Power battery module anti-sparking method, device, equipment and medium

Abstract

The invention relates to a power battery module anti-sparking method, device, equipment and medium, the power battery module anti-sparking method comprises the following steps: if a power battery module is in a non-charging state, current change information is detected in real time, whether effective load access exists or not is judged, and if the effective load access exists, the power battery module is prevented from sparking. If yes, a corresponding temporary starting instruction is generated, environment information is obtained in real time based on the slow channel, format standardization processing is conducted on the environment information and the current change information, and a corresponding structured information set is generated; assigning a corresponding starting weight to the structured information group; according to the method, each structured information group and the starting weight corresponding to each structured information group are substituted into an established starting identification model, and a corresponding confirmation instruction is generated, so that a current abrupt change signal caused by load access can be accurately captured, conduction is initiated after the load access state reaches a set stability standard, and the ignition risk is remarkably reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of preventing ignition in battery modules, and in particular, to a method, device, equipment and medium for preventing ignition in power battery modules. Background Art

[0002] At present, with the rapid popularization of new energy transportation vehicles, light power battery modules, as the core power supply units, play key functions in equipment such as electric bicycles and electric motorcycles. However, during use, due to improper electrical design or irregular user operations, arcing is likely to occur when the connector is plugged and unplugged, that is, the so-called "ignition", which not only affects the service life of the battery module, but also brings potential safety risks. Although existing battery management systems already have basic control capabilities in aspects such as charge and discharge protection and current detection, the mechanism for judging load access in the non-charging state is not yet mature. For example, in the non-charging state, simply judging whether the load is accessed by the change in voltage will cause the discharge switch to be mis-conducted due to instantaneous fluctuations or false signals, thus triggering the ignition phenomenon. Summary of the Invention

[0003] In order to solve the problem that the traditional judgment mechanism only judges whether the load is accessed by the simple change in voltage, which in turn affects the service life of the battery module and also brings potential safety risks, the present application provides a method, device, equipment and medium for preventing ignition in power battery modules.

[0004] A method for preventing ignition in a power battery module, the method for preventing ignition in a power battery module includes: Detect the working state of the power battery module, where the working state at least includes a charging state and a non-charging state; If the power battery module is in the non-charging state, then based on the fast channel with microsecond-level response, detect the current change information in real time, and then judge whether there is an effective load access. The effective load is a load that maintains a stable current characteristic for more than a preset time threshold after the current mutation signal is detected in the current change information; If there is an effective load access, generate a corresponding temporary start instruction, and the temporary start instruction is used to start the conduction of the discharge switch conduction element to supply power to the effective load; Based on the slow channel, obtain the environmental information in real time, perform format standardization processing on the environmental information and the current change information, and generate a corresponding structured information group; Call the historical data set cached locally, and assign a corresponding start weight to the structured information group according to the historical data set; Substitute each of the structured information groups and the corresponding startup weights of each of the structured information groups into the established startup recognition model to generate a corresponding confirmation instruction, where the confirmation instruction is used to control the startup discharge switch conducting element to continue to conduct.

[0005] By adopting the above technical solution, by detecting the working state of the power battery module and distinguishing between charging and non-charging modes, it is ensured that the subsequent discharge control logic is only triggered under non-charging conditions, thereby avoiding control interference in unnecessary scenarios. Combining the microsecond-level real-time monitoring of current changes in the fast channel enables the system to accurately capture the current mutation signal caused by the load access, and initiate conduction only after the load access state reaches the set stability standard, significantly reducing the risk of sparking caused by mis-conduction due to transient interference.

[0006] In a preferred example of the present application, it can be further configured that: in the step of the fast channel based on microsecond-level response detecting current change information in real time and then determining whether there is an effective load access, it includes: The fast channel based on microsecond-level response detects current change information in real time, and detects whether a corresponding current mutation signal appears in the current change information; If the current mutation signal is detected, start a timer, and continue to collect current change information through the fast channel within the preset time threshold set by the timer, and determine whether the current change information continuously maintains within the stable current characteristic range after the current mutation signal; If it maintains within the stable current characteristic range after the current mutation signal, it is determined that there is an effective load access.

[0007] By adopting the above technical solution, through continuous dynamic detection of the load current and introducing a timer mechanism to judge stability, the system not only relies on the mutation characteristics of the current in the process of determining an "effective load", but also requires it to remain stable for a period of time, effectively avoiding misjudgment caused by jitter, loose connection or false load, thereby improving the decision-making accuracy and safety of discharge control.

[0008] In a preferred example of the present application, it can be further configured that: the temporary startup instruction includes a multi-stage conduction control mechanism, and the multi-stage conduction control mechanism includes: Execute the first-stage conduction operation, where the first-stage conduction operation controls the discharge switch conducting element to conduct with a preset conduction duty ratio, so that the discharge switch conducting element is placed in a preset current-limiting conduction state; Receive the confirmation instruction in real time; If the confirmation instruction is received, control the startup discharge switch conducting element to continue to conduct; If the confirmation instruction is not received after exceeding the preset response time, the control starts to turn off the discharge switch conducting element.

