Data Analysis-Based Performance Testing Methods and Systems for All-Solid-State Batteries

Abstract

This invention discloses a data analysis-based method and system for testing the performance of all-solid-state batteries. This invention relates to the field of all-solid-state battery testing technology, addressing the problems of existing all-solid-state battery performance testing methods, such as the inability to dynamically optimize the testing scheme, susceptibility to anomaly interference, low efficiency, and poor data consistency. The method includes: a sample station confirmation end, which confirms the initial station and target test station of the all-solid-state battery. The target test station is a preset station. The confirmed initial station and target test station are transmitted to the test scheme selection end. This invention determines the optimal test scheme through station positioning, scheme screening, cycle calculation, historical data comparison, hierarchical anomaly elimination, and weighted scoring. Simultaneously, it intelligently schedules and assists in the test sequence, avoiding historical anomaly intervals and improving data consistency and repeatability; hierarchically eliminating invalid anomaly segments reduces invalid test time and improves equipment utilization and test throughput.

Description

Technical Field

[0001] This invention relates to the field of all-solid-state battery testing technology, specifically to an all-solid-state battery performance testing method and system based on data analysis. Background Technology

[0002] All-solid-state batteries have advantages such as high energy density, high safety, and long cycle life, and are the core technology direction for the next generation of power batteries and energy storage batteries. Their performance testing needs to cover key indicators such as charge and discharge, interface impedance, rate performance, cycle stability, thermal stability, and safety performance.

[0003] Existing publicly available technologies only collect basic parameters such as voltage, current, and temperature and perform fixed-process testing. They do not design data analysis logic for the characteristics of all-solid-state batteries, such as interface impedance sensitivity, poor low-temperature performance, and significant abnormal fluctuations. Conventional testing platforms use fixed timing and fixed paths, which cannot dynamically avoid abnormal intervals by combining historical test data. This results in poor test repeatability, long cycles, and low resource utilization, making it difficult to meet the high-efficiency and high-precision testing requirements of R&D and mass production scenarios.

[0004] Existing systems cannot dynamically optimize test schemes and execution sequences based on battery sample status, test station load, and historical abnormal data. This can easily cause samples to enter parameter fluctuation ranges, resulting in data distortion, test failures, and wasted time, thus hindering the industrialization process of all-solid-state batteries. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a data analysis-based method and system for testing the performance of all-solid-state batteries, solving the problems of existing all-solid-state battery performance testing methods, such as the inability to dynamically optimize the solution, susceptibility to abnormal interference, low efficiency, and poor data consistency.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a data analysis-based method and system for testing the performance of all-solid-state batteries, comprising:

[0007] The sample station confirmation end confirms the initial station and target test station for the all-solid-state battery. The target test station is a preset station. The confirmed initial station and target test station are transmitted to the test plan selection end.

[0008] At the test scheme selection end, the battery test parameter map is extracted from the cloud, and the point is calibrated based on the confirmed initial work station and target test work station. Starting from the initial work station and ending at the target test work station, multiple test schemes are determined, and then the scheme to be analyzed is selected from the multiple test schemes.

[0009] The performance data analysis end analyzes the selected solution to be analyzed, determines the test cycle corresponding to different solutions to be analyzed based on the current time, and determines whether there are any performance anomalies in the corresponding solution to be analyzed within the test cycle based on the test cloud data. Based on the anomaly duration of different anomalies, the optimal test solution is determined and transmitted to the test result output end.

[0010] The multi-dimensional performance analysis center determines the current battery position based on the sample station confirmation terminal, then determines the travel time based on the current position and the target positions of each auxiliary test item, and finally determines the optimal auxiliary test item based on the processing data of the corresponding test station and the test cloud data, and displays the results through the test result output terminal.

[0011] Preferably, the method by which the test scheme selection end selects the scheme to be analyzed from multiple sets of test schemes includes:

[0012] Select the shortest test scheme from multiple test schemes, mark it as the benchmark scheme, and determine the benchmark scheme duration Lz;

[0013] The duration values ​​for other test schemes were then determined and denoted as J. i Where i represents different other test schemes, J i Other test schemes with values ​​≤1.5×Lz are designated as candidate schemes, and J is... i Other test schemes with values ​​>1.5×Lz are not calibrated;

[0014] If there are multiple alternative solutions, select the two alternative solutions with the shortest duration and the original baseline solution as the solutions to be analyzed.

