Power battery safety risk judgment method and system, vehicle and electronic equipment

Abstract

The invention relates to a power battery safety risk judgment method and system, a vehicle and electronic equipment, relates to the technical field of new energy vehicles, in particular to the technical field of power battery safety, and aims to accurately judge the safety of a power battery. Based on the target knowledge base, performing information supplementation on the operation information to obtain target information; wherein the supplementary information comprises charging and discharging strategy information of the power battery and user portrait information; based on the target information and the base model group, performing safety risk judgment on the power battery to obtain a safety risk judgment result; wherein the base model group comprises a plurality of base model groups corresponding to a plurality of risk types of the power battery; the multiple target base models in the target base model group jointly carry out safety risk judgment of the target risk type on the power battery, and risk judgment of the power battery is carried out through the knowledge base and the base model group, so that the safety risk judgment accuracy of the power battery is improved.

Description

Technical Field

[0001] This application relates to the field of new energy vehicle technology, and more particularly to the field of power battery safety technology, specifically to a method, system, vehicle, and electronic equipment for determining the safety risks of power batteries. Background Technology

[0002] Currently, with the significant increase in the market share of new energy vehicles, the performance and safety issues of power batteries have attracted widespread attention. To effectively improve the performance and safety of power batteries, it is necessary to conduct effective safety risk assessments.

[0003] In existing technologies, the effective method for assessing the safety risks of power batteries generally involves collecting various operational data from the vehicle's power battery and using preset rules and this data to determine whether the power battery poses a risk of thermal runaway. However, this method is prone to inaccurate assessments, therefore, a more precise approach to assessing the risks of power batteries is needed. Summary of the Invention

[0004] This application provides a method, system, vehicle, and electronic device for determining the safety risks of power batteries, to at least address the technical problem of insufficient accuracy in risk assessment of power batteries in related technologies. The technical solution adopted in this application is as follows:

[0005] Firstly, this application provides a method for determining the safety risks of a power battery, comprising: acquiring the operating information of the power battery; supplementing the operating information based on a target knowledge base to obtain target information; wherein the supplemented information includes the charging and discharging strategy information and user profile information of the power battery; determining the safety risks of the power battery based on the target information and a base model group to obtain a safety risk determination result; wherein the base model group includes multiple base model sets corresponding to multiple risk types of the power battery; multiple target base models in the target base model group jointly determine the safety risks of the power battery for the target risk type; the target base model group is one of the multiple base model sets; and the target risk type is one of the multiple risk types.

[0006] Based on the aforementioned technical means, after collecting the operating information of the power battery, this application refines this operating information using the power battery's charging and discharging strategy information and user profile information in the database. On this basis, the application uses a base model group and the refined information to determine the safety risks of the power battery, making the risk assessment more comprehensive and accurate. Furthermore, the base model group is divided into different base model sets for different risk types, thus improving the risk assessment performance for different risk types of power batteries.

[0007] In one possible implementation, the target knowledge base includes historical operating information of multiple power batteries and their corresponding historical charging and discharging strategy information and historical user profile information; the process of determining supplementary information includes: determining target historical operating information that matches the operating information from multiple historical operating information; and determining the historical charging and discharging strategy information and historical user profile information corresponding to the target historical operating information as supplementary information.

[0008] Based on the above technical means, this application matches the operating information with each piece of information in the historical operating information, and uses the historical charging and discharging strategy information and historical user profile information corresponding to the matched historical operating information as supplementary information to the operating information. This makes the determination of the supplementary information more in line with the actual needs of the operating information, so that the target information of the power battery obtained after the supplementation is more comprehensive.

[0009] In one possible implementation, the safety risk of the power battery is determined based on the target information and the base model group, and the safety risk determination result is obtained. This includes: determining the safety risk determination results of multiple target base models on the target risk type based on the target information; and determining the safety risk determination result of the power battery on the target risk type based on the safety risk determination results of multiple target base models on the target risk type and the confidence level of multiple target base models.

[0010] Based on the aforementioned technical means, this application combines the safety risk assessment results of each base model on the target risk type with the corresponding confidence level of each base model to obtain the safety risk assessment results of the power battery on the target risk type. This can improve the impact of the safety risk assessment results of base models with higher confidence levels and reduce the impact of the safety risk assessment results of base models with lower confidence levels, thereby improving the assessment performance of the safety risk assessment results of the power battery.

[0011] In one possible implementation, the safety risk assessment result of the power battery on the target risk type is determined based on the safety risk assessment results of multiple target basis models on the target risk type and the confidence levels of multiple target basis models. This includes: when the confidence levels of multiple target basis models are all less than the confidence level threshold, the safety risk assessment result of the power battery on the target risk type is obtained by integrating the safety risk assessment results of multiple target basis models on the target risk type based on a first type of integration strategy; wherein, the first type of integration strategy includes an integration strategy based on Bayesian inference.

[0012] Based on the above technical means, when the confidence levels of the base models used for determining the safety risks of power batteries are all less than the confidence threshold, the safety risk determination performance of the base models is poor. By integrating the safety risk determination results through the Bayesian inference integration strategy, the output influence of low-confidence models can be suppressed and the output influence of high-confidence results can be improved. At the same time, by fusing multi-source heterogeneous data through posterior distribution, the data bias problem of a single model can be reduced and the efficiency of risk determination can be improved.

[0013] In one possible implementation, the safety risk assessment result of the power battery on the target risk type is determined based on the safety risk assessment results of multiple target basis models on the target risk type and the confidence levels of the multiple target basis models. This includes: when there is at least one candidate target basis model, the safety risk assessment result of the power battery on the target risk type is obtained by combining the safety risk assessment results of at least one candidate target basis model on the target risk type based on a second type of integration strategy; wherein, the candidate target basis model is the target basis model among the multiple target basis models whose confidence level is greater than the confidence level threshold; the second type of integration strategy includes hard voting integration strategy, soft voting integration strategy, and rule integration strategy.