[0009] By adopting the above technical solution, by introducing a multi-stage conduction control mechanism, the system will not directly enter the fully conducting state after detecting the payload, but initially supplies power in a current-limiting manner while waiting for the feedback of the confirmation signal. If the confirmation cannot be completed within the specified time, the discharge path is automatically disconnected, thereby constructing a soft-start protection logic with self-checking ability, effectively reducing the impact on the device and potential safety hazards in the case of mis-conduction.

[0010] In a preferred example of the present application, it can be further configured that: in the step of formatting and standardizing the environmental information and the current change information to generate a corresponding structured information group, the environmental information at least includes humidity information and temperature information, and the step further includes: Converting the temperature information into a relative temperature difference coefficient in degrees Celsius; Mapping the humidity information to a humidity safety coefficient between 0 and 1; Calculating a corresponding short-term stable value of the current change information through a dynamic mean filtering algorithm based on a sliding window; Encoding the relative temperature difference coefficient, the humidity safety coefficient, and the short-term stable value into a structured information group, and the structured information group is used as an input vector to be uniformly input into the subsequent start recognition model to participate in the conduction adjustment judgment.

[0011] By adopting the above technical solution, in terms of environmental information processing, the system standardizes and converts key parameters such as temperature and humidity, and combines them with current information to encode them into a unified structured information group as input data, thereby realizing the standardization and fusion of multi-source perception information, making the model input data have stronger expression ability and versatility, and further improving the system's recognition ability and robustness to complex environmental conditions.

[0012] In a preferred example of the present application, it can be further configured that: in the step of assigning a corresponding start weight to the structured information group according to the historical data set, it includes: Parsing the parameter tags of the structured information group; Calling the cumulative statistical information of the corresponding fields in the historical data set based on the parameter tags; Determining the start weight of the structured information group by looking up the table according to the mapping relationship between the parameter change trend and the conduction risk level in the cumulative statistical information.

[0013] By adopting the above technical solution, by invoking the statistical information in the local historical dataset and combining the label matching relationships of the parameters in the structured information group, the system can generate weight parameters by looking up the table and assigning values. This process reflects the knowledge-guided judgment of the system based on the operating experience and risk mapping relationship, so as to assign reasonable decision-making influence to each group of input states and improve the credibility and differentiation ability of the model scoring results.

[0014] In a preferred example of the present application, it can be further configured as follows: in the step of substituting each of the structured information groups and the startup weights corresponding to each of the structured information groups into the established startup recognition model to generate corresponding confirmation instructions, it includes: Substitute each of the structured information groups and the startup weights corresponding to each of the structured information groups into the established startup recognition model. The established startup recognition model at least includes a weighting sub-model, a scoring sub-model, and a determination sub-model; Based on the weighting sub-model, perform weighting processing on the structured information group and the corresponding startup weight to generate a corresponding startup permission; Based on the scoring sub-model, add up each of the startup permissions to generate a corresponding scoring result; Based on the determination sub-model, perform a threshold judgment on the scoring result to determine whether to generate a corresponding confirmation instruction.

[0015] By adopting the above technical solution, in terms of the decision-making mechanism, the startup recognition model introduced by the system includes three sub-modules: weighting, scoring, and determination. It can aggregate and calculate the structured information after weighting processing, and output the final control instruction based on the threshold judgment logic. This scoring and feedback mechanism has high structure and interpretability, significantly improving the intelligent discrimination ability and adjustability of the entire control system, and ensuring that the anti-spark control target can still be stably achieved under complex loads and changing environments.

[0016] The second above-mentioned invention object of the present application is achieved by the following technical solutions: An anti-spark device for a power battery module, the anti-spark device for a power battery module includes: A detection module, used to detect the working state of the power battery module, and the working state at least includes a charging state and a non-charging state; A judgment module, used to, if the power battery module is in the non-charging state, detect current change information in real time based on a fast channel with microsecond-level response, and then judge whether there is an effective load access. The effective load is a load that maintains a stable current characteristic for more than a preset time threshold after the current mutation signal is detected in the current change information; A first generation module, configured to generate a corresponding temporary start instruction if there is a payload access, where the temporary start instruction is used to start a discharge switch conducting element to conduct, so as to supply power to the payload; A second generation module, configured to obtain environment information in real time based on a slow channel, perform format standardization processing on the environment information and the current change information, and generate a corresponding structured information group; A calling module, configured to call a locally cached historical data set, and assign a corresponding start weight to the structured information group according to the historical data set; A third generation module, configured to substitute each of the structured information groups and the start weights corresponding to the structured information groups into a pre-established start recognition model to generate a corresponding confirmation instruction, where the confirmation instruction is used to control the discharge switch conducting element to continue to conduct;

[0017] The above-mentioned third object of the present application is achieved by the following technical solutions: A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the above-mentioned method for preventing ignition of a power battery module are implemented.

