Battery safety multi-model assessment early warning method and early warning system

Abstract

The present disclosure relates to the field of battery safety early warning technology, and discloses a battery safety multi-model assessment early warning method and early warning system. The present disclosure obtains the operating data of a target battery, and determines the fault severity, fault timing, and identification credibility corresponding to at least one safety early warning model respectively based on the operating data; uses a preset assessment formula to conduct a comprehensive assessment of the target battery, determines at least one assessment result, and determines the final fault assessment result of the target battery based on the at least one assessment result; when the final fault assessment result meets the early warning condition, generates safety early warning information, and the safety early warning information is used to prompt the safety risk of the target battery. In the embodiments of the present disclosure, the output format of the fault assessment results of different models can be unified, and the battery can be effectively quantitatively assessed, thereby improving the early warning effect for battery failures.

Description

Technical Field

[0001] The present disclosure relates to the technical field of battery safety early warning, and in particular to a battery safety multi-model assessment early warning method and early warning system. Background Art

[0002] With the development of new energy technology and the popularity of electric vehicles, lithium-ion batteries are widely used in electric vehicles. As the battery runs for a long time, faults will inevitably accumulate and occur, which may cause electric vehicles to spontaneously combust.

[0003] Due to the complexity of the fault triggering mechanism of the battery system, a single early warning model is difficult to cover all fault scenarios. In the relevant battery fault diagnosis methods, multiple models are often used to jointly evaluate the fault. However, due to the different principles of different models, it is difficult to unify the form of the fault evaluation results. It may be difficult to conduct effective quantitative fault evaluation, resulting in poor early warning effects for battery faults. A better solution is needed. Summary of the invention

[0004] In view of this, the present disclosure provides a battery safety multi-model assessment and early warning method and an early warning system to solve the problem that the output forms of multiple early warning models are different during the battery fault diagnosis process, resulting in poor early warning effects for battery faults.

[0005] On the one hand, the present disclosure provides a battery safety multi-model assessment and early warning method, the method comprising: obtaining operating data of a target battery, and determining, based on the operating data, an intermediate assessment result corresponding to at least one safety early warning model; the intermediate assessment data comprises: fault severity, fault timing and recognition credibility; based on at least one intermediate assessment data, a preset assessment formula is used to comprehensively assess the target battery, determine at least one assessment result, and determine a final fault assessment result of the target battery based on the at least one assessment result; when the final fault assessment result meets the early warning condition, a safety early warning message is generated, and the safety early warning message is used to indicate the safety risk of the target battery.

[0006] On the other hand, the present disclosure also provides an early warning system, which includes: a data acquisition module for acquiring operating data of a target battery; an intermediate processing module for acquiring at least one intermediate evaluation data, the intermediate evaluation data including: fault severity, fault timing and identification credibility; the intermediate evaluation data is determined by at least one safety early warning model based on the operating data; a risk assessment module for performing a comprehensive evaluation of the target battery based on at least one intermediate evaluation data and using a preset evaluation formula to determine at least one evaluation result, and the at least one evaluation result is used to determine a final fault evaluation result of the target battery; a safety early warning module for generating safety early warning information in response to the final fault evaluation result meeting the early warning condition, and the safety early warning information is used to indicate the safety risk of the target battery.

[0007] On the other hand, the present disclosure further provides a computer device, which includes: a memory and a processor, the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor implements the above-mentioned battery safety multi-model assessment and early warning method by executing the computer instructions.

[0008] On the other hand, the present disclosure further provides a computer-readable storage medium, on which computer instructions are stored, and the computer instructions are used to enable a computer to implement the above-mentioned battery safety multi-model assessment and early warning method.

[0009] On the other hand, the present disclosure further provides a computer program product, including computer instructions, which are used to enable a computer to execute the battery safety multi-model assessment and early warning method of the first aspect or any corresponding embodiment thereof.

[0010] In this process, the operating data of the target battery of the faulty vehicle is obtained, and the fault severity, fault timing and identification credibility corresponding to each safety warning model are determined based on the operating data, thereby unifying the output dimensions of different models; a preset evaluation formula is used to determine the evaluation result of each safety warning model for the fault information in the operating data, and the final fault evaluation result of the target battery is determined based on the evaluation result, thereby generating a reliable quantitative evaluation result for the fault evaluation of the target battery, thereby improving the warning effect of the target battery fault. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the specific embodiments of the present disclosure or the technical solutions in the related technologies, the drawings required for use in the specific embodiments or the related technical descriptions will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0012] Figure 1 is a flowchart of a battery safety multi-model assessment and early warning method provided by an embodiment of the present disclosure;

[0013] Figure 2 It is a flowchart of determining the intermediate evaluation results corresponding to at least one safety warning model according to the operation data in the battery safety multi-model evaluation and warning method provided by the embodiment of the present disclosure;

[0014] Figure 3 is a schematic diagram of an architecture of an early warning system provided by an embodiment of the present disclosure;

[0015] Figure 4 It is a structural diagram of the early warning system provided by the embodiment of the present disclosure. DETAILED DESCRIPTION

[0016] In recent years, with the continuous expansion of the new energy vehicle market and the rapid growth of electric vehicle ownership, lithium-ion batteries have also received widespread attention. The widespread application of lithium-ion batteries in electric vehicles and energy storage is mainly due to their superior performance characteristics, including high energy density, long service life and environmental friendliness. However, with the increasing popularity of electric vehicles, related safety hazards have become increasingly prominent. According to statistics from relevant departments, the spontaneous combustion rate of new energy vehicles increased significantly in the first quarter of 2023, reflecting the severity of battery system failures and fire risks.

