A battery failure early warning detection method and device, and a computer storage medium

Abstract

This invention relates to the field of battery safety technology and discloses a battery failure early warning detection method and device, and a computer storage medium. Through real-time monitoring, this invention can automatically collect parameter records corresponding to each preset pressure-related parameter of a first battery module. Then, using a preset battery pack pressure model, it performs target pressure calculation on the first battery module, improving the accuracy of calculating the upper limit pressure range of the first battery module. The calculated module pressure information serves as a benchmark for judging whether the first battery module meets the failure early warning conditions, providing a basis for subsequent failure early warning condition judgment and improving the accuracy of setting the failure early warning conditions. Subsequently, the detected real-time module pressure value is compared with the module pressure information to achieve intelligent judgment of whether the first battery module meets the failure early warning conditions, improving the speed of detecting battery failure in a battery module and the timeliness of executing failure early warning operations.

Description

Technical Field

[0001] This invention relates to the field of battery safety technology, and in particular to a battery failure early warning detection method and device, and a computer storage medium. Background Technology

[0002] In electric vehicles, energy storage systems, and other electronic devices, batteries serve as the core energy storage unit, and their safety and reliability are crucial for the normal operation of the entire system. Battery failure warning technology, as one of the key technologies for ensuring battery safety, has received widespread attention and research in recent years. However, despite the existence of various battery failure warning technologies, some significant shortcomings and deficiencies still exist.

[0003] First, traditional battery failure early warning technologies primarily rely on monitoring external battery characteristics, such as voltage, current, and temperature. However, these external characteristics often only reflect the overall state of the battery and cannot accurately capture subtle changes within it. For example, when a minor short circuit or abnormal increase in internal resistance occurs, traditional external characteristic monitoring methods may fail to issue a timely warning, thus increasing the risk of battery failure. Second, existing battery failure early warning technologies are often inaccurate in determining the timing of battery failure. Because battery failure is a complex process involving multiple physical and chemical changes, it is difficult to accurately determine the timing of battery failure using a single monitoring parameter. Therefore, providing a corresponding solution to the problems of slow warning speed and low accuracy in existing battery failure early warning technologies is particularly important. Summary of the Invention

[0004] This invention provides a battery failure early warning detection method and device, and a computer storage medium, which can realize battery failure early warning based on cell pressure value, improve the detection speed and accuracy of battery failure, and improve the early warning speed and accuracy of battery failure.

[0005] To address the aforementioned technical problems, the first aspect of this invention discloses a battery failure early warning detection method, the method comprising:

[0006] The first battery module to be warned is identified, and parameter recording data corresponding to the first battery module and each preset pressure correlation parameter is collected.

[0007] Based on the pre-constructed battery pack pressure model, target pressure calculation is performed on the parameter recording data to obtain the module pressure information corresponding to the first battery module. The module pressure information includes the upper limit pressure range corresponding to the first battery module. The upper limit pressure range is the pressure value range corresponding to the pressure value indicating that the first battery module has reached the battery failure state.

[0008] Based on a preset pressure sensor, pressure detection is performed on the first battery module to obtain the real-time module pressure value corresponding to the first battery module.

[0009] Based on the module pressure information, determine whether the real-time module pressure value indicates that the first battery module meets the preset failure warning conditions; when it is determined that the real-time module pressure value indicates that the first battery module meets the preset failure warning conditions, perform a failure warning operation on the first battery module according to the set warning process to indicate that the first battery module is in a preset battery failure state.

[0010] As an optional implementation, in the first aspect of the present invention, the method further includes:

[0011] The second battery module to be analyzed is identified, and the second battery module includes multiple battery cells;

[0012] Collect historical usage data of the second battery module within a preset monitoring period, including pressure change data of all cells of the second battery module during use;

[0013] Based on the historical usage data, influence parameters that are related to the cell pressure between all the cells in the second battery module are determined, resulting in multiple pressure correlation parameters.

[0014] Based on all the pressure-related parameters and the set battery pressure analysis algorithm, a battery pack pressure model is constructed.

[0015] The pressure change data is input into the battery pack pressure model to train the battery pack pressure model.

[0016] As an optional implementation, in the first aspect of the present invention, the module pressure information further includes a failure warning pressure value of the first battery module and pressure value change information corresponding to the real-time module pressure value; the upper limit pressure range corresponding to the first battery module includes a first range endpoint value and a second range endpoint value; the first range endpoint value is less than the second range endpoint value; the failure warning pressure value corresponds to the first range endpoint value;

[0017] The step of determining whether the real-time module pressure value indicates that the first battery module meets the preset failure warning conditions based on the module pressure information includes:

[0018] Determine whether the real-time module pressure value has reached the range value corresponding to the upper limit pressure range; when it is determined that the real-time module pressure value has reached the range value corresponding to the upper limit pressure range, determine that the real-time module pressure value indicates that the first battery module meets the preset failure warning conditions;

[0019] When it is determined that the real-time module pressure value has not reached the range value corresponding to the upper limit pressure range, the numerical difference between the failure warning pressure value and the real-time module pressure value is calculated.

[0020] Based on the battery pack pressure model, the numerical difference and the pressure value change information are analyzed to obtain the estimated time for the first battery module to reach the upper limit pressure range and its corresponding target prediction time.

[0021] When the target predicted time is reached at the current time, the real-time module pressure value indicates that the first battery module meets the preset failure warning conditions.

[0022] As an optional implementation, in the first aspect of the present invention, the pressure-related parameters include at least a time parameter and a temperature parameter;

[0023] Before inputting the pressure change data into the battery pack pressure model to train the battery pack pressure model, the method further includes:

[0024] Based on cell charging and cell discharging, the historical usage data is divided into primary data segments to obtain primary data segmentation results corresponding to the historical usage data; the primary data segmentation results include charging datasets corresponding to cell charging and discharging datasets corresponding to cell discharging.

[0025] Based on the pressure correlation parameters, a second-level partitioning result corresponding to the first-level partitioning result is obtained from the first-level partitioning result; the second-level partitioning result includes multiple sub-datasets, which include a charging-time dataset and a charging-temperature dataset corresponding to the charging dataset, and a discharging-time dataset and a discharging-temperature dataset corresponding to the discharging dataset;

[0026] The pressure change data is updated based on the secondary partitioning results, and the operation of inputting the pressure change data into the battery pack pressure model to train the corresponding battery pack pressure model is triggered.

[0027] As an optional implementation, in the first aspect of the present invention, the specific training method for training the battery pack pressure model using the pressure change data includes:

[0028] Based on the battery pressure analysis algorithm, the battery pack pressure model performs cell pressure change analysis on each of the subsets to obtain analysis results corresponding to each subset; each analysis result includes at least one division interval and the number of intervals corresponding to all the division intervals; each division interval corresponds to a cell pressure value interval.

[0029] For each analysis result, a pressure analysis sub-algorithm matching the analysis result is determined based on the number of pressure intervals corresponding to the analysis result;

[0030] Based on the stress analysis sub-algorithm adapted to each analysis result, a preset training operation is performed on the sub-dataset corresponding to the analysis result. The preset training operation includes at least determining the algorithm parameters and optimizing the algorithm parameters for the stress analysis sub-algorithm.

[0031] After each stress analysis sub-algorithm is determined to have converged during training, the stress analysis sub-algorithm is considered to have completed training.

[0032] As an optional implementation, in the first aspect of the present invention, the step of performing target pressure calculation on the parameter recording data according to a pre-constructed battery pack pressure model to obtain module pressure information corresponding to the first battery module includes:

[0033] Based on the pre-constructed battery pack pressure model, the data type corresponding to the parameter recording data is determined. The data type includes those representing charging type or discharging type. When the data type is the charging type, the parameter recording data indicates that the first battery module is in a cell charging state; when the data type is the discharging type, the parameter recording data indicates that the first battery module is in a cell discharging state.