[0018] The above-mentioned fourth object of the present application is achieved by the following technical solutions: A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above-mentioned method for preventing ignition of a power battery module are implemented.

[0019] In summary, the present application includes at least one of the following beneficial technical effects: The present application does not rely on traditional voltage mutation or contactor logic judgment, but introduces a fast detection path with microsecond-level response to capture the dynamic changes of load access at the current level, ensuring that the power supply action is triggered only when the load is truly stably connected, and avoiding misjudgment caused by instantaneous interference or miscontact. At the same time, the system also introduces a low-frequency slow channel to fuse and collect environmental parameters (such as temperature, humidity, etc.) and current change characteristics, standardize them to form a unified structured data group, then assign values in combination with a weight model constructed based on historical operation data, and finally input the weighted information into a structured decision model to generate a confirmation feedback for determining whether to maintain the conduction of the discharge switch. This solution realizes a mode of "temporary conduction first + multi-source decision-making later" in the control logic, improves the adaptability of the system to complex operating conditions, not only effectively avoids the arc risk brought by unstable load access, but also improves the fault tolerance of the entire discharge control system to environmental disturbances and load type changes, thereby realizing higher-reliability dynamic anti-ignition control. Description of the Drawings

[0020] Figure 1 is a flowchart of a method for preventing sparking in a power battery module according to an embodiment of the present application.

[0021] Figure 2 is a flowchart for implementing step S20 in a method for preventing sparking in a power battery module according to an embodiment of the present application; Figure 3 is a flowchart for implementing step S30 in a method for preventing sparking in a power battery module according to an embodiment of the present application; Figure 4 is a flowchart for implementing step S40 in a method for preventing sparking in a power battery module according to an embodiment of the present application; Figure 5 is a flowchart for implementing step S50 in a method for preventing sparking in a power battery module according to an embodiment of the present application; Figure 6 is a flowchart for implementing step S60 in a method for preventing sparking in a power battery module according to an embodiment of the present application; Figure 7 is a schematic block diagram of a principle of a device for preventing sparking in a power battery module according to an embodiment of the present application; Figure 8 is a schematic diagram of a device according to an embodiment of the present application. Detailed Description of the Embodiment

[0022] The present application will be further described in detail below with reference to the accompanying drawings.

[0023] In one embodiment, as Figure 1 shown, the present application discloses a method for preventing sparking in a power battery module. A method for preventing sparking in a power battery module includes: S10. Detect the working state of the power battery module, where the working state includes at least a charging state and a non-charging state; S20. If the power battery module is in a non-charging state, then detect the current change information in real time based on a fast channel with microsecond-level response, and then determine whether there is an effective load connected. An effective load is a load that maintains a stable current characteristic for more than a preset time threshold after a current mutation signal is detected in the current change information; the fast channel refers to a signal acquisition path with high sampling rate and low-latency response ability, usually composed of a microsecond-level current sensor and a fast processing circuit, used to immediately capture the rapid change of the current at the battery output terminal, mainly for determining whether a load is suddenly connected.

[0024] S30. If there is an effective load connected, then generate a corresponding temporary start instruction, and the temporary start instruction is used to start the discharge switch conducting element to conduct, so as to supply power to the effective load; S40. Obtain environmental information in real time based on the slow channel, perform format standardization processing on the environmental information and current change information, and generate corresponding structured information groups. The slow channel is a low-frequency information acquisition path relative to the fast channel, mainly used to collect environmental variables such as temperature and humidity. It has a long period and a slow change rate, and its purpose is to provide a reference basis at the environmental level for subsequent determination of conduction safety. The structured information group is a data set in a unified format formed by the perception data from different channels after standardization processing. These data usually have completed operations such as normalization, denoising, and encoding, and have clear fields and types, and can be directly used as the input vector of the model.

[0025] S50. Call the historical data set cached locally, and assign corresponding start weights to the structured information groups according to the historical data set. The historical data set is a collection of various acquisition parameters and event result data recorded by the system during operation for a long time, including current fluctuations, conduction results, abnormal trigger situations, etc. under different environmental conditions and load states in the past, and is used to provide data support or empirical mapping basis in the current determination. The start weight is a reference importance index assigned to each type of structured data, reflecting its influence degree in the conduction decision. This weight is usually set based on historical statistical laws or risk levels, and is used to highlight the influence of key features on the model output.

[0026] S60. Substitute each structured information group and the corresponding start weight of each structured information group into the established start recognition model to generate corresponding confirmation instructions. The confirmation instructions are used to control the start discharge switch conduction element to continue to maintain conduction. The start recognition model is a multi-layer structure calculation model with weighted processing, scoring calculation, and decision-making capabilities. It can receive the weighted structured information group and generate an evaluation result of the discharge behavior through the preset internal logic to determine whether to maintain or abort the conduction state. The confirmation instruction is a control result generated based on the comprehensive evaluation of the above model, and is used to feedback whether the current system should continue to maintain the discharge operation on the load. Its role is to establish a safe closed-loop for the conversion from "temporary conduction" to "continuous conduction".