[0015] If there is a set of candidate solutions, then this candidate solution and the original benchmark solution will be used as the solutions to be analyzed.

[0016] If no alternative solution exists, this benchmark solution will be used as the optimal test solution and displayed directly through the test result output.

[0017] The determined analysis scheme is transmitted to the performance data analysis terminal.

[0018] Preferably, the specific method by which the performance data analysis terminal determines the optimal test plan includes:

[0019] Determine the current time T1 and the duration L of the scheme to be analyzed. q Where q represents different schemes to be analyzed, and L is used. q ÷Vs=T q Determine the test duration T for different scenarios to be analyzed. q Where Vs is the preset test rate, and then T1+T is used. q=T2 q Determine the end time T2 of the corresponding scheme to be analyzed. q Determine the test cycle [T1, T2] for the corresponding solution to be analyzed. q ];

[0020] The test period [T1, T2] is set. q As a periodic time period, historical data from the previous day's same period period is retrieved from the test cloud data. The historical data consists of parameter data for each test segment of the solution to be analyzed. The performance fluctuation parameter RL for the corresponding test segment at the corresponding time is determined from the data of different test segments. t-k Where t represents different times and k represents different test segments, if RL t-k ≥Y1, where Y1 is a preset value, this moment is marked as an abnormal moment, and the abnormal period of the corresponding test segment is determined based on the abnormal moment;

[0021] Test segments exhibiting abnormal cycles are identified as abnormal segments. The abnormal cycles of all abnormal segments within the analyzed scheme are merged to determine a unified cycle, and [T1, T2] are defined as follows: q Compare with a unified cycle:

[0022] If [T1, T2] q If the time interval is the same as the unified period, the duration JL of the corresponding abnormal segment within the scheme to be analyzed is directly determined. If there are multiple sets of abnormal segments, the durations of the multiple sets of abnormal segments are summed to obtain JL, using the formula: JL×C1+L. q ×C2=PD q Determine the performance score (PD) of the corresponding scheme to be analyzed. q C1 and C2 are both preset coefficient factors.

[0023] Preferably, [T1, T2] q The comparison with the unified cycle also includes:

[0024] If [T1, T2] q Unlike the unified period, the abnormal segment corresponding to the initial abnormal time CS1 of the unified period is determined. Based on the preset Vs, the time T3 for the battery to reach this abnormal segment is determined. If T3 ∈ the unified period, then [T1, T2] is used. q The performance score PD is calculated in the same way as the uniform cycle. q If T3 ∉ a unified period, remove the first group of abnormal segments, then merge the abnormal periods of the remaining abnormal segments to determine the subsequent period, and recalculate the time T4 to reach the new first abnormal segment. If T4 ∈ the subsequent period, calculate the corresponding abnormal segment duration JL and obtain PD. q If T4 ∉ subsequent cycles, then continue to remove the current first abnormal segment, and so on, until PD is completed. q calculate;

[0025] PD from multiple sets of schemes to be analyzed q Among them, the group with the smallest value is selected as the optimal test plan.

[0026] Preferably, the multi-dimensional performance analysis center determines the optimal auxiliary test items in the following way:

[0027] Based on the current battery station and the destination station, determine the shortest path and path distance, and based on the preset speed Vs, determine the travel time SCv, where v represents different auxiliary test stations.

[0028] From the test cloud data, determine the number of new samples Rv per unit time and the number of samples currently in the queue Dv for the corresponding test station, and from the processed data, determine the processing volume Clv per unit time for this test station;

[0029] The waiting time DPv for the corresponding test station is determined by the formula: [Dv + (Rv - Clv) × SCv] ÷ Clv = DPv. The minimum value is selected from multiple waiting times, and the auxiliary test item corresponding to the minimum value is marked as the optimal auxiliary test item.

[0030] A data analysis-based method for testing the performance of all-solid-state batteries includes the following steps:

[0031] The initial testing station and target testing station for the all-solid-state battery were confirmed, with the target testing station being the preset station.

[0032] Extract battery test parameter maps from the cloud, generate multiple test schemes based on the initial workstation and the target test workstation, and filter the schemes to be analyzed according to the duration threshold.