[0014] Based on the aforementioned technical means, this application, when a target base model with a confidence level greater than the confidence threshold exists, exhibits high model certainty. By integrating the security risk assessment results through hard voting, the interference of anomalies in a single model on the security risk assessment results can be reduced. Alternatively, by integrating the risk assessment results through soft voting, a weighted average of the security risk assessment results can be performed using probability, thereby enhancing the stability of the risk assessment results. Or, by integrating the risk assessment results through rule-based methods, the systematic biases of the base model can be corrected through rules, improving the accuracy of the security risk assessment results.

[0015] In one possible implementation, the target knowledge base also includes a knowledge graph used to characterize the inherent causal relationship between abnormal operation information and risk types; the above method also includes using the knowledge graph to verify the security risk assessment results.

[0016] Based on the aforementioned technical means, this application utilizes knowledge graphs to verify the safety risk assessment results of power batteries in the target risk type. This can verify the logical consistency of the risk assessment results, reduce the risk of misjudgment of the safety risk assessment results of power batteries, and at the same time, the knowledge graph based on causal relationships also increases the interpretability of the safety risk assessment results.

[0017] In one possible implementation, the security risk assessment result includes: the security risk assessment approach and the security risk assessment probability; the target knowledge base also includes risk handling measures corresponding to multiple risk types; the above method also includes: using the security risk assessment approach, the security risk assessment probability, and the risk handling measures as prompt information; and outputting the prompt information.

[0018] Based on the aforementioned technical means, this application outputs the safety risk assessment approach, the probability of safety risk assessment, and risk handling measures as prompt information, which can more clearly explain the reasoning logic of the power battery safety risk assessment results; at the same time, it quantifies the risk level for safety risk assessment; and by outputting the risk handling measures, it can shorten the risk handling time of the power battery and improve the quality of risk management of the power battery.

[0019] In one possible implementation, the above method further includes: selecting target base models from a base model library based on base model performance; different base models in the base model library are constructed using different power battery safety risk assessment algorithms; and constructing a base model group using the target base models; wherein, the base model performance includes model complexity and judgment error rate.

[0020] Based on the above technical means, this application selects target base models from the base model library by model complexity and decision error rate, which can eliminate overfitted or overly simple base models as well as base models with high decision error rates, thereby improving the risk judgment performance of base models.

[0021] In one possible implementation, the charging and discharging strategy information of the power battery includes: discharge power strategy information and charging current strategy information; the user profile information includes: user fast charging status and driving speed distribution.

[0022] Based on the aforementioned technical means, this application supplements the power battery's operational information by acquiring the power battery's charging and discharging strategy information and user profile information. This allows the supplemented target information to be upgraded from the power battery's operational information to multi-dimensional information including equipment and users, thereby more quickly and accurately identifying the power battery's risk information, improving the accuracy of risk assessment, and also recommending more effective risk handling measures as prompts.

[0023] Secondly, this application provides a power battery safety risk assessment system, comprising: a data acquisition module for acquiring the operating information of the power battery; an information supplementation module for supplementing the operating information based on a target knowledge base to obtain target information; wherein the supplemented information includes the power battery's charging and discharging strategy information and user profile information; and a risk assessment module for assessing the safety risks of the power battery based on the target information and a base model group to obtain a safety risk assessment result; wherein the base model group includes multiple base model sets corresponding to multiple risk types of the power battery; multiple target base models in the target base model group jointly assess the safety risks of the power battery for the target risk type; the target base model group is one of multiple base model groups; and the target risk type is one of multiple risk types.

[0024] In one possible implementation, the target knowledge base includes historical operating information of multiple power batteries and their corresponding historical charging and discharging strategy information and historical user profile information; the information supplementation module is specifically used to determine the target historical operating information that matches the operating information from the multiple historical operating information; and to determine the historical charging and discharging strategy information and historical user profile information corresponding to the target historical operating information as supplementary information.

[0025] In one possible implementation, the risk assessment module is used to determine the safety risk assessment results of multiple target base models on the target risk type based on target information; and to determine the safety risk assessment results of the power battery on the target risk type based on the safety risk assessment results of multiple target base models on the target risk type and the confidence levels of multiple target base models.

[0026] In one possible implementation, the risk assessment module is specifically used to, when the confidence levels of multiple target base models are all less than the confidence threshold, combine the safety risk assessment results of multiple target base models on the target risk type based on a first type of integration strategy to obtain the safety risk assessment result of the power battery on the target risk type; wherein, the first type of integration strategy includes an integration strategy based on Bayesian inference.

[0027] In one possible implementation, the risk determination module is specifically used to, when there is at least one candidate target base model, based on a second type of integration strategy, integrate the safety risk determination results of at least one candidate target base model on the target risk type to obtain the safety risk determination result of the power battery on the target risk type; wherein, the candidate target base model is the target base model with a confidence level greater than a confidence threshold among multiple target base models; the second type of integration strategy includes hard voting integration strategy, soft voting integration strategy and rule integration strategy.

[0028] In one possible implementation, the target knowledge base also includes: a knowledge graph used to characterize the inherent causal relationship between abnormal operation information and risk types; the risk assessment module is also used to use the knowledge graph to verify the safety risk assessment results.

[0029] In one possible implementation, the security risk assessment result includes: the security risk assessment approach and the security risk assessment probability; the target knowledge base also includes risk handling measures corresponding to multiple risk types; the risk assessment module is also used to provide the security risk assessment approach, security risk assessment probability, and risk handling measures as prompt information; and output the prompt information.

[0030] In one possible implementation, the system further includes a model filtering module for filtering target base models from a base model library based on the performance of the base models; different base models in the base model library are constructed using different power battery safety risk assessment algorithms; a base model group is constructed using the target base models; wherein, the performance of the base models includes model complexity and judgment error rate.

[0031] In one possible implementation, the charging and discharging strategy information of the power battery includes: discharge power strategy information and charging current strategy information; the user profile information includes: user fast charging status and driving speed distribution.