[0027] Specifically, although a single current signal can reflect the load connection behavior, it may be affected by non-genuine load factors such as instantaneous disturbances, poor contact, and electromagnetic interference under complex operating conditions, resulting in a relatively high risk of misjudgment. Environmental information such as temperature and humidity parameters can reflect the current external physical state of the system and has a direct impact on the stability of electrical connections, the thermal tolerance of discharge devices, and the risk of arcing. If only relying on the characteristics of current mutation for judgment, the system may wrongly trigger discharge under conditions such as high temperature, humidity, or low battery power that are not suitable for conduction, increasing the probability of device burnout or arcing. Integrating environmental information with current change information can jointly judge the rationality of conduction from two dimensions of "behavior trigger" and "condition constraint", enabling the judgment system to shift from single-signal trigger to multi-dimensional information collaborative decision-making, thus significantly improving the recognition accuracy of abnormal states and the protection reliability of the system. Especially in application scenarios where light power systems are highly sensitive to cost and safety, this integration strategy can enhance the stability and safety boundary perception ability of the control system without increasing excessive hardware costs; Moreover, the purpose of setting up the fast channel and the slow channel is to balance the system response speed and the determination accuracy, and solve the differences in the processing requirements for different types of information in the battery management process. The fast channel is used to capture high-dynamic change characteristics such as current mutation and can respond to the connection behavior of external loads within milliseconds or even microseconds, which is the guarantee for the timeliness of the system to identify load changes; while the slow channel is responsible for collecting environmental parameters such as temperature and humidity that change slowly but are highly relevant to discharge safety. This type of information does not require high-frequency sampling but has key reference value for whether continuous discharge is ultimately allowed. Designing the two channels in parallel helps to form a two-layer control architecture with fast event trigger as the premise and slow environmental perception as the correction basis, enabling the system to have environmental adaptability while maintaining real-time performance, avoiding mis-triggering and enhancing the safety fault tolerance ability of discharge decision-making.

[0028] In one embodiment, as Figure 2 shown, in step S20, that is, in the step of real-time detecting current change information based on the fast channel with microsecond-level response and then judging whether there is an effective load access, it includes: S201. The fast channel based on microsecond-level response real-time detects current change information and detects whether there is a corresponding current mutation signal in the current change information; the current mutation signal refers to the phenomenon that the current amplitude undergoes a drastic jump in a very short time during the real-time acquisition of current change data by the fast channel, usually manifested as a rapid transition from a low-current state to a high-current state, or the current fluctuates beyond the set noise tolerance interval within a short period. This signal is often used as the first judgment basis for load access behavior.

[0029] S202. If a current mutation signal is detected, start a timer. Within the preset time threshold set by the timer, continue to collect current change information through the fast channel, and determine whether the current change information continuously remains within the stable current characteristic range after the current mutation signal. The preset time threshold refers to a fixed time window set by the system for confirming the persistence of the load connection. Its time scale is usually in the range of milliseconds to seconds. The purpose of setting this time is to avoid misjudging short-term interference or non-steady-state behavior as a valid load.

[0030] S203. If it remains within the stable current characteristic range after the current mutation signal, it is determined that there is a valid load connection. The stable current characteristic range refers to the range where the current data collected in a subsequent period after the current mutation occurs remains in a relatively stable numerical interval, and the fluctuation range is within the set error range, indicating that the load has established a stable connection and continues to consume power. This judgment is used to verify whether the load truly and reliably exists and prevent the system from generating misoperations due to false current change signals.

[0031] In one embodiment, as Figure 3 shown, in step S30, that is, the temporary start instruction includes a multi-stage conduction control mechanism. The multi-stage conduction control mechanism includes: S301. Execute the first-stage conduction operation. The first-stage conduction operation controls the discharge switch conducting element to conduct with a preset conduction duty cycle, so that the discharge switch conducting element is placed in a preset current-limiting conduction state. The conduction duty cycle refers to the time ratio of the discharge switch conducting element being in the conduction state within a unit time, which is a key parameter for measuring the conduction intensity or power supply intensity of this element. By setting the duty cycle, the output current size can be effectively controlled, thereby achieving soft start and current-limiting protection. The current-limiting conduction state means that the discharge switch conducting element does not output current at full power in the initial conduction stage, but controls the current flowing through the load to be within a safe and controlled range by restricting its conduction duty cycle. This state can not only verify that the load is indeed connected, but also prevent the current from suddenly increasing and impacting the battery system or the load itself.

[0032] S302. Receive the confirmation instruction in real time. The confirmation instruction is a control signal generated after the start recognition model fuses parameters such as the structured information group and the start weight, indicating that the current environment and current conditions meet the determination result of continuous conduction, and it is a necessary prerequisite for switching from the temporary conduction state to the continuous conduction state.