[0017] Lithium-ion battery system failures in electric vehicles can be mainly classified into three categories: battery failure, management system failure, and line connection failure. Battery failure is the most common failure mode, including internal short circuit, external short circuit, overcharge, overdischarge, and inhomogeneity. These failures may gradually accumulate during the long-term operation of the battery, especially in the case of frequent charge and discharge cycles, resulting in a decrease in battery performance and an increased risk of thermal runaway. Once thermal runaway occurs, it may cause a more serious fire accident, posing a threat to passengers and the surrounding environment.

[0018] At present, there are various battery fault diagnosis methods for power batteries, including model-based methods, signal analysis methods, data-driven methods, and statistical analysis methods. Each method has its own specific advantages and disadvantages, but due to the complexity and diversity of battery failures, a single early warning model is often difficult to cover all failure scenarios. Therefore, more and more studies have begun to explore the method of multi-model joint evaluation to improve the accuracy and reliability of fault warning.

[0019] However, the following problems often exist in related methods:

[0020] 1. Different models output different forms of results based on different principles and methods, which may include numerical and non-numerical results. These results also differ in terms of dimension, frequency, etc. These differences make it difficult to conduct effective fault quantitative assessment during battery risk assessment, resulting in poor early warning effects for battery failures.

[0021] 2. Related methods often only evaluate battery failures from the perspective of the severity of the failures, lacking completeness and comprehensiveness of the evaluation, which may lead to errors in the evaluation results.

[0022] To solve at least one of the above problems, various embodiments of the present disclosure provide a battery safety multi-model assessment and early warning method, including: obtaining operating data of a target battery, and determining, based on the operating data, an intermediate assessment result corresponding to at least one safety early warning model; the intermediate assessment data includes: fault severity, fault timing, and identification credibility; based on at least one intermediate assessment data, a preset assessment formula is used to conduct a comprehensive assessment of the target battery, determine at least one assessment result, and determine a final fault assessment result of the target battery based on the at least one assessment result; when the final fault assessment result meets the early warning condition, a safety early warning message is generated, and the safety early warning message is used to prompt the safety risk of the target battery.

[0023] In order to make the purpose, technical solution and advantages of the embodiments of the present disclosure clearer, the technical solution in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are part of the embodiments of the present disclosure, rather than all the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of the present disclosure.

[0024] According to an embodiment of the present disclosure, a battery safety multi-model assessment and early warning method is provided. Figure 1 is a flowchart of a battery safety multi-model assessment and early warning method provided by an embodiment of the present disclosure, such as Figure 1 As shown, the process includes the following steps:

[0025] Step S101, obtaining the operating data of the target battery, and determining the intermediate evaluation results corresponding to at least one safety warning model according to the operating data; the intermediate evaluation data includes: fault severity, fault sequence and recognition credibility;

[0026] In this embodiment, the target battery may refer to the battery in the target faulty vehicle that needs to be safety evaluated; the operating data of the target battery may refer to various monitoring information and status information collected during the battery operation process, for example, the operating data may include at least one of the following: temperature, voltage, current, fault warning information, fault type. The safety warning model may refer to a system for identifying and evaluating potential safety risks in the target battery.

[0027] Here, the type of target battery can be determined according to the actual application scenario, and can be a verification method in existing technology or future development technology, which is not limited by the present disclosure. For example, it can be a lithium-ion battery. The output form of the safety warning model can include data type results and non-data type results.

[0028] As an example, data-type results may include, but are not limited to: fault risk level score, fault occurrence probability, and non-data-type results may include, but are not limited to: fault type description, target battery safety status.

[0029] After obtaining the operating data of the target battery, the operating data may be preprocessed and feature extracted, which may include: removing abnormal values ​​and noise in the operating data summary, processing default values, and extracting key features.

[0030] According to the key features extracted from the operating data, the corresponding safety warning model is selected for evaluation to obtain the intermediate evaluation results of the safety warning model.

[0031] As an example, if the operating data of the target battery indicates that the target battery is overcharged, the corresponding current safety warning model can be selected for evaluation first. When the overcharge causes the target current temperature to rise, the temperature safety warning model can be selected for secondary evaluation.

[0032] As a possible implementation of the above embodiment, the fault severity can be used to characterize the impact of the fault on the performance and safety of the target battery, the fault timing can be used to characterize the time pattern of the fault occurrence to reflect whether there is regularity or early warning signal of the fault, and the recognition credibility can be used to evaluate the accuracy and reliability of fault identification.

[0033] Here, the output forms of different types of security warning models are unified into the form of intermediate evaluation data, and the dimensions of different security data models are unified into the dimensionless form of intermediate evaluation data.

[0034] Step S102, based on at least one intermediate evaluation data, using a preset evaluation formula, comprehensively evaluating the target battery, determining at least one evaluation result, and determining a final fault evaluation result of the target battery according to the at least one evaluation result;

[0035] In this embodiment, each of the at least one intermediate evaluation result obtained is input into a preset evaluation formula, and the preset evaluation formula is used to calculate the evaluation result corresponding to each security warning model.

[0036] Here, the preset evaluation formula may refer to a mathematical model or algorithm for converting intermediate evaluation data into evaluation results. As an example, the preset evaluation formula may include but is not limited to: a weighted average formula, a logistic regression formula, and a discriminant analysis formula.