[0033] Determine the current time T1, and calculate the test duration T for each scheme to be analyzed. q The test period [T1, T2] is obtained. q ];

[0034] Retrieve historical test cloud data for the same period, identify abnormal performance parameters for each test segment, determine the abnormal period, and merge them into a unified period;

[0035] The test cycle is compared with a unified cycle, invalid and abnormal segments are eliminated step by step, and the performance score PD of each scheme to be analyzed is calculated. q ;

[0036] Select PD q The minimum solution is selected as the optimal test solution and the test is executed. Based on the current workstation and each auxiliary test workstation, the movement time and waiting time are calculated to determine the optimal auxiliary test sequence and complete the full-dimensional performance test.

[0037] Preferably, the selection of the scheme to be analyzed includes: selecting the scheme with the shortest duration as the benchmark scheme, with a duration of Lz;

[0038] Other options duration J i Options ≤1.5×Lz are considered as candidate options; those exceeding this value are not included.

[0039] When there are multiple candidate schemes, the two shortest schemes plus the benchmark scheme are taken as the schemes to be analyzed.

[0040] If there is a set of candidate solutions, the set plus the benchmark solution is taken as the solution to be analyzed; if there are no candidate solutions, the benchmark solution is taken as the optimal solution.

[0041] Preferably, the step-by-step removal of invalid outliers includes:

[0042] Calculate the time T3 to reach the first anomaly segment. If it falls within a uniform period, calculate the total anomaly duration JL and obtain PD. q ;

[0043] If it does not fall into the category, the first abnormal segment is removed, the remaining abnormal cycles are merged, the time to reach the new first abnormal segment is recalculated, and the judgment is repeated until the scoring is completed.

[0044] Preferably, the optimal auxiliary test item is determined by minimizing the waiting time DPv, which is calculated by the number of queues, the rate of new entries, the processing efficiency, and the movement time.

[0045] This invention provides a data analysis-based method and system for testing the performance of all-solid-state batteries. Compared with existing technologies, it has the following advantages:

[0046] This invention determines the optimal test plan through workstation positioning, plan selection, cycle calculation, historical data comparison, step-by-step anomaly elimination, and weighted scoring. Simultaneously, intelligent scheduling assists in the test sequence, resulting in the following beneficial effects: 1. Dynamically optimizing test plans, avoiding historical anomaly intervals, and improving data consistency and repeatability; 2. Step-by-step elimination of invalid anomaly segments, reducing invalid test time, and improving equipment utilization and test throughput; 3. Multi-dimensional collaborative scheduling, covering all performance indicators, adapting to R&D and mass production scenarios; 4. Full-process automation, reducing human intervention, and improving test stability and reliability. Attached Figure Description

[0047] Figure 1 This is a schematic diagram of the principle framework of the present invention. Detailed Implementation

[0048] 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.

[0049] A data-driven all-solid-state battery performance testing system includes:

[0050] The sample station confirmation end confirms the initial station and target test station for the all-solid-state battery. The target test station is a preset station. The confirmed initial station and target test station are transmitted to the test plan selection end.

[0051] The test scheme selection end extracts battery test parameter maps from the cloud and performs point calibration based on the confirmed initial and target test stations. Starting from the initial station and ending at the target station, multiple test schemes are determined, and then the scheme to be analyzed is selected from these schemes. The method by which the test scheme selection end selects the scheme to be analyzed from the multiple test schemes includes: selecting the scheme with the shortest duration from the multiple test schemes, calibrating it as the baseline scheme, and determining the duration Lz of the baseline scheme; then determining the duration values ​​of the other test schemes and calibrating them as J. i Where i represents different other test schemes, J i Other test schemes with values ​​≤1.5×Lz are designated as candidate schemes, and J is... i Other test schemes with a duration greater than 1.5×Lz are not calibrated; if there are multiple candidate schemes, the two candidate schemes with the shortest durations and the original benchmark scheme are selected as the schemes to be analyzed; if there is only one candidate scheme, the candidate scheme and the original benchmark scheme are used as the schemes to be analyzed; if there is no candidate scheme, the benchmark scheme is directly used as the optimal test scheme and displayed directly through the test result output terminal; the determined scheme to be analyzed is transmitted to the performance data analysis terminal.

