Battery test scheme rapid generation method and system

Standardized demand vectors are constructed through hierarchical feature extraction and nonlinear mapping, combined with binary feature code matching and multi-objective optimization algorithms, and the problem that existing battery test systems cannot be dynamically adjusted is solved, achieving high-precision, adaptability and safety battery testing.

CN120009737APending Publication Date: 2025-05-16SHENZHEN ENTRY EXIT INSPECTION & QUARANTINE BUREAU INDAL PROD INSPECTION TECH CENT
View PDF 0 Cites 4 Cited by

Patent Information

Application Number
CN202510249208.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-03-04
Publication Date
2025-05-16

AI Technical Summary

Technical Problem

The existing battery testing system cannot dynamically adjust the test process based on real-time battery parameters, resulting in missed detection of key indicators, hardware resource preemption, data errors, and mismatch of safety detection strategies and battery material characteristics.

Method used

Through layered feature extraction and nonlinear mapping, key indicators are disassembled and normalized to build a standardized demand vector that reflects the real-time battery state. The binary feature code encoding and cosine similarity matching screening module are used to combine the multi-objective optimization algorithm to automatically eliminate the conflict module under the dual constraints of test coverage gain and resource occupation penalty coefficients, and the continuous reliability of the test data is ensured through independent thread task scheduling and real-time monitoring mechanism of temperature rise rate.

Benefits of technology

It significantly improves the accuracy, adaptability and safety of battery testing, reduces data errors and resource seizing risks, and can effectively adapt to the needs of multiple scenarios such as electric vehicles, energy storage systems and consumer electronics.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120009737A_ABST
    Figure CN120009737A_ABST
Patent Text Reader

Abstract

The invention discloses a method and a system for quickly generating a battery test scheme, particularly relates to the field of battery test optimization, is used for solving the problem of test process stiffness, and is characterized in that key is disassembled and normalized through hierarchical feature extraction and nonlinear mapping, and is converted into dynamic weight and embedded with priority marks; therefore, a standardized demand vector reflecting the real-time battery state is constructed; the candidate modules are coded by adopting binary feature codes, the modules meeting core test requirements are accurately screened through cosine similarity matching, and comprehensive coverage of key indexes is realized; a multi-objective optimization algorithm is utilized, conflict modules are automatically eliminated under the dual restrictions of test coverage rate gain and resource occupation penalty coefficient, the test precision and hardware resource allocation are balanced, and data errors and resource preemption risks are effectively reduced; and an independent thread task scheduling and temperature rise rate real-time monitoring mechanism is used for rapidly switching a standby module to complete data supplementary collection when abnormal temperature rise is detected, so that continuous and reliable test data is ensured.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of battery test optimization, and more specifically, to a method and system for quickly generating a battery test plan. Background Art

[0002] At present, battery testing needs to cope with the differentiated needs of multiple scenarios such as electric vehicles, energy storage systems, and consumer electronics. Different scenarios have strict requirements on the dynamic coupling of test indicators. The existing test system adopts a fixed script control mode. The trigger conditions, execution order, and resource allocation of the test modules are all statically preset, resulting in the inability to dynamically adjust the test process according to real-time battery parameters. The key indicator missed detection rate is high, hardware resource preemption causes data errors, and the safety detection strategy and battery material characteristics are mismatched. These problems frequently occur, seriously restricting the test accuracy and scenario adaptability.

[0003] The defect of the existing technology is the lack of a dynamic closed-loop adaptation mechanism for battery characteristics and test processes: the test module cannot automatically adjust the execution node according to parameters such as real-time charge state, temperature changes, and material expansion rate; no hardware resource occupancy priority model is established when multiple modules are executed in parallel, resulting in key instruments and equipment being preempted by low-priority tasks; the safety detection threshold is not associated with the characteristics of the battery chemical system, and the unified threshold strategy cannot adapt to the intrinsic parameter differences of new materials such as silicon-based negative electrodes and high-nickel positive electrodes, resulting in a high rate of missed detection of abnormal conditions.

[0004] In order to solve the above problems, a technical solution is now provided. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a method and system for quickly generating a battery test scheme. Through hierarchical feature extraction and nonlinear mapping, the key is disassembled and normalized, converted into dynamic weights and embedded with priority tags, so as to construct a standardized demand vector reflecting the real-time battery status. The candidate modules are encoded with binary feature codes, and the modules that meet the core test requirements are accurately screened through cosine similarity matching to achieve comprehensive coverage of key indicators; a multi-objective optimization algorithm is used to automatically exclude conflicting modules under the dual constraints of test coverage gain and resource occupancy penalty coefficient, balance test accuracy and hardware resource allocation, and effectively reduce data errors and resource preemption risks. The independent thread task scheduling and temperature rise rate real-time monitoring mechanism quickly switches to the standby module to complete data acquisition when an abnormal temperature rise is detected, ensuring that the test data is continuous and reliable, so as to solve the problems raised in the above background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] S1: Decompose the original demand parameters layer by layer and complete nonlinear mapping and normalization through morphological functions and piecewise functions, merging key indicators into standardized demand vectors with priority tags;

