Vehicle battery feed risk identification method and device

Abstract

The invention discloses a vehicle battery feed risk identification method and device. The method relates to the field of new energy vehicles, and comprises the following steps: when it is detected that a target vehicle is in a standing state, obtaining an accumulated standing time length when the target vehicle is in the standing state; obtaining the total discharge rate of the vehicle battery of the target vehicle in the target time period; the actual SOC capacity of the vehicle battery in the standing state is determined; based on the total discharge rate and the actual SOC capacity, the residual discharge duration of the vehicle battery is determined; and determining a feed risk identification result of the vehicle battery based on the residual discharge duration and the accumulated standing duration. According to the method and the device, the technical problem that the accuracy and the efficiency of feed risk prediction are relatively low when the feed risk of the vehicle battery is identified in the vehicle standing state in the related technology is solved.

Description

Technical Field

[0001] The present invention relates to the field of new energy vehicles, and in particular to a method and device for identifying vehicle battery feeding risks. Background Art

[0002] With the rapid development and popularization of new energy vehicles, battery management has become a key issue. In particular, the production and sales of new energy vehicles are constantly increasing, and more and more new energy vehicles are shelved for a long time. Due to the self-discharge of the battery and the consumption of the on-board system during the long-term static process, the battery cell power supply is abnormal, and the positive and negative electrodes and electrolyte components inside the battery cell undergo chemical reactions, which leads to an irreversible decrease in the battery cell capacity. In addition, the detection of static state and the power supply warning often rely on complex model predictions, lacking direct and efficient calculation methods. The warning results require a lot of manpower to review and verify, resulting in low efficiency and accuracy in identifying the power supply risk of the vehicle battery when the vehicle is static.

[0003] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0004] The embodiments of the present invention provide a method and device for identifying the risk of power feeding of a vehicle battery, so as to at least solve the technical problem in the related art of identifying the risk of power feeding of a vehicle battery when the vehicle is stationary, and the low accuracy and efficiency of power feeding risk prediction.

[0005] According to one aspect of an embodiment of the present invention, a method for identifying the risk of power feeding of a vehicle battery is provided, comprising: obtaining the cumulative static time of the target vehicle in the static state when detecting that the target vehicle is in a static state; obtaining the total discharge rate of the vehicle battery of the target vehicle in the target time period; determining the actual SOC capacity of the vehicle battery in the static state; determining the remaining discharge time of the vehicle battery based on the total discharge rate and the actual SOC capacity; and determining the power feeding risk identification result of the vehicle battery based on the remaining discharge time and the cumulative static time. The present invention solves the technical problem in the related art that the accuracy and efficiency of power feeding risk prediction are low when identifying the risk of power feeding of a vehicle battery when the vehicle is in a static state.

[0006] Through the above method, the limitations of static state detection and power supply warning relying on complex model prediction and the warning results requiring a large amount of manpower to review and verify can be effectively solved, providing a more efficient and accurate method for identifying static power supply risks of vehicle batteries.

[0007] Optionally, in the case where it is detected that the target vehicle is in a stationary state, before obtaining the cumulative stationary time of the target vehicle in the stationary state, the method also includes: obtaining multiple frames of operating data of the target vehicle collected in the current time period, wherein the multiple frames of operating data correspond to different sampling moments, and each frame of operating data at least includes: a voltage array, a charging current, a discharging current, a vehicle mileage, and a vehicle speed of the target vehicle at the corresponding moment, wherein the voltage array includes voltage values ​​corresponding to multiple single cells in the vehicle battery; detecting whether the multiple frames of operating data meet predetermined operating conditions; and determining that the target vehicle is in the stationary state when the number of operating data frames that meet the predetermined operating conditions in the multiple frames of operating data is greater than a predetermined number of consecutive frames.

[0008] Through the above method, false static states caused by misoperation or signal interference can be effectively filtered out, avoiding misjudgment of a single data point, and ensuring the reliability of static state judgment.