[0033] S303. If the confirmation instruction is received, control the start discharge switch conducting element to continue to remain conductive. S304. If the confirmation instruction is not received after exceeding the preset response time, control the discharge switch conducting element to disconnect. The preset response time refers to the maximum time window between the system issuing the temporary start instruction and waiting for the confirmation instruction to be completed. This time range is usually set according to the load type, system processing delay, and environmental safety requirements. If the confirmation action is not completed within this time, it will be automatically regarded as a risk state and trigger the disconnection control.

[0034] Specifically, in an electric bicycle system, when the user turns on the power switch, the system determines that there is an effective load connected and controls the discharge switch conducting element to conduct initially with a conduction duty cycle of 30%, so that the motor is in a light-load pre-drive state. Subsequently, the system enters the confirmation waiting stage. If the confirmation instruction generated by the start recognition model is received within 1 second, the discharge switch conducting element is increased to 100% duty cycle to enter continuous power supply; otherwise, if the confirmation instruction is not received within 1 second, the system immediately disconnects the current path to avoid the risk of sparking or overheating caused by forced conduction under abnormal load or harsh environmental conditions.

[0035] In one embodiment, as Figure 4 shown, in step S40, in the step of formatting and standardizing the environmental information and current change information to generate the corresponding structured information group, the environmental information at least includes humidity information and temperature information, and the step further includes: S401. Convert the temperature information into a relative temperature difference coefficient according to degrees Celsius. The relative temperature difference coefficient is the result obtained by standardizing the currently collected temperature information. It calculates the relative difference by comparing the current temperature value with the temperature reference or safety reference temperature set by the system, and converts it into a dimensionless coefficient with consistent dimensions and suitable for subsequent algorithm processing, which is used to characterize the degree of deviation of the current temperature from the safe operating range.

[0036] S402. Map the humidity information to a humidity safety coefficient between 0 and 1. The humidity safety coefficient is the value obtained by mapping the humidity information to an interval. Generally, the humidity is converted from its original unit (such as %RH) to a continuous variable between 0 and 1. The value close to 1 usually represents a higher humidity risk level, and the value close to 0 indicates that the environment is dry and relatively safe. This mapping method is convenient for unified participation in model input and reflects the influence intensity of humidity on the conduction decision.

[0037] S403. Calculate the corresponding short-term stable value of the current change information through a dynamic mean filtering algorithm based on a sliding window. The short-term stable value is a representative value obtained by dynamically averaging the current sampling data within a period of time using the sliding window mechanism. This value can effectively filter out high-frequency noises such as jitter and interference and reflect the steady-state characteristics of the current load current.

[0038] S404. Encode the relative temperature difference coefficient, humidity safety factor, and short-term stability value into a structured information group, which is used as an input vector to be uniformly input into the subsequent startup recognition model to participate in the conduction adjustment judgment; encoding into a structured information group means encapsulating the above-mentioned processed physical quantities into a data structure with a fixed field order and unified data format to provide a standardized input vector for the startup recognition model, enabling the model to stably receive a consistent data structure in different execution cycles and complete the subsequent scoring and judgment process.

[0039] In one embodiment, as Figure 5 shown, in step S50, that is, in the step of assigning corresponding startup weights to the structured information group according to the historical data set, it includes: S501. Analyze the parameter tags of the structured information group; the parameter tags are field markers in the structured information group used to identify the meanings of various data, used to distinguish different types of input features, such as temperature, humidity, current value, etc. They usually have predefined formats and encoding rules so that the system can accurately match the source and attributes of each data item.

[0040] S502. Call the cumulative statistical information of the corresponding fields in the historical data set based on the parameter tags; the cumulative statistical information is the statistical summary result formed by long-term operation records in the historical data set, usually including data such as the mean, extreme value, standard deviation, or frequency distribution of a certain parameter in different time periods and different working conditions, used to reflect the correlation between the parameter and specific events (such as abnormal discharge, sparking, overheating) during actual operation.

[0041] S503. According to the mapping relationship between the parameter change trend and the conduction risk level recorded in the cumulative statistical information within a specific interval, look up the table to determine the startup weight of the structured information group; the mapping relationship is the corresponding rule between the parameter change trend and the conduction risk level extracted based on the cumulative statistical information. Its form is mostly a look-up table structure or interval division rule, that is, different risk levels or control weights are assigned within different numerical intervals. This mapping method reflects the influence degree of the parameter on the system stability. Looking up the table to determine the startup weight means that the system retrieves the mapping rule table in the historical data set item by item according to the parameter tags in the structured information group, extracts the risk level corresponding to the current input value from it, and then assigns a quantified weight value to this field. This weight will be used in subsequent model calculations to determine the relative influence of this parameter in the overall judgment process.

[0042] In one embodiment, as Figure 6 shown, in step S60, that is, in the step of substituting each structured information group and the corresponding startup weight of each structured information group into the established startup recognition model to generate a corresponding confirmation instruction, it includes: S601. Substitute each structured information group and its corresponding startup weight into the established startup recognition model. The established startup recognition model includes at least a weighting sub-model, a scoring sub-model, and a determination sub-model. The startup recognition model is a multi-layer structured determination framework composed of multiple functional sub-models, aiming to quantitatively process the input structured information group and its corresponding startup weight and output a conduction decision, where each sub-model undertakes specific data processing and judgment functions.

