaining capacity of the retired battery.

[0061] The steps of the neural network algorithm are as follows: a1. Initialize the connection weights and node thresholds; a2. Input the historical operation data of the battery; a3. Take a set of operation data as a sample; a4. Calculate the output of the hidden layer nodes; a5. Calculate the output of the output layer; a6. Update the connection weights and node thresholds; a7. Determine whether all data are included; if so, proceed to the next step; if not, return to step a2; a8. Determine whether the error is less than the lower limit of 3% or whether the number of learning times reaches the upper limit of 50,000 times; if so, proceed to the next step; if not, return to step a4; a9. Obtain the estimated value of the current capacity of the retired battery.

[0062] For the multivariate nonlinear regression to construct an estimation model, input the real part impedance and imaginary part impedance of each characteristic frequency in the electrochemical impedance test, and the output is the remaining capacity of the to-be-tested retired lithium battery. The expression of the estimation model is: C = f(Z1, Z2); where Z1 is the real part impedance of each characteristic frequency in the electrochemical impedance test; Z1 is the imaginary part impedance of each characteristic frequency in the electrochemical impedance test; C is the remaining capacity of the to-be-tested retired lithium battery.

[0063] In S7, the prediction results are fused to obtain the final prediction result of the remaining capacity of the retired battery, including: using the weighted combination prediction method to calculate the weight coefficients of the two prediction results respectively, and according to the calculated weights, output the final prediction result of the remaining capacity of the retired battery; y3 = Ay1 + By2, where y1 and y2 respectively represent the results predicted based on the historical operation data and the results predicted based on the impedance feature database, y3 is the final prediction result, and A and B respectively represent the weight coefficients of y1 and y2, and A + B = 1.

[0064] When y1 > y2, A is 0.4 and B is 0.6; when y1 < y2, A is 0.7 and B is 0.3; when y1 = y2, A is 0.5 and B is 0.5. In this way, through the adjustment of the weight coefficients, different weight analyses and fusions are carried out on the historical estimated value of the current capacity of the retired battery and the existing estimated value of the current capacity of the retired battery, comprehensively considering the changes in external parameters of the battery under actual working conditions and the changes in its internal structure and materials during the battery aging process, screening the features of the historical operation data and characteristic frequency impedance of the battery, and fusing the historical estimated value and the existing estimated value to obtain the remaining capacity of the retired battery, which can improve the accurate identification of the remaining energy detection of retired lithium batteries and improve the accuracy of the results.

[0065] The present invention also provides a system for rapidly detecting the remaining capacity of retired batteries, which is characterized by comprising:

[0066] A data acquisition module, used to acquire historical operation data sets and electrochemical impedance spectroscopy data sets of retired batteries;

[0067] A data processing module, used for identifying and correcting the historical operation data set and the electrochemical impedance spectroscopy data set to obtain an optimized data set;

[0068] Feature selection module, used to conduct quantitative analysis of data using the Spearman correlation coefficient method and screen data sets with strong correlation;

[0069] Model building module, used to build estimation models based on neural network algorithms and multivariate nonlinear regression to build battery capacity combination prediction models;

[0070] The prediction module is used to take a data set with strong correlation as the model input, extract and predict the model features, and then fuse the prediction results to obtain the battery remaining capacity prediction result.

[0071] An electronic device is provided for an embodiment of the present application. The electronic device may include: at least one processor, at least one network interface, a user interface, a memory, and at least one communication bus.

[0072] The communication bus is used to realize the connection and communication between these components. Among them, the user interface may include a display screen (Display), a camera (Camera), and the optional user interface may also include a standard wired interface and a wireless interface. Among them, the network interface may optionally include a standard wired interface and a wireless interface (such as a WI-FI interface).

[0073] Among them, the processor may include one or more processing cores. The processor uses various interfaces and lines to connect various parts within the entire server, and executes various functions of the server and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory, and calling data stored in the memory. Optionally, the processor can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor can integrate one or more combinations of central processing unit (Central Processing Unit, CPU), image processor (Graphics Processing Unit, GPU) and modem. Among them, the CPU mainly processes the operating system, user interface and application program, etc.; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communication. It can be understood that the above-mentioned modem may not be integrated into the processor, but implemented by a single chip.

[0074] Among them, the memory may include a random access memory (Random Access Memory, RAM) and may also include a read-only memory (Read-Only Memory). Optionally, the memory includes a non-transitory computer-readable storage medium. The memory can be used to store instructions, programs, codes, code sets or instruction sets. The memory may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory may also be optionally at least one storage device located away from the aforementioned processor.

[0075] In the several embodiments provided in this application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some service interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0076] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0077] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0078] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a memory and includes several instructions for a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the various embodiments of the present application. The aforementioned memory includes: various media that can store program codes, such as USB flash drives, mobile hard drives, magnetic disks or optical disks.

[0079] The above are only exemplary embodiments of the present disclosure, and the scope of the present disclosure cannot be limited thereto. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the disclosure of the specification and practice, those skilled in the art will easily think of other embodiments of the present disclosure. This application is intended to cover any variation, use or adaptive change of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary techniques in the technical field that are not recorded in the present disclosure.