[0032] Thirdly, this application provides a vehicle that includes the power battery safety risk assessment system described in the second aspect.

[0033] Fourthly, this application provides an electronic device, including: a processor and a memory, wherein the memory stores at least one computer program, and the at least one computer program is loaded and executed by the processor to implement the method described in the first aspect and any possible implementation thereof.

[0034] Fifthly, this application provides a computer-readable storage medium that, when the instructions in the computer-readable storage medium are executed by a processor of an electronic device, enables the electronic device to perform the methods described in the first aspect and any possible implementation thereof.

[0035] In a sixth aspect, this application provides a computer program product comprising computer instructions that, when executed on an electronic device, cause the electronic device to perform the method described in the first aspect and any of its possible implementations.

[0036] It should be noted that the technical effects of any of the implementation methods in aspects two through six can be found in the technical effects of the corresponding implementation methods in aspect one, and will not be repeated here.

[0037] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0038] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application, and do not constitute an undue limitation of this application.

[0039] Figure 1 This is a schematic diagram of an implementation environment shown in an embodiment of this application;

[0040] Figure 2 This is a flowchart illustrating a method for determining the safety risks of a power battery, as shown in an embodiment of this application.

[0041] Figure 3 This is a schematic diagram illustrating an embodiment of this application for dividing a standard information set into intervals;

[0042] Figure 4 This is a schematic diagram illustrating a sample information classification according to an embodiment of this application;

[0043] Figure 5 This is a block diagram illustrating a power battery safety risk assessment system according to an embodiment of this application;

[0044] Figure 6 This is another block diagram of a power battery safety risk assessment system shown in the embodiments of this application;

[0045] Figure 7 This is a block diagram illustrating an electronic device according to an embodiment of this application. Detailed Implementation

[0046] To enable those skilled in the art to better understand the technical solutions of this application, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings.

[0047] It should be noted that the terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0048] In the embodiments of this application, the words "exemplary," "for example," or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as "exemplary," "for example," or "for instance" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the words "exemplary," "for example," or "for instance" is intended to present the relevant concepts in a specific manner.

[0049] The technical solutions of the embodiments of this application will be described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0050] The power battery safety risk assessment system provided in this application is used to assess the safety risks of power batteries in vehicles (especially intelligent driving vehicles). Vehicles can also be referred to as vehicles, mobile carriers, electric vehicles (EVs), hybrid electric vehicles (HEVs), plug-in hybrid electric vehicles (PHEVs), fuel cell vehicles (FCVs), autonomous vehicles, intelligent and connected vehicles (ICVs), driverless vehicles, etc.

[0051] In this application's embodiments, the vehicle can be a sedan, a sport utility vehicle (SUV), a truck, a special vehicle (such as an ambulance, fire truck, police car, etc.), a driverless taxi, an intelligent connected bus, an autonomous logistics vehicle, an electric truck, etc. Furthermore, this method is also applicable to various special-purpose vehicles, such as agricultural vehicles, mining vehicles, forestry vehicles, airport vehicles, port vehicles, etc. This application does not impose specific limitations in this regard.

[0052] like Figure 1 As shown, the implementation environment for determining the safety risks of power batteries in this application includes: a power battery safety risk determination system 101 and a feature information acquisition system 102; the power battery safety risk determination system 101 and the feature information acquisition system 102 establish a communication connection. The power battery safety risk determination system 101 can be deployed in or outside the vehicle.

[0053] The feature information acquisition system 102 is used to collect the operating information of the power battery and transmit it to the power battery safety risk assessment system 101.

[0054] The feature information acquisition system 102 is also used to store the historical operating information of the power battery and to obtain the charging and discharging strategy information, user profile information and battery field knowledge of the power battery from the terminal equipment;

[0055] The feature information acquisition system 102 is also used to build a target knowledge base based on the historical operation information of the power battery, the historical charging and discharging strategy information of the power battery, user profile information and battery field knowledge.

[0056] The power battery safety risk assessment system 101 is used to receive the power battery's operating information from the feature information acquisition system 102. At the same time, it receives the power battery's charging and discharging strategy information and user profile information from the feature information acquisition system 102 to supplement the power battery's operating information, thereby obtaining target information.

[0057] The power battery safety risk assessment system 101 is used to assess the safety risks of the power battery based on the target information and basic model group of the power battery.

[0058] In practical applications, the power battery safety risk assessment system 101 can communicate with one or more feature information acquisition systems 102.

[0059] For ease of understanding, this application uses the communication connection between a power battery safety risk assessment system 101 and a feature information acquisition system 102 as an example for illustration.

[0060] As a feasible approach, Figure 1 The power battery safety risk assessment system 101 and the feature information acquisition system 102 are installed in the vehicle. The power battery safety risk assessment system 101 and the feature information acquisition system 102 can be functional modules integrated into the same device, or they can be independently installed devices. This application does not impose any limitations on the comparison.

[0061] It is easy to understand that when the power battery safety risk assessment system 101 and the feature information acquisition system 102 are functional modules integrated within the same device, the communication method between the power battery safety risk assessment system 101 and the feature information acquisition system 102 is the same as the communication method between modules within the device. In this case, the communication process between the two is the same as the communication process when the power battery safety risk assessment system 101 and the feature information acquisition system 102 are set up independently. For ease of understanding, this application mainly uses the example of the power battery safety risk assessment system 101 and the feature information acquisition system 102 being set up independently for explanation.

[0062] As a feasible approach, Figure 1 The power battery safety risk assessment system 101 or feature information acquisition system 102 can be set in a terminal, a server, or other types of electronic devices.

[0063] When the power battery safety risk assessment system 101 or the feature information acquisition system 102 is installed at the terminal, the terminal can be a device that provides data connectivity to vehicle users or vehicle owners, a handheld device with wireless connectivity, or other processing devices connected to a wireless modem. The terminal can communicate with one or more core networks via a radio access network (RAN). The terminal can be a mobile terminal, such as a computer with a mobile terminal, or a mobile device that exchanges voice and / or data with the radio access network, such as a mobile phone, tablet computer, laptop computer, netbook, or personal digital assistant (PDA). This application does not impose any limitations on this.