[0034] Determine the target pressure analysis sub-algorithm that is compatible with the data type from all the pressure analysis sub-algorithms in the battery pressure module;

[0035] Analysis indicators are extracted from the parameter recording data. The analysis indicators include at least one of the following: real-time moment corresponding to the time parameter, real-time time node, real-time temperature corresponding to the temperature parameter, and real-time temperature node.

[0036] The target pressure is calculated based on the target pressure analysis sub-algorithm to obtain target pressure data that matches the analysis index, which is used as the module pressure information corresponding to the first battery module.

[0037] As an optional implementation, in the first aspect of the present invention, all the pressure analysis sub-algorithms include a charging-time algorithm, a charging-temperature algorithm, a discharging-time algorithm, and a discharging-temperature algorithm;

[0038] The step of performing target pressure calculation on the analysis index according to the target pressure analysis sub-algorithm to obtain target pressure data adapted to the analysis index includes:

[0039] When the quantity corresponding to the target pressure analysis sub-algorithm is 1, the target pressure calculation is performed on the analysis index according to the target pressure analysis sub-algorithm to obtain target pressure data that matches the analysis index.

[0040] When the number of the target pressure analysis sub-algorithms is greater than 1, the analysis index that is adapted to each target pressure analysis sub-algorithm is determined as the target analysis index.

[0041] For each target pressure analysis sub-algorithm, target pressure calculation is performed on the target analysis index corresponding to the target pressure analysis sub-algorithm to obtain graded pressure data corresponding to the target analysis index corresponding to the target pressure analysis sub-algorithm.

[0042] Based on the weight values ​​set for each target pressure analysis sub-algorithm, a preset weight calculation is performed on all the graded pressure data to obtain the weight calculation results corresponding to all the graded pressure data, which are used as target pressure data adapted to the analysis indicators.

[0043] A second aspect of the present invention discloses a battery failure early warning detection device, the device comprising:

[0044] The determination module is used to identify the first battery module that needs to be warned.

[0045] The data acquisition module is used to collect parameter record data corresponding to the first battery module and each preset pressure-related parameter;

[0046] The pressure calculation module is used to perform target pressure calculation on the parameter recording data according to the pre-built battery pack pressure model to obtain the module pressure information corresponding to the first battery module. The module pressure information includes the upper limit pressure range corresponding to the first battery module. The upper limit pressure range is the pressure value range corresponding to the pressure value indicating that the first battery module has reached the battery failure state.

[0047] The pressure detection module is used to perform pressure detection on the first battery module according to the preset pressure sensor, and obtain the real-time module pressure value corresponding to the first battery module.

[0048] The failure warning module is used to determine whether the real-time module pressure value indicates that the first battery module meets the preset failure warning conditions based on the module pressure information; when it is determined that the real-time module pressure value indicates that the first battery module meets the preset failure warning conditions, the module performs a failure warning operation on the first battery module according to the set warning process to indicate that the first battery module is in a preset battery failure state.

[0049] As an optional implementation, in a second aspect of the present invention, the determining module is further configured to determine a second battery module to be analyzed, the second battery module comprising a plurality of battery cells;

[0050] The data acquisition module is also used to collect historical usage data of the second battery module during a preset monitoring period. The historical usage data includes pressure change data of all cells of the second battery module during use.

[0051] The determining module is further configured to determine, based on the historical usage data, an influence parameter that is related to the cell pressure between all the cells in the second battery module, thereby obtaining multiple pressure correlation parameters;

[0052] The device further includes:

[0053] The model building module is used to construct a battery pack pressure model based on all the pressure-related parameters and the set battery pressure analysis algorithm.

[0054] The model training module is used to input the pressure change data into the battery pack pressure model to train the battery pack pressure model.

[0055] As an optional implementation, in a second aspect of the present invention, the module pressure information further includes a failure warning pressure value of the first battery module and pressure value change information corresponding to the real-time module pressure value; the upper limit pressure range corresponding to the first battery module includes a first range endpoint value and a second range endpoint value; the first range endpoint value is less than the second range endpoint value; the failure warning pressure value corresponds to the first range endpoint value;

[0056] The failure warning module determines whether the real-time module pressure value indicates that the first battery module meets the preset failure warning conditions based on the module pressure information in the following ways:

[0057] Determine whether the real-time module pressure value has reached the range value corresponding to the upper limit pressure range; when it is determined that the real-time module pressure value has reached the range value corresponding to the upper limit pressure range, determine that the real-time module pressure value indicates that the first battery module meets the preset failure warning conditions;

[0058] When it is determined that the real-time module pressure value has not reached the range value corresponding to the upper limit pressure range, the numerical difference between the failure warning pressure value and the real-time module pressure value is calculated.

[0059] Based on the battery pack pressure model, the numerical difference and the pressure value change information are analyzed to obtain the estimated time for the first battery module to reach the upper limit pressure range and its corresponding target prediction time.

[0060] When the target predicted time is reached at the current time, the real-time module pressure value indicates that the first battery module meets the preset failure warning conditions.

[0061] As an optional implementation, in the second aspect of the present invention, the pressure-related parameters include at least a time parameter and a temperature parameter;

[0062] The device further includes:

[0063] The data partitioning module is used to perform first-level data partitioning on the historical usage data based on cell charging and cell discharging before the model training module inputs the pressure change data into the battery pack pressure model to train the battery pack pressure model, thereby obtaining first-level data partitioning results corresponding to the historical usage data; the first-level data partitioning results include charging datasets corresponding to cell charging and discharging datasets corresponding to cell discharging.

[0064] The data partitioning module is further configured to use the pressure correlation parameters as a benchmark to obtain a secondary partitioning result corresponding to the primary partitioning result; the secondary partitioning result includes multiple sub-datasets, which include a charging-time dataset and a charging-temperature dataset corresponding to the charging dataset, and a discharging-time dataset and a discharging-temperature dataset corresponding to the discharging dataset;

[0065] The data update module is used to update the pressure change data according to the secondary division result, and trigger the model training module to perform the operation of inputting the pressure change data into the battery pack pressure model to train the corresponding battery pack pressure model.

[0066] As an optional implementation, in the second aspect of the present invention, the specific training method for training the battery pack pressure model using the pressure change data includes:

[0067] Based on the battery pressure analysis algorithm, the battery pack pressure model performs cell pressure change analysis on each of the subsets to obtain analysis results corresponding to each subset; each analysis result includes at least one division interval and the number of intervals corresponding to all the division intervals; each division interval corresponds to a cell pressure value interval.

[0068] For each analysis result, a pressure analysis sub-algorithm matching the analysis result is determined based on the number of pressure intervals corresponding to the analysis result;

[0069] Based on the stress analysis sub-algorithm adapted to each analysis result, a preset training operation is performed on the sub-dataset corresponding to the analysis result. The preset training operation includes at least determining the algorithm parameters and optimizing the algorithm parameters for the stress analysis sub-algorithm.

[0070] After each stress analysis sub-algorithm is determined to have converged during training, the stress analysis sub-algorithm is considered to have completed training.

[0071] As an optional implementation, in the second aspect of the present invention, the pressure calculation module performs target pressure calculation on the parameter recording data based on a pre-constructed battery pack pressure model to obtain the module pressure information corresponding to the first battery module. Specifically, this includes:

[0072] Based on the pre-constructed battery pack pressure model, the data type corresponding to the parameter recording data is determined. The data type includes those representing charging type or discharging type. When the data type is the charging type, the parameter recording data indicates that the first battery module is in a cell charging state; when the data type is the discharging type, the parameter recording data indicates that the first battery module is in a cell discharging state.

[0073] Determine the target pressure analysis sub-algorithm that is compatible with the data type from all the pressure analysis sub-algorithms in the battery pressure module;

[0074] Analysis indicators are extracted from the parameter recording data. The analysis indicators include at least one of the following: real-time moment corresponding to the time parameter, real-time time node, real-time temperature corresponding to the temperature parameter, and real-time temperature node.

[0075] The target pressure is calculated based on the target pressure analysis sub-algorithm to obtain target pressure data that matches the analysis index, which is used as the module pressure information corresponding to the first battery module.