[0008] S2: Calculate the projection coincidence degree based on the standardized demand vector and the binary feature code of the test module and generate a candidate queue in descending order of matching degree;

[0009] S3: First, the rule engine is used to eliminate indicator conflicts under the same timestamp, and then a multi-objective optimization algorithm is used to evaluate the comprehensive effectiveness of test coverage gain and resource occupation penalty coefficient, and the module combination at the Pareto frontier is selected;

[0010] S4: Split the optimal module combination into parallel thread tasks, and monitor the comparison results of the temperature rise rate and the safety threshold in real time during the execution process. Once the limit is exceeded, terminate the corresponding thread and enable the backup module to complete data acquisition.

[0011] In a preferred embodiment, step S1 includes the following contents:

[0012] The original demand parameters are obtained by extracting steady-state and transient features from historical test data, analyzing the dynamic response of real-time monitoring data, and calculating the working condition adaptation required by the application scenario, and hierarchical feature extraction is performed on each parameter. The discharge rate range is mapped and an exponential decay function is used to convert its numerical range into a value between 0 and 1. The temperature sensitivity is mapped using a piecewise function. The low temperature range is converted using the inverse tangent function. When the temperature exceeds the set threshold, an exponential decay function is used to make it quickly approach 1.

[0013] In a preferred embodiment, step S1 further includes the following contents:

[0014] Different custom mapping functions are used for other test requirement parameters, and differentiated processing is performed according to the physical characteristics of each parameter; all processed parameter values ​​form a vector, and each component is between 0 and 1; in order to eliminate the numerical differences of each parameter caused by different units or dimensions, the converted parameters are normalized uniformly; finally, a priority tag is attached to the normalized parameter vector, and the corresponding tag information is retained in the data structure. After processing, a standardized requirement vector carrying a priority tag is obtained.

[0015] In a preferred embodiment, step S2 includes the following contents:

[0016] Firstly, the input parameters and output parameters of each module in the test module library are characterized and encoded, and each characteristic code set is integrated into a binary characteristic code sequence; the standardized demand vector and each corresponding component in the binary characteristic code sequence are first subjected to controllable transformation, and then all component results are multiplied and added item by item to obtain a comprehensive value regarded as the projection overlap. The screening is achieved by setting a judgment threshold at the numerical level. When the overlap is lower than the judgment threshold, the corresponding module is judged to be unable to meet the demand and is eliminated from the subsequent queue; all the modules that pass the screening are sorted from high to low according to the overlap score, and finally a candidate queue is formed in descending order of matching degree.

[0017] In a preferred embodiment, step S3 includes the following contents:

[0018] First, the modules in the candidate queue are searched one by one, and whether there are any non-parallel test operations is determined according to the timestamp rule. If any two modules are found to involve measurement activities at the same time node and there are hardware ports that cannot be shared or switched, the two modules are marked as mutually exclusive. The candidate module set after conflict elimination is input into the multi-objective optimization algorithm, and the test coverage gain and resource occupation penalty coefficient are introduced to measure the overall utility of the candidate combination.

[0019] In a preferred embodiment, when calculating the test coverage gain, the new parameter sets that each candidate module can monitor are first counted, and the number of independent parameters covered by all modules is accumulated to measure the degree of coverage; then a nonlinear transformation is introduced to map the parameter coverage number to between 0 and 1 through an adjustable function.

[0020] In a preferred embodiment, when calculating the resource occupancy penalty coefficient, a resource occupancy degree is first assigned to each candidate module, and an occupancy value based on a nonlinear scale is given according to a combination of multiple factors. The overall resource occupancy penalty coefficient is obtained by accumulating the occupancy values ​​of all module combinations.

[0021] In a preferred embodiment, the test coverage gain and the resource occupancy penalty coefficient are jointly solved as objective functions to find a Pareto front solution set that can both improve coverage and suppress excessive resource consumption; finally, the module combination with the highest comprehensive utility is selected from the Pareto front solution set.

[0022] In a preferred embodiment, step S4 includes the following contents:

[0023] Firstly, according to the module characteristics of the optimal module combination, combined with the actual connection conditions and parallel capabilities of the instrument port, each module or mutually compatible module group is assigned to a separate thread to ensure that the available hardware resources can be fully utilized during concurrent execution; then, the monitoring task of each thread is initialized, and the temperature sensitivity mapping function is reversely calculated to calculate the temperature rise rate value; then, the temperature rise rate is continuously compared with the pre-set safety threshold in each thread. If the temperature rise rate is detected to exceed the threshold at any time, an alarm is issued to the thread management unit and the data collection of the corresponding thread is terminated. Then, the corresponding replacement module is started according to the pre-arranged spare module list to re-collect the interrupted key data indicators.