[0064] When the power battery safety risk assessment system 101 or the feature information acquisition system 102 is located on a server, the server can be a single server or a server cluster consisting of multiple servers. In some embodiments, the server cluster can also be a distributed cluster. This application does not impose any limitations on this.

[0065] It should be noted that the structure illustrated in the embodiments of this application does not constitute a limitation on the power battery safety risk assessment system 101. It may include more or fewer components than illustrated, or combine some components, or split some components, or have different component arrangements. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.

[0066] For ease of understanding, the following section provides a detailed description of the power battery safety risk assessment method provided in this application, in conjunction with the accompanying drawings.

[0067] Figure 2 This is a flowchart illustrating a method for determining the safety risks of a power battery, as shown in an embodiment of this application. Figure 2 The method includes:

[0068] S201. Obtain the operating information of the power battery.

[0069] The aforementioned power battery is a power source that provides power to vehicles and other equipment. The power battery is a power source that continuously provides energy support to the vehicle's drive system by storing and releasing electrical energy.

[0070] The aforementioned operational information refers to data used to characterize the inherent properties of the power battery, including current value, battery module temperature difference value, and battery cell voltage difference value.

[0071] The above operational information also needs to undergo outlier removal, filtering, and filling to complete data standardization in order to achieve data processing.

[0072] S202. Based on the target knowledge base, supplement the operation information to obtain the target information; the supplementary information includes the charging and discharging strategy information of the power battery and the user profile information.

[0073] The aforementioned target knowledge base includes operating information of power batteries, charging and discharging strategy information, user profile information, and risk handling measures corresponding to each safety risk assessment result. Among them, the risk handling measures are determined based on knowledge in the battery field.

[0074] The aforementioned risk management measures include technical protection measures, management and control measures, and emergency response measures. For example, technical protection measures may include using a BMS to monitor battery status in real time; management and control measures may include setting up alarm devices; and emergency response measures may include installing fireproof slots and heat insulation materials inside the battery pack.

[0075] The aforementioned knowledge in the field of batteries is a synthesis of existing battery fundamentals books, analysis process system documents, fault handling guides, battery algorithm documents, and other resources.

[0076] The aforementioned charging and discharging strategy information is characteristic information used to characterize the strategy application state of the power battery, including discharge power strategy information, charging current strategy information, etc.

[0077] The aforementioned discharge power strategy information refers to the upper limit and change rules of the output power dynamically adjusted by the battery during the discharge process based on real-time status (such as remaining charge, temperature, and load demand).

[0078] The aforementioned charging current strategy information refers to the input current magnitude and change logic that are dynamically adjusted according to the battery status (such as SOC, temperature, and health) during the charging process.

[0079] The aforementioned user profile information is characteristic information used to represent the user's usage status, including the user's fast charging status and driving speed distribution. The user's fast charging status and driving speed distribution correlate the user's driving behavior with battery performance, providing data support for safety risk assessment.

[0080] The above-mentioned user fast charging situation refers to the frequency, duration, and charging period (such as daily charging and emergency charging) of users using high-power fast charging (such as DC fast charging, with a power of ≥50kW).

[0081] The above-mentioned vehicle speed distribution refers to the statistical characteristics of vehicle speed during driving, including average speed, maximum speed, and speed fluctuation frequency (such as frequent rapid acceleration / deceleration).

[0082] In one possible implementation, the target knowledge base includes historical operating information of multiple power batteries and their corresponding historical charging and discharging strategy information and historical user profile information; the process of determining supplementary information includes: determining target historical operating information that matches the operating information from multiple historical operating information; and determining the historical charging and discharging strategy information and historical user profile information corresponding to the target historical operating information as supplementary information.

[0083] As one feasible approach, the matching process described above includes: First, extracting key features (such as current, temperature difference, and pressure difference) from the operational information, and then comparing them one by one with historical operational information in the target knowledge base. By calculating feature similarity (such as Euclidean distance or cosine similarity) or setting a threshold for filtering, the most suitable historical operational information is found as the target operational information. Subsequently, extracting corresponding historical charging and discharging strategy information and user profile information from the associated information of the matched target operational information as supplementary information.

[0084] S203. Based on the target information and the base model group, the safety risk of the power battery is determined, and the safety risk determination result is obtained; wherein, the base model group includes multiple base model groups corresponding to multiple risk types of the power battery; multiple target base models in the target base model group jointly determine the safety risk of the power battery for the target risk type; the target base model group is one of the multiple base model groups; the target risk type is one of the multiple risk types.

[0085] The aforementioned base model group includes a base model set for determining the safety risks of multiple risk types of power batteries. Each base model set contains multiple base models for determining the safety risks of one type of risk of power batteries.

[0086] The aforementioned base model set includes multiple base models for determining the safety risks of a certain type of power battery risk. For example, in the base model set of SelfDischarge-AI for battery self-discharge management, there are current identification models, voltage identification models, and temperature identification models.

[0087] The aforementioned risk types include battery overcharging or over-discharging, battery short circuit (internal or external), battery overheating, battery capacity decay, increased battery internal resistance, battery aging, battery casing cracking, battery cell imbalance, main positive and negative contactor sintering failure, single cell voltage and temperature sampling failure, leakage insulation failure, CAN communication failure, battery thermal runaway, battery management system (BMS) failure, and wiring or connector failure (such as loose connecting bolts or loose electrical connectors). The target risk type is one of the risk types; therefore, the aforementioned base model group includes power battery aging detection models and power battery communication failure models.

[0088] The above risk assessment results are used to describe the probability of battery thermal runaway risk and battery status within the current or future preset time period; the above safety risk assessment results include, but are not limited to, the safety risk assessment approach and the safety risk assessment probability.

