Energy storage module multi-station parallel EOL comprehensive test system

Abstract

This invention discloses a multi-station parallel EOL comprehensive testing system for energy storage modules. This invention relates to the field of testing technology. Through this invention, a module testing module is set up at each testing station to collect test data for each energy storage module during the testing process of the deployed energy storage modules for each test item. Simultaneously, the system determines in real time whether the energy storage modules meet stability conditions. Based on the determination results and the test results of each item, it determines whether to continue testing the energy storage module for the next test item. In this way, based on the different disturbances caused by different test items to the energy storage modules, the system accurately determines the waiting time required for the disturbance to disappear, eliminating the need for passive waiting to adapt to the maximum disturbance dissipation time. Furthermore, by combining stability condition determination with test results, this system effectively avoids problems such as data deviation, result failure, and module detection malfunctions caused by continuing testing when module parameters are unstable or test results are unqualified.

Description

Technical Field

[0001] This invention relates to the field of testing technology, specifically to a multi-station parallel EOL integrated testing system for energy storage modules. Background Technology

[0002] With the large-scale development of the energy storage industry, energy storage modules, as the core unit of energy storage systems, directly determine the operational safety and service life of energy storage power stations through the consistency and reliability of their factory performance and safety indicators. EOL comprehensive testing is the core quality inspection process at the end of the energy storage module manufacturing process. It is a key link to intercept module performance defects and avoid subsequent operation and maintenance risks. In order to adapt to the needs of mass production, the industry generally adopts a multi-station parallel EOL comprehensive testing mode. By operating multiple stations simultaneously, the batch testing efficiency of energy storage modules is greatly improved, meeting the needs of large-scale factory testing. The multi-station parallel EOL comprehensive testing of energy storage modules is not a single testing procedure, but a systematic and comprehensive testing process composed of multiple performance and safety test items. It mainly covers multiple test contents such as module voltage acquisition, internal resistance testing, insulation withstand voltage testing, charge and discharge performance testing, temperature consistency testing, and communication function verification. In actual testing operations, due to the constraints of testing principles, electrical characteristics, and testing standards, each test item must be executed in a strictly preset fixed order. It cannot be carried out in a disorderly or parallel manner, and has a very strong process sequence. At the same time, after completing several test items, the electrical parameters and physical state of the module, such as voltage, current, and temperature, are still in a state of disturbance and fluctuation, and cannot quickly return to a stable steady state. If subsequent tests are carried out directly, it is very easy to cause test data deviation, test result failure, or even problems such as abnormal testing equipment and module testing failure. Therefore, in the existing EOL comprehensive testing process, a fixed-length waiting period is set after several test items are completed and before the next stage of testing is started. This period is used to wait for the disturbances of various parameters of the energy storage module to completely dissipate and the equipment operating conditions to return to stability, so as to ensure the detection accuracy and testing safety of subsequent test items. However, in the existing multi-station parallel EOL testing system, the fixed waiting period is only used as a simple working condition rest buffer period. It is idle and wasteful throughout. The system only passively waits for the time to end and does not reuse the idle time resource in any way, resulting in a waste of test timing resources. Moreover, the disturbance dissipation time is different for different test items. The existing fixed duration is set to be too long to adapt to the maximum disturbance dissipation time of all test items. Most modules that have completed steady-state recovery in most working conditions still need to wait passively, which will cause a lot of invalid test time loss, lengthen the overall test cycle of a single module, and offset some of the efficiency advantages of multi-station parallel testing. Especially in large-scale production scenarios, it will significantly reduce the throughput of the entire test production line, increase production energy consumption and equipment idling loss, and raise the quality inspection cost of the module at the factory. Meanwhile, the EOL comprehensive test of energy storage modules will generate a massive amount of raw test data, including various test parameters, operating condition data, abnormal logs, equipment operation data, etc. In accordance with the industry requirements of production traceability, quality control and cloud big data analysis, the above test data must be completely uploaded to the cloud service terminal for storage, archiving, analysis and traceability management. At present, in the industry's conventional test architecture, cloud data transmission operations can only be executed centrally after the entire EOL comprehensive test process of a single energy storage module is completed, and cannot be carried out in the test intervals. To address the above problems, this invention proposes a solution. Summary of the Invention

[0003] The purpose of this invention is to provide a multi-station parallel EOL comprehensive testing system for energy storage modules, in order to solve the problems mentioned in the background art.

[0004] This invention provides a multi-station parallel EOL comprehensive testing system for energy storage modules, comprising: The module testing module is used to control each test station to conduct EOL comprehensive testing on the energy storage modules that are already in place in a preset test sequence. The EOL comprehensive test includes several test items. Each test station runs the pre-configured test script corresponding to each test item according to the preset test sequence to test the energy storage modules that are already in place in a preset test sequence. The module testing module is also used to perform a data acquisition operation on any energy storage module based on several pre-selected benchmark parameters for the test item when performing any test item test on any energy storage module. Based on the execution result, it generates test data of the energy storage module for the test item. At the same time, after the data acquisition operation is completed, it determines in real time whether the energy storage module meets the stability conditions. Based on the determination result and the test result of the energy storage module in the test item fed back by the test data by the test result determination module, it determines whether to continue to perform the next test item test on the energy storage module. The test result determination module is used to output the test result of the energy storage module in any test item after receiving the test data of any energy storage module in any test item. The test result is divided into two types: qualified and unqualified. The module testing module is also used to perform a fixed number of similarity matching iterations on all the comparison datasets stored, based on the test data of all the test items that the energy storage module has completed, after completing a fixed number of test items on any energy storage module. It outputs a unique comparison dataset as the adaptation dataset of the energy storage module and selects a number of test items from all the test items that the energy storage module has not yet been tested. Subsequently, whenever the selected test item of the energy storage module is completed, a priority transmission operation is performed to select a number of test data from the test data of all the test items that have been tested and transmit them first. The data analysis module is used to determine whether any energy storage module meets the sample analysis conditions after all test items have been completed, and to select and store the test analysis data of the energy storage module based on the determination result. The data analysis module is also used to analyze test analysis data according to preset analysis steps after the amount of data stored in it reaches a preset fixed amount, generating several comparison datasets.

[0005] Furthermore, it also includes a cloud service terminal, which is used to store the test data received from any energy storage module for any test item.

