Ex-service power battery echelon utilization health assessment system and method

Abstract

The invention discloses a decommissioned power battery echelon utilization health assessment system and method, and relates to the technical field of battery health management, and the method comprises a preliminary full inspection module which is used for disassembling a decommissioned power battery pack to be assessed into Q battery blocks for preliminary full inspection and screening, and obtaining P qualified battery blocks; the detection feature acquisition module is used for analyzing and determining a preliminary detection feature according to the preliminary detection result; the sampling battery distribution obtaining module is used for determining the adaptive sampling detection number and obtaining the optimal sampling battery distribution; and the health state evaluation module is used for carrying out sampling detection on the P qualified battery blocks according to the optimal sampling battery distribution to obtain a sampling detection result, and evaluating and obtaining the health state of the recombined battery pack according to the sampling detection result and the optimal sampling battery distribution. According to the invention, the technical problem that the detection precision and efficiency cannot be effectively balanced by the battery health state evaluation in the prior art is solved.

Description

Technical Field

[0001] This invention relates to the field of battery health management technology, specifically to a health assessment system and method for the secondary use of retired power batteries. Background Technology

[0002] With the rapid development of the global new energy industry, the disposal of power batteries after their service life has become increasingly prominent. A large number of retired batteries still retain considerable remaining capacity. Accurate health status assessment and grading screening, followed by their tiered application in energy storage systems, low-speed electric vehicles, and other areas with relatively lower performance requirements, would greatly promote resource recycling and reduce environmental pollution. However, the core prerequisite for realizing this green vision lies in the rapid, accurate, and low-cost health status assessment of retired batteries, which is currently the main technological bottleneck facing the industry. Existing assessment methods have significant shortcomings. High-precision testing methods, such as complete electrochemical impedance spectroscopy or charge-discharge testing, while providing rich information, are extremely time-consuming. Testing a single cell can take several minutes to several hours, and assessing the entire battery pack can easily take more than ten hours, resulting in low efficiency. Furthermore, coarse random sampling testing may lead to unreliable results. Summary of the Invention

[0003] This application provides a health assessment system and method for the secondary use of retired power batteries, which is intended to address the technical problem that existing battery health status assessments cannot effectively balance detection accuracy and efficiency.

[0004] In view of the above problems, this application provides a health assessment system and method for the secondary use of retired power batteries.

[0005] Firstly, this application provides a health assessment system for the secondary use of retired power batteries, the method of which includes: The preliminary full inspection module is used to disassemble the retired power battery pack to be evaluated into Q battery blocks, perform a preliminary full inspection on the Q battery blocks, and screen the Q battery blocks to obtain P qualified battery blocks based on the preliminary inspection results. Each battery block has a spatial coordinate mark. The detection feature acquisition module is used to analyze and determine the proportion of abnormal battery blocks, the distribution dispersion of abnormal battery blocks, the average performance of normal battery blocks, and the performance fluctuation coefficient of normal battery blocks based on the preliminary detection results, as preliminary detection features. The sampling battery distribution acquisition module is used to determine the appropriate sampling inspection quantity based on the historical usage characteristics and preliminary detection characteristics of the retired power battery pack, and to iteratively optimize the sampling battery distribution scheme based on the appropriate sampling inspection quantity and the spatial coordinate marking of the battery block to obtain the optimal sampling battery distribution. The health status assessment module is used to perform sampling inspection on the P qualified battery blocks according to the optimal sampling battery distribution to obtain the sampling inspection results, and to assess the health status of the recombined battery pack based on the sampling inspection results and the optimal sampling battery distribution.

[0006] Secondly, this application provides a health assessment method for the secondary use of retired power batteries, including: The retired power battery pack to be evaluated is disassembled into Q battery blocks. A preliminary full inspection is performed on the Q battery blocks. Based on the preliminary inspection results, P qualified battery blocks are selected from the Q battery blocks. Each battery block has a spatial coordinate mark. Based on the preliminary test results, the proportion of abnormal battery blocks, the distribution dispersion of abnormal battery blocks, the average performance of normal battery blocks, and the performance fluctuation coefficient of normal battery blocks are analyzed and determined as preliminary test characteristics. Based on the historical usage characteristics of the retired power battery pack and the preliminary detection characteristics, the appropriate sampling inspection quantity is determined. Then, based on the appropriate sampling inspection quantity, the sampling battery distribution scheme is iteratively optimized in combination with the spatial coordinate marking of the battery block to obtain the optimal sampling battery distribution. The P qualified battery blocks are sampled and tested according to the optimal sampling battery distribution to obtain the sampling test results. The health status of the recombined battery pack is evaluated based on the sampling test results and the optimal sampling battery distribution.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages: This application proposes a health assessment system and method for the cascade utilization of retired power batteries. By constructing a hierarchical and progressive technical architecture encompassing preliminary full-inspection for rapid screening, intelligent decision-making based on multi-dimensional features, and optimized sampling for precise assessment, it significantly improves the overall efficiency and accuracy of health status assessment for retired battery packs. Compared to traditional methods, the technical solution provided in this application significantly overcomes the technical problems of excessively high costs for full-inspection and insufficient representativeness of sampling, achieving the technical effect of obtaining highly reliable test results at a lower cost and time than comprehensive and precise testing. Attached Figure Description

[0008] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0009] Figure 1 This is a schematic diagram of the structure of the health assessment system for the secondary use of retired power batteries provided in the embodiments of this application; Figure 2 This is a flowchart illustrating the health assessment method for the secondary use of retired power batteries provided in this application embodiment.

[0010] The components represented by each number in the attached diagram are explained below: The system includes a preliminary full inspection module (100), a detection feature acquisition module (200), a sampled battery distribution acquisition module (300), and a health status assessment module (400). Detailed Implementation

[0011] This application provides a health assessment system and method for the secondary use of retired power batteries, which addresses the technical problem that existing battery health status assessments cannot effectively balance detection accuracy and efficiency.

[0012] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0013] It should be noted that the terms "comprising" and "having" are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to these processes, methods, products, or devices.

[0014] Example 1, as Figure 1 As shown, this application provides a health assessment system for the secondary use of retired power batteries, wherein the system includes: The preliminary full inspection module 100 is used to disassemble the retired power battery pack to be evaluated into Q battery blocks, perform a preliminary full inspection on the Q battery blocks, and select P qualified battery blocks from the Q battery blocks based on the preliminary inspection results. Each battery block has spatial coordinate markings.

[0015] In the evaluation process for the cascade utilization of retired power batteries, the core challenge in the initial screening stage lies in the imbalance between efficiency and information completeness. Simply measuring superficial parameters such as overall terminal voltage and total internal resistance of retired battery packs is quick, but it completely fails to capture key differences such as uneven performance degradation and localized aging defects among the battery cells within the pack, leading to inaccurate evaluation results.