[0089] The output of the above security risk assessment also includes risk management measures for each risk type. These risk management measures are provided by the target knowledge base and output the security risk assessment approach, the probability of security risk assessment, and the risk management measures.

[0090] Based on the aforementioned technical means, after collecting the operating information of the power battery, this application refines this operating information using the power battery's charging and discharging strategy information and user profile information in the database. On this basis, the application uses a base model group and the refined information to determine the safety risks of the power battery, making the risk assessment more comprehensive and accurate. Furthermore, the base model group is divided into different base model sets for different risk types, thus improving the risk assessment performance for different risk types of power batteries.

[0091] For example, the operational information, charging / discharging strategy information, and user profile information in the aforementioned target knowledge base are selected separately according to vehicle model and battery type; for instance, the total number of selected power batteries exceeds 10,000. The sampling period for the selected power battery data is no less than one year. Outlier handling, data field filtering, and null value imputation are also required for the operational information, charging / discharging strategy information, and user profile information to obtain a standard information set. Based on the aforementioned standard information set, intervals are divided, and data aggregation and feature extraction are performed within different intervals.

[0092] For example, Figure 3 This is a schematic diagram illustrating the division of a standard information set into intervals, as shown in an embodiment of this application. Figure 3As shown, the interval division includes seven division methods: division by a certain moment, division by a time sliding window consisting of a certain moment and a specified time interval, division by charging trip, division by discharging trip, division by resting trip, division by full life cycle data of a specified vehicle, and division by data of all vehicles of a certain model. The above division methods include power battery operation information, charging and discharging strategy information, and user profile information.

[0093] Secondly, feature extraction is performed on the segmented data. For example... Figure 3 As shown, feature extraction methods can be divided into four categories: frame-based extraction, trip-based extraction, vehicle-based extraction, and model-based extraction. Among them, information divided by a certain moment or by a time window consisting of a certain moment and a specified time interval is extracted by frame; information divided by charging trip, discharging trip, and stationary trip is extracted by trip; information divided by the full life cycle data of a specified vehicle is extracted by the specified vehicle; and information divided by all vehicle data of a model is extracted by the specified model.

[0094] Finally, the extracted feature information is the sample information, which can be divided into three categories. For example... Figure 4 As shown, Figure 4 This is a schematic diagram illustrating a sample information classification method according to an embodiment of this application.

[0095] The sample information is divided into operational information (including current, temperature difference, pressure difference, etc.), charging and discharging strategy information (including discharge power strategy, charging current strategy, etc.), and user profile data (including user fast charging status, driving speed distribution, etc.).

[0096] The aforementioned target knowledge base stores multiple sample information entries. Each sample information entry includes operational information (including current, temperature difference, pressure difference, etc.), charging and discharging strategy information (including discharge power strategy, charging current strategy, etc.), user profile data (including user fast charging status, driving speed distribution, etc.).

[0097] As an achievable approach, the data processing of the aforementioned target knowledge base also includes: cleaning existing documents, mainly deduplication (merging duplicate question-and-answer content) and error correction (correcting or deleting potentially erroneous content); document templated formatting, defining standard formats to convert various types of document data into a unified format, such as defining it as a problem description + solution + related guide format; importing and deploying documents using a knowledge management platform; and conducting retrieval tests to confirm whether the correct documents are returned by searching for questions (such as the analysis process of battery AI algorithms and the invitation mechanism for battery faults).

[0098] In one possible implementation, the safety risk of the power battery is determined based on the target information and the base model group, and the safety risk determination result is obtained. This includes: determining the safety risk determination results of multiple target base models on the target risk type based on the target information; and determining the safety risk determination result of the power battery on the target risk type based on the safety risk determination results of multiple target base models on the target risk type and the confidence level of multiple target base models.

[0099] The aforementioned confidence level refers to the reliability or credibility assessment of the target base model's judgment on the safety risks of the power battery. It is usually expressed in the form of probability or percentage. For example, if model A determines that power battery B has a "high risk of thermal runaway" with a confidence level of 90%, it means that model A has a 90% probability that the risk judgment of "high risk of thermal runaway" is correct. If model A determines that power battery B has a "high risk of thermal runaway" with a confidence level of 50%, it means that model A has a 50% probability that the risk judgment of "high risk of thermal runaway" is correct. That is, the certainty of the risk judgment result is poor, and it is necessary to combine the risk judgment results of other models or use other models for further judgment.

[0100] The confidence level mentioned above is related to the information quality of the target information input to the base model, the complexity of the model, the correlation between the input target information and the risk type, and the robustness of the algorithm in the base model.

[0101] The confidence level mentioned above can be determined using methods such as probability output method, uncertainty estimation method, distance metric method and ensemble learning method.

[0102] In one possible implementation, the safety risk assessment result of the power battery on the target risk type is determined based on the safety risk assessment results of multiple target basis models on the target risk type and the confidence levels of multiple target basis models. This includes: when the confidence levels of multiple target basis models are all less than the confidence level threshold, the safety risk assessment result of the power battery on the target risk type is obtained by integrating the safety risk assessment results of multiple target basis models on the target risk type based on a first type of integration strategy; wherein, the first type of integration strategy includes an integration strategy based on Bayesian inference.

[0103] The aforementioned confidence threshold is determined based on the safety standards for power batteries. For example, ISO 26262 ASIL D requires a confidence level of ≥99% for determining the thermal runaway risk of power batteries, so the confidence threshold needs to be set to 0.99.

[0104] The aforementioned base model is used to determine battery safety risks, including the risk of thermal runaway of power batteries. It has high requirements for safety performance. Therefore, the confidence threshold is generally in the range of [0.9, 0.99].

[0105] The aforementioned Bayesian inference-based fusion strategy combines Bayesian probability theory and model fusion methods to dynamically fuse the safety risk assessment results of multiple target base models within the same base model group. The fusion process based on Bayesian inference involves quantifying the uncertainty of each base model using Bayes' theorem and adjusting the weights of the risk assessment results of the base models through posterior probability distributions, thereby obtaining the final risk assessment result for the power battery.