[0006] Furthermore, when performing any test item on any energy storage module that is already in place at any test station, a data acquisition operation is performed on the energy storage module to generate the test data of the energy storage module for the test item, as follows: The module test module obtains several benchmark parameters pre-selected by the administrator for the test item, and uses the start time of the test item test as the start time of the acquisition cycle corresponding to the acquisition, and the end time of the test item test completion as the end time of the acquisition cycle corresponding to the acquisition, to determine the acquisition cycle for the energy storage module to perform the acquisition. During the acquisition period, the parameter values ​​of all pre-selected benchmark parameters for the test item are acquired in real time, and the test data of the energy storage module for the test item are generated based on these values.

[0007] Furthermore, the determination of whether the energy storage module meets the stability conditions at any time after the completion of the acquisition operation is as follows: Based on the test indicators and test scripts pre-selected by the management personnel, the parameter values ​​of all state parameters at the specified time are collected, and the parameter value of each state parameter collected at the specified time is compared with its corresponding preset stable threshold range. If and only if the parameter values ​​of all state parameters at the specified time simultaneously fall within their respective preset stable threshold ranges, it is determined that the energy storage module has reached a stable condition at the specified time. At this time, the time period from the time when the first acquisition operation is completed to the specified time is defined as the static buffer period of the energy storage module in the test item, and the difference between the specified time and the time when the first acquisition operation is completed is taken as the waiting time of the static buffer period. If the value of any state parameter collected at the specified time does not fall within its corresponding preset stability threshold range, it is determined that the energy storage module at the specified time does not meet the stability condition. At this time, the parameter values ​​of all state parameters are collected in chronological order at the next time until the energy storage module at a certain time meets the stability condition.

[0008] Furthermore, based on the test data of all completed test items of the energy storage module, a fixed number of similarity matching iterations are performed on all stored comparison datasets, and a unique comparison dataset is output as the adaptation dataset of the energy storage module, as follows: The first round of similarity matching iteration is carried out, which includes: obtaining the test data of the energy storage module in test item A1, and simultaneously obtaining the data capacity of the test data; The test data is compared with the comparison data of test item A1 in all the stored comparison datasets. The comparison dataset to which all the comparison data with the test data have a similarity greater than or equal to the preset similarity comparison threshold belongs is taken as the output of the first round of similarity matching iteration. This completes the first round of similarity matching iteration. Based on the above, the 2nd, 3rd, ..., g-1th similarity matching iterations are performed sequentially, and the test data of the energy storage module in test items A2, A3, ..., Ag-1 are obtained sequentially. For each similarity matching iteration, the similarity between the obtained test data and the output of the previous similarity matching iteration is calculated. This process continues until the g-1th similarity matching iteration is completed. The output of the g-1th similarity matching iteration is used as the adaptation dataset of the energy storage module, where g is a pre-selected fixed quantity.

[0009] Furthermore, after each selected test item of the energy storage module is completed, a priority transmission operation is performed. The following data is selected from the test data of all completed test items and prioritized for transmission to the cloud service terminal: Obtain the predicted matching duration configured for the test item within the adaptation dataset, and determine the maximum data capacity that can be transmitted at present, taking into account the current network bandwidth and wireless transmission environment. The system acquires all test data and their data sizes for all test items currently collected by the energy storage module but not yet transmitted to the cloud service terminal. It then accumulates the data sizes of the corresponding test data in descending order of their data sizes. When the accumulated data size is within the maximum transmittable data size, the test data corresponding to all accumulated data sizes is transmitted to the cloud service terminal. If the accumulated data size for a certain test item exceeds the maximum data size, the test data for that test item is not accumulated or transmitted; only the test data corresponding to all accumulated data sizes that do not exceed the maximum data size is transmitted.

[0010] Furthermore, when any energy storage module completes all test items, the data analysis module determines whether it meets the sample analysis conditions. If it does, the specific content of the test analysis data for the energy storage module is collected and generated as follows: If any energy storage module passes all test items after completing all test items, then the test data of the energy storage module in all test items, as well as the waiting time of the energy storage module in the static buffer period of each test item, are obtained and stored to generate test analysis data of the energy storage module.

[0011] Furthermore, the data analysis module analyzes and generates several comparison datasets, and the analysis steps are as follows: S11: In the order of execution, label all test items included in the EOL comprehensive test as A1, A2, ..., Aa, where a is the total number of test items included in the EOL comprehensive test; S12: Label all stored test analysis data as B1, B2, ..., Bb, where b is the total number of stored test analysis data; S13: Create a similarity tag set for test item A1. First, select the test data of test item A1 in test analysis data B1 as analog data in ascending order of tag index. Initially, the similarity tag set is an empty set. Calculate the similarity between the analog data and the test data of test item A1 in test analysis data B2, B3, ..., Bb respectively. Determine whether test analysis data B2, B3, ..., Bb meet the addition conditions based on the similarity. Add all test analysis data that meet the addition conditions to the similarity tag set and establish the association between the similarity tag set and the test analysis data containing the test data selected as analog data, i.e., test analysis data B1. S14: According to the order of the mark subscripts from small to large, among the test data of test item A1 in the test analysis data B2, B3, ..., Bb, the test data that has not been selected as analogy data and has not been added to any of the similar mark sets of test item A1 is selected as analogy data. Each time one test data is selected as analogy data, a similar mark set of test item A1 is created, and all test analysis data that meet the addition condition are added to said similar mark set according to step S13. Based on the above, several similar mark sets of test item A1 are obtained; S15: According to the preset generation steps, f similar mark sets of test items A2, A3, ..., Aa are obtained in sequence; S16: Acquire the test analysis data corresponding to the f similar mark sets of test items A1, A2, ..., Aa and perform deduplication thereon, and mark all the remaining test analysis data after deduplication as G1, G2, ..., Gf respectively; S17: Construct one comparison data set I1 of the energy storage module according to the test analysis data G1, and the construction content is as follows: SS31: Extract the similar mark sets corresponding to the test analysis data G1 from the f similar mark sets of test items A1, A2, ..., Aa in sequence, and mark them correspondingly as H1, H2, ..., Ha; SS32: Traverse the marks contained in the similar mark sets H1, H2, ..., Ha. For each mark that is present in all of the similar mark sets H1, H2, ..., Ha, acquire the test analysis data corresponding to said mark. From all the acquired test data of test items A1, A2, ..., Aa in the obtained test analysis data, select one piece of test data respectively as the comparison data of test items A1, A2, ..., Aa, add them all to an empty set, and define the set after the addition is completed as one comparison data set of the energy storage module, marked as I1; S18: According to the preset configuration rule, configure the predicted matching duration for the comparison data of several test items among test items Ag, Ag+1, ..., Aa-1 in the comparison data set I1, where 1 ≤ g < a, g is a preselected fixed value, and the configuration rule is specifically as follows: Extract the waiting duration of test item Ag from all the test analysis data obtained in step SS32, and use a discrete point filtering algorithm to perform data processing on all the extracted waiting durations of test item A1; obtain the average value of the waiting durations remaining after the data processing, compare the average value with the preset minimum allowable reuse duration in terms of size, if the average value is greater than or equal to the minimum allowable reuse duration, configure the average value as the predicted matching duration of test item Ag in the comparison data set I1, otherwise no operation is performed; Configure the prediction matching duration for test items Ag+1, ..., Aa-1 within the comparison dataset as described above; S19: Based on test analysis data G2, ..., Gf, construct f-1 comparison datasets for the energy storage module according to S17 to S18, and configure the prediction matching time for the comparison data of several test items within it.