[0052] The performance data analysis terminal analyzes the selected analysis scheme, determines the test period corresponding to different analysis schemes based on the current time, and determines whether there are performance anomalies in the corresponding analysis scheme within the current test period based on test cloud data. Based on the anomaly duration of different anomalies, the optimal test scheme is determined and transmitted to the test result output terminal. The specific method by which the performance data analysis terminal determines the optimal test scheme includes: determining the current time T1 and determining the duration value L of the analysis scheme. q Where q represents different schemes to be analyzed, and L is used. q ÷Vs=T q Determine the test duration T for different scenarios to be analyzed. q Where Vs is the preset test rate, and then T1+T is used. q =T2 q Determine the end time T2 of the corresponding scheme to be analyzed. q Determine the test cycle [T1, T2] for the corresponding solution to be analyzed. q ]; Set the test period [T1, T2] qAs a periodic time period, historical data from the previous day's same period period is retrieved from the test cloud data. The historical data consists of parameter data for each test segment of the solution to be analyzed. The performance fluctuation parameter RL for the corresponding test segment at the corresponding time is determined from the data of different test segments. t-k Where t represents different times and k represents different test segments, if RL t-k ≥Y1, where Y1 is a preset value, this moment is marked as an abnormal moment, and the abnormal period of the corresponding test segment is determined based on the abnormal moment; the test segment with an abnormal period is marked as an abnormal segment, and the abnormal periods of all abnormal segments in this analysis scheme are merged to determine a unified period, and [T1, T2] are used as the basis for determining the abnormal period. q Compare with a uniform period: If [T1, T2] q If the time interval is the same as the unified period, the duration JL of the corresponding abnormal segment within the scheme to be analyzed is directly determined. If there are multiple sets of abnormal segments, the durations of the multiple sets of abnormal segments are summed to obtain JL, using the formula: JL×C1+L. q ×C2=PD q Determine the performance score (PD) of the corresponding scheme to be analyzed. q Where C1 and C2 are preset coefficient factors. Where [T1, T2] q The comparison with the unified period also includes: if [T1, T2] q Unlike the unified period, the abnormal segment corresponding to the initial abnormal time CS1 of the unified period is determined. Based on the preset Vs, the time T3 for the battery to reach this abnormal segment is determined. If T3 ∈ the unified period, then [T1, T2] is used. q The performance score PD is calculated in the same way as the uniform cycle. q If T3 ∉ a unified period, remove the first group of abnormal segments, then merge the abnormal periods of the remaining abnormal segments to determine the subsequent period, and recalculate the time T4 to reach the new first abnormal segment. If T4 ∈ the subsequent period, calculate the corresponding abnormal segment duration JL and obtain PD. q If T4 ∉ subsequent cycles, then continue to remove the current first abnormal segment, and so on, until PD is completed. q Calculation; from multiple sets of PDs of the schemes to be analyzed q Among them, the group with the smallest value is selected as the optimal test plan.

[0053] The multi-dimensional performance analysis center determines the current battery station based on the sample station confirmation terminal, then determines the travel time based on the current station and the destination station of each auxiliary test item, and finally determines the optimal auxiliary test item based on the processing data of the corresponding test station and the test cloud data, and displays the results through the test result output terminal. The specific method for determining the optimal auxiliary test item by the multi-dimensional performance analysis center is as follows: Based on the current battery station and the destination station, the shortest path and path distance are determined; based on the preset speed Vs, the travel time SCv is determined, where v represents different auxiliary test stations; from the test cloud data, the number of new samples Rv and the number of currently queued samples Dv for the corresponding test station are determined; from the processing data, the processing volume Clv for this test station per unit time is determined; the waiting time DPv for the corresponding test station is determined using: [Dv + (Rv - Clv) × SCv] ÷ Clv = DPv; the minimum value is selected from multiple waiting times, and the auxiliary test item corresponding to the minimum value is marked as the optimal auxiliary test item.

[0054] A data analysis-based method for testing the performance of all-solid-state batteries includes the following steps:

[0055] The initial and target testing stations for the all-solid-state battery were confirmed, with the target station being a preset station. Battery test parameter maps were extracted from the cloud, and multiple test schemes were generated based on the initial and target testing stations. Schemes to be analyzed were selected according to a duration threshold. The selection process included: choosing the scheme with the shortest duration as the baseline scheme (duration Lz); other schemes had durations J. i ≤1.5×Lz is considered a candidate solution; solutions exceeding this value are not included. When there are multiple candidate solutions, the two shortest groups plus the baseline solution are selected as the solutions to be analyzed. When there is only one candidate solution, the group plus the baseline solution is selected as the solutions to be analyzed. When there are no candidate solutions, the baseline solution is directly selected as the optimal solution.