[0076] As an optional implementation, in the second aspect of the present invention, all the pressure analysis sub-algorithms include a charging-time algorithm, a charging-temperature algorithm, a discharging-time algorithm, and a discharging-temperature algorithm;

[0077] The pressure calculation module performs target pressure calculation on the analysis index according to the target pressure analysis sub-algorithm to obtain target pressure data that matches the analysis index. Specifically, this includes the following methods:

[0078] When the quantity corresponding to the target pressure analysis sub-algorithm is 1, the target pressure calculation is performed on the analysis index according to the target pressure analysis sub-algorithm to obtain target pressure data that matches the analysis index.

[0079] When the number of the target pressure analysis sub-algorithms is greater than 1, the analysis index that is adapted to each target pressure analysis sub-algorithm is determined as the target analysis index.

[0080] For each target pressure analysis sub-algorithm, target pressure calculation is performed on the target analysis index corresponding to the target pressure analysis sub-algorithm to obtain graded pressure data corresponding to the target analysis index corresponding to the target pressure analysis sub-algorithm.

[0081] Based on the weight values ​​set for each target pressure analysis sub-algorithm, a preset weight calculation is performed on all the graded pressure data to obtain the weight calculation results corresponding to all the graded pressure data, which are used as target pressure data adapted to the analysis indicators.

[0082] A third aspect of the present invention discloses another battery failure early warning detection device, the device comprising:

[0083] Memory containing executable program code;

[0084] A processor coupled to the memory;

[0085] The processor calls the executable program code stored in the memory to execute the battery failure early warning detection method disclosed in the first aspect of the present invention.

[0086] The fourth aspect of the present invention discloses a computer storage medium storing computer instructions, which, when invoked, are used to execute the battery failure early warning detection method disclosed in the first aspect of the present invention.

[0087] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0088] This invention provides a battery failure early warning detection method, which includes: determining a first battery module to be warned, and simultaneously collecting parameter recording data corresponding to the first battery module and each preset pressure association parameter; performing target pressure calculation on the parameter recording data according to a pre-constructed battery pack pressure model to obtain module pressure information corresponding to the first battery module, wherein the module pressure information includes an upper limit pressure range corresponding to the first battery module; the upper limit pressure range is a pressure value range corresponding to the pressure value indicating that the first battery module has reached a battery failure state; performing pressure detection on the first battery module according to a preset pressure sensor to obtain a real-time module pressure value corresponding to the first battery module; determining whether the real-time module pressure value indicates that the first battery module meets a preset failure early warning condition based on the module pressure information; when it is determined that the real-time module pressure value indicates that the first battery module meets the preset failure early warning condition, performing a failure early warning operation on the first battery module according to the set early warning process to indicate that the first battery module is in a preset battery failure state. As can be seen, by implementing this invention, an early warning scheme is set up for abnormal battery cell pressure values ​​(excessive pressure). Upon identifying the first battery module requiring an early warning, the system automatically collects parameter records corresponding to each preset pressure-related parameter for that first battery module. Then, using a pre-constructed battery pack pressure model, the system performs target pressure calculations on the first battery module to obtain module pressure information. This improves the accuracy of calculating and determining the upper limit pressure range for each battery module. This module pressure information includes the calculated upper limit pressure range for the first battery module, serving as a benchmark for judging whether the first battery module has reached the failure warning condition. This provides a basis for subsequent failure warning condition judgment and improves the accuracy of failure warning condition setting. Then, real-time pressure detection is performed on the first battery module to obtain the real-time module pressure value. The real-time module pressure value is compared with the module pressure information to realize intelligent judgment of whether the first battery module meets the failure warning condition. If the first battery module meets the failure warning condition, the corresponding failure warning operation is executed. This improves the accuracy of the condition judgment of a certain battery module meeting the failure warning condition, and helps to improve the detection speed when a battery module has battery failure, as well as the timeliness and accuracy of the failure warning operation. Attached Figure Description

[0089] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0090] Figure 1This is a schematic flowchart of a battery failure early warning detection method disclosed in an embodiment of the present invention;

[0091] Figure 2 This is a schematic flowchart of another battery failure early warning detection method disclosed in an embodiment of the present invention;

[0092] Figure 3 This is a schematic diagram of the structure of a battery failure early warning detection device disclosed in an embodiment of the present invention;

[0093] Figure 4 This is a schematic diagram of another battery failure early warning detection device disclosed in an embodiment of the present invention;

[0094] Figure 5 This is a schematic diagram of the structure of another battery failure early warning detection device disclosed in an embodiment of the present invention. Detailed Implementation

[0095] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. 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.

[0096] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product, or end that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or ends.

[0097] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0098] This invention discloses a battery failure early warning detection method and device, and a computer storage medium. It sets up an early warning scheme for abnormal (excessively high) pressure values ​​in battery cells. Upon identifying a first battery module requiring early warning, it automatically collects parameter records corresponding to each preset pressure-related parameter for that first battery module. Then, using a pre-constructed battery pack pressure model, it performs target pressure calculations on the first battery module to obtain module pressure information. This improves the accuracy of calculating and determining the upper limit pressure range for each battery module. The module pressure information includes the calculated upper limit pressure range for the first battery module, serving as a criterion for judging whether the first battery module has reached the failure threshold. The baseline conditions for failure warning are established to provide a basis for subsequent judgments on failure warning conditions, improving the accuracy of setting failure warning conditions. Subsequently, real-time pressure detection is performed on the first battery module to obtain the real-time module pressure value. This real-time module pressure value is compared with the module pressure information to intelligently determine whether the first battery module meets the failure warning conditions. If the first battery module meets the failure warning conditions, the corresponding failure warning operation is executed. This improves the accuracy of determining whether a battery module meets the failure warning conditions, and helps to improve the speed of detecting battery failure in a battery module, as well as the timeliness and accuracy of the failure warning operation. These will be explained in detail below.

[0099] Example 1

[0100] Please see Figure 1 , Figure 1 This is a schematic flowchart of a battery failure early warning detection method disclosed in an embodiment of the present invention. Wherein, Figure 1 The described battery failure early warning detection method can be applied to battery failure early warning detection devices, and the embodiments of the present invention are not limited thereto. Figure 1 As shown, the battery failure early warning detection method may include the following operations:

[0101] 101. Determine the first battery module to be warned, and at the same time collect parameter recording data corresponding to the first battery module and each preset pressure correlation parameter.

[0102] In this embodiment of the invention, the first battery module may consist of at least one battery cell.

[0103] In this embodiment of the invention, the first battery module may include multiple battery modules. That is, the first battery module may be a descriptive name for multiple battery modules. In fact, the first battery module refers to a set of battery modules composed of multiple battery modules. This embodiment of the invention does not limit this.

[0104] 102. Based on the pre-constructed battery pack pressure model, perform target pressure calculation on the parameter recording data to obtain the module pressure information corresponding to the first battery module.

[0105] In this embodiment of the invention, the module pressure information includes the upper limit pressure range corresponding to the first battery module; the upper limit pressure range is the pressure value range corresponding to the pressure value indicating that the first battery module has reached the battery failure state.

[0106] In this embodiment of the invention, when the first battery module refers to a set of battery modules consisting of multiple battery modules, and the set of battery modules includes battery modules of different models, the module pressure information includes sub-module pressure information corresponding to each model of battery module.

[0107] 103. Based on the preset pressure sensor, perform pressure detection on the first battery module to obtain the real-time module pressure value corresponding to the first battery module.

[0108] In this embodiment of the invention, the pressure value of the first battery module can be collected and detected in real time by a pressure sensor installed on the first battery module, thereby obtaining the real-time module pressure value.

[0109] 104. Determine whether the real-time module pressure value indicates that the first battery module meets the preset failure warning conditions based on the module pressure information; when it is determined that the real-time module pressure value indicates that the first battery module meets the preset failure warning conditions, perform a failure warning operation on the first battery module according to the set warning process to indicate that the first battery module is in a preset battery failure state.