[0012] Furthermore, in S15, the steps for generating the f similarity tag sets of test items A2, A3, ..., Aa in sequence are as follows: SS21: Obtain all similarity tag sets for test item A1, and label them as F1, F2, ..., Ff, where f is the total number of similarity tag sets for test item A1 created through steps S13 to S14; SS22: Create a similarity tag set for test item A2, select the test data of test item A2 within the test analysis data corresponding to the similarity tag set F1 as analog data, and establish an association between the similarity tag set and the test analysis data; SS23: Calculate the similarity between the analogy data in SS22 and the test data of test item A2 in the test analysis data corresponding to each marker in the similarity marker set F1. Determine whether each marker meets the addition condition based on the similarity. Add all markers that meet the addition condition to the similarity marker set of test item A2 created in SS22. The content for determining whether any marker meets the addition condition is as follows: If the similarity between the test data of test item A2 in the test analysis data corresponding to the marker and the analogy data is greater than or equal to the preset similarity percentage threshold, then the marker is determined to meet the addition conditions; otherwise, the marker is determined not to meet the addition conditions. SS24: Following SS21 to SS22, create a similar tag set for test item A2 in sequence, creating a total of f-1 similar tag sets. For each similar tag set created for test item A2, establish an association relationship between it and the corresponding test analysis data. Select the corresponding similar tag set from similar tag sets F2, F3, ..., Ff in ascending order of tag index. Select the test data of test item A2 in the test analysis data corresponding to the similar tag set as analog data. Then, according to SS23, determine whether all tags in the similar tag set meet the conditions. Based on the determination results, add all tags in the similar tag set that meet the addition conditions to the corresponding created similar tag set for test item A2. SS25: Obtain f similarity marker sets for test items A3, A4, ..., Aa in sequence from SS21 to SS24.

[0013] Furthermore, in SS32, the comparison data for selected test item A1 is as follows: Test data for test item A1 is extracted from all the acquired test analysis data. One test data is randomly selected as the baseline data. The similarity between the baseline data and each test data of the extracted test item A1 is calculated. The mean value is calculated using the summation and averaging formula. The mean value is used as the average similarity when the test data is selected as the baseline data. Similarly, each test data of the extracted test item A1 is selected as the baseline data, and the average similarity when it is selected as the baseline data is calculated. All extracted test data were selected as baseline data and the corresponding average similarity was calculated. The test data with the highest average similarity was selected as the baseline data and used as the comparison data for test item A1.

[0014] Compared with existing technologies, it has the following advantages: This invention employs a module testing module at each testing station to conduct EOL (Effective End-of-Life) comprehensive testing on pre-positioned energy storage modules. During the testing process, test data for each energy storage module is collected for each test item. Simultaneously, the stability conditions of the energy storage modules are determined in real time. Based on the determination results and the test results of the energy storage modules for each test item fed back by the test data by the test result determination module, it is determined whether to continue testing the energy storage module for the next test item. In this way, based on the differentiated state disturbances (such as cell polarization, thermal inertia, and residual charge) caused by different test items to the energy storage modules, the waiting time required for the disturbance to disappear can be accurately determined. It eliminates the need for passive waiting to adapt to the maximum disturbance dissipation time, effectively eliminating a large amount of invalid test time loss, significantly shortening the overall test cycle of a single energy storage module, fully releasing the efficiency advantages of multi-station parallel testing, significantly improving the throughput capacity of the test production line in large-scale mass production scenarios, reducing equipment idling losses and production energy consumption, and reducing module factory quality inspection costs. At the same time, through the dual verification logic of stability condition judgment combined with test results, it completely avoids problems such as data deviation, result failure, equipment abnormality and module detection failure caused by continuing to test under unstable module parameters or unqualified test conditions, and comprehensively ensures the detection accuracy and operational safety of EOL comprehensive testing. This invention, after completing a fixed number of test items on any energy storage module, performs a fixed number of rounds of similarity matching iterations on all stored comparison datasets based on the test data of all completed test items for the energy storage module. A unique comparison dataset is output as the adaptation dataset for the energy storage module. Based on this dataset, several test items are selected from all untested test items for the energy storage module. Subsequently, after each selected test item is completed, a priority transmission operation is performed, selecting several test data points from all completed test items for priority transmission. This method avoids the need to upload data centrally only after all module testing is completed. This approach effectively utilizes idle time-series resources during the static buffer phases of several test items to achieve fragmented data transmission during test intervals. Simultaneously, it dynamically adapts data transmission volume by combining predicted matching time for the matching dataset, real-time network bandwidth, and wireless transmission environment. Test data with suitable transmission capacity is selected and uploaded to the cloud. Without occupying test job time or affecting test process efficiency, it achieves batch, efficient, and orderly uploading of massive test data, meeting industry needs for production traceability, quality control, and real-time cloud big data analysis. This improves the timeliness of energy storage module test data archiving, traceability, and data analysis, maximizing the utilization of test time-series resources and eliminating resource waste during idle test periods. Attached Figure Description