[0056] The stepwise elimination of invalid outlier segments includes: calculating the time T3 to reach the first outlier segment; if it falls within a uniform period, calculating the total outlier duration JL and obtaining PD. q If it does not fall into the category, the first abnormal segment is removed, the remaining abnormal cycles are merged, the time to reach the new first abnormal segment is recalculated, and the judgment is repeated until the scoring is completed.

[0057] Determine the current time T1, and calculate the test duration T for each scheme to be analyzed. q The test period [T1, T2] is obtained. q ];

[0058] Retrieve historical test cloud data for the same period, identify abnormal performance parameters for each test segment, determine the abnormal period, and merge them into a unified period;

[0059] The test cycle is compared with a unified cycle, invalid and abnormal segments are eliminated step by step, and the performance score PD of each scheme to be analyzed is calculated.q ;

[0060] Select PD q The minimum solution is selected as the optimal test solution and the test is executed. Based on the current workstation and each auxiliary test workstation, the movement time and waiting time are calculated to determine the optimal auxiliary test sequence and complete the full-dimensional performance test. The optimal auxiliary test item is determined by minimizing the waiting time DPv, which is calculated jointly by the number of queues, the addition rate, the processing efficiency, and the movement time.

[0061] Example 1

[0062] This embodiment discloses a data analysis-based all-solid-state battery performance testing system, including a sample station confirmation end, a test scheme selection end, a performance data analysis end, a multi-dimensional performance analysis center, and a test result output end; the sample station confirmation end is electrically connected to the test scheme selection end, the test scheme selection end is electrically connected to the performance data analysis end, the performance data analysis end is electrically connected to the test result output end, and the multi-dimensional performance analysis center is electrically connected to the sample station confirmation end and the test result output end, respectively.

[0063] The sample station confirmation end confirms the initial station and target test station of the all-solid-state battery. The target test station is the preset calibration station, and the positioning information is transmitted to the test plan selection end.

[0064] The test scheme selection end retrieves battery test parameter maps from the cloud, calibrates the initial and target test stations, and generates multiple test schemes; the scheme with the shortest duration is selected as the benchmark scheme, with a duration of Lz; the durations of other schemes are J. i ≤1.5×Lz is a candidate solution; when there are ≥2 candidate solutions, the two shortest groups plus the benchmark solution are taken as the solution to be analyzed; when there is 1 candidate solution, the group plus the benchmark solution are taken as the solution to be analyzed; when there are no candidate solutions, the benchmark solution is taken as the optimal solution.

[0065] The performance data analysis terminal determines the current time T1, and presses L. q ÷Vs=T q Calculate the test duration to obtain the test cycle [T1, T2] q Retrieve historical test cloud data from the previous day for the same period to identify the performance fluctuation parameter RL. t-k If the time exceeds Y1, mark the abnormal moment and determine the abnormal period; merge the abnormal periods to obtain a unified period, and compare it with the test period; if they overlap, calculate the total abnormal duration JL, and then calculate JL×C1+L. q ×C2=PD q Calculate the score; if there is no overlap, remove the first abnormal segment step by step, re-merge the periods and calculate, until the PD is obtained. q Select PD q The minimum solution is the optimal test solution.

[0066] The multi-dimensional performance analysis center calculates the travel time SCv from the current workstation to each auxiliary test workstation; obtains the queue number Dv, the number of new entries Rv, and the processing volume Clv, and calculates the waiting time DPv according to the formula; selects the item with the smallest DPv as the optimal auxiliary test item, and displays it through the output terminal.

[0067] Example 2

[0068] This embodiment discloses a data analysis-based method for testing the performance of all-solid-state batteries, with the following steps:

[0069] 1. Locate the initial loading station and target charge / discharge test station for all-solid-state batteries;

[0070] 2. Retrieve test parameter graphs, generate multiple test schemes, and filter the schemes to be analyzed according to a 1.5 times baseline duration threshold;

[0071] 3. Calculate the test cycle [T1, T2q] for each scheme;

[0072] 4. Compare historical data to identify abnormal voltage, impedance, and temperature ranges, and merge them to obtain a unified abnormal cycle;

[0073] 5. Time series comparison, eliminate non-affecting abnormal segments step by step, and calculate the performance score PDq;

[0074] 6. Select the minimum PDq solution and execute the test;

[0075] 7. Calculate the waiting time for auxiliary items such as interface impedance, thermal stability, cycle life, and safety testing, sort them by the shortest time, and complete the tests in sequence.