[0110] It is evident that implementation Figure 1 The described battery failure early warning detection method includes an early warning scheme for abnormal battery cell pressure values ​​(excessive pressure). Upon identifying the first battery module requiring an early warning, it automatically collects parameter records corresponding to each preset pressure-related parameter for that module. Then, using a pre-constructed battery pack pressure model, it performs target pressure calculations on the first battery module to obtain module pressure information. This improves the accuracy of calculating and determining the upper limit pressure range for each battery module. This module pressure information includes the calculated upper limit pressure range for the first battery module, serving as the basis for judging whether the first battery module meets the failure early warning conditions. The system establishes accurate conditions to provide a basis for subsequent failure warning condition judgments, improving the accuracy of failure warning condition settings. Then, real-time pressure detection is performed on the first battery module to obtain the real-time module pressure value. This real-time module pressure value is compared with the module pressure information to intelligently determine whether the first battery module meets the failure warning conditions. If the first battery module meets the failure warning conditions, the corresponding failure warning operation is executed. This improves the accuracy of condition judgment for a specific battery module meeting the failure warning conditions, and helps to improve the speed of detecting battery failure in a specific battery module, as well as the timeliness and accuracy of failure warning operations.

[0111] In an optional embodiment, the module pressure information further includes a failure warning pressure value of the first battery module and pressure value change information corresponding to the real-time module pressure value; the upper limit pressure range corresponding to the first battery module includes a first range endpoint value and a second range endpoint value; the first range endpoint value is less than the second range endpoint value; the failure warning pressure value corresponds to the first range endpoint value;

[0112] The method for determining whether the real-time module pressure value indicates that the first battery module meets the preset failure warning conditions based on the module pressure information in step 104 above specifically includes:

[0113] Determine whether the real-time module pressure value has reached the range value corresponding to the upper limit pressure range; when it is determined that the real-time module pressure value has reached the range value corresponding to the upper limit pressure range, determine that the real-time module pressure value indicates that the first battery module meets the preset failure warning conditions;

[0114] When it is determined that the real-time module pressure value has not reached the range corresponding to the upper limit pressure range, the numerical difference between the failure warning pressure value and the real-time module pressure value is calculated.

[0115] Based on the battery pack pressure model, the numerical difference and pressure value change information are analyzed to obtain the estimated time for the first battery module to reach the upper limit pressure range and the corresponding target prediction time.

[0116] When the target predicted time is reached at the current moment, the real-time module pressure value indicates that the first battery module meets the preset failure warning conditions.

[0117] As can be seen, in this optional embodiment, at least two judgment methods are set for whether the first battery module meets the failure warning conditions. The first method directly compares the real-time module pressure value of the first battery module with the range value of the upper limit pressure range, thereby quickly and easily determining whether the first battery module meets the failure warning conditions, improving the determination speed of whether the battery module meets or does not meet the failure warning conditions, and this judgment method is more convenient, fast and direct. The second method can calculate the difference between the failure warning pressure value and the real-time module pressure value, calculate the estimated time for the difference between the two values, and calculate the target prediction time corresponding to the estimated time. Then, it is determined that the failure warning conditions are met when the target prediction time is reached at the current time. That is, the second method provides a failure warning prediction mechanism, which can calculate and predict the future time (target prediction time) when the first battery module meets the failure warning in advance. Unlike comparing the real-time module pressure value with the upper limit pressure range, this prediction mechanism can provide a battery failure warning in advance, improving the intelligence level of battery failure warning, enriching the warning methods of battery failure warning, and improving the prediction accuracy of battery failure warning.

[0118] In another alternative embodiment, the method further includes:

[0119] When it is determined that the current time has reached the target prediction time, the real-time module pressure value obtained by the pressure sensor after performing pressure detection on the first battery module at the target prediction time is recorded as the pressure value at the prediction time.

[0120] Determine whether the predicted pressure value at the predicted time is consistent with the value at the endpoint of the first interval; if it is determined that the predicted time and the target predicted time are correct.

[0121] When it is determined that the pressure value at the predicted time is inconsistent with the endpoint value of the first interval, the numerical difference between the pressure value at the predicted time and the endpoint value of the first interval is calculated to obtain the error difference value and its corresponding difference attribute; the difference attribute includes a first attribute or a second attribute, the first attribute indicating that the pressure value at the predicted time is greater than the endpoint value of the first interval; the second attribute indicating that the pressure value at the predicted time is less than the endpoint value of the first interval.

[0122] Based on the error difference and its attribute, perform at least one of the following operations on the battery pack pressure model: model parameter correction, model parameter update, and model parameter optimization.

[0123] As can be seen, in this optional embodiment, a model update and optimization mechanism for the battery pack pressure model is set up. In the case where the pressure value at the predicted time is inconsistent with the value at the endpoint of the first interval, the error difference can be automatically calculated and the difference attribute can be determined. Based on the error difference and the difference attribute, the model parameters are optimized and iterated, thereby improving the applicability of the battery pack pressure model.

[0124] In another optional embodiment, the method by which step 104 above performs a failure warning operation on the first battery module according to the set warning process specifically includes:

[0125] Based on module pressure information, real-time module pressure value, and failure warning conditions, generate failure warning information for the first battery module.

[0126] Determine the warning method and feedback terminal corresponding to the failure warning information; the warning method includes one of the following: warning sound prompt, warning light prompt, and warning light and sound prompt; the feedback terminal includes a server and / or a user terminal; the user terminal includes a mobile phone terminal, a platform terminal, and a computer terminal;

[0127] The feedback end is used as the data receiving terminal to transmit failure warning information to the feedback end; after confirming that the feedback end has received the failure warning information, the feedback end is controlled to perform failure warning operation according to the warning method.

[0128] When the warning method is an audible warning, the failure warning operation triggers the sound playback device / equipment set in the feedback terminal to play the warning sound. When the warning method is a light warning, the failure warning operation triggers the warning light installed in the feedback terminal to display different warning lights according to the set light display program, such as controlling the warning light to be constantly on or controlling the warning light to display flashing lights. When the warning method is a warning light and audible warning, the corresponding failure warning operations for both the audible and light warnings are triggered simultaneously.

[0129] As can be seen, in this optional embodiment, failure warning information can be personalized and intelligently transmitted according to actual warning needs, which enriches the display means of failure warning information, improves the applicability of failure warning information, and enhances the convenience for users to receive failure warning information on different terminals.

[0130] Example 2

[0131] Please see Figure 2 , Figure 2 This is a schematic flowchart of another battery failure early warning detection method disclosed in an embodiment of the present invention. Wherein, Figure 2 The described battery failure early warning detection method can be applied to battery failure early warning detection devices, and the embodiments of the present invention are not limited thereto. Figure 2 As shown, the battery failure early warning detection method may include the following operations:

[0132] 201. Identify the second battery module to be analyzed. The second battery module includes multiple battery cells.

[0133] 202. Collect historical usage data of the second battery module within a preset monitoring period. The historical usage data includes the pressure change data of all cells of the second battery module during use.

[0134] 203. Based on historical usage data, determine the influencing parameters that are related to the cell pressure between all cells in the second battery module, and obtain multiple pressure correlation parameters.

[0135] 204. Based on all pressure-related parameters and the set battery pressure analysis algorithm, construct a battery pack pressure model.

[0136] 205. Input the pressure change data into the battery pack pressure model to train the battery pack pressure model.

[0137] 206. Determine the first battery module to be warned, and at the same time collect parameter recording data corresponding to the first battery module and each preset pressure correlation parameter.

[0138] 207. Based on the pre-constructed battery pack pressure model, perform target pressure calculation on the parameter recording data to obtain the module pressure information corresponding to the first battery module.

[0139] 208. Based on the preset pressure sensor, perform pressure detection on the first battery module to obtain the real-time module pressure value corresponding to the first battery module.