[0009] Optionally, the detecting whether the multiple frames of operating data meet the predetermined operating conditions includes: detecting whether any frame of operating data in the multiple frames of operating data meets the predetermined operating conditions in the following manner: based on the voltage data of the target vehicle at any moment, detecting whether the voltage change rate of the vehicle battery is less than a preset change rate threshold, whether there is a charging current for the target vehicle at any moment, whether the discharge current of the target vehicle at any moment is less than a preset current, whether the vehicle mileage of the target vehicle at any moment changes, and whether the vehicle speed of the target vehicle at any moment is less than a preset speed, to obtain a detection result, wherein the voltage change rate includes the overall voltage change rate of the vehicle battery and the single cell voltage change rate of each single cell in the vehicle battery; determining whether any frame of operating data meets the predetermined operating conditions based on the detection result; detecting whether the multiple frames of operating data meet the predetermined operating conditions by detecting whether any frame of operating data meets the predetermined operating conditions.

[0010] Through the above method, whether the multi-frame operating data meets the predetermined operating conditions is evaluated from multiple dimensions, which can further confirm whether the vehicle is actually in a stationary state and avoid misidentification of the vehicle's stationary state.

[0011] Optionally, determining whether any frame operating data satisfies the predetermined operating condition based on the detection result includes: when the voltage change rate is less than the preset change rate threshold, and the target vehicle has no charging current at any time, and the discharge current is less than the preset current, and the vehicle mileage has not changed, and the vehicle speed is less than the preset speed, determining that any frame operating data satisfies the predetermined operating condition; or when the voltage change rate is greater than or equal to the preset change rate threshold, or the target vehicle has a charging current at any time, or the discharge current is greater than or equal to the preset current, or the vehicle mileage changes, or the vehicle speed is greater than or equal to the preset speed, determining that any frame operating data does not satisfy the predetermined operating condition.

[0012] In the above manner, the judgment logic requires that multiple conditions be met simultaneously, which can ensure the consistency and validity of the data when making static state judgments and avoid making erroneous conclusions about the vehicle state due to accidental fluctuations in individual parameters.

[0013] Optionally, when the number of frames of operating data that meet the predetermined operating condition in the multiple frames of operating data is greater than a predetermined number of consecutive frames, determining that the target vehicle is in the stationary state includes: when the number of frames of operating data that meet the predetermined operating condition in the multiple frames of operating data is greater than the predetermined number of consecutive frames, obtaining a target time corresponding to the first frame of operating data that meets the predetermined operating condition in the multiple frames of operating data, and an end time of the current time period; determining an initial stationary time based on the target time and the end time; obtaining multiple frames of first feature data of the target vehicle collected in the current time period, wherein In the method, the multiple frames of first feature data correspond to different sampling moments, and each frame of feature data is used to indicate the running characteristics of the target vehicle at the corresponding moment; based on the multiple frames of first feature data, a long short-term memory network model is adopted to obtain a first predicted state and a first predicted stationary time of the target vehicle, wherein the first predicted state is used to indicate whether the target vehicle is in a stationary state during a period corresponding to the initial stationary time; when the first predicted state is the stationary state, and the difference between the initial stationary time and the first predicted stationary time is less than a preset difference threshold, it is determined that the target vehicle is in the stationary state.

[0014] Through the above method, when it is determined in the real-time calculation that the target vehicle is in a stationary state, a request is issued to call the long short-term memory network model to make an accuracy judgment on the stationary state. If the long short-term memory network model returns a result showing that the target vehicle does not meet the stationary condition, the stationary state count will be reset to zero in the real-time calculation, and the stationary state judgment will be made again to prevent the real-time calculation from making an error in the stationary state judgment due to the inability to process the reissued data.

[0015] Optionally, obtaining the total discharge rate of the vehicle battery of the target vehicle within the target time period includes: obtaining the current discharge rate of the vehicle battery during the target time period; obtaining the self-discharge rate of the vehicle battery during the target time period; and obtaining the total discharge rate based on the current discharge rate and the self-discharge rate.