[0037] The final fault assessment result of the target battery is determined according to at least one assessment result to improve the interpretability of the fault assessment result for the target battery. Here, interpretability may mean that the final fault assessment result is not just a number or classification, but data of a specific assessment process can be obtained, so that the tester can know how to derive the final fault assessment result from the intermediate assessment results.

[0038] Step S103: when the final fault assessment result meets the warning condition, safety warning information is generated, and the safety warning information is used to prompt the safety risk of the target battery.

[0039] In this embodiment, the warning condition may refer to a preset condition for determining whether to generate safety warning information based on the final fault assessment result. For example, the warning condition may include but is not limited to: the final fault assessment result is greater than a preset threshold.

[0040] The safety warning information may include at least one of the following: evaluation results, fault description, fault indicators, and recommended measures.

[0041] Furthermore, historical fault data corresponding to the fault description may be obtained as reference data for analyzing the current fault situation.

[0042] In the battery safety multi-model assessment and early warning method and early warning system of the above-mentioned embodiments of the present invention, the output dimensions of different models are unified by setting the fault severity, fault timing and recognition credibility as the intermediate assessment data output by each safety early warning model; a preset assessment formula is used to determine the assessment result of each safety early warning model for the fault information in the operating data, and the final fault assessment result of the target battery is determined based on the assessment result, thereby generating a reliable quantitative assessment result for the fault assessment of the target battery, thereby improving the early warning effect of the fault of the target battery.

[0043] For further information, please refer to Figure 2 , Figure 2 is a flow chart of determining the intermediate evaluation results corresponding to at least one safety warning model according to the operation data in the battery safety multi-model evaluation and warning method provided by the embodiment of the present disclosure, such as Figure 2 As shown, the process includes:

[0044] Step S201, determining the fault severity of a fault signal corresponding to each safety warning model in at least one safety warning model according to the operation data;

[0045] Step S202, determining the fault timing of the fault signal corresponding to each safety warning model in at least one safety warning model according to the operation data;

[0046] Step S203: determining the recognition credibility of the fault signal corresponding to each safety warning model in at least one safety warning model according to the operation data.

[0047] In this embodiment, the fault signal is determined by the safety warning model based on the operating data. The fault severity is used to characterize the degree of hazard of the fault signal; the fault timing is used to characterize the timeliness of the safety warning model in identifying the fault signal; and the recognition credibility is used to characterize the credibility of the safety warning model in identifying the fault signal.

[0048] Determining the fault severity of a fault signal corresponding to each safety warning model in at least one safety warning model based on operating data may include: obtaining key information from the operating data of the target battery, analyzing the operating data using the safety warning model, identifying the fault signal, determining the degree of hazard of each fault signal based on a preset fault severity assessment standard and corresponding influencing factors, and outputting the fault severity corresponding to each fault signal.

[0049] Determining the fault timing of a fault signal corresponding to each safety warning model in at least one safety warning model based on operating data may include: acquiring timing data of the fault signal, performing timing analysis on the timing data, determining a timing score for each fault signal, and outputting the fault timing corresponding to the fault signal.

[0050] Here, the timing data may include but is not limited to: occurrence frequency, sending time interval, and occurrence regularity.

[0051] Determining the recognition credibility of a fault signal corresponding to each safety warning model in at least one safety warning model based on operating data may include: establishing a recognition credibility assessment model based on credibility factors, using the recognition credibility assessment model to make credibility judgments on the safety warning model that identifies the fault signal, and determining the recognition credibility of each safety warning model.

[0052] Here, credibility factors may include, but are not limited to: past fault identification accuracy, model training accuracy, and reliability of data sources.

[0053] The battery safety multi-model assessment and early warning method in this implementation can more accurately identify high-risk faults by evaluating the degree of hazard of each fault signal, prioritize potential threats, and reduce the possibility of accidents; it combines fault severity, timing, and recognition credibility to form a comprehensive assessment framework, making the monitoring of battery status more comprehensive; through clear evaluation standards and indicators, the interpretability of the system is improved, allowing users to understand the source and impact of fault signals.

[0054] In some optional implementations of this embodiment, determining, according to the operating data, the fault severity of the fault signal corresponding to each safety warning model in at least one safety warning model includes:

[0055] Based on the operation data, determine the degree of harm of the fault signal corresponding to the safety warning model and whether there is alarm information in the fault signal;

[0056] When the hazard level satisfies the first condition and there is no warning information in the fault signal, determining that the fault severity of the fault signal is a first severity value;

[0057] When the hazard level meets the first condition and there is warning information in the fault signal, determining that the fault severity of the fault signal is a second severity value;

[0058] When the hazard level meets the second condition, determining the fault severity of the fault signal to be a third severity value;

[0059] When the degree of harm meets the third condition, determining the fault severity of the fault signal to be a fourth severity value;

[0060] When the degree of harm meets the fourth condition, determining the fault severity of the fault signal to be a fifth severity value;

[0061] Among them, the degree of harm of the first condition, the second condition, the third condition and the fourth condition decreases in sequence; the fault severity of the first severity value, the second severity value, the third severity value, the fourth severity value and the fifth severity value decreases in sequence.

[0062] Determining the degree of hazard of a fault signal corresponding to a safety warning model and whether there is alarm information in the fault signal based on operating data may include: obtaining the fault signal in the operating data, determining whether there is alarm information in the fault signal, and using a preset fault severity assessment standard to determine the degree of hazard of the fault signal.