[0024] A battery test scheme rapid generation system, comprising: a parameter mapping module, a feature screening module, a combination optimization module and a task execution module;

[0025] Parameter mapping module: decomposes the original demand parameters layer by layer, uses morphological functions and piecewise functions for nonlinear mapping, and performs normalization to generate a standardized demand vector with priority tags;

[0026] Feature screening module: Based on the standardized demand vector output by the parameter mapping module, the projection coincidence is calculated with the binary feature code of the test module, the modules with low matching degree are eliminated, and the candidate queue is generated in descending order of matching degree;

[0027] Combination optimization module: Using the candidate queue output by the feature screening module, first eliminate the timestamp conflict module through the rule engine, then perform multi-objective optimization based on the test coverage gain and resource occupation penalty coefficient, and select the optimal module combination on the Pareto frontier;

[0028] Task execution module: Receives the optimal module combination output by the combination optimization module, distributes it to parallel threads according to the physical connection status of the instrument port, monitors the temperature rise rate in real time during the execution process and compares it with the safety threshold. Once the limit is exceeded, the relevant threads are terminated and the backup module is started to collect additional data.

[0029] The technical effects and advantages of a method and system for quickly generating a battery test solution of the present invention are as follows:

[0030] The present invention uses hierarchical feature extraction and nonlinear mapping to disassemble and normalize the key, convert it into dynamic weights and embed priority tags, so as to construct a standardized demand vector reflecting the real-time battery status. The candidate modules are encoded with binary feature codes, and the modules that meet the core test requirements are accurately screened through cosine similarity matching to achieve comprehensive coverage of key indicators; a multi-objective optimization algorithm is used to automatically exclude conflicting modules under the dual constraints of test coverage gain and resource occupancy penalty coefficient, balance test accuracy and hardware resource allocation, and effectively reduce data errors and resource preemption risks. The independent thread task scheduling and temperature rise rate real-time monitoring mechanism quickly switches to the standby module to complete data acquisition when an abnormal temperature rise is detected, ensuring continuous and reliable test data. The overall solution significantly improves test efficiency, safety and adaptability in scenarios such as electric vehicles, energy storage systems and consumer electronics, and provides strong technical support for high-reliability battery testing. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 The figure is a flow chart of a method for quickly generating a battery test scheme according to the present invention.

[0032] Figure 2 The present invention is a structural schematic diagram of a battery test solution rapid generation system. DETAILED DESCRIPTION

[0033] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0034] Embodiment 1: Figure 1 The present invention provides a method for quickly generating a battery test scheme, comprising:

[0035] S1: The original demand parameters are decomposed layer by layer and nonlinear mapping and normalization are completed through morphological functions and piecewise functions, and the key indicators are merged into a standardized demand vector carrying priority tags.

[0036] S2: Calculate the projection coincidence based on the standardized demand vector and the binary feature code of the test module and generate a candidate queue in descending order of matching degree.

[0037] S3: First, the rule engine is used to eliminate indicator conflicts under the same timestamp, and then a multi-objective optimization algorithm is used to evaluate the comprehensive effectiveness of test coverage gain and resource occupation penalty coefficient, and the module combination on the Pareto frontier is selected.

[0038] S4: Split the optimal module combination into parallel thread tasks, and monitor the comparison results of the temperature rise rate and the safety threshold in real time during the execution process. Once the limit is exceeded, terminate the corresponding thread and enable the backup module to complete data acquisition.

[0039] Step S1 includes the following contents:

[0040] In the multi-scenario battery testing process, the testing requirements are highly differentiated due to the differences in battery material systems, working environments, and application scenarios, which makes it difficult to adapt the fixed process testing method, leading to problems such as missed detection, resource competition, and failure of safety policies. To solve this problem, it is necessary to first hierarchically decompose the test requirement parameters and construct a standardized requirement vector that can adapt to dynamic test process adjustments, so that the matching, scheduling, and optimization of subsequent test modules can be calculated under a unified dimension, thereby improving the adaptability and accuracy of the test process.

[0041] The original demand parameters are derived from the key influencing factors of the battery test scenario, which are usually determined jointly by historical test data, real-time monitoring data and application scenario requirements. Specifically, they include but are not limited to the discharge rate range, temperature sensitivity, material expansion rate, voltage stability, current response characteristics, etc. These parameters directly affect the execution conditions and adaptability of the test module. Among them, the discharge rate range is derived from the battery load characteristic requirements, which can be obtained by fitting the discharge current curve under different working conditions; the temperature sensitivity is based on the battery temperature rise rate and internal heat conduction characteristics, and is calculated and extracted from the steady-state and transient temperature rise curves in the historical test data; the material expansion rate relies on the characteristics of the battery electrode microstructure change, and is obtained through online or offline expansion test data analysis; the voltage stability reflects the voltage fluctuation at the output end of the battery, which can be calculated based on the steady-state voltage fluctuation range under different workloads; the current response characteristics are used to describe the battery's adaptability to transient load changes, which is obtained through pulse current response data processing. These parameters constitute the basic input data for battery test requirements, and are subsequently processed through feature extraction and standardization to form a demand vector suitable for dynamic test process optimization.