[0140] 209. Determine whether the real-time module pressure value indicates that the first battery module meets the preset failure warning conditions based on the module pressure information; when it is determined that the real-time module pressure value indicates that the first battery module meets the preset failure warning conditions, perform a failure warning operation on the first battery module according to the set warning process to indicate that the first battery module is in a preset battery failure state.

[0141] For further descriptions of steps 206-209 in this embodiment of the invention, please refer to the other specific descriptions of steps 101-104 in Embodiment 1. These descriptions will not be repeated in this embodiment of the invention.

[0142] It is evident that implementation Figure 2 The described battery failure early warning detection method establishes a mechanism for constructing and training a battery pack pressure model. This includes determining the second battery module, collecting its historical usage data, determining multiple pressure-related parameters, constructing the battery pack pressure model, and intelligently training the model. Through deep learning and analysis of historical usage data using the battery pack pressure model, the influence / correlation information between the cell pressure change data and pressure-related parameters of the battery module is determined. Subsequently, using the trained battery pack pressure model, the module pressure information of the first battery module to be analyzed can be accurately determined. In other words, the construction and training of this battery pack pressure model helps improve the accuracy of subsequent determination / calculation of module pressure information.

[0143] In an optional embodiment, the pressure-related parameters include at least a time parameter and a temperature parameter;

[0144] Before inputting the pressure change data into the battery pack pressure model to train the battery pack pressure model, the method further includes:

[0145] Based on cell charging and cell discharging, a first-level data partitioning is performed on historical usage data to obtain the first-level data partitioning results corresponding to the historical usage data; the first-level data partitioning results include the charging dataset corresponding to cell charging and the discharging dataset corresponding to cell discharging.

[0146] Based on the pressure correlation parameters, the second-level partitioning results are obtained from the first-level partitioning results. The second-level partitioning results include multiple sub-datasets, which include the charging-time dataset and charging-temperature dataset corresponding to the charging dataset, and the discharging-time dataset and discharging-temperature dataset corresponding to the discharging dataset.

[0147] The pressure change data is updated based on the secondary partitioning results, and the operation of inputting the pressure change data into the battery pack pressure model is triggered to train the corresponding operation of the battery pack pressure model.

[0148] As can be seen, in this optional embodiment, before training the battery pack pressure model using historical usage data including pressure change data, a data preprocessing process can be performed on the pressure change data: First-level data partitioning is performed based on two different operation types, cell charging and cell discharging, thereby dividing the historical usage data into different charging and discharging datasets. This facilitates targeted data learning and analysis during subsequent battery pack pressure model training, avoiding situations where individual differences in cell pressure values ​​due to cell charging and discharging increase the training difficulty and reduce training efficiency of the battery pack pressure model. Furthermore, after partitioning the data based on cell charging and discharging, a second-level data partitioning can be performed on the first-level partitioning result based on refined pressure correlation parameters. Based on the concept of controlled variables, the number of pressure correlation parameters that need to be calculated is reduced, thereby improving the training speed and reducing the training difficulty of the battery pack pressure model. Simultaneously, the final second-level partitioning result includes multiple subset datasets, achieving accurate subset partitioning while also expanding the amount of training data to a certain extent.

[0149] In another optional embodiment, the specific training method for training the battery pack pressure model using pressure change data includes:

[0150] Based on the battery pressure analysis algorithm, the battery pack pressure model performs cell pressure change analysis on each subset of data to obtain analysis results corresponding to each subset of data. Each analysis result includes at least one division interval and the number of intervals corresponding to all division intervals. Each division interval corresponds to a cell pressure value interval.

[0151] For each analysis result, a pressure analysis sub-algorithm matching the analysis result is determined based on the number of pressure intervals corresponding to that analysis result;

[0152] Based on the stress analysis sub-algorithm adapted for each analysis result, a preset training operation is performed on the corresponding subset of the analysis result. The preset training operation includes at least the determination and optimization of algorithm parameters for the stress analysis sub-algorithm.

[0153] After each stress analysis sub-algorithm has been trained to convergence, the stress analysis sub-algorithm is considered to have completed training.

[0154] In this optional embodiment, after the battery pack stress model training is completed, at least four types of stress analysis sub-algorithms are included: cell charging-temperature, cell charging-time, cell discharging-temperature, and cell discharging-time, totaling four types of stress analysis sub-algorithms; furthermore, for different cell models / battery module models, there are four matching types of stress analysis sub-algorithms.

[0155] As can be seen, in this optional embodiment, after the battery pack pressure model detects the input subset, it can perform preliminary data analysis on each subset to obtain the division intervals corresponding to the pressure value intervals of each subset and the number of intervals. Then, it determines the pressure analysis sub-algorithm suitable for processing the subset based on the different number of intervals. Thus, each subset is used to intelligently train its suitable pressure analysis sub-algorithm, improving the utilization rate and accuracy of subset utilization; at the same time, it improves the accuracy of algorithm training for different subsets; and improves the training accuracy and reliability of the finally trained pressure analysis sub-algorithm.

[0156] In another optional embodiment, the method of performing target pressure calculation on parameter recording data based on a pre-built battery pack pressure model to obtain the module pressure information corresponding to the first battery module specifically includes:

[0157] Based on the pre-built battery pack pressure model, the data type corresponding to the parameter recording data is determined. The data type includes the charging type or the discharging type. When the data type is charging, the parameter recording data indicates that the first battery module is in the cell charging state; when the data type is discharging, the parameter recording data indicates that the first battery module is in the cell discharging state.

[0158] Determine the target pressure analysis sub-algorithm that is compatible with the data type from all pressure analysis sub-algorithms in the battery pressure module;

[0159] The analysis indicators are extracted from the parameter recording data. The analysis indicators include at least one of the following: real-time moment corresponding to the time parameter, real-time time node, real-time temperature corresponding to the temperature parameter, and real-time temperature node.

[0160] The target pressure calculation is performed on the analysis index according to the target pressure analysis sub-algorithm to obtain the target pressure data that matches the analysis index, which is used as the module pressure information corresponding to the first battery module.

[0161] As can be seen, in this optional embodiment, when the battery pack pressure model is trained in the front end, various pressure analysis sub-algorithms can be trained. After obtaining the parameter recording data, the target pressure analysis sub-algorithm suitable for analyzing the current parameter recording data can be accurately determined from the various pressure analysis sub-algorithms based on the data type and analysis index corresponding to the parameter recording data, thereby performing target pressure calculation and improving the operational accuracy of target pressure calculation.

[0162] In another alternative embodiment, all pressure analysis sub-algorithms include a charge-time algorithm, a charge-temperature algorithm, a discharge-time algorithm, and a discharge-temperature algorithm;

[0163] The above-mentioned method of performing target pressure calculations on the analysis indicators based on the target pressure analysis sub-algorithm to obtain target pressure data that matches the analysis indicators specifically includes:

[0164] When the quantity corresponding to the target pressure analysis sub-algorithm is 1, the target pressure calculation is performed on the analysis index according to the target pressure analysis sub-algorithm to obtain the target pressure data that matches the analysis index.

[0165] When the number of target pressure analysis sub-algorithms is greater than 1, the analysis index that is adapted to each target pressure analysis sub-algorithm will be determined as the target analysis index.

[0166] For each target pressure analysis sub-algorithm, target pressure calculation is performed on the target analysis index corresponding to the target pressure analysis sub-algorithm to obtain the graded pressure data corresponding to the target analysis index of the target pressure analysis sub-algorithm.

[0167] Based on the weight values ​​set for each target pressure analysis sub-algorithm, a preset weight calculation is performed on all graded pressure data to obtain the weight calculation results corresponding to all graded pressure data, which are used as target pressure data adapted to the analysis indicators.

[0168] In this optional embodiment, the data processing method for actually processing all the graded pressure data after determining multiple graded pressure data may include, in addition to the weight calculation, data processing based on a preset data transformation algorithm, big data processing based on a deep learning model, etc., and the embodiments of the present invention are not limited thereto.