[0015] Figure 1 This is a system block diagram of the present invention. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] Please see Figure 1 This application provides a multi-station parallel EOL comprehensive testing system for energy storage modules, including a module testing module, a test result judgment module, a data analysis module, and a cloud service terminal; The module testing module is used to control each test station to conduct EOL comprehensive testing on the energy storage module that has been placed in place in a preset test order. The EOL comprehensive test includes several test items, each test item corresponds to a test index and a pre-configured test script. Each test station runs the test script of each test item in a preset test order to perform the corresponding test item test on the energy storage module that has been placed in place. In this application, the test sequence is preset by the management personnel based on the test principles, electrical characteristics and testing standard constraints of all test items included in the EOL comprehensive test, in order to indicate the test sequence of each test item in the EOL comprehensive test. Each test station executes the test script of the corresponding test item in sequence according to the test sequence to complete the corresponding test index test of the energy storage module. In this application, several test stations are set up to conduct EOL comprehensive testing on the energy storage modules that have been delivered to and placed therein in parallel. In this application, the test indicators are divided into four categories: electrical indicators, thermal performance indicators, high voltage safety indicators, and BMS functional indicators. Among them, the module total voltage, individual cell series voltage, cell voltage difference, open circuit voltage (OCV), charge and discharge capacity, charge and discharge efficiency, loop impedance, and residual voltage belong to the electrical indicators. Overall module temperature, individual cell temperature, temperature difference, temperature change rate, and temperature rise rate are classified as thermal performance indicators; insulation resistance, withstand voltage performance, high-voltage circuit switching characteristics, contactor and relay operation status are classified as high-voltage safety indicators; BMS sampling accuracy, SOC calculation error, protection threshold, fault alarm function, and register status are classified as BMS functional indicators. The module testing module is also used to perform a data acquisition operation on any energy storage module that is already in place at any test station, based on several pre-selected benchmark parameters for the test item. After the acquisition operation is completed, it generates test data for the energy storage module for the test item based on all the data acquired during the operation. Furthermore, it determines in real time whether the energy storage module meets the stability conditions after the acquisition operation is completed, until the energy storage module meets the stability conditions at a certain point. At this point, the test result determination module returns the result based on the test data. The test results of the aforementioned test items determine whether to proceed to the next test item for the energy storage module. Specifically: if the test result is qualified, it is determined that the energy storage module will proceed to the next test item, and the corresponding test station will conduct the next test item test on the energy storage module. If the test result is unqualified, it is determined that the energy storage module will stop the next test item test, and the corresponding test station will stop the further testing of the energy storage module and output an alarm command to the management personnel. The management personnel will then handle the energy storage module, including fault verification, anomaly isolation, defect location, rework and debugging, retesting and verification, and scrapping determination. The module testing module transmits the test data to the test result determination module; In this application, the module test module stores several benchmark parameters selected by the administrator for each test item based on the test indicators corresponding to each test item in the EOL comprehensive test, including but not limited to the total voltage of the energy storage module, the voltage of each individual string, the voltage difference between strings, the open circuit voltage, the charging and discharging current, the charging and discharging capacity, the charging and discharging efficiency, the circuit impedance, the residual voltage, the temperature of each monitoring point of the module, the temperature difference, the temperature change rate, the insulation resistance value, the withstand voltage leakage current, the high voltage circuit on / off status, the contactor action feedback signal, the BMS sampling voltage, the BMS sampling temperature, the actual calculated value of SOC, the BMS fault flag bit, etc. When performing any test item on any energy storage module that is already in place at any test station, a data acquisition operation is performed on the energy storage module, and the test data of the energy storage module for the test item is generated as follows: The module test module obtains several benchmark parameters pre-selected by the administrator for the test item, and uses the start time of the test item test as the start time of the acquisition cycle corresponding to the acquisition, and the end time of the test item test completion as the end time of the acquisition cycle corresponding to the acquisition. The acquisition cycle for the energy storage module to perform the acquisition is determined. It should be noted that the acquisition cycle is determined only to make the content of the test data generation clearer. In fact, the end time can only be determined when the test item test is completed. During the acquisition period, the parameter values ​​of all pre-selected benchmark parameters for the test item are acquired in real time, and the test data of the energy storage module for the test item are generated based on them. The test result determination module is used to, at any test station, after completing each test item test of the energy storage module in the test order, output the test result of the energy storage module in the test item based on the received test data of the energy storage module in the test item, and transmit the test result to the module test module. The test result determination module has a set of qualified thresholds configured for each test item in the EOL comprehensive test, wherein the test result is divided into qualified and unqualified, and the set of qualified thresholds contains a preset qualified threshold range of several benchmark parameters. Based on the received test data of any energy storage module in any test item, the specific content of the test result of the energy storage module in that test item is as follows: For each parameter value of each benchmark parameter in the test data, compare it with the preset qualified threshold range of the benchmark parameter; If and only if all parameter values ​​of all benchmark parameters in the test data simultaneously fall within their respective preset qualified threshold ranges, then the test result of the energy storage module in the test item is output as qualified, indicating that the energy storage module has passed the test item; otherwise, the test result of the energy storage module in the test item is output as unqualified, indicating that the energy storage module has not passed the test item. The following is the content of determining whether the energy storage module meets the stability conditions at any time after the completion of the data acquisition operation: Based on the test indicators and test scripts pre-selected by the management personnel, the parameter values ​​of all state parameters at the specified time are collected, and the parameter value of each state parameter collected at the specified time is compared with its corresponding preset stable threshold range. If and only if the parameter values ​​of all state parameters at the specified time simultaneously fall within their respective preset stable threshold ranges, it is determined that the energy storage module has reached a stable condition at the specified time. At this time, the time period from the time when the first acquisition operation is completed to the specified time is defined as the static buffer period of the energy storage module in the test item, and the difference between the specified time and the time when the first acquisition operation is completed is taken as the waiting time of the static buffer period. If the value of any state parameter collected at the specified time does not fall within its corresponding preset stability threshold range, it is determined that the energy storage module at the specified time does not meet the stability condition. At this time, the parameter values ​​of all state parameters are collected in chronological order at the next time until the energy storage module at a certain time meets the stability condition. In this application, the administrator pre-selects several state parameters for each test item in the EOL comprehensive test and sets a corresponding stable threshold range. After setting, the parameters are pre-input into the test result judgment module for storage. If, after a preset maximum data collection period has elapsed since the completion of the previous test, the energy storage module fails to meet the stability conditions at any given time, then the next test item for the energy storage module will be stopped. The corresponding test station will cease testing the energy storage module and issue an alarm command to the management personnel, who will then handle the issue. It should be noted that, in the test order, once the last test item is completed, the energy storage module will no longer be judged on whether to continue to the next test item. It should be noted that after any data acquisition operation is completed on any energy storage module, the stability condition is determined in real time to eliminate the state disturbances caused by the test items to the energy storage module. The state disturbances include at least one of the following: cell polarization effect, module thermal inertia, residual charge in the circuit, BMS register refresh lag, residual arc in the high voltage circuit, and transient disturbance of fixture contact. The module testing module is further configured to determine the total number of test items that have been tested for the energy storage module after each test item is completed for any energy storage module. If the total number is less than or equal to a preset fixed number, no processing is performed. If the total number is equal to the fixed number, a fixed number of similarity matching iterations are performed on all the comparison datasets stored based on the test data of all the test items that the energy storage module has completed. In each round of similarity matching iteration, several comparison datasets are selected from all the comparison datasets stored as the output of the corresponding round of similarity matching iteration, until the last round of similarity matching iteration is completed and a unique comparison dataset is output as the adaptation dataset of the energy storage module. Based on the adaptation dataset, select several test items from all test items that have not yet been tested for the energy storage module. Subsequently, after each test of any selected test item of the energy storage module is completed, perform a priority transmission operation to select several test data from the test data of all completed test items and transmit them to the cloud service terminal first, where the fixed amount is g. Based on the test data of all completed test items of the energy storage module, a fixed number of similarity matching iterations are performed on all stored comparison datasets. In each round of similarity matching iteration, several comparison datasets are selected from all stored comparison datasets as the output of the corresponding round of similarity matching iteration, until the last round of similarity matching iteration is completed and a unique comparison dataset is output as the adaptation dataset of the energy storage module. The content is as follows: The first round of similarity matching iteration is carried out, which includes: obtaining the test data of the energy storage module in test item A1, and simultaneously obtaining the data capacity of the test data; The test data is compared with the comparison data of test item A1 in all the stored comparison datasets. The comparison dataset to which all the comparison data with the test data have a similarity greater than or equal to the preset similarity comparison threshold belongs is taken as the output of the first round of similarity matching iteration. This completes the first round of similarity matching iteration. Based on the above, the 2nd, 3rd, ..., g-1th similarity matching iterations are performed sequentially, corresponding to the acquisition of test data of the energy storage module in test items A2, A3, ..., Ag-1. For each similarity matching iteration, the