[0042] The original demand parameters are obtained by extracting steady-state and transient features of historical test data, analyzing the dynamic response of real-time monitoring data, and calculating the working condition adaptation required by the application scenario.

[0043] First, the original demand parameter set is recorded as {p1, p2, ..., p n}, where the discharge rate range is denoted by x d , the temperature sensitivity is denoted by x t(If there are other parameters, they can be further subdivided and named), and hierarchical feature extraction is performed on each parameter. Mapping is performed on the discharge rate range, and an exponential decay function is used to convert its value range into a value between 0 and 1, ensuring that the larger the discharge rate, the closer the mapping value is to 1, to reflect its importance to the test requirements. The mapping process is: where λ d It is used to control the conversion speed to ensure that the output range gradually approaches 1 as the discharge rate range increases.

[0044] The temperature sensitivity is mapped using a piecewise function. The low temperature range is converted using the inverse tangent function to ensure that the value increases slowly, so that low temperature changes will not have too much impact on the final demand priority; when the temperature exceeds the set threshold, an exponential decay function is used to make it quickly approach 1 to reflect the strong impact of high temperature changes on test requirements. This piecewise mapping method allows the weight of the impact of different temperature ranges on the test process to be dynamically adapted, rather than a single linear transformation. The mapping process of the piecewise function is as follows:

[0045]

[0046] where β t , α1, α2, γ t The slope of the mapping curve, the base magnification, the exponential decay rate, and the segmentation threshold are controlled respectively. Such nonlinear functions can be used to map the discharge rate range and temperature sensitivity differently, avoiding the limitations of common methods such as variance and standard deviation.

[0047] Next, different custom mapping functions are used for other test requirement parameters, and each parameter is processed differently according to its physical characteristics, so that the converted value can accurately reflect the importance of the original parameter. All processed parameter values ​​form a vector, each component is between 0 and 1, and the numerical distribution between the parameters conforms to the actual importance of the test requirements.

[0048] Then, in order to eliminate the numerical differences of each parameter due to different units or dimensions, the converted parameters are normalized uniformly so that the values ​​of all parameters are comparable under the same scale. Normalization uses the maximum adjustment method or dynamic scale adjustment so that the converted parameter values ​​can reflect their relative influence in the entire test requirement set without causing information loss or misleading due to different physical units.

[0049] Finally, a priority tag is attached to the normalized parameter vector. The priority setting is classified according to the urgency and risk level of the test requirements, and the tag information is retained in the data structure. After processing, a standardized requirement vector carrying a priority tag is obtained. Its data structure is directly referenced in subsequent steps for projection overlap calculation with the module feature code and further optimization and screening.

[0050] Priority is set based on the urgency and risk level of the test requirements, where the urgency is quantified based on the impact of the test requirements on battery performance, life and safety. First, a preliminary classification is made based on whether the test requirements involve key performance parameters (such as capacity decay rate, internal resistance growth trend), and test requirements involving core performance are set to high urgency, while test requirements involving secondary performance are set to low urgency; further, based on the requirements of demand response time, requirements that require real-time monitoring are set to the highest urgency, while periodic monitoring requirements are set to a lower urgency; for the division of risk levels, first, the impact of the target parameters of the test requirements on battery safety is classified. If the target parameters involve thermal runaway, short circuit or lithium plating risks, they are set to a high risk level. Otherwise, they are divided into medium or low risk levels based on their impact on battery performance decay. Finally, the final priority mark is determined through a joint classification matrix of urgency and risk level, where test requirements with high urgency and high risk are set to the highest priority, those with high urgency but low risk or those with low urgency but high risk are set to the second highest priority, and those with low urgency and low risk are set to the lowest priority.

[0051] In the previous step, a standardized requirement vector with a priority tag has been obtained to characterize the core features of multi-dimensional requirements such as discharge rate range and temperature sensitivity. Now it is necessary to match this standardized requirement vector with the test module library one by one to screen out candidate modules that can meet the requirements of key parameter collection and analysis. The subsequent steps will further eliminate conflicts and optimize combinations based on these candidate modules.