[0063] Here, determining whether there is alarm information in the fault signal can be achieved based on the following steps: collecting real-time data from the battery management system and / or the vehicle-mounted monitoring system to determine whether there is an alarm signal in the real-time data; aligning the real-time data with the timestamp of the alarm information to determine whether the alarm information is triggered within a preset time window when the fault signal occurs; if the alarm information is detected within the preset time range of the occurrence of the fault signal, it is determined that there is alarm information in the fault signal.

[0064] When the degree of hazard meets the first condition and there is no warning information in the fault signal, determining the fault severity of the fault signal to be the first severity value may include: when the potential fault affects the safe operation of the vehicle and / or does not comply with relevant regulations, determining that the degree of hazard of the fault meets the first condition; when the degree of hazard meets the first condition and there is no warning information in the fault signal, determining that the fault severity of the fault signal is the first severity value.

[0065] Here, a potential fault may refer to a type of fault or failure that may occur during the operation of the vehicle, such as a short circuit of a battery, abnormal temperature, etc. If the degree of hazard meets the first condition and there is no warning information in the fault signal, it can be indicated that the potential fault has not been identified and handled, which may result in the passenger being unable to take any preventive measures when the fault occurs, seriously endangering the passenger's life safety. As an example, the first severity value can be represented by 5.

[0066] When the degree of hazard meets the first condition and there is warning information in the fault signal, determining the fault severity of the fault signal to be the second severity value may include: when the degree of hazard of the vehicle's potential fault meets the first condition, but the potential fault can be detected and warned, determining the fault severity of the fault signal to be the second severity value.

[0067] Here, if the degree of harm of the potential fault of the vehicle satisfies the first condition, but the potential fault can be detected and an alarm is given, it can be represented that although the potential fault will affect the safe driving of the vehicle, the passengers can take relevant preventive measures when the fault occurs, and the life safety of the passengers will not be seriously affected. As an example, the second severity can be represented by 4.

[0068] When the degree of hazard meets the second condition, determining the fault severity of the fault signal to be the third severity value may include: when the potential fault causes the loss of a major function of the vehicle and does not pose a direct threat to the life safety of passengers, determining the fault severity of the fault signal to be the third severity value.

[0069] Here, the main functions of the vehicle may include: driving, steering, when the main functions are lost, the vehicle cannot be operated, but will not affect the life safety of passengers. As an example, the third severity can be represented by 3.

[0070] When the degree of hazard meets the third condition, determining the fault severity of the fault signal to be the fourth severity value may include: when the potential fault causes the vehicle performance to be degraded, resulting in the degradation of the main function of the vehicle, determining the fault severity of the fault signal to be the third severity value. Here, the degradation of the main function may mean that the vehicle can operate, but the performance is degraded. As an example, the fourth severity can be represented by 2.

[0071] When the degree of hazard meets the fourth condition, determining the fault severity of the fault signal as the fifth severity value may include: the potential fault has no impact on the operational safety and performance of the vehicle, the fault has no identifiable consequences on the vehicle, and determining the fault severity of the fault signal as the fifth severity value. As an example, the fifth severity may be represented by 1.

[0072] In a specific example, the content shown in the following Table 1 may be used as a basis for evaluating the severity of the fault.

[0073]

[0074] Table 1

[0075] The battery safety multi-model assessment and early warning method in this implementation makes the risk assessment process more systematic and standardized by dividing the fault severity into clear levels, which facilitates different models to perform consistent analysis in different situations; through the assessment of fault severity, potential faults with the greatest impact on safety and compliance can be effectively identified, thereby optimizing the allocation and management of resources.

[0076] In some optional implementations of this embodiment, determining, according to the operating data, the fault timing of the fault signal corresponding to each safety warning model in at least one safety warning model includes:

[0077] Based on the operation data, determine the advance warning time of the safety warning model. The advance warning time is used to represent the time when the safety warning model identifies the fault signal in advance.

[0078] The advance warning time standard corresponding to the safety warning model is obtained, the ratio of the advance warning time to the advance warning time standard is determined, and the ratio is used as the fault timing of the fault signal corresponding to the safety warning model.

[0079] In this embodiment, the advance warning time standard can be used to characterize the length of time that the safety warning model should issue a warning before a fault occurs. The advance warning time standards of different safety warning models can be stored in the form of a standard library for subsequent query and comparison.

[0080] Determining the advance warning time of the safety warning model based on the operating data may include: extracting the actual alarm timestamp and fault occurrence timestamp of the safety warning model from the operating data, the actual alarm timestamp is used to characterize the timestamp when the model detects the fault signal in advance and issues an alarm, and the difference between the fault occurrence timestamp and the actual alarm timestamp is used as the advance warning time.

[0081] As an example, when the fault occurrence timestamp indicates that the battery spontaneous combustion occurred at 10:30, and the actual alarm timestamp indicates that the safety warning model detected the abnormal temperature fault at 10 o'clock and issued an alarm message, the advance alarm time can be 30 minutes.

[0082] Obtaining the advance warning time standard corresponding to the safety warning model, determining the ratio of the advance warning time to the advance warning time standard, and using the ratio as the fault timing of the fault signal corresponding to the safety warning model, may include: obtaining the advance warning time standard corresponding to the safety warning model in a standard library, and using the ratio of the advance warning time to the advance warning time standard as the fault timing of the fault signal corresponding to the safety warning model.