[0016] Through the above method, the total discharge rate is calculated by combining the current discharge rate and the self-discharge rate, which can comprehensively evaluate the energy loss rate of the battery in a static state, providing a more comprehensive perspective on energy consumption.

[0017] Optionally, determining the actual SOC capacity of the vehicle battery in the stationary state includes: obtaining a voltage array of the vehicle battery collected in the stationary state, wherein the voltage array includes voltage values ​​of a plurality of single cells included in the vehicle battery in the stationary state; identifying a maximum voltage value and a minimum voltage value in the voltage array; based on a corresponding relationship between voltage values ​​and SOC capacities, determining a maximum SOC capacity corresponding to the maximum voltage value, and a minimum SOC capacity corresponding to the minimum voltage value; and determining the actual SOC capacity based on the maximum SOC capacity and the minimum SOC capacity.

[0018] In this way, a SOC value that is closer to the actual state of the battery can be provided. Using the maximum and minimum voltage values ​​in the voltage array to calculate the SOC capacity can avoid overall SOC evaluation errors caused by abnormal data of a single cell.

[0019] Optionally, determining the power feeding risk identification result of the vehicle battery based on the remaining discharge time and the cumulative stationary time includes: determining a target time range to which the remaining discharge time belongs from multiple time ranges, wherein the multiple time ranges are determined based on different proportions of the cumulative stationary time; determining a target power feeding risk level corresponding to the target time range from multiple power feeding risk levels, wherein the multiple power feeding risk levels correspond one-to-one to the multiple time ranges; and determining the power feeding risk identification result based on the target power feeding risk level.

[0020] Through the above method, the relationship between the remaining discharge time and the accumulated static time is converted into different risk levels, realizing the quantitative representation of risk. By dividing multiple time ranges and corresponding risk levels, a multi-level early warning mechanism can be realized.

[0021] Optionally, the method also includes: when the target vehicle is in the stationary state, obtaining multiple frames of second feature data of the target vehicle collected in a first time period according to a preset timing task, wherein the multiple frames of second feature data correspond to different sampling moments, and each frame of feature data is used to indicate the operating characteristics of the target vehicle at the corresponding moment, and the first time period is the time period after the current time period; based on the multiple frames of second feature data, a long short-term memory network model is used to obtain a second predicted state of the target vehicle and a second predicted stationary time, wherein the second predicted state is used to indicate whether the target vehicle is in a stationary state during the first time period; when the second predicted state is the stationary state, and the difference between the cumulative stationary time and the second predicted stationary time is less than a preset difference threshold, the cumulative stationary time is corrected to obtain a corrected cumulative stationary time.

[0022] In the above manner, the static time of a vehicle is determined by accumulating the static time in real time, and the long short-term memory network model is called every certain period of time to correct the static time to ensure the accuracy of the static time. This prevents the real-time computing thread from being stuck when requesting to call the model because the model calling the application interface is disconnected or reports an error, and the result is not returned for a long time, which greatly ensures the availability of the model calling service.

[0023] According to another aspect of an embodiment of the present invention, a vehicle battery power feeding risk identification device is also provided, including: when a target vehicle is detected to be in a stationary state, obtaining the cumulative stationary time of the target vehicle in the stationary state; obtaining the total discharge rate of the vehicle battery of the target vehicle within a target period; determining the actual SOC capacity of the vehicle battery in the stationary state; based on the total discharge rate and the actual SOC capacity, determining the remaining discharge time of the vehicle battery; based on the remaining discharge time and the cumulative stationary time, determining the power feeding risk identification result of the vehicle battery.