[0016] The preliminary full inspection module 100 in the system provided in this application embodiment includes: The retired power battery pack to be evaluated is disassembled into Q battery blocks, and the positions of the battery blocks are marked according to the three-dimensional spatial coordinates of the battery blocks within the retired power battery pack, where Q is an integer greater than 1; Open-circuit voltage and AC internal resistance are detected for Q battery cells respectively, and Q voltage detection results and Q internal resistance detection results are obtained. Based on Q voltage detection results and Q internal resistance detection results, Q battery blocks are screened. Battery blocks whose voltage detection results meet the preset voltage threshold and whose internal resistance detection results meet the preset internal resistance threshold are selected as qualified battery blocks, and P qualified battery blocks are obtained, where P is an integer greater than or equal to 1 and less than or equal to Q.

[0017] In this embodiment of the application, the preliminary full inspection module 100 is used to disassemble the retired power battery pack to be evaluated into Q battery blocks, perform a preliminary full inspection on the Q battery blocks, and screen the Q battery blocks to obtain P qualified battery blocks based on the preliminary inspection results, wherein each battery block has a spatial coordinate mark.

[0018] Specifically, firstly, the retired power battery pack to be evaluated is disassembled into Q battery blocks, and the positions of the battery blocks are marked according to their three-dimensional spatial coordinates within the retired power battery pack, where Q is an integer greater than 1. For example, using standard battery pack disassembly tools, the outer casing and electrical connections of the retired power battery pack to be evaluated are removed, physically disassembling it into several independent battery modules, i.e., battery blocks. For example, after disassembling a battery pack, 100 such battery blocks are obtained, meaning the variable Q equals 100. Further, during the disassembly process, the position of each battery block in the original battery pack's three-dimensional space is recorded. For example, using a corner point of the battery pack as the origin of the three-dimensional coordinate system, such as setting the lower left corner as the origin (0,0,0), the X-axis (length direction), Y-axis (width direction), and Z-axis (height direction) coordinates of the geometric center of each battery block relative to this origin are recorded.

[0019] Furthermore, open-circuit voltage and AC internal resistance are measured for each of the Q battery cells, yielding Q voltage and Q internal resistance results. For example, using a high-precision digital multimeter, after the battery cells have been left to stand for a sufficient time to stabilize, the potential difference across their terminals is measured, yielding an open-circuit voltage result for each battery cell, such as 3.65 volts. Further, a dedicated battery internal resistance tester is used to measure the AC internal resistance of each battery cell at a specific frequency, such as 1 kHz, yielding a result of 15 milliohms.

[0020] Further, based on Q voltage detection results and Q internal resistance detection results, Q battery cells are screened. Battery cells whose voltage detection results meet a preset voltage threshold and whose internal resistance detection results meet a preset internal resistance threshold are considered qualified battery cells, resulting in P qualified battery cells, where P is an integer greater than or equal to 1 and less than or equal to Q. For example, automatic screening is performed based on preset threshold conditions. The preset voltage threshold can be set according to the actual needs of the battery's secondary use scenario, and is usually a range. For example, in low-speed electric vehicles, the voltage threshold is set to ±5% of the nominal voltage. That is, for a battery cell with a nominal voltage of 3.7 volts, its qualified voltage range may be set to 3.515 volts to 3.885 volts. Further, the preset internal resistance threshold is usually an upper limit value, for example, set to 1.5 times the factory nominal internal resistance. Assuming the factory nominal internal resistance is 10 milliohms, the internal resistance threshold is set to 15 milliohms. Iterate through all 100 battery blocks, check the test results of each battery block. When its voltage value is between 3.515 volts and 3.885 volts and its internal resistance is less than or equal to 15 milliohms, the battery block is judged as a qualified battery block. Count the number of qualified battery blocks and obtain P qualified battery blocks.

[0021] The battery pack is disassembled into battery blocks and subjected to a rapid preliminary full inspection with spatial coordinate markings. All battery blocks are screened using basic parameters that can be measured quickly, such as voltage and internal resistance. Battery blocks that are obviously failed or seriously deviate from the standard are removed, which greatly reduces the number of battery blocks that need to enter the subsequent precision testing stage, thereby reducing the time and resource consumption of the overall evaluation from the source.

[0022] The detection feature acquisition module 200 is used to analyze and determine the proportion of abnormal battery blocks, the distribution dispersion of abnormal battery blocks, the average performance of normal battery blocks, and the performance fluctuation coefficient of normal battery blocks based on the preliminary detection results, as preliminary detection features.

[0023] The important information contained in the distribution patterns of anomalous battery cells is often overlooked in traditional methods. Whether anomalous cells are randomly dispersed or clustered reflects different historical operating conditions and potential risks of the battery pack, but traditional methods cannot quantify this characteristic. Furthermore, for the majority of qualified battery cells, focusing solely on their average performance is far from sufficient, as the degree of fluctuation within the group is also crucial in determining whether the battery pack can operate stably.

[0024] The detection feature acquisition module 200 in the system provided in this application embodiment includes: Battery blocks whose voltage detection results do not meet the preset voltage threshold and / or whose internal resistance detection results do not meet the preset internal resistance threshold are identified as abnormal battery blocks. The spatial distribution of abnormal battery blocks is obtained, and the proportion of abnormal battery blocks is calculated. The dispersion of abnormal battery block distribution is calculated based on the spatial distribution of abnormal battery blocks. Obtain P voltage test results and P internal resistance test results for P qualified battery cells, and calculate the average voltage and average internal resistance. Calculate the voltage variation coefficient of P voltage detection results as the voltage fluctuation coefficient; The coefficient of variation of internal resistance for P internal resistance test results is calculated as the internal resistance fluctuation coefficient.

[0025] In this embodiment of the application, the detection feature acquisition module 200 is used to analyze and determine the proportion of abnormal battery blocks, the distribution dispersion of abnormal battery blocks, the average performance of normal battery blocks, and the performance fluctuation coefficient of normal battery blocks based on the preliminary detection results, as preliminary detection features.

[0026] Specifically, firstly, battery blocks whose voltage detection results do not meet a preset voltage threshold and / or whose internal resistance detection results do not meet a preset internal resistance threshold are identified as abnormal battery blocks. The spatial distribution of these abnormal battery blocks is then obtained, and the percentage of abnormal battery blocks is calculated. For example, 15 abnormal battery blocks are identified out of 100. The three-dimensional spatial coordinates corresponding to these abnormal battery blocks are recorded, forming the spatial distribution of the abnormal battery blocks. Further, the percentage of abnormal battery blocks is calculated: percentage of abnormal battery blocks = number of abnormal battery blocks / total number of battery blocks. For example, percentage of abnormal battery blocks = 15 / 100 = 0.15.