[0083] As an example, the advance warning time is 30 minutes, and the advance warning time standard is 15 minutes, then the fault sequentiality may be 30 / 15=2.

[0084] Here, when the fault sequentiality is greater than 1, it can be characterized that the safety warning model can give an early warning. The larger the ratio, the better the sequentiality. When the fault sequentiality is less than 1, it can be characterized that the safety warning model does not meet the early warning time standard and cannot give an early warning.

[0085] The above fault sequence determination step can be used for post-warning when the fault occurrence timestamp is known. For pre-warning, since the fault occurrence time is unknown, the historical time difference statistical results can be used for alternative calculation. Here, the historical time difference statistical results can refer to the statistical results of the model warning and actual fault time difference of a large number of faulty vehicle samples under different models, different battery systems, and different fault scenarios.

[0086] The fault sequence of each safety warning model corresponding to different fault signals is recorded in the database as one of the analysis bases for the credibility of the safety warning model.

[0087] The battery safety multi-model assessment and early warning method in this implementation can objectively evaluate the early warning performance of different models through the quantitative indicators provided by the fault timing score; by comparing the actual failure time with the predicted timing, a data feedback mechanism is established to promote the continuous improvement and optimization of the model, forming a virtuous circle.

[0088] In some optional implementations of this embodiment, determining, according to the operating data, the recognition credibility of the fault signal corresponding to each safety warning model in at least one safety warning model includes:

[0089] Determine the data quality score of the operating data based on whether the data quality of the operating data meets the input requirements of the safety early warning model;

[0090] Obtain the historical fault recognition accuracy of the safety early warning model, and use the historical fault recognition accuracy as the recognition reliability score of the fault signal determined by the safety early warning model based on the operation data;

[0091] Based on the preset scoring standard library, determine the compatibility score of the security early warning model for identifying the operating data;

[0092] The product of the data quality score, the recognition reliability score and the compatibility score is taken as the recognition credibility of the safety warning model for identifying fault signals.

[0093] In this embodiment, the data quality of the acquired operation data often has quality problems, which may include but are not limited to: missing values, abnormal values, duplicate values, erroneous values, etc. Therefore, the data quality of the operation data needs to be evaluated.

[0094] The data quality score of the operating data is determined based on whether the data quality of the operating data meets the input requirements of the safety early warning model, which may include:

[0095] When the data quality of the operating data meets the input requirements of the safety early warning model, determining that the data quality score of the operating data is a preset non-zero integer;

[0096] When the data quality of the operating data does not meet the input requirements of the security early warning model, the data quality score of the operating data is determined to be zero. Here, the operating data that meets the input requirements of the security early warning model may refer to operating data that does not need to be repaired or has been repaired. As an example, for operating data with missing values, the repair method may include: deleting records containing missing values ​​and interpolating and filling missing values; for operating data with duplicate values, the repair method may include: deleting duplicate records and merging duplicate records; for operating data with outliers, the repair method may include: standardizing outliers and replacing them with the median of adjacent values; for operating data with erroneous values, the repair method may include: correcting erroneous values ​​and performing logical checks on erroneous values.

[0097] Exemplarily, when the operating data that does not need to be repaired or has been repaired meets the input requirements of the security early warning model, the data quality score of the operating data can be determined to be 1; when the data quality of the operating data does not meet the input requirements of the security early warning model, the data quality score of the operating data is determined to be 0.

[0098] Obtaining the historical fault recognition accuracy of the safety warning model, and using the historical fault recognition accuracy as the recognition reliability score of the fault signal determined by the safety warning model based on the operation data, may include: obtaining the historical fault recognition accuracy of the safety warning model based on a preset number of samples, and using the historical fault recognition accuracy as the recognition reliability score of the safety warning model. For example, if the overvoltage fault recognition rate of the first model is 95%, 95% may be used as the recognition reliability score of the first model.

[0099] Determining the compatibility score of the security warning model for identifying the operating data based on a preset scoring standard library may include: determining whether the security warning model is adapted to the scenario of the operating data according to the preset scoring standard library; when the degree of adaptation of the security warning model to the scenario of the operating data is greater than a first adaptation threshold, determining the compatibility score to be 1; when the degree of adaptation of the security warning model to the scenario of the operating data is greater than a second adaptation threshold and less than the first adaptation threshold, determining the compatibility score to be 0.5; when the degree of adaptation of the security warning model to the scenario of the operating data is less than the second adaptation threshold, determining the compatibility score to be 0.

[0100] Here, a degree of adaptation greater than the first adaptation threshold can indicate that the security warning model is fully adapted to the scenario; a degree of adaptation greater than the second adaptation threshold and less than the first adaptation threshold can indicate that the security warning model is not fully adapted to the scenario; a degree of adaptation between the security warning model and the scenario of operating data less than the second adaptation threshold can indicate that the security warning model is not adapted to the scenario.

[0101] As an example, if the data quality score of the model is 1, the recognition reliability score is 80%, and the compatibility score is 0.5, the recognition credibility of the model is 1*0.8*0.5=0.4.

[0102] The battery safety multi-model assessment and early warning method in this implementation uses the data quality of operating data as one of the measurement dimensions of recognition credibility, which can enable the model to support more accurate fault signal recognition, so that the model can effectively detect potential faults in practical applications; through the quantification of indicators such as the true fault recognition rate, it can effectively improve the reliability of the model in real scenarios, improve the fault recognition rate and reduce false alarms; through the adaptability of the model to the specific battery system hardware and operating conditions, the effectiveness and reliability of the model in different scenarios are ensured.