[0043] S602. Based on the weighting sub-model, perform weighting processing on the structured information group and its corresponding startup weight to generate a corresponding startup permission. The startup recognition model is a multi-layer structured determination framework composed of multiple functional sub-models, aiming to quantitatively process the input structured information group and its corresponding startup weight and output a conduction decision, where each sub-model undertakes specific data processing and judgment functions.

[0044] S603. Based on the scoring sub-model, add up each startup permission to generate a corresponding scoring result. The function of the scoring sub-model is to aggregate all startup permissions, usually calculating a total score value by means of linear weighted summation or weighted average, etc. This score value is used to comprehensively reflect the overall evaluation result of the current system state in terms of conduction safety.

[0045] S604. Based on the determination sub-model, perform a threshold judgment on the scoring result to determine whether to generate a corresponding confirmation instruction. The determination sub-model takes the scoring result as input and performs a judgment operation according to a preset threshold value. The threshold value can be set statically or adjusted dynamically. Once the scoring result exceeds this threshold, the system will generate a confirmation instruction, indicating that the current discharge path is allowed to continue conducting, otherwise no confirmation instruction will be given to interrupt the subsequent power supply process. The confirmation instruction is the final conduction control signal of the system, used to connect the temporary conduction and the formal conduction states, and is the key output that plays a closed-loop role in the control logic.

[0046] It should be understood that the magnitudes of the sequence numbers of the steps in the above embodiments do not mean the order of execution. The execution order of each process should be determined according to its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0047] In one embodiment, a fire prevention device for a power battery module is provided. This fire prevention device for a power battery module corresponds one-to-one with the fire prevention method for a power battery module in the above embodiment. As Figure 7 shown, this fire prevention device for a power battery module includes a detection module, a judgment module, a first generation module, a second generation module, a call module, and a third generation module. The detailed description of each functional module is as follows: A detection module for detecting the working state of a power battery module, where the working state at least includes a charging state and a non-charging state; A judgment module for, if the power battery module is in the non-charging state, detecting current change information in real time based on a fast channel with microsecond-level response, and then judging whether there is an effective load access. The effective load is a load that maintains a stable current characteristic for more than a preset time threshold after a current mutation signal is detected in the current change information; A first generation module for, if there is an effective load access, generating a corresponding temporary start instruction, where the temporary start instruction is used to start the conduction of a discharge switch conducting element to supply power to the effective load; A second generation module for obtaining environmental information in real time based on a slow channel, performing format standardization processing on the environmental information and the current change information, and generating a corresponding structured information group; A calling module for calling a locally cached historical data set and assigning a corresponding start weight to the structured information group according to the historical data set; A third generation module for substituting each of the structured information groups and the start weights corresponding to the structured information groups into a established start recognition model to generate a corresponding confirmation instruction, where the confirmation instruction is used to control the discharge switch conducting element to continue to maintain conduction;

[0048] Optionally, the judgment module includes: A detection unit for detecting current change information in real time based on a fast channel with microsecond-level response and detecting whether a corresponding current mutation signal appears in the current change information; A first judgment unit for, if the current mutation signal is detected, starting a timer, collecting current change information through the fast channel within a preset time threshold set by the timer, and judging whether the current change information continuously maintains within the stable current characteristic range after the current mutation signal; A determination unit for, if it maintains within the stable current characteristic range after the current mutation signal, determining that there is an effective load access; Optionally, in the first generation module, the temporary start instruction includes a multi-stage conduction control mechanism, and the multi-stage conduction control mechanism includes: An execution unit for performing a first-stage conduction operation, where the first-stage conduction operation controls the discharge switch conducting element to conduct with a preset conduction duty ratio so that the discharge switch conducting element is placed in a preset current-limiting conduction state; A receiving unit for receiving the confirmation instruction in real time; A first control unit for, if the confirmation instruction is received, controlling the discharge switch conducting element to continue to maintain conduction; A second control unit, configured to control the discharge switch conducting element to disconnect if the confirmation instruction has not been received after exceeding a preset response time; Optionally, in the second generation module, the environmental information at least includes humidity information and temperature information, and the third generation module further includes: A conversion unit, configured to convert the temperature information into a relative temperature difference coefficient in degrees Celsius; A mapping unit, configured to map the humidity information to a humidity safety coefficient between 0 and 1; A calculation unit, configured to calculate a corresponding short-term stability value of the current change information through a dynamic mean filtering algorithm based on a sliding window; An encoding unit, configured to encode the relative temperature difference coefficient, the humidity safety coefficient, and the short-term stability value into a structured information group, and the structured information group is used as an input vector to be uniformly input into a subsequent startup recognition model to participate in conduction adjustment judgment; Optionally, the calling module includes: An analysis unit, configured to analyze parameter tags of the structured information group; A calling unit, configured to call cumulative statistical information of corresponding fields in the historical dataset based on the parameter tags; A determination unit, configured to look up a table to determine a startup weight of the structured information group according to a mapping relationship between a parameter change trend and a conduction risk level recorded in the cumulative statistical information within a specific interval; Optionally, the third generation module includes: A substitution unit, configured to substitute each structured information group and a startup weight corresponding to each structured information group into a established startup recognition model, and the established startup recognition model at least includes a weighted sub-model, a scoring sub-model, and a determination sub-model; A weighted processing unit, configured to perform weighted processing on the structured information group and the corresponding startup weight based on the weighted sub-model to generate a corresponding startup permission; A generation unit, configured to add up each startup permission based on the scoring sub-model to generate a corresponding scoring result; A second judgment unit, configured to perform a threshold judgment on the scoring result based on the determination sub-model to judge whether to generate a corresponding confirmation instruction.