[0027] Furthermore, the dispersion of the abnormal battery blocks is calculated based on their spatial distribution. For example, the geometric center point of all abnormal battery block coordinates is calculated. The X-coordinate of the center point is equal to the average of the X-coordinates of all abnormal blocks; the Y and Z coordinates are calculated similarly. Further, the straight-line distance from the spatial coordinates of each abnormal battery block to this center point is calculated. The dispersion of the abnormal battery block distribution is the average of the straight-line distances from the spatial coordinates of the abnormal battery blocks to this center point. Abnormal battery block dispersion = (sum of distances from all abnormal blocks to their center points) / number of abnormal battery blocks. If the abnormal blocks are densely clustered in space, the dispersion of the abnormal battery block distribution is small; if the abnormal blocks are scattered, the dispersion of the abnormal battery block distribution is large.

[0028] Furthermore, obtain P voltage test results and P internal resistance test results for P qualified battery cells, and calculate the average voltage and average internal resistance. Average voltage = sum of voltage test results for P qualified battery cells / P; Average internal resistance = sum of internal resistance test results for P qualified battery cells / P.

[0029] Furthermore, the voltage variation coefficient of the P voltage detection results is calculated as the voltage fluctuation coefficient, and the internal resistance variation coefficient of the P internal resistance detection results is calculated as the internal resistance fluctuation coefficient. The standard deviations of the P voltage detection results and the P internal resistance detection results are also calculated. The voltage fluctuation coefficient = standard deviation of voltage detection results / mean voltage; the internal resistance fluctuation coefficient = standard deviation of internal resistance detection results / mean internal resistance. The obtained fluctuation coefficient is a dimensionless relative value, which more purely reflects the dispersion of the data.

[0030] By calculating a series of characteristics, including the proportion of abnormal battery cells, the dispersion of abnormal battery cell distribution, the average performance of normal battery cells, and the performance fluctuation coefficient of normal battery cells, the overall degradation level and pattern of the battery pack can be accurately quantified. The proportion of abnormal battery cells directly reflects the overall healthy proportion of the battery pack, while the dispersion of abnormal battery cell distribution reveals the distribution characteristics of abnormal phenomena.

[0031] The sampling battery distribution acquisition module 300 is used to determine the appropriate sampling inspection quantity based on the historical usage characteristics and preliminary inspection characteristics of retired power battery packs, and to iteratively optimize the sampling battery distribution scheme by using the appropriate sampling inspection quantity as a benchmark and combining the spatial coordinate markings of the battery blocks to obtain the optimal sampling battery distribution.

[0032] Determining the number of battery cells to be sampled and their spatial distribution is a core decision-making step for achieving low-cost and accurate assessment. However, existing technologies suffer from significant rigidity and blind spots in this step. Using fixed-ratio sampling may lead to over- or under-testing, and randomly sampled battery cells may not be representative, thus affecting the accuracy of the overall assessment results.

[0033] The sampled battery distribution acquisition module 300 in the system provided in this application embodiment includes: Obtain the historical usage characteristics of retired power battery packs, including the battery pack's service life, current capacity retention rate, and current peak power retention rate; The ratio of the battery pack's service life to the preset standard service life is used as the first local compensation coefficient; The ratio of the preset standard capacity retention rate to the current capacity retention rate is used as the second local compensation coefficient; The ratio of the preset standard peak power retention rate to the current peak power retention rate is used as the third local compensation coefficient; The characteristic compensation coefficient is calculated by weighting the first local compensation coefficient, the second local compensation coefficient, and the third local compensation coefficient. The ratio of the percentage of abnormal battery cells to the percentage of abnormal battery cells according to the preset standard is used as the fourth local compensation coefficient. The ratio of the pre-set standard abnormal battery block distribution dispersion to the abnormal battery block distribution dispersion is used as the fifth local compensation coefficient. Calculate the voltage deviation between the average voltage and the preset standard voltage, and use the ratio of the voltage deviation to the preset standard voltage deviation as the sixth local compensation coefficient. Calculate the deviation of the mean internal resistance from the preset standard internal resistance, and use the ratio of the deviation to the preset standard internal resistance as the seventh local compensation coefficient. The ratio of the voltage fluctuation coefficient to the preset standard voltage fluctuation coefficient is used as the eighth local compensation coefficient. The ratio of the internal resistance fluctuation coefficient to the preset standard internal resistance fluctuation coefficient is used as the ninth local compensation coefficient. The detection feature compensation coefficient is calculated by weighting the fourth, fifth, sixth, seventh, eighth, and ninth local compensation coefficients. The overall feature compensation coefficient is calculated by weighting the feature compensation coefficient and the detection feature compensation coefficient, and the product of the overall feature compensation coefficient and the initial sampling detection quantity is rounded to obtain the appropriate sampling detection quantity. Based on the appropriate sampling inspection quantity, and combined with the spatial coordinate marking of the battery blocks, random sampling schemes are enumerated within P qualified battery blocks to generate several sampling battery distribution schemes. Combining the P voltage detection results and P internal resistance detection results of P qualified battery blocks, the sampling battery detection features are configured for several sampling battery distribution schemes to generate several sampling battery detection schemes. Using the historical usage characteristics of retired power battery packs as constraints, a sample dataset is collected to train a deep learning model, and a sampling detection scheme evaluation plugin is constructed. Among them, constrained by the historical usage characteristics of retired power battery packs, a sample dataset is collected to train a deep learning model, and a sampling detection scheme evaluation plugin is constructed, including: Based on historical battery testing records, information retrieval is performed using the historical usage characteristics of retired power battery packs as constraints, and a sample battery testing scheme set is collected. Obtain the historical detection error ratio of different sample battery detection schemes in the historical detection process, and use 1 to subtract the historical detection error ratio to obtain the sample scheme detection accuracy, and obtain the sample scheme detection accuracy set; Using the sampled battery detection scheme set as input data and the sample scheme detection accuracy set as supervision data, a deep learning model is trained until convergence, generating a sampling detection scheme evaluation plugin. Using the sampling detection scheme evaluation plugin, several sampling battery detection schemes are evaluated, and the prediction accuracy of several schemes is output. The sampling battery distribution scheme corresponding to the scheme with the highest prediction accuracy is selected as the optimal sampling battery distribution.

[0034] In this embodiment of the application, the sampling battery distribution acquisition module 300 is used to determine the appropriate sampling inspection quantity based on the historical usage characteristics and preliminary detection characteristics of the retired power battery pack, and to iteratively optimize the sampling battery distribution scheme by combining the spatial coordinate markings of the battery blocks with the appropriate sampling inspection quantity as a benchmark, so as to obtain the optimal sampling battery distribution.

[0035] Specifically, firstly, the historical usage characteristics of the retired power battery pack are obtained. These characteristics include the battery pack's service life, current capacity retention rate, and current peak power retention rate. The battery pack's service life is the total number of years from the date of manufacture to the current evaluation date, for example, 5 years. The current capacity retention rate is obtained by dividing the current maximum usable capacity measured in the most recent complete charge-discharge test by its factory rated capacity; for example, a capacity retention rate might be 78%. The current peak power retention rate is obtained by dividing the maximum output power measured by its factory rated peak power; for example, a current peak power retention rate might be 82%.