[0103] In some optional implementations of this embodiment, based on at least one intermediate evaluation data, a preset evaluation formula is used to comprehensively evaluate the target battery, determine at least one evaluation result, and determine a final fault evaluation result of the target battery according to the at least one evaluation result, including:

[0104] Obtaining the fault severity, fault timing and identification credibility corresponding to each safety warning model in at least one safety warning model;

[0105] A preset evaluation formula is used to comprehensively evaluate the target battery. The evaluation results corresponding to each safety warning model are determined according to the fault severity, fault timing and identification credibility corresponding to each safety warning model. The largest evaluation result among at least one evaluation result is taken as the final fault evaluation result of the target battery.

[0106] In this embodiment, the evaluation result may refer to a risk priority number (PRN), which may be a custom evaluation formula used to unify the three different evaluation dimensions of the safety warning model, namely, fault severity, fault timing, and identification credibility, into a comprehensive indicator to simplify the complexity of multi-model outputs so that the results of different models can be uniformly compared.

[0107] Here, S can be used to represent the severity of the fault, R can be used to represent the recognition reliability, and T can be used to represent the timing of the fault. The following formula can be used to represent PRN:

[0108]

[0109] Here, the higher the PRN value, the greater the risk of the fault signal detected by the safety early warning model.

[0110] Taking the largest evaluation result among at least one evaluation result as the final fault evaluation result of the target battery may include: determining the PRN value of each safety warning model, and taking the data with the largest PRN value as the final fault evaluation result.

[0111] As an example, the fault severity, fault sequence and identification credibility corresponding to the three models are provided, as shown in Table 2 below:

[0112] Model Fault severity Identify credibility Fault sequence RPN 1 5 95％ 1 4.75 2 4 90％ 0.5 7.2 3 1 80％ 2 0.4

[0113] Table 2

[0114] Among them, the RPN of model 1 is 4.75, the RPN of model 2 is 7.2, and the RPN of model 3 is 0.4. The evaluation result of model 2 with the largest PRN value is taken as the final fault evaluation result of the target battery.

[0115] Furthermore, in order to improve the authenticity and reliability of the final fault assessment results determined based on RPN, the RPN formula can be optimized by combining the experience of domain experts and the rules of real fault model identification results. For example, the contribution of a certain part of the fault severity (S), identification credibility (R), and fault timing (T) to RPN can be increased by exponentialization. This can be shown in the following formula:

[0116]

[0117] Among them, m, n, and p can be specifically set and are not limited here.

[0118] The battery safety multi-model assessment and early warning method in this implementation combines recognition credibility and fault timing in the battery fault assessment, thereby improving the comprehensiveness and rigor of the results; by unifying the fault severity (S), recognition credibility (R) and fault timing (T) into RPN, a multi-dimensional analysis of the target battery failure risk is achieved, thereby improving the comprehensiveness and accuracy of the assessment; the assessment results of different model outputs are converted into RPN values, thereby simplifying the complexity of multi-model outputs, so that the results of different safety early warning models can be intuitively compared; the calculation method of RPN clearly shows the impact of each component on the final result, thereby improving the interpretability of the result.

[0119] Figure 3 is a schematic diagram of an architecture of an early warning system provided by an embodiment of the present disclosure, such as Figure 3 As shown, the system can be applied to intelligent electronic devices such as servers and computers; the system includes: a data acquisition module 301, an intermediate processing module 302, a risk assessment module 303 and a security warning module 304;

[0120] Wherein, the data acquisition module 301 is used to acquire the operating data of the target battery;

[0121] The intermediate processing module 302 is used to obtain at least one intermediate evaluation data, the intermediate evaluation data including: fault severity, fault sequence and recognition credibility; the intermediate evaluation data is determined by at least one safety warning model based on the operation data;

[0122] A risk assessment module 303 is used to perform a comprehensive assessment on the target battery based on at least one intermediate assessment data and using a preset assessment formula to determine at least one assessment result, wherein the at least one assessment result is used to determine a final fault assessment result of the target battery;

[0123] The safety warning module 304 is used to generate safety warning information in response to the final fault assessment result meeting the warning condition, and the safety warning information is used to indicate the safety risk of the target battery.

[0124] In a specific embodiment, the intermediate processing module 302 includes a severity calculation module 3021, a temporal calculation module 3022 and a credibility calculation module 3023;

[0125] The severity calculation module 3021 is used to determine the fault severity of the fault signal corresponding to each safety warning model in at least one safety warning model; the fault signal is determined by the safety warning model based on the operation data, and the fault severity is used to characterize the degree of harm of the fault signal;

[0126] A timing calculation module 3022 is used to determine the fault timing of the fault signal corresponding to each safety warning model in at least one safety warning model, where the fault timing is used to characterize the timeliness of the safety warning model identifying the fault signal;

[0127] The credibility calculation module 3023 is used to determine the recognition credibility of the fault signal corresponding to each safety warning model in at least one safety warning model, and the recognition credibility is used to characterize the credibility of the safety warning model for identifying the fault signal.