[0076] Some of the data in the above formulas are numerical calculations with dimensions removed, and the contents not described in detail in this specification are all prior art known to those skilled in the art.

[0077] 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 data analysis-based all-solid-state battery performance testing system, characterized in that, include: The sample station confirmation end confirms the initial station and target test station for the all-solid-state battery. The target test station is a preset station. The confirmed initial station and target test station are transmitted to the test plan selection end. At the test scheme selection end, the battery test parameter map is extracted from the cloud, and the point is calibrated based on the confirmed initial work station and target test work station. Starting from the initial work station and ending at the target test work station, multiple test schemes are determined, and then the scheme to be analyzed is selected from the multiple test schemes. The performance data analysis end analyzes the selected solution to be analyzed, determines the test cycle corresponding to different solutions to be analyzed based on the current time, and determines whether there are any performance anomalies in the corresponding solution to be analyzed within the test cycle based on the test cloud data. Based on the anomaly duration of different anomalies, the optimal test solution is determined and transmitted to the test result output end. The multi-dimensional performance analysis center determines the current battery position based on the sample station confirmation terminal, then determines the travel time based on the current position and the target positions of each auxiliary test item, and finally determines the optimal auxiliary test item based on the processing data of the corresponding test station and the test cloud data, and displays the results through the test result output terminal.

2. The all-solid-state battery performance testing system based on data analysis according to claim 1, characterized in that, The method by which the test plan selection end selects the test plan to be analyzed from multiple sets of test plans includes: Select the shortest test scheme from multiple test schemes, mark it as the benchmark scheme, and determine the benchmark scheme duration Lz; The duration values ​​for other test schemes were then determined and denoted as J. i Where i represents different other test schemes, J i Other test schemes with values ​​≤1.5×Lz are designated as candidate schemes, and J is... i Other test schemes with values ​​>1.5×Lz are not calibrated; If there are multiple alternative solutions, select the two alternative solutions with the shortest duration and the original baseline solution as the solutions to be analyzed. If there is a set of candidate solutions, then this candidate solution and the original benchmark solution will be used as the solutions to be analyzed. If no alternative solution exists, this benchmark solution will be used as the optimal test solution and displayed directly through the test result output. The determined analysis scheme is transmitted to the performance data analysis terminal.

3. The all-solid-state battery performance testing system based on data analysis according to claim 2, characterized in that, The specific methods by which the performance data analysis terminal determines the optimal test plan include: Determine the current time T1 and the duration L of the scheme to be analyzed. q Where q represents different schemes to be analyzed, and L is used. q ÷Vs=T q Determine the test duration T for different scenarios to be analyzed. q Where Vs is the preset test rate, and then T1+T is used. q =T2 q Determine the end time T2 of the corresponding scheme to be analyzed. q Determine the test cycle [T1, T2] for the corresponding solution to be analyzed. q ]; The test period [T1, T2] is set. q As a periodic time period, historical data from the previous day's same period period is retrieved from the test cloud data. The historical data consists of parameter data for each test segment of the solution to be analyzed. The performance fluctuation parameter RL for the corresponding test segment at the corresponding time is determined from the data of different test segments. t-k Where t represents different times and k represents different test segments, if RL t-k ≥Y1, where Y1 is a preset value, this moment is marked as an abnormal moment, and the abnormal period of the corresponding test segment is determined based on the abnormal moment; Test segments exhibiting abnormal cycles are identified as abnormal segments. The abnormal cycles of all abnormal segments within the analyzed scheme are merged to determine a unified cycle, and [T1, T2] are defined as follows: q Compare with a unified cycle: If [T1, T2] q If the time interval is the same as the unified period, the duration JL of the corresponding abnormal segment within the scheme to be analyzed is directly determined. If there are multiple sets of abnormal segments, the durations of the multiple sets of abnormal segments are summed to obtain JL, using the formula: JL×C1+L. q ×C2=PD q Determine the performance score (PD) of the corresponding scheme to be analyzed. q C1 and C2 are both preset coefficient factors.