[0036] Furthermore, the ratio of the battery pack's service life to the preset standard service life is used as the first local compensation coefficient; the ratio of the preset standard capacity retention rate to the current capacity retention rate is used as the second local compensation coefficient; and the ratio of the preset standard peak power retention rate to the current peak power retention rate is used as the third local compensation coefficient. The preset standard service life, preset standard capacity retention rate, and preset standard peak power can be obtained based on the application scenario of the retired power battery pack. For example, for retired power battery packs of electric vehicles, the standard service life can be set to 8 years, the standard capacity retention rate to 80%, and the standard peak power retention rate to 70%, to meet safety requirements such as multiple charge-discharge conditions. For other application scenarios, the preset standard service life, preset standard capacity retention rate, and preset standard peak power can be adaptively adjusted. The calculation is as follows: First local compensation coefficient = Battery pack service life / Preset standard service life; Second local compensation coefficient = Preset standard capacity retention rate / Current capacity retention rate; Third local compensation coefficient = Preset standard peak power retention rate / Current peak power retention rate.

[0037] Furthermore, the characteristic compensation coefficient is calculated by weighting the first local compensation coefficient, the second local compensation coefficient, and the third local compensation coefficient. For example, the weights in the weighted calculation can be obtained based on the application scenario of the retired power battery pack. For instance, if the application scenario has high safety requirements for the retired power battery, then a longer service life is required. In this case, the weight of the first local compensation coefficient is set to 0.4, and the weights of the second and third local compensation coefficients are set to 0.3. Therefore, the characteristic compensation coefficient = 0.4 × first compensation coefficient + 0.3 × second compensation coefficient + 0.3 × third compensation coefficient.

[0038] Furthermore, the ratio of the percentage of abnormal battery cells to the preset standard percentage of abnormal battery cells is used as the fourth local compensation coefficient; the ratio of the preset standard distribution dispersion of abnormal battery cells to the distribution dispersion of abnormal battery cells is used as the fifth local compensation coefficient; the voltage deviation between the average voltage and the preset standard voltage is calculated, and the ratio of the voltage deviation to the preset standard voltage deviation is used as the sixth local compensation coefficient; the internal resistance deviation between the average internal resistance and the preset standard internal resistance is calculated, and the ratio of the internal resistance deviation to the preset standard internal resistance deviation is used as the seventh local compensation coefficient; the ratio of the voltage fluctuation coefficient to the preset standard voltage fluctuation coefficient is used as the eighth local compensation coefficient; and the ratio of the internal resistance fluctuation coefficient to the preset standard internal resistance fluctuation coefficient is used as the ninth local compensation coefficient. Specifically, the fourth local compensation coefficient = percentage of abnormal battery blocks / preset standard percentage of abnormal battery blocks; the fifth local compensation coefficient = preset standard distribution dispersion of abnormal battery blocks / distribution dispersion of abnormal battery blocks; the sixth local compensation coefficient = voltage deviation amplitude / preset standard voltage deviation amplitude; the seventh local compensation coefficient = internal resistance deviation amplitude / preset standard internal resistance deviation amplitude; the eighth local compensation coefficient = voltage fluctuation coefficient / preset standard voltage fluctuation coefficient; and the ninth local compensation coefficient = internal resistance fluctuation coefficient / preset standard internal resistance fluctuation coefficient. The preset standard percentage of abnormal battery blocks and the preset standard distribution dispersion of abnormal battery blocks are obtained based on the application scenario of the retired power battery pack. For example, if the application scenario is a retired automotive power battery, considering safety requirements, the preset standard percentage of abnormal battery blocks can be 0.1, and the preset standard distribution dispersion of abnormal battery blocks can be 0.6. Furthermore, the preset standard voltage deviation, preset standard current deviation, preset standard internal resistance deviation, preset standard voltage fluctuation coefficient, and preset standard internal resistance fluctuation coefficient are obtained based on the factory design parameters of the retired power battery pack. Among them, the preset standard voltage deviation, preset standard current deviation, and preset standard internal resistance deviation can be obtained based on the average deviation of the preset standard voltage, current, and internal resistance parameters in the same model and application scenario of power batteries and the factory settings. For example, after statistical analysis of power batteries of the same model and application scenario, if the average voltage deviation is 0.05, the average current deviation is 0.07, and the average internal resistance deviation is 0.1, then the preset standard voltage deviation can be set to 0.05, the preset standard current deviation to 0.07, and the preset standard internal resistance deviation to 0.1. The deviation is obtained by dividing the absolute difference between the actual value and the preset standard value by the preset standard value. Similarly, the preset standard voltage fluctuation coefficient and the preset standard internal resistance fluctuation coefficient can be obtained based on the average values ​​of the preset standard voltage and internal resistance fluctuation coefficients in the same model and application scenario of power batteries and factory settings.

[0039] Furthermore, the detection feature compensation coefficient is calculated by weighting the fourth, fifth, sixth, seventh, eighth, and ninth local compensation coefficients. The weights are determined based on the application scenario and factory design parameters of the retired power battery pack. For example, for retired power battery packs of electric vehicles, the weights of the fourth and fifth local compensation coefficients can be set to be relatively large (0.2), while the weights of the other local compensation coefficients can be relatively small (0.15). Therefore, the detection feature compensation parameter = 0.2 × fourth local compensation coefficient + 0.2 × fifth local compensation coefficient + 0.15 × sixth local compensation coefficient + 0.15 × seventh local compensation coefficient + 0.15 × eighth local compensation coefficient + 0.15 × ninth local compensation coefficient.

[0040] Furthermore, the overall feature compensation coefficient is calculated by weighting the usage feature compensation coefficient and the detection feature compensation coefficient. The product of the overall feature compensation coefficient and the initial sampling detection quantity is then rounded to obtain the suitable sampling detection quantity. For example, the weights of both the usage feature compensation coefficient and the detection feature compensation coefficient can be set to 0.5, then the overall feature compensation coefficient = 0.5 × usage feature compensation coefficient + 0.5 × detection feature compensation coefficient. The initial sampling detection quantity can be set to 20, and the suitable sampling detection quantity is calculated as: overall feature compensation coefficient × initial sampling detection quantity. For example, if the overall feature compensation coefficient is 0.8, then the suitable sampling detection quantity = 0.8 × 20 = 16.

[0041] Furthermore, based on the number of suitable sampling tests, and combined with the spatial coordinate markings of the battery blocks, random sampling schemes are enumerated within the P qualified battery blocks to generate several sampled battery distribution schemes. For example, when the number of suitable sampling tests is 16, the coordinates of 16 battery blocks are randomly selected from the P qualified battery blocks to form a sampled battery distribution scheme, which includes the spatial coordinates of the selected 16 battery blocks. The above random sampling process is repeated to generate several sampled battery distribution schemes.