[0024] In an embodiment of the present invention, by detecting that a target vehicle is in a stationary state, obtaining the cumulative stationary time of the target vehicle in the stationary state; obtaining the total discharge rate of the vehicle battery of the target vehicle within the target time period; determining the actual SOC capacity of the vehicle battery in the stationary state; determining the remaining discharge time of the vehicle battery based on the total discharge rate and the actual SOC capacity; and determining the power feeding risk identification result of the vehicle battery based on the remaining discharge time and the cumulative stationary time, the purpose of accurately identifying the power feeding risk of the vehicle battery dynamically based on the remaining discharge time and the cumulative stationary time of the vehicle battery obtained in real time is achieved, thereby achieving the technical effect of improving the efficiency and accuracy of power feeding risk identification of the vehicle battery when the vehicle is in a stationary state, and further solving the technical problem of low accuracy and efficiency of power feeding risk prediction in the related art for power feeding risk identification of the vehicle battery when the vehicle is in a stationary state. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0026] Figure 1 is a flow chart of a vehicle battery feeding risk identification method according to an embodiment of the present invention;

[0027] Figure 2 is a flow chart of an optional vehicle battery feeding risk identification method according to an embodiment of the present invention;

[0028] Figure 3 is a schematic diagram of a vehicle battery feeding risk identification device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0029] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0030] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0031] First, to facilitate understanding of the embodiments of the present invention, some terms or nouns involved in the present invention are explained below:

[0032] Static Power Feed Risk (SFR) refers to the phenomenon in the power system that the feeder line continues to supply power when there is no load or the load is very small. This phenomenon may lead to some risks, such as: equipment overheating, power waste, voltage fluctuation, equipment damage, safety risks, etc.

[0033] Battery Cell refers to the basic unit that makes up a battery and is the part of the battery that stores electrical energy. A battery cell can be a single battery cell or a combination of multiple battery cells. There are many types of battery cells, including lithium-ion batteries, nickel-metal hydride batteries, lead-acid batteries, etc. They differ in voltage, capacity, shape, and size. The battery cell is the core component of the battery system, and its performance directly affects the overall performance of the battery.

[0034] The Long Short-Term Memory (LSTM) model is a special type of Recurrent Neural Network (RNN), which is mainly used to solve the gradient vanishing and gradient exploding problems of traditional RNN when processing long sequence data. LSTM controls the flow of information by introducing a gating mechanism, so that it can learn long-term dependencies.

[0035] Battery Management System (BMS) is an electronic system used to monitor and manage battery cells or battery packs. Its main functions include: battery status monitoring, battery status evaluation, battery balancing, safety protection, data reporting and recording, etc.

[0036] Microservices Cluster refers to the application of microservice architecture in a cluster environment, which means deploying multiple microservice instances on a group of servers to form a highly scalable, maintainable and fault-tolerant system. This architecture can improve the flexibility and scalability of the system while reducing the impact of a single service failure on the entire system.

[0037] Application Programming Interface (API) is a set of predefined functions, protocols and tools for building software and applications. API allows different software systems to interact with each other, enabling developers to access the functions or data of a service or application.

[0038] The battery state of charge (SOC) is a percentage value that represents the ratio of the current remaining power of the battery to the total capacity of the battery. The SOC value usually ranges from 0% (fully discharged) to 100% (fully charged).

[0039] According to an embodiment of the present invention, a method embodiment for identifying vehicle battery feeding risks is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0040] Figure 1 is a flow chart of a vehicle battery feeding risk identification method according to an embodiment of the present invention. Figure 1 As shown, the method comprises the following steps:

[0041] Step S102, when it is detected that the target vehicle is in a stationary state, obtaining the cumulative stationary time of the target vehicle in the stationary state;

[0042] Optionally, the target vehicle can be an electric vehicle. It can be, but is not limited to, timely identifying whether the vehicle has stopped running and maintained a static state by real-time monitoring of key indicators such as the battery voltage, current, speed and mileage of the vehicle, which provides a basis for subsequent power supply risk assessment and ensures that the vehicle is in a static state; the cumulative static time can reflect the length of time the vehicle has been in a static state. The vehicle battery will gradually lose power due to self-discharge and vehicle system consumption in a static state. The longer the cumulative static time, the higher the possibility of battery power supply; by obtaining the cumulative static time, a direct basis can be provided for risk quantification. By monitoring and accumulating the static time, the energy loss of the vehicle battery in a static state can be more accurately calculated, and then the remaining battery discharge time can be predicted, providing quantitative indicators for early warning of power supply risks.