[0169] As can be seen, in this optional embodiment, when the number of target pressure analysis sub-algorithms is singular (specifically, the number is 1), target pressure calculation is only performed on the target analysis indicators through this adapted and unique target pressure analysis sub-algorithm, which improves the convenience and accuracy of target pressure data analysis. When the number of target pressure analysis sub-algorithms is greater than 1, after each target pressure analysis sub-algorithm performs target pressure calculation on its corresponding target analysis indicators to obtain graded pressure data, weight calculation is set for all graded pressure data, realizing the comprehensive utilization of multiple graded pressure data and improving the calculation accuracy of the final target pressure data.

[0170] Example 3

[0171] Please see Figure 3 , Figure 3 This is a schematic diagram of the structure of a battery failure early warning detection device disclosed in an embodiment of the present invention. The battery failure early warning detection device can be a battery failure early warning detection terminal, equipment, system, or server. The server can be a local server, a remote server, or a cloud server (also known as a cloud-based server). When the server is not a cloud server, it can communicate with the cloud server; this embodiment of the present invention does not impose any limitations. Figure 3 As shown, the battery failure early warning detection device may include a determination module 301, a data acquisition module 302, a pressure calculation module 303, a pressure detection module 304, and a failure early warning module 305, wherein:

[0172] Module 301 is used to determine the first battery module currently under warning;

[0173] Data acquisition module 302 is used to acquire parameter record data corresponding to the first battery module and each preset pressure correlation parameter;

[0174] The pressure calculation module 303 is used to perform target pressure calculation on the parameter recording data according to the pre-built battery pack pressure model to obtain the module pressure information corresponding to the first battery module. The module pressure information includes the upper limit pressure range corresponding to the first battery module; the upper limit pressure range is the pressure value range corresponding to the pressure value indicating that the first battery module has reached the battery failure state.

[0175] The pressure detection module 304 is used to perform pressure detection on the first battery module according to the preset pressure sensor, and obtain the real-time module pressure value corresponding to the first battery module.

[0176] The failure warning module 305 is used to determine whether the real-time module pressure value indicates that the first battery module meets the preset failure warning conditions based on the module pressure information. When it is determined that the real-time module pressure value indicates that the first battery module meets the preset failure warning conditions, the failure warning operation is performed on the first battery module according to the set warning process to indicate that the first battery module is in a preset battery failure state.

[0177] It is evident that implementation Figure 3 The described battery failure early warning detection device is equipped with an early warning scheme for abnormal battery cell pressure values ​​(excessive pressure). Upon identifying a first battery module requiring an early warning, it automatically collects parameter records corresponding to each preset pressure-related parameter for that first battery module. Then, using a pre-constructed battery pack pressure model, it performs target pressure calculations on the first battery module to obtain module pressure information. This improves the accuracy of calculating and determining the upper limit pressure range for each battery module. This module pressure information includes the calculated upper limit pressure range for the first battery module, serving as the basis for judging whether the first battery module has reached the failure early warning condition. The system establishes accurate conditions to provide a basis for subsequent failure warning condition judgments, improving the accuracy of failure warning condition settings. Then, real-time pressure detection is performed on the first battery module to obtain the real-time module pressure value. This real-time module pressure value is compared with the module pressure information to intelligently determine whether the first battery module meets the failure warning conditions. If the first battery module meets the failure warning conditions, the corresponding failure warning operation is executed. This improves the accuracy of condition judgment for a specific battery module meeting the failure warning conditions, and helps to improve the speed of detecting battery failure in a specific battery module, as well as the timeliness and accuracy of failure warning operations.

[0178] In an optional embodiment, the determining module 301 is further configured to determine a second battery module to be analyzed, the second battery module comprising a plurality of battery cells;

[0179] The data acquisition module 302 is also used to collect historical usage data of the second battery module during a preset monitoring period. The historical usage data includes pressure change data of all cells of the second battery module during use.

[0180] The determination module 301 is also used to determine, based on historical usage data, the influence parameters that are related to the cell pressure between all cells in the second battery module, and obtain multiple pressure correlation parameters;

[0181] like Figure 4 As shown, the device also includes a model building module 306 and a model training module 307, wherein:

[0182] Model building module 306 is used to build a battery pack pressure model based on all pressure-related parameters and the set battery pressure analysis algorithm.

[0183] The model training module 307 is used to input pressure change data into the battery pack pressure model in order to train the battery pack pressure model.

[0184] As can be seen, in this optional embodiment, a mechanism for constructing and training a battery pack pressure model is set up, including determining the second battery module, collecting its historical usage data, determining multiple pressure-related parameters, constructing the battery pack pressure model, and intelligently training the battery pack pressure model. Through deep learning and analysis of historical usage data using the battery pack pressure model, the influence / correlation information between the cell pressure change data and pressure-related parameters of the battery module is determined. Subsequently, using the trained battery pack pressure model, the module pressure information of the first battery module to be analyzed can be accurately analyzed. That is, the construction and training of this battery pack pressure model helps to improve the accuracy of subsequent determination / calculation of module pressure information.

[0185] In another optional embodiment, the module pressure information further includes a failure warning pressure value of the first battery module and pressure value change information corresponding to the real-time module pressure value; the upper limit pressure range corresponding to the first battery module includes a first range endpoint value and a second range endpoint value; the first range endpoint value is less than the second range endpoint value; the failure warning pressure value corresponds to the first range endpoint value;

[0186] The failure warning module 305 determines whether the real-time module pressure value indicates that the first battery module meets the preset failure warning conditions based on the module pressure information in the following ways:

[0187] Determine whether the real-time module pressure value has reached the range value corresponding to the upper limit pressure range; when it is determined that the real-time module pressure value has reached the range value corresponding to the upper limit pressure range, determine that the real-time module pressure value indicates that the first battery module meets the preset failure warning conditions;

[0188] When it is determined that the real-time module pressure value has not reached the range corresponding to the upper limit pressure range, the numerical difference between the failure warning pressure value and the real-time module pressure value is calculated.

[0189] Based on the battery pack pressure model, the numerical difference and pressure value change information are analyzed to obtain the estimated time for the first battery module to reach the upper limit pressure range and the corresponding target prediction time.

[0190] When the target predicted time is reached at the current moment, the real-time module pressure value indicates that the first battery module meets the preset failure warning conditions.

[0191] As can be seen, in this optional embodiment, at least two judgment methods are set for whether the first battery module meets the failure warning conditions. The first method directly compares the real-time module pressure value of the first battery module with the range value of the upper limit pressure range, thereby quickly and easily determining whether the first battery module meets the failure warning conditions, improving the determination speed of whether the battery module meets or does not meet the failure warning conditions, and this judgment method is more convenient, fast and direct. The second method can calculate the difference between the failure warning pressure value and the real-time module pressure value, calculate the estimated time for the difference between the two values, and calculate the target prediction time corresponding to the estimated time. Then, it is determined that the failure warning conditions are met when the target prediction time is reached at the current time. That is, the second method provides a failure warning prediction mechanism, which can calculate and predict the future time (target prediction time) when the first battery module meets the failure warning in advance. Unlike comparing the real-time module pressure value with the upper limit pressure range, this prediction mechanism can provide a battery failure warning in advance, improving the intelligence level of battery failure warning, enriching the warning methods of battery failure warning, and improving the prediction accuracy of battery failure warning.

[0192] In yet another optional embodiment, the pressure-related parameters include at least a time parameter and a temperature parameter;

[0193] like Figure 4 As shown, the device also includes a data partitioning module 308 and a data updating module 309, wherein:

[0194] The data partitioning module 308 is used to input pressure change data into the battery pack pressure model in the model training module 307 before training the battery pack pressure model. Based on cell charging and cell discharging, it performs first-level data partitioning on historical usage data to obtain first-level data partitioning results corresponding to historical usage data. The first-level data partitioning results include charging datasets corresponding to cell charging and discharging datasets corresponding to cell discharging.