[0042] Furthermore, by combining the P voltage detection results and P internal resistance detection results of P qualified battery blocks, sampling battery detection features are configured for several sampling battery distribution schemes, generating several sampling battery detection schemes. Specifically, by combining the P voltage detection results and P internal resistance detection results of P qualified battery blocks, several sampling battery detection schemes for several sampling battery distribution schemes are obtained. For example, a sampling battery detection scheme may include the three-dimensional spatial coordinates of 16 battery blocks distributed at different locations and the detection schemes executed.

[0043] Furthermore, by using the historical usage characteristics of retired power battery packs as constraints, a sample dataset is collected to train a deep learning model, and a sampling detection scheme evaluation plugin is constructed.

[0044] Specifically, firstly, based on historical battery testing records, information retrieval is performed using the historical usage characteristics of retired power battery packs as constraints to collect a sample set of battery testing schemes. Specifically, using the historical usage characteristics of the current battery pack, such as a 5-year lifespan and a capacity retention rate of 78%, as query conditions, past battery pack testing records with similar characteristics are retrieved from the historical database, and historically implemented sampled battery testing schemes are collected as the sample set of battery testing schemes.

[0045] Furthermore, the historical detection error ratio of different sample battery testing schemes in the historical testing process is obtained, and the detection accuracy of the sample scheme is obtained by subtracting the historical detection error ratio from 1, thus obtaining the sample scheme detection accuracy set. The test results corresponding to the sampled battery testing scheme set are obtained, and by comparing its sampling evaluation results with the full inspection results of all battery blocks in the corresponding battery pack, the historical detection error ratio of this sampling is calculated, and the sample scheme detection accuracy is calculated as 1 - historical detection error ratio.

[0046] Furthermore, using a sample set of battery detection schemes as input data and a set of sample scheme detection accuracy as supervised data, a deep learning model is trained until convergence, generating a sampling detection scheme evaluation plugin. For example, a fully connected neural network model is constructed as the sampling detection scheme evaluation plugin. The number of nodes in the input layer equals the feature dimension of a sample battery detection scheme. The number of neurons in the three hidden layers are 128, 64, and 32, respectively, all using the ReLU activation function. The output layer has one neuron, using the Sigmoid activation function, with an output value between 0 and 1, representing the predicted scheme accuracy. The model uses mean squared error as the loss function and the Adam optimizer. Supervised training is performed using a sample set of battery detection schemes as input data and a set of corresponding sample scheme detection accuracy as supervised data until the loss function converges. The stable model obtained after training is the sampling detection scheme evaluation plugin.

[0047] Furthermore, using the sampling detection scheme evaluation plugin, several sampling battery detection schemes are evaluated, and the prediction accuracy of several schemes is output. The sampling battery distribution scheme corresponding to the scheme with the highest prediction accuracy is selected as the optimal sampling battery distribution.

[0048] By analyzing the initial detection features of the battery pack and performing weighted compensation calculations, a suitable sampling quantity is dynamically generated to match the specific state of the battery pack. For battery packs with complex states and high risks, the sampling quantity is increased to reduce uncertainty; for battery packs with good states and high consistency, the sampling quantity is reduced to improve efficiency. Based on the suitable quantity, and combined with the spatial coordinates of each battery block, the optimal sampling battery distribution is found from possible sampling combinations. This ensures that the selected battery blocks not only have broad spatial coverage but also target typical areas that may represent different aging modes. The detection results based on the optimal sampling battery distribution can maximize the characterization of the overall health status.

[0049] The health status assessment module 400 is used to perform sampling inspection on P qualified battery blocks according to the optimal sampling battery distribution to obtain the sampling inspection results, and to assess the health status of the recombined battery pack based on the sampling inspection results and the optimal sampling battery distribution.

[0050] Traditional evaluation methods often simply average the test results of sampled battery cells and use this as the performance indicator for the entire battery pack, ignoring the spatial non-uniformity of performance within the battery pack. The health status of a reconstructed battery pack is not a simple summation of the status of sampled battery cells, but a comprehensive reflection of the overall performance and reliability of all battery cells.

[0051] The health status assessment module 400 in the system provided in this application embodiment includes: Configure a sampling and testing process, which includes at least the following: full capacity test, DC internal resistance test, pulse power characteristic test, high and low temperature performance point test, self-discharge rate test, and charge and discharge efficiency test. P qualified battery blocks are selected according to the optimal sampling battery distribution to obtain multiple sampling battery blocks; According to the sampling and testing process, multiple sampled battery blocks were subjected to targeted precision testing, and multiple performance precision test results were output as the sampling and testing results. The P preliminary test results of P qualified battery blocks are mapped and associated with the spatial coordinates of the P qualified battery blocks. Multiple performance fine test results are mapped and associated with the spatial coordinates of the battery blocks in the optimal sampled battery distribution to generate the global test data distribution of the battery blocks. The health status of the reassembled battery pack is assessed using a battery evaluation expert system based on the global detection data distribution.

[0052] In this embodiment, the health status assessment module 400 is used to perform sampling inspection on P qualified battery blocks according to the optimal sampling battery distribution to obtain the sampling inspection results, and to assess the health status of the recombined battery pack based on the sampling inspection results and the optimal sampling battery distribution.

[0053] Specifically, firstly, a sampling inspection process is configured, which includes at least full capacity testing, DC internal resistance testing, pulse power characteristic testing, high and low temperature performance testing, self-discharge rate testing, and charge / discharge efficiency testing. This sampling inspection process defines a more in-depth and time-consuming set of performance tests to be performed on the sampled battery cells than a preliminary full inspection.

[0054] Furthermore, P qualified battery blocks are selected according to the optimal sampled battery distribution to obtain multiple sampled battery blocks. These multiple sampled battery blocks also include spatial coordinate markers for the battery blocks.

[0055] Furthermore, following the sampling and testing procedure, targeted precision testing was performed on multiple sampled battery blocks, and multiple performance precision test results were output as the sampling and testing results. Multiple sampled battery blocks were then removed, and their performance was precisely tested separately according to the sampling and testing procedure to obtain the sampling and testing results.

[0056] Furthermore, the preliminary test results of P qualified battery blocks are mapped and associated with the spatial coordinates of the P qualified battery blocks, and multiple performance precision test results are mapped and associated with the spatial coordinates of battery blocks in the optimal sampled battery distribution, generating a global test data distribution for the battery blocks. For example, a record is created for each qualified battery block, containing its spatial coordinates, corresponding preliminary voltage value, and preliminary internal resistance value. Further, each performance precision test result in the sampled test results is mapped and associated with the coordinates in the optimal sampled battery distribution scheme to obtain the global test data distribution for the battery blocks. In this distribution, each battery block location is associated with at least basic voltage and internal resistance data, while for those sampled coordinate points, a complete set of precision performance data is additionally mapped.