similarity between the acquired test data and the output of the previous similarity matching iteration is calculated. This process continues until the g-1th similarity matching iteration is completed. The output of the g-1th similarity matching iteration is used as the adaptation dataset of the energy storage module. During the g-1th similarity matching iteration, if the number of comparison datasets to which all comparison data belong is greater than or equal to 1 if the similarity calculated with the test data of test item Ag-1 is greater than or equal to the preset similarity comparison threshold, then the comparison dataset with the largest similarity calculated with the test data of test item Ag-1 is selected as the output of the g-1th similarity matching iteration. Subsequently, after each selected test item of the energy storage module is completed, a priority transmission operation is performed. The following test data is selected from the test data of all completed test items and prioritized for transmission to the cloud service terminal: Obtain the predicted matching duration configured for the test item within the adaptation dataset, and obtain the current network bandwidth and wireless transmission environment. Based on the predicted matching duration, the network bandwidth, and the wireless transmission environment, determine the maximum data capacity that can be transmitted at present. The system acquires all test data and their data sizes for all test items currently collected by the energy storage module but not yet transmitted to the cloud service terminal. It then accumulates the data sizes of the corresponding test data in descending order of their data sizes. When the accumulated data size is within the maximum transmittable data size, the test data corresponding to all accumulated data sizes is transmitted to the cloud service terminal. If the accumulated data size for a certain test item exceeds the maximum data size, the system does not accumulate or transmit the test data for that test item; instead, it only transmits the test data corresponding to all accumulated data sizes that do not exceed the maximum data size. The network bandwidth refers to the maximum theoretical data transmission rate that the communication link between the current test station and the cloud service terminal can carry per unit time; the wireless transmission environment refers to the set of dynamic factors that affect data transmission efficiency, including at least the real-time load rate, signal strength and packet loss rate of the current link. The module testing module is used to transmit all remaining test data that have not been transmitted to the cloud service terminal to the cloud service terminal after any energy storage module has completed the collection of test data for all test items. In this application, several test stations independently and in parallel perform EOL comprehensive testing, and each test station performs each test item according to the same set of preset test order. The data analysis module is used to determine whether any energy storage module meets the sample analysis conditions after completing all test items. If it does, the module's test analysis data is collected and stored. Specifically: If any energy storage module passes all test items after completing all test items, then the test data of the energy storage module in all test items, as well as the waiting time of the energy storage module in the static buffer period of each test item, are obtained and stored to generate test analysis data of the energy storage module. The data analysis module is used to analyze test and analysis data from a fixed number of energy storage modules according to preset analysis steps, generating several comparison datasets. The analysis steps are as follows: S11: In the order of execution, label all test items included in the EOL comprehensive test as A1, A2, ..., Aa, where a is the total number of test items included in the EOL comprehensive test; S12: Label all stored test analysis data as B1, B2, ..., Bb, where b is the total number of stored test analysis data; S13: Create a similarity tag set for test item A1. First, select the test data of test item A1 in test analysis data B1 as analog data in ascending order of tag index. Initially, the similarity tag set is an empty set. Calculate the similarity between the analog data and the test data of test item A1 in test analysis data B2, B3, ..., Bb respectively. Determine whether test analysis data B2, B3, ..., Bb meet the addition conditions based on the similarity. Add all test analysis data that meet the addition conditions to the similarity tag set and establish the association between the similarity tag set and the test analysis data containing the test data selected as analog data, i.e., test analysis data B1. It should be noted here that the added data is a tag for the corresponding test analysis data, used to identify the corresponding test analysis data; Specifically, for each similarity calculated, the similarity is compared with a preset similarity percentage threshold. If the similarity is greater than or equal to the similarity percentage threshold, it is determined that the test analysis data to which the test data with the similarity calculated from the analogy data belongs meets the addition condition. In this application, the method for calculating the similarity between the analogy data and any test data is as follows: The parameter values ​​of each benchmark parameter in the analog data are normalized so that the parameter values ​​of each benchmark parameter fall within the interval [0, 1]. Similarly, the same processing is performed on the parameter values ​​of each benchmark parameter in the test data. Using formula Calculate the similarity between the analogy data and the test data, where Cc ’ 、Dc ’ The terms refer to the normalized analog data and the parameter value of each benchmark parameter in the test data, respectively. c refers to each benchmark parameter. Since the analog data and the test data are collected by different energy storage modules when conducting test item A1, the benchmark data are consistent. Therefore, c can refer to each benchmark parameter in both the analog data and the test data. d represents all the benchmark parameters pre-selected by the management personnel for test item A1. S14: In order of ascending subscript, select the test data of test item A1 in test analysis data B2, B3, ..., Bb that have not been selected as analog data and have not been added to any similar tag set of test item A1 as analog data. For each selected test data as analog data, create a similar tag set of test item A1 and add all test analysis data that meet the addition conditions to the similar tag set according to step S13. Based on the above, several similar tag sets of test item A1 are obtained. It should be noted that if any test analysis data has been added to a similarity tag set of test item A1, the other similarity tag sets of test item A1 will not be added to it again. In this application, all similarity tag sets of test item A1 are bound to a test analysis data according to S13; S15: Following the preset generation steps, f similarity tag sets for test items A2, A3, ..., Aa are obtained sequentially, with the following contents: SS21: Obtain all similarity tag sets for test item A1, and label them as F1, F2, ..., Ff, where f is the total number of similarity tag sets for test item A1 created through steps S13 to S14; SS22: Create a similarity tag set for test item A2, select the test data of test item A2 within the test analysis data corresponding to the similarity tag set F1 as analog data, and establish an association between the similarity tag set and the test analysis data; SS23: Calculate the similarity between the analogy data in SS22 and the test data of test item A2 in the test analysis data corresponding to each tag in the similarity tag set F1. Determine whether each tag meets the addition conditions based on the similarity. Add all tags that meet the addition conditions to the similarity tag set of test item A2 created in SS22. The criteria for determining whether any marker meets the addition conditions are as follows: If the similarity between the test data of test item A2 in the test analysis data corresponding to the marker and the analogy data is greater than or equal to the preset similarity percentage threshold, then the marker is determined to meet the addition conditions; otherwise, the marker is determined not to meet the addition conditions. SS24: Following SS21 to SS22, create a similar tag set for test item A2 in sequence, creating a total of f-1 similar tag sets. For each similar tag set created for test item A2, establish an association relationship between it and the corresponding test analysis data. Select the corresponding similar tag set from similar tag sets F2, F3, ..., Ff in ascending order of tag index. Select the test data of test item A2 in the test analysis data corresponding to the similar tag set as analog data. Then, according to SS23, determine whether all tags in the similar tag set meet the conditions. Based on the determination results, add all tags in the similar tag set that meet the addition conditions to the corresponding created similar tag set for test item A2. SS25: Obtain f similarity marker sets for test items A3, A4, ..., Aa in sequence from SS21 to SS24; S16: Obtain the test analysis data corresponding to the f similar label sets of test items A1, A2, ..., Aa and remove duplicates from them. Label all the remaining test analysis data after deduplication as G1, G2, ..., Gf respectively. S17: Based on the test analysis data G1, construct a comparison dataset I1 for the energy storage module. The construction content is as follows: SS31: Extract the similarity set corresponding to the test analysis data G1 from the f similarity set of test items A1, A2, ..., Aa in sequence, and label them as H1, H2, ..., Ha; SS32: Traverse the tags contained in the similarity tag sets H1, H2, ..., Ha. For each tag that is contained in the similarity tag sets H1, H2, ..., Ha, obtain the test analysis data corresponding to that tag. From all the test analysis data obtained, select one test data corresponding to each of the test data for test items A1, A2, ..., Aa as the comparison data for test items A1, A2, ..., Aa, and add them to an empty set. Define the set after adding the data as a comparison dataset for the energy storage module, and label it I1. It should be noted that before the comparison data of each test item is added to the collection, the association and binding operation needs to be completed, that is, each test item is associated and bound with its comparison data; The comparison data for selected test item A1 is as follows: Extract the test data of test item A1 from all the acquired test analysis data, randomly select one piece of test data therefrom as reference data, calculate the similarity between the reference data and each piece of the extracted test data of test item A1, calculate the mean value thereof using a summation and averaging formula, and take the mean value as the average similarity when the test data is selected as the reference data. Similarly, sequentially select each piece of the extracted test data of test item A1 as the reference data, and calculate and acquire the average similarity thereof when it is selected as the reference data; After all the extracted test data have been selected as reference data and the corresponding average similarities are calculated, select the test data corresponding to the maximum average similarity when selected as the reference data as the comparison data of test item A1; S18: According to a preset configuration rule, configure predicted matching durations for the comparison data of several test items among test items Ag, Ag+1, ..., Aa-1 in the comparison dataset I1, where 1≤g<a, g is a preselected fixed value, the selection of g is based on ensuring that sufficient preorder samples are available to ensure the accuracy of prediction results, and the configuration rule is specifically as follows: Extract the waiting durations of test item Ag from all the test analysis data acquired in step SS32, and perform data processing on all the extracted waiting durations of test item A1 using a discrete point filtering algorithm; acquire the average value of the remaining waiting durations after data processing, compare the average value with a preset minimum allowable reuse duration, if the average value is greater than or equal to the minimum allowable reuse duration, configure the average value as the predicted matching duration of test item Ag in the comparison dataset I1, otherwise no processing is performed, wherein the minimum allowable reuse duration is set by management personnel based on the transmission duration of the test data of each test item in historical data to the cloud service terminal; Determine whether to configure the predicted matching durations for test items Ag+1, ..., Aa-1 in the comparison dataset according to the above manner; S19: Construct f-1 comparison datasets of the energy storage module according to test analysis data G2, ..., Gf respectively in the manner from S17 to S18, and configure predicted matching durations for the comparison data of several test items therein; The data analysis module transmits all the constructed comparison datasets to the module test module for storage.