[0043] In an optional embodiment, when it is detected that the target vehicle is in a stationary state, before obtaining the cumulative stationary time of the target vehicle in the stationary state, the method also includes: obtaining multiple frames of operating data of the target vehicle collected in the current time period, wherein the multiple frames of operating data correspond to different sampling moments, and each frame of operating data at least includes: a voltage array, a charging current, a discharging current, a vehicle mileage, and a vehicle speed of the target vehicle at the corresponding moment, wherein the voltage array includes voltage values ​​corresponding to multiple single cells in the vehicle battery; detecting whether the multiple frames of operating data meet predetermined operating conditions; and determining that the target vehicle is in a stationary state when the number of operating data frames that meet the predetermined operating conditions in the multiple frames of operating data is greater than a predetermined number of consecutive frames.

[0044] Optionally, by detecting whether multiple frames of operating data meet predetermined operating conditions, such as the vehicle speed is 0, the mileage has not changed, the charging current and the discharging current are within specific thresholds, etc., it is possible to more accurately determine whether the vehicle is truly in a stationary state, rather than a short stop or a temporary abnormality in a certain data. In the multiple frames of operating data, if the number of frames that meet the stationary conditions is greater than the predetermined number of consecutive frames, it indicates that the vehicle has been stationary for a period of time, rather than being stationary accidentally or briefly. By setting a threshold for the number of consecutive frames, false stationary states caused by misoperation or signal interference can be effectively filtered out, avoiding misjudgment of a single data point, and ensuring the reliability of the stationary state judgment.

[0045] Optionally, after obtaining the multi-frame operation data, the multi-frame operation data can be preprocessed first. Specifically, the real-time vehicle data (i.e., multi-frame operation data) can be received from the Kafka server. The real-time data collected by the vehicle from the battery management system (BMS) is sent to the monitoring server of the vehicle factory through the vehicle-borne terminal (Telematics Box, TBOX) of the vehicle. The vehicle factory packages and pushes the data to the Kafka server for consumption by the big data platform for data analysis. The system cleans and regularizes the vehicle data obtained from the big data platform. The main steps include: cleaning and replacing the default values ​​and basic abnormal values ​​(such as 0, 255, 65535, etc.); effectively filling the missing values; identifying and replacing the abnormal data caused by the abnormal acquisition module; filtering the duplicate data; evaluating the validity of the data; shunting the vehicle data of different cells in the vehicle battery, etc. This process ensures the authenticity and validity of the data and can eliminate a large number of false alarms caused by data problems.

[0046] In an optional embodiment, detecting whether multiple frames of operating data meet predetermined operating conditions includes: detecting whether any frame of operating data in the multiple frames of operating data meets the predetermined operating conditions in the following manner: based on the voltage data of the target vehicle at any time, detecting whether the voltage change rate of the vehicle battery is less than a preset change rate threshold, whether there is a charging current at any time for the target vehicle, whether the discharge current of the target vehicle at any time is less than a preset current, detecting whether the vehicle mileage of the target vehicle at any time changes, and whether the vehicle speed of the target vehicle at any time is less than a preset speed, and obtaining a detection result, wherein the voltage change rate includes the overall voltage change rate of the vehicle battery and the single cell voltage change rate of each single cell in the vehicle battery; determining whether any frame of operating data meets the predetermined operating conditions based on the detection result; detecting whether multiple frames of operating data meet the predetermined operating conditions by detecting whether any frame of operating data meets the predetermined operating conditions.