[0052] Step S2 includes the following contents:

[0053] First, the input parameters and output parameters of each module in the test module library are characterized and encoded, and each feature code set is integrated into a binary feature code sequence. Subsequently, the standardized demand vector obtained in the previous step is used to measure the similarity of each binary feature code sequence. In order to avoid the lack of generalization ability caused by the common linear weighting method, each component can be nonlinearly mapped and then summed based on the vector dot product, so that the intrinsic correlation between the two is closer to the actual impact of the demand priority. Specifically, each corresponding component in the standardized demand vector and the binary feature code sequence is first subjected to a controllable transformation such as power or logarithm, and then all component results are multiplied and added item by item to obtain a comprehensive value that can be regarded as the projection overlap. The screening is achieved by setting a judgment threshold at the numerical level. When the overlap is lower than the judgment threshold, the corresponding module is judged to be unable to meet the core requirements and is eliminated from the subsequent queue. All the modules that pass the screening are sorted from high to low according to the overlap score, and finally a candidate queue arranged in descending order of matching is formed, so that the next step can be based on the high matching module for conflict troubleshooting and multi-objective optimization.

[0054] Each module in the test module library refers to an independent functional unit that performs test tasks for specific battery performance parameters in different battery test scenarios. Each module contains specific input parameters, output parameters and execution logic, and can provide corresponding data collection, processing and analysis capabilities for different test requirements. The input parameters of the module include but are not limited to key variables affecting battery performance such as current, voltage, temperature, internal resistance, state of charge, etc., while the output parameters correspond to specific test results, such as static or dynamic voltage characteristics, the effect of discharge rate on capacity retention rate, the effect of temperature change on the evolution trend of internal resistance, etc. The execution logic of each module is based on different test methods, such as constant current discharge, step load test, pulse load test, temperature cycle test, etc., to ensure that data collection and analysis can be performed for specific working conditions during the test process. In the test process, multiple modules can be run in a specific order or in parallel to achieve coverage for complex test requirements, thereby ensuring the integrity and accuracy of the test process. In the technical solution of the present invention, the role of the test module library is to provide optional test function units, so that they can be dynamically screened and optimized after matching the demand vector, thereby improving the adaptability of the test process and avoiding the test rigidity problem caused by the traditional fixed script mode.

[0055] After this process, all optional modules in the test module library have been deeply matched with the standardized requirement vectors, modules with low overlap have been successfully eliminated, and candidate queues that meet the requirements of the test scenarios have been generated according to the degree of matching, laying a solid foundation for the subsequent rapid elimination of timestamp conflicts and performing multi-objective optimization.

[0056] In the previous step, the candidate queues arranged in descending order of matching degree have been obtained. On this basis, it is necessary to eliminate the modules with indicator collection conflicts at the same timestamp, and then perform multi-objective optimization on the remaining modules to balance the relationship between test coverage gain and resource occupation penalty coefficient, so as to find a module combination that has advantages in both coverage and resource utilization efficiency.

[0057] Step S3 includes the following contents:

[0058] First, the modules in the candidate queue are searched one by one, and the timestamp rule is used to determine whether there are non-parallel test operations. If any two modules are found to involve measurement activities at the same time node and there are hardware ports that cannot be shared or switched, the two modules are marked as mutually exclusive, and only modules with fewer conflicts or higher compatibility with high-priority requirements are retained in the candidate queue. Next, the candidate module set after conflict elimination is input into the multi-objective optimization algorithm, and the two indexes of test coverage gain and resource occupation penalty coefficient are introduced to measure the overall utility of the candidate combination.

[0059] When calculating the test coverage gain, first count the new parameter sets that each candidate module can monitor, and accumulate the number of independent parameters covered by all modules to measure the coverage; then introduce a nonlinear transformation, map the parameter coverage number to between 0 and 1 through an adjustable function, and as the coverage number increases, the gain value increases at a decreasing rate, avoiding the problem that the simple linear summation leads to the weaker gain significance as the coverage approaches full load. This mapping can be in the form of 1 minus exponential decay, and a set of adjustable control constants are used to ensure that the coverage gain is more sensitive to core requirements. For example, the test coverage gain can be calculated in the following way:

[0060] The test coverage gain measures the number of independent parameters that the current module combination can cover in the test scenario and its contribution to the test completeness. k}, each test module M i With a set of measurable parameters P i ={p i1 ,p i2 ,…,p in}, the total parameter coverage set of the entire module combination is defined as Where |P total | represents the total number of unique parameters covered by all test modules combined. Since some parameters may be measured repeatedly in multiple modules, it is necessary to define the parameter independence factor γ(p) as the contribution attenuation factor when parameter p appears in multiple modules, which is defined as Where N pis the number of repeated measurements of parameter p in different modules. This formula ensures that as the repetition of a parameter in multiple modules increases, its contribution value decreases, preventing repeated measurements of a single parameter in multiple modules from excessively affecting the coverage gain. The final test coverage gain G cov Calculated as Where f(p) is the importance mapping function of the parameter ppp in the current test scenario, which is defined by exponential mapping as f(p) = 1-e -λp , where λ is a tuning parameter used to control the decay rate of the importance gain, making the contribution to the key parameters greater while the impact of the secondary parameters is relatively small.