Claims

1. A method for quickly detecting the remaining capacity of retired batteries, characterized in that: include: S1, obtaining historical operation data of retired batteries, wherein the historical operation data of retired batteries includes operation data of the retired batteries over their entire life cycle; S2, identifying and correcting the historical operation data set of the retired battery to obtain an optimized historical operation data set of the retired battery; S3, using the Spearman correlation coefficient method to perform feature screening on the optimized battery historical operation data set and establish a historical operation data feature database; S4, performing an electrochemical impedance spectroscopy test on the retired battery to obtain an electrochemical impedance spectroscopy data set, wherein the electrochemical impedance spectroscopy test includes an electrochemical impedance spectroscopy test at different SOCs and an electrochemical impedance spectroscopy test at different open circuit voltages; S5, identifying and correcting the obtained electrochemical impedance spectroscopy data set of retired batteries to obtain an optimized electrochemical impedance spectroscopy data set of retired batteries, S6, using the Spearman correlation coefficient method to perform feature screening on the electrochemical impedance spectroscopy data of retired batteries and establish an impedance feature database; S7, input the data of the historical operation data feature database into the neural network algorithm for training, input the impedance feature data into the multivariate nonlinear regression estimation model for training, use the neural network algorithm and the multivariate nonlinear regression estimation model to respectively extract the features and predict the remaining capacity of the battery, and then merge the prediction results of the two to obtain the final prediction result of the remaining capacity of the retired battery.

2. The method for rapidly detecting the remaining capacity of retired batteries according to claim 1, characterized in that: In S2 and S5, the data set is identified and corrected, including: using the DBSCAN algorithm to identify abnormal data.

3. The method for rapidly detecting the remaining capacity of retired batteries according to claim 2, characterized in that: The specific steps of using DBSCAN algorithm to identify abnormal data are as follows: input the data set, randomly select an unvisited data point p from the data set to start clustering, take p as the access point, and start searching whether p contains Minpts data points in the Eps neighborhood. If so, p is marked as a core point, and at the same time, the core point p creates a new cluster M, and all the points in the Eps neighborhood are added to the candidate set N; for all the objects q that have not been selected in the candidate set N, check whether their Eps neighborhood contains at least Minpts data points. If so, these points are added to N. If q in the candidate set N is not added to any cluster, q is added to the new cluster M, and the unprocessed data points in the candidate set N are continued to be checked. When all the points in the candidate set N have been processed, stop, and repeat the above steps until all the points have been visited. At this time, any points that have not been added to any cluster are classified as abnormal data.

4. The method for rapidly detecting the remaining capacity of retired batteries according to claim 1, characterized in that: The operating data of the retired battery throughout its life cycle includes discharge curve data, discharge temperature, charge curve data, overcharge times, and overdischarge times.

5. The method for rapidly detecting the remaining capacity of retired batteries according to claim 1, characterized in that: The Spearman correlation coefficient calculation formula is: Where ρ is the Spearman correlation coefficient between sample X and sample Y, n is the total amount of correlation analysis data, X i is the i-th data of the sample, is the mean of sample X, Y i is the i-th data of the sample, is the mean of sample Y.

6. The method for rapidly detecting the remaining capacity of retired batteries according to claim 1, characterized in that: The steps of the neural network algorithm are: a1, initialize connection weights and node thresholds; a2, input historical operating data of retired batteries; a3, take a set of operating data as samples; a4, calculate hidden layer node outputs; a5, calculate output layer outputs; a6, update connection weights and node thresholds; a7, determine whether all data are included; if so, proceed to the next step; if not, return to step a2; a8, determine whether the error is less than the lower limit of 3% or the number of learning times reaches the upper limit of 50,000 times; if so, proceed to the next step; if not, return to step a4; a9, obtain an estimated value of the current capacity of the retired battery.

7. The method for rapidly detecting the remaining capacity of retired batteries according to claim 1, characterized in that: The estimation model was constructed by multivariate nonlinear regression, and the real impedance and imaginary impedance of each characteristic frequency in the electrochemical impedance test were input, and the output was the remaining capacity of the retired lithium battery to be tested. The estimation model expression was: C = f(Z1, Z2); where Z1 is the real impedance of each characteristic frequency in the electrochemical impedance test; Z2 is the imaginary impedance of each characteristic frequency in the electrochemical impedance test; and C is the remaining capacity of the retired lithium battery to be tested.

8. The method for rapidly detecting the remaining capacity of retired batteries according to claim 1, characterized in that: In S7, the prediction results are integrated to obtain the final prediction result of the remaining capacity of the retired battery, including: using a weighted combination prediction method to calculate the weight coefficients of the two prediction results respectively, and outputting the final prediction result of the remaining capacity of the retired battery according to the calculated weights; 3 =Ay 1 +By 2 ,y 1 and 2 , respectively represent the prediction results based on historical operation data and the prediction results based on impedance characteristic database, y 3 is the final prediction result, A and B represent y 1 and 2 The weight coefficient is A+B=1.

9. The method for rapidly detecting the remaining capacity of retired batteries according to claim 8, when y 1 >y 2 When A is 0.4 and B is 0.6, when y 1 <y 2 When A is 0.7 and B is 0.3, when y 1 =y 2 When , A is 0.5 and B is 0.

5.

10. A system for rapidly detecting the remaining capacity of retired batteries, characterized in that: include: A data acquisition module is used to acquire a historical operation data set and an electrochemical impedance spectroscopy data set of retired batteries; a data processing module is used to identify and correct the historical operation data set and the electrochemical impedance spectroscopy data set to obtain an optimized data set; Feature selection module, used to conduct quantitative analysis of data using the Spearman correlation coefficient method and screen data sets with strong correlation; Model building module, used to build estimation models based on neural network algorithms and multivariate nonlinear regression to build battery capacity combination prediction models; The prediction module is used to take a data set with strong correlation as the model input, extract and predict the model features, and then fuse the prediction results to obtain the battery remaining capacity prediction result.

Images