[0106] In one possible implementation, the safety risk assessment result of the power battery on the target risk type is determined based on the safety risk assessment results of multiple target basis models on the target risk type and the confidence levels of the multiple target basis models. This includes: when there is at least one candidate target basis model, the safety risk assessment result of the power battery on the target risk type is obtained by combining the safety risk assessment results of at least one candidate target basis model on the target risk type based on a second type of integration strategy; wherein, the candidate target basis model is the target basis model among the multiple target basis models whose confidence level is greater than the confidence level threshold; the second type of integration strategy includes hard voting integration strategy, soft voting integration strategy, and rule integration strategy.

[0107] The aforementioned candidate target basis models refer to target basis models with a confidence level greater than the confidence threshold.

[0108] For example, if the candidate target base model contains only one base model, the safety risk of the power battery is determined based on the candidate target base model; if the candidate target base model contains more than one base model, the safety risk determination results of the candidate target base model are integrated by hard voting integration, soft voting integration and rule integration respectively to obtain the safety risk determination results of the power battery.

[0109] The aforementioned hard-vote integration strategy refers to statistically analyzing the security risk assessment results of multiple base models in the same base model group, with the result that receives the most assessments being used as the final security risk assessment result.

[0110] The aforementioned soft voting integration strategy refers to obtaining the final probability values ​​of different risk types as the security risk judgment result by weighted averaging or probability fusion based on the probability distributions output by multiple base models in the same base model group.

[0111] The aforementioned rule integration strategy dynamically adjusts the weight or priority of the judgment results of each base model through manually defined rules or meta-learning models. These rules need to be determined in conjunction with battery domain knowledge in the target knowledge base.

[0112] According to the above technical solution, this application integrates the safety risk assessment results of each base model on the target risk type and the corresponding confidence level of each base model to obtain the safety risk assessment results of the power battery on the target risk type. This integrates the safety risk assessment results of base models with different confidence levels in different ways, improves the impact of the safety risk assessment results of base models with higher confidence levels, and reduces the impact of the safety risk assessment results of base models with lower confidence levels, thereby improving the assessment performance of the safety risk assessment results of the power battery.

[0113] In one possible implementation, the target knowledge base also includes a knowledge graph used to characterize the inherent causal relationship between abnormal operation information and risk types; the above method also includes using the knowledge graph to verify the security risk assessment results.

[0114] The aforementioned knowledge graph is a risk causal knowledge graph. It uses triples of entities, attributes, and relationships to express the causal relationship between abnormal operation information and risk type quality inspection in a structured and semantic way. For example, abnormal operation information can be "voltage mutation" or "temperature anomaly", risk types include "thermal runaway risk" or "short circuit risk", and causal relationships include "cause", "trigger", or "exacerbate". For example, "voltage mutation causes internal short circuit risk in battery" or "abnormal differential pressure data exacerbates self-discharge problem" are examples.

[0115] The above test refers to determining whether the safety risk assessment result of the power battery in the target risk type satisfies the causal relationship of the knowledge graph. If the safety risk assessment result of the power battery in the target risk type satisfies the causal relationship of the knowledge graph, the result is output; if the safety risk assessment result of the power battery in the target risk type does not satisfy the causal relationship of the knowledge graph, the result is not output, and the safety risk assessment of the power battery needs to be re-performed to reduce the risk of misjudgment of the safety risk assessment result of the power battery.

[0116] In one possible implementation, the above method further includes: selecting target base models from a base model library based on base model performance; different base models in the base model library are constructed using different power battery safety risk assessment algorithms; and constructing a base model group using the target base models; wherein, the base model performance includes model complexity and judgment error rate, so that the risk assessment performance of the selected base models is more accurate, thereby achieving more accurate power battery safety risk assessment.

[0117] The aforementioned base model performance refers to the base model's ability to determine security risks.

[0118] The aforementioned indicators for assessing the safety risks of base model performance include complexity, distribution characteristics, and outlier characteristics. Complexity includes model structure complexity, data dependency complexity, and interpretability complexity. Distribution characteristics include predicted value distribution and feature distribution. Outlier characteristics include data anomalies, training process anomalies, and judgment error rate.

[0119] The level of model complexity affects the fitting ability and generalization of the base model. The determination of model complexity includes parameter complexity, computational complexity, and structural complexity. Parameter complexity refers to the total number of parameters that can be used for learning in the base model; computational complexity refers to the computational resources required for the model to perform inference; and structural complexity refers to the depth of the model architecture or the number of branches, such as the number of layers in a neural network.

[0120] The aforementioned error rate is the ratio of the number of incorrect judgments to the total number of security risk judgments when the base model is used to determine security risks.

[0121] Figure 5 This is a block diagram illustrating a power battery safety risk assessment system according to an embodiment of this application. (Refer to...) Figure 5 The system includes: a data acquisition module 501, an information supplementation module 502, and a risk assessment module 503.

[0122] The data acquisition module 501 is used to acquire the operating information of the power battery.

[0123] The information supplementation module 502 is used to supplement the operation information based on the target knowledge base to obtain the target information; the supplemented information includes the charging and discharging strategy information of the power battery and user profile information.

[0124] The risk assessment module 503 is used to assess the safety risks of the power battery based on target information and a base model group, and obtain the safety risk assessment result. The base model group includes multiple base model groups corresponding to multiple risk types of the power battery. Multiple target base models in the target base model group jointly assess the safety risks of the power battery for the target risk type. The target base model group is one of the multiple base model groups. The target risk type is one of the multiple risk types.

[0125] In one possible implementation, the target knowledge base includes historical operating information of multiple power batteries and their corresponding historical charging and discharging strategy information and historical user profile information; the information supplementation module is specifically used to determine the target historical operating information that matches the operating information from the multiple historical operating information; and to determine the historical charging and discharging strategy information and historical user profile information corresponding to the target historical operating information as supplementary information.