[0061] Where k represents the total number of test modules in the candidate queue that have passed the matching screening; M i Represents the i-th test module in the test module library; n represents the test module M i The total number of measurable parameters of different test modules may be different.

[0062] When calculating the resource occupancy penalty coefficient, it is necessary to consider the degree of demand for the same hardware resource by different modules and the possible port usage competition. First, a resource occupancy is assigned to each candidate module, and an occupancy value based on a nonlinear scale is given based on factors such as execution time, number of ports required, and parallel feasibility. The occupancy value of each module is then appropriately amplified to reflect the potential impact of resource contention on test accuracy and security under high load conditions. The overall resource occupancy penalty coefficient is obtained by accumulating or multiplying the occupancy values ​​of all module combinations, and by dynamically adjusting the constant, it produces a stronger penalty feedback for higher occupancy, prompting the optimization process to tend to select combinations with relatively balanced resource requirements. For example, the resource occupancy penalty coefficient can be calculated in the following way:

[0063] The resource occupation penalty coefficient measures the degree of competition of the current test module combination for hardware resources (including ports, computing power, data throughput, etc.), and gives a negative weight to suppress excessive occupation. i The required hardware resource set is R i = {r i1 ,r i2 ,…,r im}, then the combined total resource demand set is defined as Where m represents the test module M i The total number of hardware resources required. The hardware resource requirements of different test modules may be different. Each resource r may be occupied repeatedly in multiple modules. Therefore, the resource conflict factor δ(r) is defined as the calculation formula for the degree of competition when resource r appears in different modules: δ(r) = 1 + log(1 + N r ), where Nr Represents the number of times resource r is called by different modules in the current combination. This formula ensures that as resources are occupied by multiple modules, their conflict coefficient grows logarithmically, avoiding excessively amplifying the impact of sharing a small amount of resources. The final resource occupation penalty coefficient C res Calculated as Among them, g(r) is the importance function of resource r, which is defined in a similar way to the parameter importance function, using a nonlinear mapping method g(r) = e μr -1, where μ is a tuning parameter used to control the increase in the impact of resource occupancy, so that high-occupancy resources contribute significantly to the penalty coefficient, while the impact of low-occupancy resources is relatively mild.

[0064] The multi-objective optimization algorithm uses the above two indexes - test coverage gain and resource occupation penalty coefficient - as objective functions for joint solution, looking for the Pareto front solution set that can both improve coverage and suppress excessive resource consumption. Finally, the module combination with the highest comprehensive utility is selected from the Pareto front solution set, which not only ensures extensive coverage of multi-scenario requirements, but also avoids excessive occupation of hardware ports, thereby maximizing overall efficiency and reducing conflict risks in subsequent test execution.

[0065] By performing conflict detection and multi-objective optimization on the modules in the candidate queue, a dynamic balance between test coverage gain and resource occupation penalty coefficient is achieved, thereby achieving higher test accuracy and resource utilization under a variety of battery materials and working conditions, laying a solid foundation for subsequent test thread allocation and safety threshold monitoring.

[0066] After the third step, the module combination with the best comprehensive utility has been obtained, and the coverage and resource usage have been balanced. In order to ensure that the combination has high reliability and flexibility in actual execution, each module needs to be assigned to an independent thread and compared with the physical connection status of the instrument port, and the comparison result of the temperature rise rate and the safety threshold is monitored in real time. Once the temperature rise is detected to be too fast, the relevant thread is cut off and the spare module is enabled for data acquisition, so as to ensure that the entire test process continues within the safety boundary.

[0067] Step S4 includes the following contents:

[0068] First, according to the module characteristics in the optimal module combination, combined with the actual connection conditions and parallel capabilities of the instrument port, each module or compatible module group is assigned to a separate thread to ensure that the available hardware resources can be fully utilized during concurrent execution. Then, the monitoring task of each thread is initialized, including real-time temperature acquisition, recording parameter data during the measurement process, and reverse calculation of the temperature sensitivity mapping function mentioned above to calculate the temperature rise rate value. If the temperature sequence is recorded as T(t), T(t+1),..., the temperature rise rate R(t) can be defined as:

[0069]

[0070] Where φ(·) represents a controllable function that denoises or smoothes the temperature value to reduce the interference of spike data; Δt is the sampling time interval between adjacent moments. Subsequently, the temperature rise rate is continuously compared with the pre-set safety threshold in each thread. If the temperature rise rate is detected to exceed the threshold at any time, an alert is immediately issued to the thread management unit and the data collection of the thread is terminated. Then, the corresponding replacement module is started according to the list of spare modules arranged in the previous article, and the interrupted key data indicators are re-collected to ensure the continuity of subsequent data. In order to improve the synchronization performance, it is necessary to add a timestamp recording mechanism to the thread management to align the sampling information before termination with the starting sampling information of the replacement module to avoid abnormal analysis results caused by data misalignment.