[0195] The data partitioning module 308 is also used to obtain a second-level partitioning result corresponding to the first-level partitioning result based on the pressure correlation parameters. The second-level partitioning result includes multiple sub-datasets, which include a charging-time dataset and a charging-temperature dataset corresponding to the charging dataset, and a discharging-time dataset and a discharging-temperature dataset corresponding to the discharging dataset.

[0196] The data update module 309 is used to update the pressure change data according to the secondary division results and trigger the model training module 307 to input the pressure change data into the battery pack pressure model to train the corresponding operation of the battery pack pressure model.

[0197] As can be seen, in this optional embodiment, before training the battery pack pressure model using historical usage data including pressure change data, a data preprocessing process can be performed on the pressure change data: First-level data partitioning is performed based on two different operation types, cell charging and cell discharging, thereby dividing the historical usage data into different charging and discharging datasets. This facilitates targeted data learning and analysis during subsequent battery pack pressure model training, avoiding situations where individual differences in cell pressure values ​​due to cell charging and discharging increase the training difficulty and reduce training efficiency of the battery pack pressure model. Furthermore, after partitioning the data based on cell charging and discharging, a second-level data partitioning can be performed on the first-level partitioning result based on refined pressure correlation parameters. Based on the concept of controlled variables, the number of pressure correlation parameters that need to be calculated is reduced, thereby improving the training speed and reducing the training difficulty of the battery pack pressure model. Simultaneously, the final second-level partitioning result includes multiple subset datasets, achieving accurate subset partitioning while also expanding the amount of training data to a certain extent.

[0198] In another optional embodiment, the specific training method for training the battery pack pressure model using pressure change data includes:

[0199] Based on the battery pressure analysis algorithm, the battery pack pressure model performs cell pressure change analysis on each subset of data to obtain analysis results corresponding to each subset of data. Each analysis result includes at least one division interval and the number of intervals corresponding to all division intervals. Each division interval corresponds to a cell pressure value interval.

[0200] For each analysis result, a pressure analysis sub-algorithm matching the analysis result is determined based on the number of pressure intervals corresponding to that analysis result;

[0201] Based on the stress analysis sub-algorithm adapted for each analysis result, a preset training operation is performed on the corresponding subset of the analysis result. The preset training operation includes at least the determination and optimization of algorithm parameters for the stress analysis sub-algorithm.

[0202] After each stress analysis sub-algorithm has been trained to convergence, the stress analysis sub-algorithm is considered to have completed training.

[0203] As can be seen, in this optional embodiment, after the battery pack pressure model detects the input subset, it can perform preliminary data analysis on each subset to obtain the division intervals corresponding to the pressure value intervals of each subset and the number of intervals. Then, it determines the pressure analysis sub-algorithm suitable for processing the subset based on the different number of intervals. Thus, each subset is used to intelligently train its suitable pressure analysis sub-algorithm, improving the utilization rate and accuracy of subset utilization; at the same time, it improves the accuracy of algorithm training for different subsets; and improves the training accuracy and reliability of the finally trained pressure analysis sub-algorithm.

[0204] In another optional embodiment, the pressure calculation module 303 performs target pressure calculation on the parameter recording data based on the pre-built battery pack pressure model to obtain the module pressure information corresponding to the first battery module. Specifically, this includes:

[0205] Based on the pre-built battery pack pressure model, the data type corresponding to the parameter recording data is determined. The data type includes the charging type or the discharging type. When the data type is charging, the parameter recording data indicates that the first battery module is in the cell charging state; when the data type is discharging, the parameter recording data indicates that the first battery module is in the cell discharging state.

[0206] Determine the target pressure analysis sub-algorithm that is compatible with the data type from all pressure analysis sub-algorithms in the battery pressure module;

[0207] The analysis indicators are extracted from the parameter recording data. The analysis indicators include at least one of the following: real-time moment corresponding to the time parameter, real-time time node, real-time temperature corresponding to the temperature parameter, and real-time temperature node.

[0208] The target pressure calculation is performed on the analysis index according to the target pressure analysis sub-algorithm to obtain the target pressure data that matches the analysis index, which is used as the module pressure information corresponding to the first battery module.

[0209] As can be seen, in this optional embodiment, when the battery pack pressure model is trained in the front end, various pressure analysis sub-algorithms can be trained. After obtaining the parameter recording data, the target pressure analysis sub-algorithm suitable for analyzing the current parameter recording data can be accurately determined from the various pressure analysis sub-algorithms based on the data type and analysis index corresponding to the parameter recording data, thereby performing target pressure calculation and improving the operational accuracy of target pressure calculation.

[0210] In this optional embodiment, all pressure analysis sub-algorithms further include a charge-time algorithm, a charge-temperature algorithm, a discharge-time algorithm, and a discharge-temperature algorithm;

[0211] The aforementioned pressure calculation module 303 performs target pressure calculations on the analysis indicators according to the target pressure analysis sub-algorithm, and the specific methods for obtaining target pressure data that matches the analysis indicators include:

[0212] When the quantity corresponding to the target pressure analysis sub-algorithm is 1, the target pressure calculation is performed on the analysis index according to the target pressure analysis sub-algorithm to obtain the target pressure data that matches the analysis index.

[0213] When the number of target pressure analysis sub-algorithms is greater than 1, the analysis index that is adapted to each target pressure analysis sub-algorithm will be determined as the target analysis index.

[0214] For each target pressure analysis sub-algorithm, target pressure calculation is performed on the target analysis index corresponding to the target pressure analysis sub-algorithm to obtain the graded pressure data corresponding to the target analysis index of the target pressure analysis sub-algorithm.

[0215] Based on the weight values ​​set for each target pressure analysis sub-algorithm, a preset weight calculation is performed on all graded pressure data to obtain the weight calculation results corresponding to all graded pressure data, which are used as target pressure data adapted to the analysis indicators.

[0216] As can be seen, in this optional embodiment, when the number of target pressure analysis sub-algorithms is singular (specifically, the number is 1), target pressure calculation is only performed on the target analysis indicators through this adapted and unique target pressure analysis sub-algorithm, which improves the convenience and accuracy of target pressure data analysis. When the number of target pressure analysis sub-algorithms is greater than 1, after each target pressure analysis sub-algorithm performs target pressure calculation on its corresponding target analysis indicators to obtain graded pressure data, weight calculation is set for all graded pressure data, realizing the comprehensive utilization of multiple graded pressure data and improving the calculation accuracy of the final target pressure data.

[0217] Example 4

[0218] Please see Figure 5 , Figure 5 This is a schematic diagram of the structure of another battery failure early warning detection device disclosed in an embodiment of the present invention. For example... Figure 5 As shown, the battery failure early warning detection device may include:

[0219] Memory 401 storing executable program code;

[0220] Processor 402 coupled to memory 401;

[0221] The processor 402 calls the executable program code stored in the memory 401 to execute the steps in the battery failure early warning detection method described in Embodiment 1 or Embodiment 2 of the present invention.

[0222] Example 5

[0223] This invention discloses a computer storage medium storing computer instructions. When these computer instructions are invoked, they are used to execute the steps in the battery failure early warning detection method described in Embodiment 1 or Embodiment 2 of this invention.

[0224] Example 6

[0225] This invention discloses a computer program product, which includes a non-transitory computer storage medium storing a computer program, and the computer program is operable to cause a computer to perform the steps in the battery failure early warning detection method described in Embodiment 1 or Embodiment 2.