[0057] Furthermore, a battery evaluation expert system is used to assess the health status of the reassembled battery pack based on the global detection data distribution. The expert system is a neural network integrating multiple evaluation rules and logical judgments. By performing a comprehensive evaluation of the global detection data distribution using this expert system, the health status of the reassembled battery pack can be accurately assessed without precisely testing all battery cells. For example, if a battery cell has a low measured capacity, surrounding battery cells with low initial voltages will be considered to have a higher capacity risk. Based on limited but optimized sampling data, the battery evaluation expert system analyzes spatial correlations and global detection data to achieve a reliable inference of the overall health status.

[0058] First, a series of precise tests are performed on the selected battery blocks according to the optimal sampling distribution to obtain their depth performance profile. Then, the module does not use this depth data in isolation, but maps and correlates it with the preliminary test results of the battery blocks and their respective spatial coordinates, forming a global test data distribution that includes extensive preliminary data and depth test data. Finally, a battery evaluation expert system is used to comprehensively analyze this global test data distribution, taking into account multiple factors such as consistency, worst values, and average trends, to obtain a comprehensive and weighted evaluation conclusion on the overall health status of the recombined battery pack, serving as a reliable evaluation result for the overall performance of the battery pack.

[0059] Example 2, as Figure 2 As shown, based on the same inventive concept as the health assessment system for the cascade utilization of retired power batteries provided in Embodiment 1, this embodiment of the invention also provides a health assessment method for the cascade utilization of retired power batteries, including: The retired power battery pack to be evaluated is disassembled into Q battery blocks. A preliminary full inspection is carried out on the Q battery blocks. Based on the preliminary inspection results, P qualified battery blocks are selected from the Q battery blocks. Each battery block has a spatial coordinate mark. Based on the analysis of the preliminary test results, the proportion of abnormal battery blocks, the dispersion of the distribution of abnormal battery blocks, the average performance of normal battery blocks, and the performance fluctuation coefficient of normal battery blocks were determined as preliminary test characteristics. Based on the historical usage characteristics and preliminary testing characteristics of retired power battery packs, the appropriate sampling inspection quantity is determined. Then, based on the appropriate sampling inspection quantity, the sampling battery distribution scheme is iteratively optimized by combining the spatial coordinate markings of the battery blocks to obtain the optimal sampling battery distribution. According to the optimal sampling battery distribution, P qualified battery blocks are sampled and tested to obtain the sampling test results. Based on the sampling test results and the optimal sampling battery distribution, the health status of the recombined battery pack is evaluated.

[0060] In one embodiment, the retired power battery pack to be evaluated is disassembled into Q battery blocks, a preliminary full inspection is performed on the Q battery blocks, and P qualified battery blocks are obtained from the Q battery blocks based on the preliminary inspection results. Each battery block has spatial coordinate markers, and the method further includes: The retired power battery pack to be evaluated is disassembled into Q battery blocks, and the positions of the battery blocks are marked according to the three-dimensional spatial coordinates of the battery blocks within the retired power battery pack, where Q is an integer greater than 1; Open-circuit voltage and AC internal resistance are detected for Q battery cells respectively, and Q voltage detection results and Q internal resistance detection results are obtained. Based on Q voltage detection results and Q internal resistance detection results, Q battery blocks are screened. Battery blocks whose voltage detection results meet the preset voltage threshold and whose internal resistance detection results meet the preset internal resistance threshold are selected as qualified battery blocks, and P qualified battery blocks are obtained, where P is an integer greater than or equal to 1 and less than or equal to Q.

[0061] In one embodiment, based on the preliminary detection results, the percentage of abnormal battery blocks, the dispersion of abnormal battery block distribution, the average performance of normal battery blocks, and the performance fluctuation coefficient of normal battery blocks are determined as preliminary detection features. The method further includes: Battery blocks whose voltage detection results do not meet the preset voltage threshold and / or whose internal resistance detection results do not meet the preset internal resistance threshold are identified as abnormal battery blocks. The spatial distribution of abnormal battery blocks is obtained, and the proportion of abnormal battery blocks is calculated. The dispersion of abnormal battery block distribution is calculated based on the spatial distribution of abnormal battery blocks. Obtain P voltage test results and P internal resistance test results for P qualified battery cells, and calculate the average voltage and average internal resistance. Calculate the voltage variation coefficient of P voltage detection results as the voltage fluctuation coefficient; The coefficient of variation of internal resistance for P internal resistance test results is calculated as the internal resistance fluctuation coefficient.

[0062] In one embodiment, the appropriate sampling inspection quantity is determined based on the historical usage characteristics and preliminary inspection characteristics of the retired power battery pack. Using this appropriate sampling inspection quantity as a benchmark, and combining iterative optimization of the sampling battery distribution scheme with the spatial coordinate markings of the battery blocks, the optimal sampling battery distribution is obtained. The method further includes: Obtain the historical usage characteristics of retired power battery packs, including the battery pack's service life, current capacity retention rate, and current peak power retention rate; The ratio of the battery pack's service life to the preset standard service life is used as the first local compensation coefficient; The ratio of the preset standard capacity retention rate to the current capacity retention rate is used as the second local compensation coefficient; The ratio of the preset standard peak power retention rate to the current peak power retention rate is used as the third local compensation coefficient; The characteristic compensation coefficient is calculated by weighting the first local compensation coefficient, the second local compensation coefficient, and the third local compensation coefficient. The ratio of the percentage of abnormal battery cells to the percentage of abnormal battery cells according to the preset standard is used as the fourth local compensation coefficient. The ratio of the pre-set standard abnormal battery block distribution dispersion to the abnormal battery block distribution dispersion is used as the fifth local compensation coefficient. Calculate the voltage deviation between the average voltage and the preset standard voltage, and use the ratio of the voltage deviation to the preset standard voltage deviation as the sixth local compensation coefficient. Calculate the deviation of the mean internal resistance from the preset standard internal resistance, and use the ratio of the deviation to the preset standard internal resistance as the seventh local compensation coefficient. The ratio of the voltage fluctuation coefficient to the preset standard voltage fluctuation coefficient is used as the eighth local compensation coefficient. The ratio of the internal resistance fluctuation coefficient to the preset standard internal resistance fluctuation coefficient is used as the ninth local compensation coefficient. The detection feature compensation coefficient is calculated by weighting the fourth, fifth, sixth, seventh, eighth, and ninth local compensation coefficients. The overall feature compensation coefficient is calculated by weighting the feature compensation coefficient and the detection feature compensation coefficient, and the product of the overall feature compensation coefficient and the initial sampling detection quantity is rounded to obtain the appropriate sampling detection quantity. Based on the appropriate sampling inspection quantity, and combined with the spatial coordinate marking of the battery blocks, random sampling schemes are enumerated within P qualified battery blocks to generate several sampling battery distribution schemes. Combining the P voltage detection results and P internal resistance detection results of P qualified battery blocks, the sampling battery detection features are configured for several sampling battery distribution schemes to generate several sampling battery detection schemes. Using the historical usage characteristics of retired power battery packs as constraints, a sample dataset is collected to train a deep learning model, and a sampling detection scheme evaluation plugin is constructed. Among them, constrained by the historical usage characteristics of retired power battery packs, a sample dataset is collected to train a deep learning model, and a sampling detection scheme evaluation plugin is constructed, including: Based on historical battery testing records, information retrieval is performed using the historical usage characteristics of retired power battery packs as constraints, and a sample battery testing scheme set is collected. Obtain the historical detection error ratio of different sample battery detection schemes in the historical detection process, and use 1 to subtract the historical detection error ratio to obtain the sample scheme detection accuracy, and obtain the sample scheme detection accuracy set; Using the sampled battery detection scheme set as input data and the sample scheme detection accuracy set as supervision data, a deep learning model is trained until convergence, generating a sampling detection scheme evaluation plugin. Using the sampling detection scheme evaluation plugin, several sampling battery detection schemes are evaluated, and the prediction accuracy of several schemes is output. The sampling battery distribution scheme corresponding to the scheme with the highest prediction accuracy is selected as the optimal sampling battery distribution.