[0071] By decomposing the optimal module combination into parallel threads, using the temperature rise rate threshold monitoring to promptly interrupt threads with safety hazards, and then dynamically enabling the backup module to continue data acquisition, the test process can efficiently deal with unexpected risks such as temperature anomalies and port resource constraints in actual execution, thereby improving the adaptability and safety of the overall test.

[0072] The present invention utilizes hierarchical feature extraction and nonlinear mapping methods to convert the original battery test requirements into standardized demand vectors carrying priority tags, and then accurately matches and screens the test modules using binary feature codes and cosine similarity. Subsequently, a rule engine is used to eliminate conflicts with the same timestamp, and a multi-objective optimization is used to balance the test coverage gain and resource occupancy penalty coefficient, and select the module combination with the best comprehensive utility in the Pareto frontier. Finally, the optimal module combination is decomposed into independent thread tasks, and real-time monitoring is performed according to the connection status of the instrument port. When the temperature rise rate exceeds the safety threshold, the backup module is enabled in time to collect supplementary data. The dynamic adaptability and data acquisition accuracy of the test process are improved, the missed detection rate of key indicators and the risk of hardware resource preemption are effectively reduced, and the requirements for high-precision, real-time and safety of battery testing in multiple scenarios such as electric vehicles, energy storage systems and consumer electronics are met.

[0073] Embodiment 2: Figure 2 The invention provides a battery test scheme rapid generation system, comprising: a parameter mapping module, a feature screening module, a combination optimization module and a task execution module;

[0074] Parameter mapping module: decomposes the original demand parameters layer by layer, uses morphological functions and piecewise functions for nonlinear mapping, and performs normalization to generate a standardized demand vector with priority tags;

[0075] Feature screening module: Based on the standardized demand vector output by the parameter mapping module, the projection coincidence is calculated with the binary feature code of the test module, the modules with low matching degree are eliminated, and the candidate queue is generated in descending order of matching degree;

[0076] Combination optimization module: Using the candidate queue output by the feature screening module, first eliminate the timestamp conflict module through the rule engine, then perform multi-objective optimization based on the test coverage gain and resource occupation penalty coefficient, and select the optimal module combination on the Pareto frontier;

[0077] Task execution module: Receives the optimal module combination output by the combination optimization module, distributes it to parallel threads according to the physical connection status of the instrument port, monitors the temperature rise rate in real time during the execution process and compares it with the safety threshold. Once the limit is exceeded, the relevant threads are terminated and the backup module is started to collect additional data.

[0078] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0079] The above description is only by way of illustration of certain exemplary embodiments of the present invention. It is undoubted that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

[0080] It should be noted that, in this article, if there are relational terms such as first and second, etc., they are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "including a..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0081] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for quickly generating a battery test scheme, characterized in that: Includes steps: S1: Decompose the original demand parameters layer by layer and complete nonlinear mapping and normalization through morphological functions and piecewise functions, merging key indicators into standardized demand vectors with priority tags; S2: Calculate the projection coincidence degree based on the standardized demand vector and the binary feature code of the test module and generate a candidate queue in descending order of matching degree; S3: First, the rule engine is used to eliminate indicator conflicts under the same timestamp, and then a multi-objective optimization algorithm is used to evaluate the comprehensive effectiveness of test coverage gain and resource occupation penalty coefficient, and the module combination at the Pareto frontier is selected; S4: Split the optimal module combination into parallel thread tasks, and monitor the comparison results of the temperature rise rate and the safety threshold in real time during the execution process. Once the limit is exceeded, terminate the corresponding thread and enable the backup module to complete data acquisition.

2. The method for quickly generating a battery test solution according to claim 1, characterized in that: Step S1 includes the following contents: The original demand parameters are obtained by extracting steady-state and transient features from historical test data, analyzing the dynamic response of real-time monitoring data, and calculating the working condition adaptation required by the application scenario, and hierarchical feature extraction is performed on each parameter. The discharge rate range is mapped and an exponential decay function is used to convert its value range into a value between 0 and 1. The temperature sensitivity is mapped using a piecewise function. The low temperature range is converted using the inverse tangent function; when the temperature exceeds the set threshold, an exponential decay function is used to make it quickly approach 1.

3. The method for quickly generating a battery test scheme according to claim 2, characterized in that: Step S1 also includes the following contents: Different custom mapping functions are used for other test requirement parameters, and differentiated processing is performed according to the physical characteristics of each parameter; all processed parameter values ​​form a vector, and each component is between 0 and 1; in order to eliminate the numerical differences of each parameter caused by different units or dimensions, the converted parameters are normalized uniformly; finally, a priority tag is attached to the normalized parameter vector, and the corresponding tag information is retained in the data structure. After processing, a standardized requirement vector carrying a priority tag is obtained.