[0226] The device embodiments described above are merely illustrative. The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0227] Through the detailed description of the above embodiments, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer storage medium, including read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0228] Finally, it should be noted that the battery failure early warning detection method and device and computer storage medium disclosed in the embodiments of the present invention are only preferred embodiments of the present invention, and are only used to illustrate the technical solutions of the present invention, and not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A battery failure early warning detection method, characterized in that, The method includes: The first battery module to be warned is identified, and parameter recording data corresponding to the first battery module and each preset pressure correlation parameter is collected. Based on the pre-constructed battery pack pressure model, target pressure calculation is performed on the parameter recording data to obtain the module pressure information corresponding to the first battery module. The module pressure information includes the upper limit pressure range corresponding to the first battery module. The upper limit pressure range is the pressure value range corresponding to the pressure value indicating that the first battery module has reached the battery failure state. Based on a preset pressure sensor, pressure detection is performed on the first battery module to obtain the real-time module pressure value corresponding to the first battery module. Based on the module pressure information, determine whether the real-time module pressure value indicates that the first battery module meets the preset failure warning conditions; when it is determined that the real-time module pressure value indicates that the first battery module meets the preset failure warning conditions, perform a failure warning operation on the first battery module according to the set warning process to indicate that the first battery module is in a preset battery failure state. The module pressure information also includes the failure warning pressure value of the first battery module and pressure value change information corresponding to the real-time module pressure value; the upper limit pressure range corresponding to the first battery module includes a first interval endpoint value and a second interval endpoint value; the first interval endpoint value is less than the second interval endpoint value; the failure warning pressure value corresponds to the first interval endpoint value. The step of determining whether the real-time module pressure value indicates that the first battery module meets the preset failure warning conditions based on the module pressure information includes: Determine whether the real-time module pressure value has reached the range value corresponding to the upper limit pressure range; when it is determined that the real-time module pressure value has reached the range value corresponding to the upper limit pressure range, determine that the real-time module pressure value indicates that the first battery module meets the preset failure warning conditions; When it is determined that the real-time module pressure value has not reached the range value corresponding to the upper limit pressure range, the numerical difference between the failure warning pressure value and the real-time module pressure value is calculated. Based on the battery pack pressure model, the numerical difference and the pressure value change information are analyzed to obtain the estimated time for the first battery module to reach the upper limit pressure range and its corresponding target prediction time. When the target predicted time is reached at the current time, the real-time module pressure value indicates that the first battery module meets the preset failure warning conditions.

2. The battery failure early warning detection method according to claim 1, characterized in that, The method further includes: The second battery module to be analyzed is identified, and the second battery module includes multiple battery cells; Collect historical usage data of the second battery module within a preset monitoring period, including pressure change data of all cells of the second battery module during use; Based on the historical usage data, influence parameters that are related to the cell pressure between all the cells in the second battery module are determined, resulting in multiple pressure correlation parameters. Based on all the pressure-related parameters and the set battery pressure analysis algorithm, a battery pack pressure model is constructed. The pressure change data is input into the battery pack pressure model to train the battery pack pressure model.

3. The battery failure early warning detection method according to claim 2, characterized in that, The pressure-related parameters include at least time parameters and temperature parameters; Before inputting the pressure change data into the battery pack pressure model to train the battery pack pressure model, the method further includes: Based on cell charging and cell discharging, the historical usage data is divided into first-level data segments to obtain first-level data segmentation results corresponding to the historical usage data; the first-level data segmentation results include charging datasets corresponding to cell charging and discharging datasets corresponding to cell discharging. Based on the pressure correlation parameters, a second-level data partitioning is performed on the first-level data partitioning result to obtain a second-level data partitioning result corresponding to the first-level data partitioning result; the second-level data partitioning result includes multiple sub-data sets, which include a charging-time dataset and a charging-temperature dataset corresponding to the charging dataset, and a discharging-time dataset and a discharging-temperature dataset corresponding to the discharging dataset; The pressure change data is updated based on the secondary data partitioning results, and the operation of inputting the pressure change data into the battery pack pressure model to train the corresponding battery pack pressure model is triggered.

4. The battery failure early warning detection method according to claim 3, characterized in that, The specific training methods for training the battery pack pressure model using the pressure change data include: Based on the battery pressure analysis algorithm, the battery pack pressure model performs cell pressure change analysis on each of the subsets to obtain analysis results corresponding to each subset; each analysis result includes at least one division interval and the number of intervals corresponding to all the division intervals; each division interval corresponds to a cell pressure value interval. For each analysis result, a pressure analysis sub-algorithm matching the analysis result is determined based on the number of intervals corresponding to the analysis result; Based on the stress analysis sub-algorithm adapted to each analysis result, a preset training operation is performed on the sub-dataset corresponding to the analysis result. The preset training operation includes at least determining the algorithm parameters and optimizing the algorithm parameters for the stress analysis sub-algorithm. After each stress analysis sub-algorithm is determined to have converged during training, the stress analysis sub-algorithm is considered to have completed training.

5. The battery failure early warning detection method according to claim 4, characterized in that, The step of performing target pressure calculation on the parameter recording data according to the pre-constructed battery pack pressure model to obtain the module pressure information corresponding to the first battery module includes: Based on the pre-constructed battery pack pressure model, the data type corresponding to the parameter recording data is determined. The data type includes charging type or discharging type. When the data type is the charging type, the parameter recording data indicates that the first battery module is in the cell charging state; when the data type is the discharging type, the parameter recording data indicates that the first battery module is in the cell discharging state. Determine the target pressure analysis sub-algorithm that is compatible with the data type from all the pressure analysis sub-algorithms in the battery pack pressure model; The analysis indicators are extracted from the parameter recording data. The analysis indicators include at least one of the following: real-time moment corresponding to the time parameter, real-time time node, real-time temperature corresponding to the temperature parameter, and real-time temperature node. The target pressure is calculated based on the target pressure analysis sub-algorithm to obtain target pressure data that matches the analysis index, which is used as the module pressure information corresponding to the first battery module.

6. The battery failure early warning detection method according to claim 5, characterized in that, All of the aforementioned pressure analysis sub-algorithms include the charging-time algorithm, the charging-temperature algorithm, the discharging-time algorithm, and the discharging-temperature algorithm; The step of performing target pressure calculation on the analysis index according to the target pressure analysis sub-algorithm to obtain target pressure data adapted to the analysis index includes: When the quantity corresponding to the target pressure analysis sub-algorithm is 1, the target pressure calculation is performed on the analysis index according to the target pressure analysis sub-algorithm to obtain target pressure data that matches the analysis index. When the number of the target pressure analysis sub-algorithms is greater than 1, the analysis index that is adapted to each target pressure analysis sub-algorithm is determined as the target analysis index. For each target pressure analysis sub-algorithm, target pressure calculation is performed on the target analysis index corresponding to the target pressure analysis sub-algorithm to obtain graded pressure data corresponding to the target analysis index corresponding to the target pressure analysis sub-algorithm. Based on the weight values ​​set for each target pressure analysis sub-algorithm, a preset weight calculation is performed on all the graded pressure data to obtain the weight calculation results corresponding to all the graded pressure data, which are used as target pressure data adapted to the analysis indicators.

7. A battery failure early warning detection device, characterized in that, The device is used to perform the battery failure early warning detection method as described in any one of claims 1-6, and the device comprises: The determination module is used to identify the first battery module that needs to be warned. The data acquisition module is used to collect parameter record data corresponding to the first battery module and each preset pressure-related parameter; The pressure calculation module is used to perform target pressure calculation on the parameter recording data according to the pre-built battery pack pressure model to obtain the module pressure information corresponding to the first battery module. The module pressure information includes the upper limit pressure range corresponding to the first battery module. The upper limit pressure range is the pressure value range corresponding to the pressure value indicating that the first battery module has reached the battery failure state. The pressure detection module is used to perform pressure detection on the first battery module according to the preset pressure sensor, and obtain the real-time module pressure value corresponding to the first battery module. The failure warning module is used to determine whether the real-time module pressure value indicates that the first battery module meets the preset failure warning conditions based on the module pressure information; when it is determined that the real-time module pressure value indicates that the first battery module meets the preset failure warning conditions, the module performs a failure warning operation on the first battery module according to the set warning process to indicate that the first battery module is in a preset battery failure state.

8. A battery failure early warning detection device, characterized in that, The device includes: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the battery failure early warning detection method as described in any one of claims 1-6.

9. A computer storage medium, characterized in that, The computer storage medium stores computer instructions, which, when invoked, are used to execute the battery failure early warning detection method as described in any one of claims 1-6.

Images