[0047] Optionally, it is possible to effectively determine whether the vehicle is truly in a stationary state by detecting whether the voltage change rate of the vehicle battery at any time is less than a preset change rate threshold. A low voltage change rate indicates that the battery voltage is relatively stable in a short period of time, and the battery voltage will change significantly during dynamic driving or charging. At the same time, by detecting the overall voltage change rate and the single cell voltage change rate of each single cell respectively, the stability of the battery pack can be comprehensively evaluated to avoid overall misjudgment due to local abnormalities; checking whether the target vehicle has a charging current at any time and whether the discharge current is less than the preset current can exclude the situation where the vehicle is in the charging process or has a large discharge current (non-stationary feeding). The selection of the preset current takes into account the common system power consumption of the vehicle in a stationary state, such as the standby consumption of electronic equipment, etc., to avoid misjudging these normal consumption as non-stationary states; detecting whether the mileage of the target vehicle at any time changes and whether the vehicle speed is less than the preset speed can preliminarily infer whether the vehicle is in a stationary state. Combined with the voltage and current detection results, the evaluation of whether the multi-frame operation data meets the predetermined operation conditions from multiple dimensions can further confirm whether the vehicle is truly in a stationary state and avoid misidentification of the vehicle's stationary state.

[0048] Optionally, it can be based on the voltage array {V0,V1,V2,V3,…,V n}, through the state management mapWithState method of the real-time computing framework, the upper and lower frame data are analyzed to determine whether the voltage change trend of each single cell and the overall battery of the vehicle battery is smooth, and whether the voltage change rate has experienced large fluctuations. For example, if the voltage change rate is greater than the preset change rate threshold, it indicates that a large fluctuation has occurred, otherwise it indicates that the voltage change rate has not experienced a large fluctuation. Among them, the overall voltage change rate is used to indicate the overall voltage change of the voltage data of the vehicle battery at any moment relative to the voltage data at the previous moment; the single cell voltage change rate is used to indicate the change of the voltage value of each single cell at any moment relative to the voltage value at the previous moment.

[0049] Optionally, the real-time computing framework of the real-time data platform can be used to monitor whether the vehicle meets the stationary conditions. Through this computing framework, the millisecond-level real-time detection of vehicles that meet the stationary conditions can be achieved, and the changes in the discharge of the vehicle battery can be detected in real time, which provides better accuracy for calculating the remaining discharge time of the stationary vehicle. In addition, the use of real-time computing can timely improve the risk level of the vehicle's stationary power supply. This timeliness of real-time computing cannot be provided by the offline computing framework. By using the state management in the real-time computing framework, the state counting of multiple frames of running data can be performed to ensure the accuracy of the real-time calculation results; in order to greatly reduce or eliminate the human verification of the warning results, offline computing can be further used to pre-process the vehicle's historical data. By training a long short-term memory network (Long Short-Term Memory, LSTM) model suitable for time series analysis, the model is used to check and match the vehicle's stationary time, ensuring the accuracy of the vehicle's stationary state judgment and vehicle stationary time calculation, and can also fill the shortcoming that the real-time computing framework is not suitable for processing vehicle reissue data.

[0050] In an optional embodiment, based on the detection results, determining whether any frame of operating data meets the predetermined operating condition includes: when the voltage change rate is less than a preset change rate threshold, and there is no charging current for the target vehicle at any time, and the discharge current is less than the preset current, and the vehicle mileage has not changed, and the vehicle speed is less than the preset speed, determining that any frame of operating data meets the predetermined operating condition; or when the voltage change rate is greater than or equal to the preset change rate threshold, or there is a charging current for the target vehicle at any time, or the discharge current is greater than or equal to the preset current, or the vehicle mileage changes, or the vehicle speed is greater than or equal to the preset speed, determining that any frame of operating data does not meet the predetermined operating condition.

[0051] Optionally, by setting specific conditions (voltage change rate less than a preset threshold, no charging current, discharge current less than a preset threshold, mileage unchanged, driving speed less than a preset threshold), it is possible to more accurately determine whether the vehicle is in a stationary state, avoid misjudgment due to abnormal fluctuations in a single indicator, and improve the accuracy of battery feeding risk identification. The judgment logic requires that multiple conditions be met at the same time, which can ensure the consistency and validity of the data when judging the stationary state, and avoid making incorrect conclusions about the vehicle state due to accidental fluctuations in individual parameters. By comprehensively considering multiple parameters, it is ensured that any key factors that may affect the health of the battery are not ignored when judging the stationary state, thereby reducing false alarms and omissions and improving the reliability of early warnings. When any frame of operating data meets the predetermined stationary operating conditions, the data will be used for subsequent total discharge rate calculations, actual SOC capacity determination, remaining discharge time prediction, and formulation of feeding risk identification results, thereby ensuring that the entire process is based on accurate and consistent data and improving the credibility of the results.