4. The method for quickly generating a battery test solution according to claim 3, characterized in that: Step S2 includes the following contents: Firstly, the input parameters and output parameters of each module in the test module library are characterized and encoded, and each characteristic encoding set is integrated into a binary characteristic code sequence; A controllable transformation is first performed on each corresponding component of the standardized demand vector and the binary feature code sequence, and then all component results are multiplied and added item by item to obtain a comprehensive value regarded as the projection overlap. Screening is achieved by setting a judgment threshold at the numerical level. When the overlap is lower than the judgment threshold, the corresponding module is judged to be unable to meet the demand and is eliminated from the subsequent queue; all modules that pass the screening are sorted from high to low according to the overlap score, and finally a candidate queue is formed in descending order of matching degree.

5. The method for quickly generating a battery test scheme according to claim 4, characterized in that: Step S3 includes the following contents: First, the modules in the candidate queue are searched one by one, and whether there are any non-parallel test operations is determined according to the timestamp rule. If any two modules are found to involve measurement activities at the same time node and there are hardware ports that cannot be shared or switched, the two modules are marked as mutually exclusive. The candidate module set after conflict elimination is input into the multi-objective optimization algorithm, and the test coverage gain and resource occupation penalty coefficient are introduced to measure the overall utility of the candidate combination.

6. A method for quickly generating a battery test scheme according to claim 5, characterized in that: When calculating the test coverage gain, we first count the new parameter sets that each candidate module can monitor, and then add up the number of independent parameters covered by all modules to measure the coverage; Then a nonlinear transformation is introduced to map the parameter coverage quantity to between 0 and 1 through an adjustable function.

7. The method for quickly generating a battery test solution according to claim 5, characterized in that: When calculating the resource occupancy penalty coefficient, a resource occupancy degree is first assigned to each candidate module, and an occupancy value based on a nonlinear scale is given according to a combination of multiple factors. The overall resource occupancy penalty coefficient is obtained by accumulating the occupancy values ​​of all module combinations.

8. The method for quickly generating a battery test solution according to claim 5, characterized in that: The test coverage gain and resource occupation penalty coefficient are jointly solved as objective functions to find the Pareto frontier solution set that can both improve coverage and inhibit excessive resource consumption; finally, the module combination with the highest comprehensive utility is selected from the Pareto frontier solution set.

9. A method for quickly generating a battery test scheme according to claim 8, characterized in that: Step S4 includes the following contents: Firstly, according to the module characteristics of the optimal module combination, combined with the actual connection conditions and parallel capabilities of the instrument port, each module or mutually compatible module group is assigned to a separate thread to ensure that the available hardware resources can be fully utilized during concurrent execution; then, the monitoring task of each thread is initialized, and the temperature sensitivity mapping function is reversely calculated to calculate the temperature rise rate value; then, the temperature rise rate is continuously compared with the pre-set safety threshold in each thread. If the temperature rise rate is detected to exceed the threshold at any time, an alarm is issued to the thread management unit and the data collection of the corresponding thread is terminated. Then, the corresponding replacement module is started according to the pre-arranged spare module list to re-collect the interrupted key data indicators.

10. A battery test scheme rapid generation system, used to implement a battery test scheme rapid generation method according to any one of claims 1 to 9, characterized in that: include: Parameter mapping module, feature screening module, combination optimization module and task execution module; Parameter mapping module: decomposes the original demand parameters layer by layer, uses morphological functions and piecewise functions for nonlinear mapping, and performs normalization to generate a standardized demand vector with priority tags; Feature screening module: Based on the standardized demand vector output by the parameter mapping module, the projection coincidence is calculated with the binary feature code of the test module, the modules with low matching degree are eliminated, and the candidate queue is generated in descending order of matching degree; Combination optimization module: Using the candidate queue output by the feature screening module, first eliminate the timestamp conflict module through the rule engine, then perform multi-objective optimization based on the test coverage gain and resource occupation penalty coefficient, and select the optimal module combination on the Pareto frontier; Task execution module: Receives the optimal module combination output by the combination optimization module, distributes it to parallel threads according to the physical connection status of the instrument port, monitors the temperature rise rate in real time during the execution process and compares it with the safety threshold. Once the limit is exceeded, the relevant threads are terminated and the backup module is started to collect additional data.

Citation Information

Cited By

  • Electric power material detection task generation method and device

    CN120634209A

  • Motor type selection method and system

    CN120723818A

  • Battery test scheduling method and device based on multi-objective optimization driving

    CN121476984A

  • A high-temperature solid oxide battery multi-level bidirectional test control method and system

    CN122652333A