[0018] Part of the data in the above formula are all subjected to dimensionality removal for numerical calculation, and all the content not described in detail in this specification belongs to the prior art well known to those skilled in the art.

[0019] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A multi-station parallel EOL comprehensive testing system for energy storage modules, characterized in that, include: The module testing module is used to control each test station to conduct EOL comprehensive testing on the energy storage modules that are already in place in a preset test sequence. The EOL comprehensive test includes several test items. Each test station runs the pre-configured test script corresponding to each test item according to the preset test sequence to test the energy storage modules that are already in place in a preset test sequence. The module testing module is also used to perform a data acquisition operation on any energy storage module based on several pre-selected benchmark parameters for the test item when performing any test item test on any energy storage module. Based on the execution result, it generates test data of the energy storage module for the test item. At the same time, after the data acquisition operation is completed, it determines in real time whether the energy storage module meets the stability conditions. Based on the determination result and the test result of the energy storage module in the test item fed back by the test data by the test result determination module, it determines whether to continue to perform the next test item test on the energy storage module. The test result determination module is used to output the test result of the energy storage module in any test item after receiving the test data of any energy storage module in any test item. The test result is divided into two types: qualified and unqualified. The module testing module is also used to perform a fixed number of similarity matching iterations on all the comparison datasets stored, based on the test data of all the test items that the energy storage module has completed, after completing a fixed number of test items on any energy storage module. It outputs a unique comparison dataset as the adaptation dataset of the energy storage module and selects a number of test items from all the test items that the energy storage module has not yet been tested. Subsequently, whenever the selected test item of the energy storage module is completed, a priority transmission operation is performed to select a number of test data from the test data of all the test items that have been tested and transmit them first. The data analysis module is used to determine whether any energy storage module meets the sample analysis conditions after all test items have been completed, and to select and store the test analysis data of the energy storage module based on the determination result. The data analysis module is also used to analyze test analysis data according to preset analysis steps after the amount of data stored in it reaches a preset fixed amount, generating several comparison datasets.