[0128] In a specific embodiment, the severity calculation module 3021 is specifically used to determine the degree of harm of the fault signal corresponding to the safety warning model and whether there is warning information in the fault signal based on the operation data;

[0129] When the hazard level satisfies the first condition and there is no warning information in the fault signal, determining that the fault severity of the fault signal is a first severity value;

[0130] When the hazard level meets the first condition and there is warning information in the fault signal, determining that the fault severity of the fault signal is a second severity value;

[0131] When the hazard level meets the second condition, determining the fault severity of the fault signal to be a third severity value;

[0132] When the degree of harm meets the third condition, determining the fault severity of the fault signal to be a fourth severity value;

[0133] When the degree of harm meets the fourth condition, determining the fault severity of the fault signal to be a fifth severity value;

[0134] Among them, the degree of harm of the first condition, the second condition, the third condition and the fourth condition decreases in sequence; the fault severity of the first severity value, the second severity value, the third severity value, the fourth severity value and the fifth severity value decreases in sequence.

[0135] In a specific embodiment, the timing calculation module 3022 is specifically used to:

[0136] Based on the operation data, the actual alarm time of the safety warning model is determined. The actual alarm time is used to represent the time when the safety warning model identifies the fault signal in advance;

[0137] The advance warning time standard corresponding to the safety warning model is obtained, the ratio of the actual warning time to the advance warning time standard is determined, and the ratio is used as the fault timing of the fault signal corresponding to the safety warning model.

[0138] In a specific embodiment, the credibility calculation module 3023 is specifically used to:

[0139] Determine the data quality score of the operating data based on whether the data quality of the operating data meets the input requirements of the safety early warning model;

[0140] Obtain the historical fault recognition accuracy of the safety early warning model, and use the historical fault recognition accuracy as the recognition reliability score of the fault signal determined by the safety early warning model based on the operation data;

[0141] Based on the preset scoring standard library, determine the compatibility score of the security early warning model for identifying the operating data;

[0142] The product of the data quality score, the recognition reliability score and the compatibility score is taken as the recognition credibility of the safety warning model for identifying fault signals.

[0143] In a specific embodiment, the credibility calculation module 3023 is specifically used to:

[0144] When the data quality of the operating data meets the input requirements of the safety early warning model, determining that the data quality score of the operating data is a preset non-zero integer;

[0145] When the data quality of the operating data does not meet the input requirements of the safety early warning model, the data quality score of the operating data is determined to be zero.

[0146] In a specific embodiment, the risk assessment module 303 is used to obtain the fault severity, fault sequence and identification credibility corresponding to each safety warning model in at least one safety warning model;

[0147] A preset evaluation formula is used to comprehensively evaluate the target battery. The evaluation results corresponding to each safety warning model are determined according to the fault severity, fault timing and identification credibility corresponding to each safety warning model. The largest evaluation result among at least one evaluation result is taken as the final fault evaluation result of the target battery.

[0148] It should be noted that: the early warning system provided in the above embodiment only uses the division of the above program modules as an example to illustrate when implementing the corresponding battery safety multi-model assessment early warning method. In actual applications, the above processing can be assigned to different program modules as needed, that is, the internal structure of the device can be divided into different program modules to complete all or part of the above-described processing. Figure 1The embodiments of the method shown belong to the same concept, and the specific implementation process is detailed in the method embodiments, which will not be repeated here.

[0149] The present disclosure also provides a computer device having the above Figure 3 Early warning system shown.

[0150] See also Figure 4 , Figure 4 is a structural diagram of an early warning system provided by an embodiment of the present disclosure, such as Figure 4 As shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components are connected to each other using different buses for communication, and can be installed on a common mainboard or installed in other ways as needed. The processor can process the instructions executed in the computer device, including instructions stored in or on the memory to display the graphical information of the GUI on an external input / output device (such as, a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 4 A processor 10 is taken as an example.

[0151] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0152] The memory 20 stores instructions executable by at least one processor 10, so that the at least one processor 10 executes the method shown in the above embodiment.

[0153] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0154] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 20 may also include a combination of the above types of memory.

[0155] The computer device also includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30 and the output device 40 may be connected via a bus or other means. Figure 4 The example of connecting through bus is taken in the following.

[0156] The input device 30 can receive input digital or character information, and generate key signal input related to the user settings and function control of the computer device, such as a touch screen, a keypad, a mouse, a track pad, a touch pad, an indicator bar, one or more mouse buttons, a trackball, a joystick, etc. The output device 40 may include a display device, an auxiliary lighting device (e.g., an LED) and a tactile feedback device (e.g., a vibration motor), etc. The above-mentioned display device includes but is not limited to a liquid crystal display, a light emitting diode, a display and a plasma display. In some optional embodiments, the display device can be a touch screen.

[0157] The computer device further comprises a communication interface 30 for the computer device to communicate with other devices or a communication network.

[0158] The embodiments of the present disclosure also provide a computer-readable storage medium. The above-mentioned method according to the embodiments of the present disclosure can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium and downloaded through a network, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state drive, etc.; further, the storage medium can also include a combination of the above-mentioned types of memory. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.

[0159] A part of the present disclosure may be applied as a computer program product, such as a computer program instruction, which, when executed by a computer, can call or provide the method and / or technical solution according to the present disclosure through the operation of the computer. Those skilled in the art should understand that the existence of computer program instructions in computer-readable media includes, but is not limited to, source files, executable files, installation package files, etc., and accordingly, the way in which computer program instructions are executed by a computer includes, but is not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to the computer.