[0063] In one embodiment, sampling and testing are performed on P qualified battery blocks according to the optimal sampling battery distribution to obtain sampling and testing results. The health status of the reassembled battery pack is then evaluated based on the sampling and testing results and the optimal sampling battery distribution. The method further includes: Configure a sampling and testing process, which includes at least the following: full capacity test, DC internal resistance test, pulse power characteristic test, high and low temperature performance point test, self-discharge rate test, and charge and discharge efficiency test. P qualified battery blocks are selected according to the optimal sampling battery distribution to obtain multiple sampling battery blocks; According to the sampling and testing process, multiple sampled battery blocks were subjected to targeted precision testing, and multiple performance precision test results were output as the sampling and testing results. The P preliminary test results of P qualified battery blocks are mapped and associated with the spatial coordinates of the P qualified battery blocks. Multiple performance fine test results are mapped and associated with the spatial coordinates of the battery blocks in the optimal sampled battery distribution to generate the global test data distribution of the battery blocks. The health status of the reassembled battery pack is assessed using a battery evaluation expert system based on the global detection data distribution.

[0064] In summary, the embodiments of this application have at least the following technical effects: This application proposes a health assessment system and method for the cascade utilization of retired power batteries. By constructing a hierarchical and progressive technical architecture—comprising preliminary full-inspection for rapid screening, intelligent decision-making based on multi-dimensional features, and optimized sampling for precise assessment—it significantly improves the overall efficiency and accuracy of health status assessment for retired battery packs. Specifically, firstly, by rapidly detecting basic electrical performance parameters and marking the spatial locations of all battery cells, preliminary screening and spatial topological relationship construction are completed. Subsequently, the historical usage characteristics of the battery pack are fused and analyzed with the statistical distribution characteristics extracted from the preliminary detection. This dynamically calculates the appropriate sampling quantity matching the individual state of the retired battery pack, ensuring that the sampling size is no longer a fixed empirical value but a variable that responds to the actual aging condition and consistency of the battery pack. This provides a scientific basis for balancing detection costs and assessment accuracy from the outset. Furthermore, using this appropriate quantity as a constraint, combined with the spatial coordinate information of the battery cells, an iterative optimization algorithm generates the optimal sampling battery distribution scheme that is most representative in spatial distribution. This ensures that the extracted samples can reflect the overall distribution and local characteristics of the battery pack's internal performance to the greatest extent, effectively avoiding the sampling bias risk caused by uneven aging. Finally, based on this optimized distribution scheme, comprehensive and precise performance tests were conducted on a small number of samples. The extensive data from the initial full inspection and the in-depth data from the sampling precision test were combined, and the overall health status of the reconstituted battery pack was evaluated through data fusion and an expert system. Compared to traditional methods, the technical solution provided in this application significantly overcomes the technical problems of excessively high full inspection costs and insufficient representativeness of sampling, achieving the technical effect of obtaining highly reliable test results at a lower cost and time than comprehensive precision testing.

[0065] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0066] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0067] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A health assessment system for the cascade utilization of retired power batteries, characterized in that, The system includes: The preliminary full inspection module is used to disassemble the retired power battery pack to be evaluated into Q battery blocks, perform a preliminary full inspection on the Q battery blocks, and screen the Q battery blocks to obtain P qualified battery blocks based on the preliminary inspection results. Each battery block has a spatial coordinate mark. The detection feature acquisition module is used to analyze and determine the proportion of abnormal battery blocks, the distribution dispersion of abnormal battery blocks, the average performance of normal battery blocks, and the performance fluctuation coefficient of normal battery blocks based on the preliminary detection results, as preliminary detection features. The sampling battery distribution acquisition module is used to determine the appropriate sampling inspection quantity based on the historical usage characteristics and preliminary detection characteristics of the retired power battery pack, and to iteratively optimize the sampling battery distribution scheme based on the appropriate sampling inspection quantity and the spatial coordinate marking of the battery block to obtain the optimal sampling battery distribution. The health status assessment module is used to perform sampling inspection on the P qualified battery blocks according to the optimal sampling battery distribution to obtain the sampling inspection results, and to assess the health status of the recombined battery pack based on the sampling inspection results and the optimal sampling battery distribution.

2. The health assessment system for the cascade utilization of retired power batteries according to claim 1, characterized in that, The retired power battery pack to be evaluated is disassembled into Q battery blocks. A preliminary full inspection is performed on the Q battery blocks. Based on the preliminary inspection results, P qualified battery blocks are selected from the Q battery blocks, including: The retired power battery pack to be evaluated is disassembled into Q battery blocks, and the positions of the battery blocks are marked according to the three-dimensional spatial coordinates of the battery blocks within the retired power battery pack, where Q is an integer greater than 1. Open-circuit voltage and AC internal resistance are detected for each of the Q battery cells, and Q voltage detection results and Q internal resistance detection results are obtained respectively. Based on the Q voltage detection results and Q internal resistance detection results, the Q battery blocks are screened. Battery blocks whose voltage detection results meet the preset voltage threshold and whose internal resistance detection results meet the preset internal resistance threshold are selected as qualified battery blocks, and P qualified battery blocks are obtained, where P is an integer greater than or equal to 1 and less than or equal to Q.

3. The health assessment system for the cascade utilization of retired power batteries according to claim 2, characterized in that, Based on the preliminary test results, the following parameters were determined: percentage of abnormal battery cells, distribution dispersion of abnormal battery cells, average performance of normal battery cells, and performance fluctuation coefficient of normal battery cells. Battery blocks whose voltage detection results do not meet the preset voltage threshold and / or whose internal resistance detection results do not meet the preset internal resistance threshold are identified as abnormal battery blocks. The spatial distribution of abnormal battery blocks is obtained, and the proportion of abnormal battery blocks is calculated. The dispersion of the abnormal battery block distribution is calculated based on the spatial distribution of the abnormal battery blocks. Obtain P voltage detection results and P internal resistance detection results for the P qualified battery cells, and calculate the average voltage and average internal resistance. The voltage variation coefficient of the P voltage detection results is calculated as the voltage fluctuation coefficient; The coefficient of variation of the internal resistance of the P internal resistance detection results is calculated as the internal resistance fluctuation coefficient.