[0049] For the specific limitations of a power battery module anti - sparking device, reference can be made to the limitations of a power battery module anti - sparking method in the above text, which will not be elaborated here. Each module in the above - mentioned power battery module anti - sparking device can be implemented in whole or in part through software, hardware, and their combination. The above - mentioned modules can be embedded in the processor of a computer device in hardware form or be independent of it, or be stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to each of the above - mentioned modules.

[0050] In one embodiment, a computer device is provided. The computer device can be a server, and its internal structure diagram can be as Figure 8 shown. The computer device includes a processor, a memory, a network interface, and a database connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non - volatile storage medium and an internal memory. The non - volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non - volatile storage medium. The network interface of the computer device is used to communicate with external terminals through a network connection. When the computer program is executed by the processor, it implements a power battery module anti - sparking method.

[0051] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the following steps are implemented: S10. Detect the working state of the power battery module, and the working state at least includes a charging state and a non - charging state; S20. If the power battery module is in a non - charging state, then based on a fast channel with micro - second - level response, detect the current change information in real - time, and then determine whether there is an effective load access. The effective load is a load that maintains a stable current characteristic for more than a preset time threshold after the current mutation signal is detected in the current change information; S30. If there is an effective load access, generate a corresponding temporary start instruction, and the temporary start instruction is used to start the discharge switch conducting element to conduct, so as to supply power to the effective load; S40. Based on a slow channel, obtain the environmental information in real - time, perform format standardization processing on the environmental information and the current change information, and generate a corresponding structured information group; S50. Call the locally cached historical data set, and according to the historical data set, assign a corresponding start weight to the structured information group; S60. Substitute each structured information group and the corresponding startup weight into the established startup recognition model to generate a corresponding confirmation instruction, where the confirmation instruction is used to control the startup discharge switch conducting element to continue to remain conducting; In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented: S10. Detect the working state of the power battery module, where the working state includes at least a charging state and a non-charging state; S20. If the power battery module is in a non-charging state, then based on a fast channel with microsecond-level response, detect the current change information in real time, and then determine whether there is a payload access. The payload is a load that, after the current mutation signal is detected in the current change information, still maintains a stable current characteristic for more than a preset time threshold; S30. If there is a payload access, then generate a corresponding temporary startup instruction, where the temporary startup instruction is used to start the startup discharge switch conducting element to conduct so as to supply power to the payload; S40. Based on the slow channel, obtain the environmental information in real time, perform format standardization processing on the environmental information and the current change information, and generate a corresponding structured information group; S50. Call the locally cached historical data set, and according to the historical data set, assign a corresponding startup weight to the structured information group; S60. Substitute each structured information group and the corresponding startup weight into the established startup recognition model to generate a corresponding confirmation instruction, where the confirmation instruction is used to control the startup discharge switch conducting element to continue to remain conducting; Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium used in the embodiments provided in the present application can include non-volatile and / or volatile memories. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or an external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), etc.

[0052] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the above division of each functional unit and module is used as an example. In actual applications, the above functions can be allocated to different functional units and modules according to needs, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0053] The above embodiments are only used to illustrate the technical solutions of the present application, not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A method for preventing ignition of a power battery module, characterized in that, The method for preventing ignition of a power battery module includes: Detect the working state of the power battery module, where the working state at least includes a charging state and a non-charging state; If the power battery module is in the non-charging state, then based on a fast channel with microsecond-level response, detect the current change information in real time, and then determine whether there is an effective load connected. The effective load is a load that maintains a stable current characteristic for more than a preset time threshold after a current mutation signal is detected in the current change information; If there is an effective load connected, generate a corresponding temporary start instruction, where the temporary start instruction is used to start the conduction of the discharge switch conducting element to supply power to the effective load; Based on a slow channel, obtain the environmental information in real time, perform format standardization processing on the environmental information and the current change information, and generate a corresponding structured information group; Call the historical data set cached locally, and assign a corresponding start weight to the structured information group according to the historical data set; Substitute each structured information group and the start weight corresponding to each structured information group into the established start recognition model to generate a corresponding confirmation instruction, where the confirmation instruction is used to control the discharge switch conducting element to continue to maintain conduction; 2. The method for preventing ignition of a power battery module according to claim 1, wherein In the step of detecting the current change information in real time based on a fast channel with microsecond-level response and then determining whether there is an effective load connected, it includes: Detect the current change information in real time based on a fast channel with microsecond-level response, and detect whether a corresponding current mutation signal appears in the current change information; If the current mutation signal is detected, start a timer, and continue to collect the current change information through the fast channel within the preset time threshold set by the timer, and determine whether the current change information continuously maintains within the stable current characteristic range after the current mutation signal; If it maintains within the stable current characteristic range after the current mutation signal, it is determined that there is an effective load connected.