2. The multi-station parallel EOL comprehensive testing system for energy storage modules according to claim 1, characterized in that, It also includes a cloud service terminal, which is used to store the test data of any energy storage module in any test item after receiving the test data.

3. The multi-station parallel EOL comprehensive testing system for energy storage modules according to claim 1, characterized in that, When performing any test item on any energy storage module that is already in place at any test station, a data acquisition operation is performed on the energy storage module, and the test data of the energy storage module for the test item is generated as follows: The module test module obtains several benchmark parameters pre-selected by the administrator for the test item, and uses the start time of the test item test as the start time of the acquisition cycle corresponding to the acquisition, and the end time of the test item test completion as the end time of the acquisition cycle corresponding to the acquisition, to determine the acquisition cycle for the energy storage module to perform the acquisition. During the acquisition period, the parameter values ​​of all pre-selected benchmark parameters for the test item are acquired in real time, and the test data of the energy storage module for the test item are generated based on these values.

4. The multi-station parallel EOL comprehensive testing system for energy storage modules according to claim 1, characterized in that, The following is the content of determining whether the energy storage module meets the stability conditions at any time after the completion of the data acquisition operation: Based on the test indicators and test scripts pre-selected by the management personnel, the parameter values ​​of all state parameters at the specified time are collected, and the parameter value of each state parameter collected at the specified time is compared with its corresponding preset stable threshold range. If and only if the parameter values ​​of all state parameters at the specified time simultaneously fall within their respective preset stable threshold ranges, it is determined that the energy storage module has reached a stable condition at the specified time. At this time, the time period from the time when the first acquisition operation is completed to the specified time is defined as the static buffer period of the energy storage module in the test item, and the difference between the specified time and the time when the first acquisition operation is completed is taken as the waiting time of the static buffer period. If the value of any state parameter collected at the specified time does not fall within its corresponding preset stability threshold range, it is determined that the energy storage module at the specified time does not meet the stability condition. At this time, the parameter values ​​of all state parameters are collected in chronological order at the next time until the energy storage module at a certain time meets the stability condition.

5. The multi-station parallel EOL comprehensive testing system for energy storage modules according to claim 1, characterized in that, Based on the test data of all completed test items of the energy storage module, a fixed number of similarity matching iterations are performed on all stored comparison datasets, and a unique comparison dataset is output as the adaptation dataset of the energy storage module, as follows: The first round of similarity matching iteration is carried out, which includes: obtaining the test data of the energy storage module in test item A1, and simultaneously obtaining the data capacity of the test data; The test data is compared with the comparison data of test item A1 in all the stored comparison datasets. The comparison dataset to which all the comparison data with the test data have a similarity greater than or equal to the preset similarity comparison threshold belongs is taken as the output of the first round of similarity matching iteration. This completes the first round of similarity matching iteration. Based on the above, the 2nd, 3rd, ..., g-1th similarity matching iterations are performed sequentially, and the test data of the energy storage module in test items A2, A3, ..., Ag-1 are obtained sequentially. For each similarity matching iteration, the similarity between the obtained test data and the output of the previous similarity matching iteration is calculated. This process continues until the g-1th similarity matching iteration is completed. The output of the g-1th similarity matching iteration is used as the adaptation dataset of the energy storage module, where g is a pre-selected fixed quantity.

6. The multi-station parallel EOL comprehensive testing system for energy storage modules according to claim 2, characterized in that, Subsequently, after each selected test item of the energy storage module is completed, a priority transmission operation is performed. The following test data is selected from the test data of all completed test items and prioritized for transmission to the cloud service terminal: Obtain the prediction matching duration configured for the test item within the adaptation dataset, and determine the maximum data capacity that can be transmitted at present by combining the current network bandwidth and wireless transmission environment. The test data and data capacity of all test items collected by the energy storage module that have not yet been transmitted to the cloud service terminal are obtained. The data capacity of the corresponding test data is accumulated in descending order of data capacity. When the accumulated data capacity is within the maximum data capacity that can be transmitted, the test data corresponding to all the accumulated data capacity is transmitted to the cloud service terminal. If the data size accumulated to a certain test item exceeds the maximum data size, then the test data for that test item will not be accumulated or transmitted; only the test data corresponding to all data sizes that have been accumulated and have not exceeded the maximum data size will be transmitted.

7. The multi-station parallel EOL integrated testing system for energy storage modules according to claim 4, characterized in that, When any energy storage module completes all test items, the data analysis module determines whether it meets the sample analysis conditions. If it does, the module collects and generates the specific content of the test analysis data for that energy storage module. If any energy storage module passes all test items after completing all test items, then the test data of the energy storage module in all test items, as well as the waiting time of the energy storage module in the static buffer period of each test item, are obtained and stored to generate test analysis data of the energy storage module.