4. The health assessment system for the cascade utilization of retired power batteries according to claim 3, characterized in that, The historical usage characteristics of the retired power battery pack are obtained, wherein the historical usage characteristics include the battery pack's service life, the battery pack's current capacity retention rate, and the current peak power retention rate.

5. The health assessment system for the cascade utilization of retired power batteries according to claim 4, characterized in that, The appropriate sampling inspection quantity is determined based on the historical usage characteristics of the retired power battery pack and the preliminary detection characteristics, including: The ratio of the battery pack's service life to the preset standard service life is used as the first local compensation coefficient; The ratio of the preset standard capacity retention rate to the current capacity retention rate is used as the second local compensation coefficient; The ratio of the preset standard peak power retention rate to the current peak power retention rate is used as the third local compensation coefficient; The characteristic compensation coefficient is calculated by weighting the first local compensation coefficient, the second local compensation coefficient, and the third local compensation coefficient. The ratio of the percentage of abnormal battery blocks to the percentage of abnormal battery blocks according to a preset standard is used as the fourth local compensation coefficient. The ratio of the preset standard abnormal battery block distribution dispersion to the abnormal battery block distribution dispersion is used as the fifth local compensation coefficient. Calculate the voltage deviation between the average voltage and the preset standard voltage, and use the ratio of the voltage deviation to the preset standard voltage deviation as the sixth local compensation coefficient. Calculate the deviation of the mean internal resistance from the preset standard internal resistance, and use the ratio of the deviation to the preset standard internal resistance as the seventh local compensation coefficient. The ratio of the voltage fluctuation coefficient to the preset standard voltage fluctuation coefficient is used as the eighth local compensation coefficient. The ratio of the internal resistance fluctuation coefficient to the preset standard internal resistance fluctuation coefficient is used as the ninth local compensation coefficient. The detection feature compensation coefficient is calculated by weighting the fourth, fifth, sixth, seventh, eighth, and ninth local compensation coefficients. The overall feature compensation coefficient is calculated by weighting the use feature compensation coefficient and the detection feature compensation coefficient, and the product of the overall feature compensation coefficient and the initial sampling detection number is rounded to obtain the appropriate sampling detection number.

6. The health assessment system for the cascade utilization of retired power batteries according to claim 1, characterized in that, Based on the aforementioned number of adaptive sampling tests, and combined with the spatial coordinate markings of the battery blocks, the sampling battery distribution scheme is iteratively optimized to obtain the optimal sampling battery distribution, including: Based on the number of suitable sampling tests, and combined with the spatial coordinate markings of the battery blocks, random sampling schemes are enumerated within the P qualified battery blocks to generate several sampling battery distribution schemes. Based on the P voltage detection results and P internal resistance detection results of the P qualified battery blocks, the sampling battery detection features are configured for the several sampling battery distribution schemes to generate several sampling battery detection schemes. Using the historical usage characteristics of the retired power battery packs as constraints, a sample dataset is collected to train a deep learning model, and a sampling detection scheme evaluation plugin is constructed. Using the sampling detection scheme evaluation plugin, the several sampling battery detection schemes are evaluated respectively, and the prediction accuracy of several schemes is output. The sampling battery distribution scheme corresponding to the scheme with the highest prediction accuracy is selected as the optimal sampling battery distribution.

7. The health assessment system for the cascade utilization of retired power batteries according to claim 6, characterized in that, Constrained by the historical usage characteristics of the retired power battery packs, a sample dataset is collected to train a deep learning model, and a sampling detection scheme evaluation plugin is constructed, including: Based on historical battery testing records, information retrieval is performed using the historical usage characteristics of the retired power battery packs as constraints, and a sampled battery testing scheme set is collected. Obtain the historical detection error ratio of different sample battery detection schemes in the historical detection process, and use 1 to subtract the historical detection error ratio to obtain the sample scheme detection accuracy, and obtain the sample scheme detection accuracy set; Using the sampled battery detection scheme set as input data and the sample scheme detection accuracy set as supervision data, a deep learning model is trained until convergence, generating a sampling detection scheme evaluation plugin.

8. The health assessment system for the cascade utilization of retired power batteries according to claim 1, characterized in that, Sampling and testing are performed on the P qualified battery blocks according to the optimal sampling battery distribution to obtain sampling and testing results. The health status of the reassembled battery pack is then evaluated based on the sampling and testing results and the optimal sampling battery distribution, including: Configure a sampling and testing process, wherein the sampling and testing process includes at least full capacity testing, DC internal resistance testing, pulse power characteristic testing, high and low temperature performance point testing, self-discharge rate testing, and charge and discharge efficiency testing; According to the optimal sampling battery distribution, the P qualified battery blocks are selected to obtain multiple sampling battery blocks; According to the sampling and testing process, the multiple sampled battery blocks are subjected to targeted precision testing, and multiple performance precision testing results are output as the sampling and testing results.

9. The health assessment system for the cascade utilization of retired power batteries according to claim 8, characterized in that, The health status of the reassembled battery pack is assessed based on the sampling test results and the optimal sampled battery distribution, including: The P preliminary test results of P qualified battery blocks are mapped and associated with the spatial coordinates of the P qualified battery blocks, and the multiple performance fine test results are mapped and associated with the spatial coordinates of the battery blocks in the optimal sampled battery distribution to generate a global test data distribution of the battery blocks. The health status of the reassembled battery pack is assessed using a battery evaluation expert system based on the global detection data distribution.

10. A health assessment method for the secondary use of retired power batteries, applied to the health assessment system for the secondary use of retired power batteries as described in any one of claims 1 to 9, the method comprising: The retired power battery pack to be evaluated is disassembled into Q battery blocks. A preliminary full inspection is performed on the Q battery blocks. Based on the preliminary inspection results, P qualified battery blocks are selected from the Q battery blocks. Each battery block has a spatial coordinate mark. Based on the preliminary test results, the proportion of abnormal battery blocks, the distribution dispersion of abnormal battery blocks, the average performance of normal battery blocks, and the performance fluctuation coefficient of normal battery blocks are analyzed and determined as preliminary test characteristics. Based on the historical usage characteristics of the retired power battery pack and the preliminary detection characteristics, the appropriate sampling inspection quantity is determined. Then, based on the appropriate sampling inspection quantity, the sampling battery distribution scheme is iteratively optimized in combination with the spatial coordinate marking of the battery block to obtain the optimal sampling battery distribution. The P qualified battery blocks are sampled and tested according to the optimal sampling battery distribution to obtain the sampling test results. The health status of the recombined battery pack is evaluated based on the sampling test results and the optimal sampling battery distribution.

Images