[0126] In one possible implementation, the risk assessment module is used to determine the safety risk assessment results of multiple target base models on the target risk type based on target information; and to determine the safety risk assessment results of the power battery on the target risk type based on the safety risk assessment results of multiple target base models on the target risk type and the confidence levels of multiple target base models.

[0127] In one possible implementation, the risk assessment module is specifically used to, when the confidence levels of multiple target base models are all less than the confidence threshold, combine the safety risk assessment results of multiple target base models on the target risk type based on a first type of integration strategy to obtain the safety risk assessment result of the power battery on the target risk type; wherein, the first type of integration strategy includes an integration strategy based on Bayesian inference.

[0128] In one possible implementation, the risk determination module is specifically used to, when there is at least one candidate target base model, based on a second type of integration strategy, integrate the safety risk determination results of at least one candidate target base model on the target risk type to obtain the safety risk determination result of the power battery on the target risk type; wherein, the candidate target base model is the target base model with a confidence level greater than a confidence threshold among multiple target base models; the second type of integration strategy includes hard voting integration strategy, soft voting integration strategy and rule integration strategy.

[0129] In one possible implementation, the target knowledge base also includes: a knowledge graph used to characterize the inherent causal relationship between abnormal operation information and risk types; the risk assessment module is also used to use the knowledge graph to verify the safety risk assessment results.

[0130] In one possible implementation, the security risk assessment result includes: the security risk assessment approach and the security risk assessment probability; the target knowledge base also includes risk handling measures corresponding to multiple risk types; the risk assessment module is also used to provide the security risk assessment approach, security risk assessment probability, and risk handling measures as prompt information; and output the prompt information.

[0131] In one possible implementation, the system further includes a model filtering module for filtering target base models from a base model library based on the performance of the base models; different base models in the base model library are constructed using different power battery safety risk assessment algorithms; a base model group is constructed using the target base models; wherein, the performance of the base models includes model complexity and judgment error rate.

[0132] In one possible implementation, the charging and discharging strategy information of the power battery includes: discharge power strategy information and charging current strategy information; the user profile information includes: user fast charging status and driving speed distribution.

[0133] For example, Figure 6 This is another block diagram illustrating a power battery safety risk assessment system according to an embodiment of this application. It includes a target knowledge base, a base model group, and a large model system.

[0134] The large model system is used to output target information, risk assessment ideas, risk assessment probabilities, and risk handling measures.

[0135] The target knowledge base is derived from the accumulation of knowledge based on battery characteristic information and battery-related knowledge.

[0136] The base model group is obtained by integrating algorithms such as DeltaSOC-AI and SelfDischarge-AI.

[0137] The target knowledge base and base model group are integrated and analyzed as the output of the large model system.

[0138] The aforementioned battery characteristic information includes the power battery's operating information, charging and discharging strategy information, and user profile information.

[0139] Regarding the methods in the above embodiments, the specific manner in which each step is performed has been described in detail in the embodiments of the power battery safety risk determination method, and will not be elaborated here.

[0140] Figure 7 This is a block diagram illustrating an electronic device according to an embodiment of this application. Figure 7 As shown, the electronic device includes, but is not limited to, a processor 701 and a memory 702.

[0141] The memory 702 described above is used to store the executable instructions of the processor 701. It is understood that the processor 701 is configured to execute instructions to implement the power battery safety risk assessment method in the above embodiments.

[0142] It should be noted that those skilled in the art will understand that Figure 7 The electronic device structure shown does not constitute a limitation on the electronic device; the electronic device may include, but is not limited to, other electronic devices. Figure 7 This may indicate more or fewer components, or combinations of certain components, or different component arrangements.

[0143] Processor 701 is the control center of the electronic device. It connects various parts of the electronic device via various interfaces and lines. By running or executing software programs and / or modules stored in memory 702, and by calling data stored in memory 702, it performs various functions and processes data, thereby providing overall monitoring of the electronic device. Processor 701 may include one or more processing units. Processor 701 may integrate an application processor and a modem processor. The application processor mainly handles the operating system, user interface, and applications, while the modem processor mainly handles wireless communication. It is understood that the modem processor may not be integrated into processor 701.

[0144] The memory 702 can be used to store software programs and various data. The memory 702 may primarily include a program storage area and a data storage area. The program storage area may store the operating system, application programs required by at least one functional module (such as deterministic components, integrated components, etc.), etc. Furthermore, the memory 702 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0145] In an exemplary embodiment, a computer-readable storage medium including instructions is also provided, such as a memory 702 including instructions, which can be executed by a processor 701 of an electronic device to implement the methods in the above embodiments.

[0146] In actual implementation, Figure 5 The functions of the data acquisition module 501, information supplementation module 502, and risk assessment module 503 can all be provided by... Figure 7 The processor 701 calls the computer program stored in the memory 702 to implement the process. The specific execution process can be found in the description of the method section in the previous embodiment, and will not be repeated here.

[0147] Optionally, the computer-readable storage medium may be a non-transitory computer-readable storage medium, such as a read-only memory (ROM), random access memory (RAM), compact disc read-only memory (CD-ROM), magnetic tape, floppy disk, and optical data storage device. In an exemplary embodiment, this application also provides a computer program product including one or more instructions, which can be executed by the processor 701 of an electronic device to perform the methods in the above embodiments.

[0148] It should be noted that when one or more instructions in the computer-readable storage medium or computer program product are executed by the processor of an electronic device, they implement the various processes of the above method embodiments and achieve the same technical effect as the above method. To avoid repetition, they will not be described again here.

[0149] Through the above description of the embodiments, those skilled in the art can clearly understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0150] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another apparatus, or some features may be ignored or not executed. Furthermore, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0151] The units described as separate components may or may not be physically separate. A component shown as a unit can be one or more physical units; that is, it can be located in one place or distributed in multiple different locations. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0152] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0153] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a readable storage medium. Based on this understanding, the technical solutions of the embodiments of this application, essentially, or the parts that contribute to the prior art, or all or part of the technical solutions, can be embodied in the form of a software product. This software product is stored in a storage medium and includes several instructions to cause a device (which may be a microcontroller, chip, etc.) or processor to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, ROM, RAM, magnetic disks, or optical disks.