4. The all-solid-state battery performance testing system based on data analysis according to claim 3, characterized in that, Where [T1, T2] q The comparison with the unified cycle also includes: If [T1, T2] q Unlike the unified period, the abnormal segment corresponding to the initial abnormal time CS1 of the unified period is determined. Based on the preset Vs, the time T3 for the battery to reach this abnormal segment is determined. If T3 ∈ the unified period, then [T1, T2] is used. q The performance score PD is calculated in the same way as the uniform cycle. q If T3 ∉ a unified period, remove the first group of abnormal segments, then merge the abnormal periods of the remaining abnormal segments to determine the subsequent period, and recalculate the time T4 to reach the new first abnormal segment. If T4 ∈ the subsequent period, calculate the corresponding abnormal segment duration JL and obtain PD. q If T4 ∉ subsequent cycles, then continue to remove the current first abnormal segment, and so on, until PD is completed. q calculate; PD from multiple sets of schemes to be analyzed q Among them, the group with the smallest value is selected as the optimal test plan.

5. The all-solid-state battery performance testing system based on data analysis according to claim 1, characterized in that, The specific method by which the multi-dimensional performance analysis center determines the optimal auxiliary test items is as follows: Based on the current battery station and the destination station, determine the shortest path and path distance, and based on the preset speed Vs, determine the movement time SCv, where v represents different auxiliary test stations. From the test cloud data, determine the number of new samples Rv per unit time and the number of samples currently in the queue Dv for the corresponding test station, and from the processed data, determine the processing volume Clv per unit time for this test station; The waiting time DPv for the corresponding test station is determined by the formula: [Dv + (Rv - Clv) × SCv] ÷ Clv = DPv. The minimum value is selected from multiple waiting times, and the auxiliary test item corresponding to the minimum value is marked as the optimal auxiliary test item.

6. A data analysis-based method for testing the performance of all-solid-state batteries, applicable to the data analysis-based all-solid-state battery performance testing system as described in any one of claims 1-5, characterized in that, Includes the following steps: The initial testing station and target testing station for the all-solid-state battery were confirmed, with the target testing station being the preset station. Extract battery test parameter maps from the cloud, generate multiple test schemes based on the initial workstation and the target test workstation, and filter the schemes to be analyzed according to the duration threshold. Determine the current time T1, and calculate the test duration T for each scheme to be analyzed. q The test period [T1, T2] is obtained. q ]; Retrieve historical test cloud data for the same period, identify abnormal performance parameters for each test segment, determine the abnormal period, and merge them into a unified period; The test cycle is compared with a unified cycle, invalid and abnormal segments are eliminated step by step, and the performance score PD of each scheme to be analyzed is calculated. q ; Select PD q The minimum solution is selected as the optimal test solution and the test is executed. Based on the current workstation and each auxiliary test workstation, the movement time and waiting time are calculated to determine the optimal auxiliary test sequence and complete the full-dimensional performance test.

7. The data analysis-based all-solid-state battery performance testing method according to claim 6, wherein the screening of the scheme to be analyzed includes: The scheme with the shortest duration is selected as the baseline scheme, with a duration of Lz; Other options duration J i Options ≤1.5×Lz are considered as candidate options; those exceeding this value are not included. When there are multiple candidate schemes, the two shortest schemes plus the benchmark scheme are taken as the schemes to be analyzed. If there is a set of candidate solutions, the set plus the benchmark solution is taken as the solution to be analyzed; if there are no candidate solutions, the benchmark solution is taken as the optimal solution.

8. The data analysis-based all-solid-state battery performance testing method according to claim 6, wherein the step-by-step elimination of invalid and abnormal segments includes: Calculate the time T3 to reach the first anomaly segment. If it falls within a uniform period, calculate the total anomaly duration JL and obtain PD. q ; If it does not fall into the category, the first abnormal segment is removed, the remaining abnormal cycles are merged, the time to reach the new first abnormal segment is recalculated, and the judgment is repeated until the scoring is completed.

9. In the data analysis-based all-solid-state battery performance testing method according to claim 6, the optimal auxiliary test item is determined by minimizing the waiting time DPv, and the waiting time is calculated by queuing number, new addition rate, processing efficiency and movement time.

Images