[0160] Although the embodiments of the present disclosure have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present disclosure, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A battery safety multi-model assessment and early warning method, characterized in that: The method comprises: Acquire operating data of the target battery, and determine, based on the operating data, an intermediate evaluation result corresponding to at least one safety warning model; the intermediate evaluation data includes: fault severity, fault sequence and recognition credibility; Based on at least one intermediate evaluation data, a preset evaluation formula is used to comprehensively evaluate the target battery, determine at least one evaluation result, and determine a final fault evaluation result of the target battery according to the at least one evaluation result; When the final fault assessment result meets the warning condition, safety warning information is generated, and the safety warning information is used to prompt the safety risk of the target battery.

2. The method according to claim 1, characterized in that Determining, based on the operation data, an intermediate evaluation result corresponding to at least one safety warning model, respectively, includes: Determine, according to the operation data, a fault severity of a fault signal corresponding to each safety warning model in at least one safety warning model; the fault signal is determined by the safety warning model based on the operation data, and the fault severity is used to characterize the degree of harm of the fault signal; Determine, according to the operation data, the fault sequence of the fault signal corresponding to each safety warning model in the at least one safety warning model; the fault sequence is used to characterize the timeliness of the safety warning model identifying the fault signal; According to the operation data, the recognition credibility of the fault signal corresponding to each safety warning model in the at least one safety warning model is determined; the recognition credibility is used to characterize the credibility of the safety warning model that recognizes the fault signal.

3. The method according to claim 2, characterized in that The step of determining the fault severity of the fault signal corresponding to each safety warning model in at least one safety warning model according to the operation data includes: Based on the operation data, determine the degree of harm of the fault signal corresponding to the safety warning model and whether there is warning information in the fault signal; When the hazard level satisfies a first condition and there is no warning information in the fault signal, determining that the fault severity of the fault signal is a first severity value; When the hazard level satisfies the first condition and the warning information exists in the fault signal, determining that the fault severity of the fault signal is a second severity value; When the hazard level satisfies the second condition, determining the fault severity of the fault signal to be a third severity value; When the hazard level satisfies the third condition, determining the fault severity of the fault signal to be a fourth severity value; When the hazard level satisfies the fourth condition, determining the fault severity of the fault signal to be a fifth severity value; Among them, the degree of harm of the first condition, the second condition, the third condition and the fourth condition decreases in sequence; the fault severity of the first severity value, the second severity value, the third severity value, the fourth severity value and the fifth severity value decreases in sequence.

4. The method according to claim 2, characterized in that: The determining, according to the operation data, the fault timing of the fault signal corresponding to each safety warning model in the at least one safety warning model comprises: Based on the operation data, determining the advance warning time of the safety warning model, wherein the advance warning time is used to represent the time when the safety warning model identifies the fault signal in advance; An advance warning time standard corresponding to the safety warning model is obtained, a ratio of the advance warning time to the advance warning time standard is determined, and the ratio is used as the fault timing of the fault signal corresponding to the safety warning model.

5. The method according to claim 2, characterized in that: The step of determining, based on the operation data, the recognition credibility of the fault signal corresponding to each safety warning model in the at least one safety warning model comprises: Determining a data quality score of the operating data according to whether the data quality of the operating data meets the input requirements of the safety early warning model; Acquire a historical fault recognition accuracy rate of the safety early warning model, and use the historical fault recognition accuracy rate as a recognition reliability score of a fault signal determined by the safety early warning model based on the operation data; Determining, based on a preset scoring standard library, a compatibility score for the security early warning model to identify the operating data; The product of the data quality score, the recognition reliability score and the compatibility score is used as the recognition credibility of the safety warning model for identifying the fault signal.

6. The method according to claim 5, characterized in that Determining the data quality score of the operation data according to whether the data quality of the operation data meets the input requirements of the safety early warning model includes: When the data quality of the operating data meets the input requirements of the safety early warning model, determining that the data quality score of the operating data is a preset non-zero integer; When the data quality of the operating data does not meet the input requirement of the safety early warning model, it is determined that the data quality score of the operating data is zero.

7. The method according to claim 1, characterized in that The step of comprehensively evaluating the target battery based on at least one intermediate evaluation data and using a preset evaluation formula to determine at least one evaluation result, and determining a final fault evaluation result of the target battery according to the at least one evaluation result includes: Obtaining the fault severity, fault timing and identification credibility corresponding to each safety warning model in at least one safety warning model; A preset evaluation formula is used to comprehensively evaluate the target battery. The evaluation results corresponding to each of the safety warning models are determined according to the fault severity, fault timing and identification credibility corresponding to each of the safety warning models. The largest evaluation result among at least one evaluation result is taken as the final fault evaluation result of the target battery.

8. Early warning system, characterized in that, The system comprises: A data acquisition module, used to acquire the operating data of the target battery; An intermediate processing module, used to obtain at least one intermediate evaluation data, the intermediate evaluation data including: fault severity, fault sequence and recognition credibility; the intermediate evaluation data is determined by at least one safety warning model based on the operation data; a risk assessment module, configured to perform a comprehensive assessment on the target battery based on the at least one intermediate assessment data and using a preset assessment formula to determine at least one assessment result, wherein the at least one assessment result is used to determine a final fault assessment result of the target battery; The safety warning module is used to generate safety warning information in response to the final fault assessment result meeting the warning condition, wherein the safety warning information is used to indicate the safety risk of the target battery.

9. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the battery safety multi-model assessment and early warning method according to any one of claims 1 to 7 by executing the computer instructions.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the battery safety multi-model assessment and early warning method according to any one of claims 1 to 7.

Images