3. A method for preventing ignition of a power battery module according to claim 1, characterized in that, The temporary start instruction includes a multi-stage conduction control mechanism, and the multi-stage conduction control mechanism includes: Execute the first-stage conduction operation, where the first-stage conduction operation controls the discharge switch conducting element to conduct with a preset conduction duty ratio, so that the discharge switch conducting element is in a preset current-limiting conduction state; Receive the confirmation instruction in real time; If the confirmation instruction is received, control the discharge switch conducting element to continue to maintain conduction; If the confirmation instruction is not received after exceeding the preset response time, control the discharge switch conducting element to disconnect.

4. A method for preventing ignition of a power battery module according to claim 1, characterized in that, In the step of performing format standardization processing on the environmental information and the current change information to generate a corresponding structured information group, the environmental information at least includes humidity information and temperature information, and the step further includes: Convert the temperature information into a relative temperature difference coefficient in degrees Celsius; Map the humidity information to a humidity safety factor between 0 and 1; Calculate the corresponding short-term stable value of the current change information through a dynamic mean filtering algorithm based on a sliding window; Encode the relative temperature difference coefficient, the humidity safety coefficient, and the short-term stability value into a structured information group, which is used as an input vector to be uniformly input into the subsequent start recognition model to participate in the conduction adjustment judgment.

5. A method for preventing ignition of a power battery module according to claim 1, characterized in that, In the step of assigning corresponding start weights to the structured information group according to the historical data set, it includes: Analyze the parameter tags of the structured information group; Based on the parameter tags, call the cumulative statistical information of the corresponding fields in the historical data set; According to the mapping relationship between the parameter change trend and the conduction risk level in the cumulative statistical information, look up the table to determine the start weight of the structured information group.

6. A method for preventing ignition of a power battery module according to claim 1, characterized in that, In the step of substituting each structured information group and the corresponding start weight of each structured information group into the established start recognition model to generate a corresponding confirmation instruction, it includes: Substitute each structured information group and the corresponding start weight of each structured information group into the established start recognition model, and the established start recognition model at least includes a weighted sub-model, a scoring sub-model, and a determination sub-model; Based on the weighted sub-model, perform weighted processing on the structured information group and the corresponding start weight to generate a corresponding start permission; Based on the scoring sub-model, add up each start permission to generate a corresponding scoring result; Based on the determination sub-model, perform a threshold judgment on the scoring result to determine whether to generate a corresponding confirmation instruction.

7. A kind of anti-sparking device for power battery modules, characterized in that, The anti-spark device for a power battery module includes: A detection module for detecting the working state of the power battery module, and the working state at least includes a charging state and a non-charging state; A judgment module for, if the power battery module is in the non-charging state, detecting current change information in real time based on a fast channel with a microsecond-level response, and then judging whether there is an effective load connected. The effective load is a load that maintains a stable current characteristic for more than a preset time threshold after the current mutation signal is detected in the current change information; A first generation module for, if there is an effective load connected, generating a corresponding temporary start instruction, and the temporary start instruction is used to start the conduction of the discharge switch conducting element to supply power to the effective load; A second generation module for obtaining environmental information in real time based on a slow channel, performing format standardization processing on the environmental information and the current change information, and generating a corresponding structured information group; A call module for calling the historical data set cached locally, and assigning corresponding start weights to the structured information group according to the historical data set; A third generation module for substituting each structured information group and the corresponding start weight of each structured information group into the established start recognition model to generate a corresponding confirmation instruction, and the confirmation instruction is used to control the discharge switch conducting element to continue to conduct.

8. The anti-sparking device for a power battery module according to claim 7, wherein, The judgment module includes: A detection unit for detecting current change information in real time based on a fast channel with a microsecond-level response, and detecting whether a corresponding current mutation signal appears in the current change information; The first judgment unit is configured to, if the current mutation signal is detected, start a timer, continue to collect current change information through the fast channel within a preset time threshold set by the timer, and judge whether the current change information continuously maintains within the stable current characteristic range after the current mutation signal; The determination unit is configured to, if it maintains within the stable current characteristic range after the current mutation signal, determine that there is a payload access.

9. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, the steps of a method for preventing ignition of a power battery module according to any one of claims 1 to 6 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, the steps of a method for preventing ignition of a power battery module according to any one of claims 1 to 6 are implemented.

Images