[0154] This application provides a computer program product containing instructions that, when run on a computer, cause the computer to perform the methods described in the above method embodiments.

[0155] This application also provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the method in the method flow shown in the above method embodiments.

[0156] The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable computer disk, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory, a register, a hard disk, an optical fiber, a portable compact disk read-only memory, an optical storage device, a magnetic storage device, or any suitable combination thereof, or any other form of computer-readable storage medium well known in the art. An exemplary storage medium is coupled to a processor, enabling the processor to read information from and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can reside in an application-specific integrated circuit (ASIC). In embodiments of this application, the computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0157] Since the power battery safety risk assessment system, computer-readable storage medium, and computer program product in the embodiments of this application can be applied to the above methods, the technical effects they can achieve can also be referred to the above method embodiments. The embodiments of this application will not be repeated here.

[0158] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for determining the safety risks of a power battery, characterized in that, The method for determining the safety risks of power batteries includes: Obtain operational information of the power battery; Based on the target knowledge base, the operation information is supplemented to obtain the target information; wherein, the supplemented information includes the charging and discharging strategy information of the power battery and user profile information; Based on the target information and the base model group, the safety risk of the power battery is determined, and a safety risk determination result is obtained; wherein, the base model group includes multiple base model groups corresponding to multiple risk types of the power battery; multiple target base models in the target base model group jointly determine the safety risk of the power battery for the target risk type; the target base model group is one of the multiple base model groups; the target risk type is one of the multiple risk types.

2. The method for determining the safety risks of power batteries according to claim 1, characterized in that, The target knowledge base includes historical operating information of multiple power batteries and their corresponding historical charging and discharging strategy information and historical user profile information. The process of determining the supplementary information includes: Determine the target historical operation information that matches the operation information from multiple historical operation information sources; The historical charging and discharging strategy information and historical user profile information corresponding to the target historical operation information are determined as the supplementary information.

3. The method for determining the safety risks of power batteries according to claim 1, characterized in that, The step of assessing the safety risks of the power battery based on the target information and the base model group, and obtaining the safety risk assessment result, includes: Based on the target information, determine the security risk assessment results of the multiple target base models on the target risk type; Based on the safety risk assessment results of the multiple target base models on the target risk type and the confidence level of the multiple target base models, the safety risk assessment result of the power battery on the target risk type is determined.

4. The method for determining the safety risks of power batteries according to claim 3, characterized in that, The determination of the safety risk assessment result of the power battery on the target risk type based on the safety risk assessment results of the multiple target basis models on the target risk type and the confidence levels of the multiple target basis models includes: When the confidence levels of the multiple target base models are all less than the confidence threshold, based on the first type of integration strategy, the safety risk assessment results of the multiple target base models on the target risk type are combined to obtain the safety risk assessment result of the power battery on the target risk type. The first type of integration strategy includes integration strategies based on Bayesian inference.

5. The method for determining the safety risks of power batteries according to claim 3, characterized in that, The determination of the safety risk assessment result of the power battery on the target risk type based on the safety risk assessment results of the multiple target basis models on the target risk type and the confidence levels of the multiple target basis models includes: In the presence of at least one candidate target base model, based on the second type of integration strategy, the safety risk assessment results of the at least one candidate target base model on the target risk type are combined to obtain the safety risk assessment result of the power battery on the target risk type. Wherein, the candidate target basis model is the target basis model among the plurality of target basis models whose confidence is greater than a confidence threshold; The second type of integration strategy includes hard voting integration strategy, soft voting integration strategy, and rule integration strategy.

6. The method for determining the safety risks of power batteries according to claim 1, characterized in that, The target knowledge base also includes: a knowledge graph used to characterize the inherent causal relationship between abnormal operation information and risk types; The method further includes: The security risk assessment results are then verified using the knowledge graph.

7. The method for determining the safety risks of power batteries according to claim 1, characterized in that, The security risk assessment result includes: the security risk assessment approach and the security risk assessment probability; the target knowledge base also includes risk handling measures corresponding to multiple risk types; the method further includes: The security risk assessment approach, the security risk assessment probability, and the risk handling measures will be used as prompt information. Output the aforementioned prompt message.

8. The method for determining the safety risk of a power battery according to any one of claims 1-7, characterized in that, The method further includes: Based on the performance of the base models, target base models are selected from the base model library; different base models in the base model library are constructed using different power battery safety risk assessment algorithms. Using the target base model, construct the base model group; The performance of the base model includes model complexity and decision error rate.

9. The method for determining the safety risks of power batteries according to claim 8, characterized in that, The charging and discharging strategy information of the power battery includes: discharge power strategy information and charging current strategy information; The user profile information includes: user fast charging status and driving speed distribution.

10. A power battery safety risk assessment system, characterized in that, The power battery safety risk assessment system includes: The data acquisition module is used to acquire the operating information of the power battery; The information supplementation module is used to supplement the operation information based on the target knowledge base to obtain the target information; wherein, the supplemented information includes the charging and discharging strategy information of the power battery and user profile information; The risk assessment module is used to assess the safety risks of the power battery based on the target information and the base model group, and obtain a safety risk assessment result; wherein, the base model group includes multiple base model sets corresponding to multiple risk types of the power battery; multiple target base models in the target base model set jointly assess the safety risks of the power battery for the target risk type; the target base model set is one of the multiple base model sets; the target risk type is one of the multiple risk types.

11. A vehicle, characterized in that, The vehicle includes the power battery safety risk assessment system as described in claim 10.

12. An electronic device, characterized in that, The electronic device includes a processor and a memory, wherein the memory stores at least one computer program, and the at least one computer program is loaded and executed by the processor to implement the power battery safety risk determination method according to any one of claims 1 to 9.

Images