8. The multi-station parallel EOL comprehensive testing system for energy storage modules according to claim 7, characterized in that, The data analysis module generates several comparison datasets, and the analysis steps are as follows: S11: In the order of execution, label all test items included in the EOL comprehensive test as A1, A2, ..., Aa, where a is the total number of test items included in the EOL comprehensive test. S12: Label all stored test analysis data as B1, B2, ..., Bb, where b is the total number of stored test analysis data; S13: Create a similarity tag set for test item A1. First, select the test data of test item A1 in test analysis data B1 as analog data in ascending order of tag index. Initially, the similarity tag set is an empty set. Calculate the similarity between the analog data and the test data of test item A1 in test analysis data B2, B3, ..., Bb respectively. Determine whether test analysis data B2, B3, ..., Bb meet the addition conditions based on the similarity. Add all test analysis data that meet the addition conditions to the similarity tag set and establish the association between the similarity tag set and the test analysis data containing the test data selected as analog data, i.e., test analysis data B1. S14: According to the order of the mark subscripts from small to large, among the test data of test item A1 in the test analysis data B2, B3, ..., Bb, the test data that has not been selected as analogy data and has not been added to any of the similar mark sets of test item A1 is selected as analogy data. Each time a piece of test data is selected as analogy data, a similar mark set of test item A1 is created, and all test analysis data that meets the addition conditions is added to the similar mark set according to step S13. Based on the above, a number of similar mark sets of test item A1 are obtained; S15: According to the preset generation steps, f similar mark sets of test items A2, A3, ..., Aa are obtained in sequence; S16: Acquire the test analysis data corresponding to the f similar mark sets of test items A1, A2, ..., Aa and perform deduplication thereon, and all the remaining test analysis data after deduplication are respectively marked as G1, G2, ..., Gf; S17: Construct one comparison data set I1 of the energy storage module according to the test analysis data G1, and the construction contents are as follows: SS31: Extract the similar mark sets corresponding to the test analysis data G1 from the f similar mark sets of test items A1, A2, ..., Aa in sequence, and mark them correspondingly as H1, H2, ..., Ha; SS32: Traverse the marks contained in the similar mark sets H1, H2, ..., Ha, and for each mark that is present in all of the similar mark sets H1, H2, ..., Ha, acquire the test analysis data corresponding to said mark. From all the acquired test analysis data, select one piece of test data from all the test data of each of test items A1, A2, ..., Aa correspondingly as the comparison data of test items A1, A2, ..., Aa, add them uniformly to an empty set, and define the set after the addition is completed as one comparison data set of the energy storage module, marked as I1; S18: According to the preset configuration rules, configure predicted matching durations for the comparison data of several test items among test items Ag, Ag+1, ..., Aa-1 in the comparison data set I1, where 1≤g<a, g is a preselected fixed value, and the configuration rules are specifically as follows: Extract the waiting duration of test item Ag from all the test analysis data acquired in step SS32, and use a discrete point filtering algorithm to perform data processing on all the extracted waiting durations of test item A1; obtain the average value of the waiting durations remaining after the data processing, compare the average value with the preset minimum allowable reuse duration in terms of size, if the average value is greater than or equal to the minimum allowable reuse duration, configure the average value in the comparison data set I1 as the predicted matching duration of test item Ag, otherwise no processing is performed; Determine whether to configure the predicted matching duration for test items Ag+1, ..., Aa-1 in the comparison data set according to the above method; S19: According to S17 to S18, construct f-1 comparison data sets of the energy storage module based on the test analysis data G2, ..., Gf respectively, and configure predicted matching durations for the comparison data of several test items therein.

9. A multi-station parallel EOL integrated testing system for energy storage modules according to claim 8, characterized in that, In S15, the steps for generating the f similarity tag sets of test items A2, A3, ..., Aa in sequence are as follows: SS21: Obtain all similarity tag sets for test item A1, and label them as F1, F2, ..., Ff, where f is the total number of similarity tag sets for test item A1 created through steps S13 to S14; SS22: Create a similarity tag set for test item A2, select the test data of test item A2 within the test analysis data corresponding to the similarity tag set F1 as analog data, and establish an association between the similarity tag set and the test analysis data; SS23: Calculate the similarity between the analogy data in SS22 and the test data of test item A2 in the test analysis data corresponding to each marker in the similarity marker set F1. Determine whether each marker meets the addition condition based on the similarity. Add all markers that meet the addition condition to the similarity marker set of test item A2 created in SS22. The content for determining whether any marker meets the addition condition is as follows: If the similarity between the test data of test item A2 in the test analysis data corresponding to the marker and the analogy data is greater than or equal to the preset similarity percentage threshold, then the marker is determined to meet the addition conditions; otherwise, the marker is determined not to meet the addition conditions. SS24: Following SS21 to SS22, create a similar tag set for test item A2 in sequence, creating a total of f-1 similar tag sets. For each similar tag set created for test item A2, establish an association relationship between it and the corresponding test analysis data. Select the corresponding similar tag set from similar tag sets F2, F3, ..., Ff in ascending order of tag index. Select the test data of test item A2 in the test analysis data corresponding to the similar tag set as analog data. Then, according to SS23, determine whether all tags in the similar tag set meet the conditions. Based on the determination results, add all tags in the similar tag set that meet the addition conditions to the corresponding created similar tag set for test item A2. SS25: Obtain f similarity marker sets for test items A3, A4, ..., Aa in sequence from SS21 to SS24.

10. A multi-station parallel EOL comprehensive testing system for energy storage modules according to claim 8, characterized in that, In SS32, the comparison data for selected test item A1 is as follows: Test data for test item A1 is extracted from all the acquired test analysis data. One test data is randomly selected as the baseline data. The similarity between the baseline data and each test data of the extracted test item A1 is calculated. The mean value is calculated using the summation and averaging formula. The mean value is used as the average similarity when the test data is selected as the baseline data. Similarly, each test data of the extracted test item A1 is selected as the baseline data, and the average similarity when it is selected as the baseline data is calculated. All extracted test data were selected as baseline data and the corresponding average similarity was calculated. The test data with the highest average similarity was selected as the baseline data and used as the comparison data for test item A1.

Images