[0052] Optionally, a real-time computing framework can be used to determine the static state and accumulate the static time of the vehicle real-time data. This includes the voltage array {V0, V1, V2, V3, …, V n Specifically, the voltage array {V0, V1, V2, V3, ..., V n}Use the state management mapWithState method of the real-time computing framework to analyze the upper and lower frame data to determine whether the change trend of each single cell and the whole is smooth, whether the change rate has fluctuated greatly, etc.; at the same time, the vehicle's charging and discharging current, vehicle mileage and vehicle speed must be monitored. The judgment of the static state must first meet the requirements that the vehicle's mileage has not changed for a long time and the vehicle speed is 0km / h; secondly, it is necessary to ensure that the vehicle has no charging current, and its discharge current is 0mA or continuously less than 8000mA (the discharge current threshold of the static conditions of different models may be slightly different). Finally, if all the above conditions are met, that is, the frame operation data meets the predetermined operation conditions, it is necessary to count and accumulate the vehicles that meet the predetermined operation conditions. If the counting threshold is met, it is judged that the vehicle has entered the static state from the start of counting, and the static time is accumulated; if there is an interruption in the counting accumulation process, it is started to re-judge whether it meets the static state.

[0053] In an optional embodiment, when the number of frames of operating data that meet the predetermined operating conditions in multiple frames of operating data is greater than the predetermined number of consecutive frames, it is determined that the target vehicle is in a stationary state, including: when the number of frames of operating data that meet the predetermined operating conditions in multiple frames of operating data is greater than the predetermined number of consecutive frames, obtaining the target time corresponding to the first frame of operating data that meets the predetermined operating conditions in the multiple frames of operating data, and the end time of the current time period; determining the initial stationary time based on the target time and the end time; obtaining multiple frames of first feature data of the target vehicle collected in the current time period, wherein the multiple frames of first feature data correspond to different sampling times, and each frame of feature data is used to indicate the operating characteristics of the target vehicle at the corresponding time; based on the multiple frames of first feature data, using a long short-term memory network model, obtain a first predicted state of the target vehicle and a first predicted stationary time, wherein the first predicted state is used to indicate whether the target vehicle is in a stationary state within the time period corresponding to the initial stationary time; when the first predicted state is a stationary state, and the difference between the initial stationary time and the first predicted stationary time is less than a preset difference threshold, it is determined that the target vehicle is in a stationary state.

[0054] Optionally, when it is determined in real-time calculation that the target vehicle is in a stationary state, a request is issued to call the long short-term memory network model to make an accuracy judgment on the stationary state. If the result returned by the long short-term memory network model shows that the target vehicle does not meet the stationary condition, the stationary state count will be reset to zero in the real-time calculation, and the stationary state judgment will be made again to prevent the real-time calculation from making an incorrect judgment on the stationary state due to the inability to process the reissued data.

[0055] Optionally, the long short-term memory network model can be obtained through machine learning based on the operating characteristic data of the target vehicle in a stationary state and the operating characteristic data of the target vehicle in a non-stationary state collected during historical periods.

[0056] Optionally, the LSTM model can be used based on offline calculation to verify the static time and static state of the vehicle. The state management in the real-time calculation framework needs to ensure that the data is time-series ordered, and it is impossible to process the reissued data that the vehicle TBOX has not sent in time, which may cause the static state judgment obtained by real-time calculation to fail or the calculated static time to be inaccurate. In the above method, an LSTM model suitable for processing time series information can be added to verify the static state and static time of the vehicle. The specific implementation includes: training data preprocessing, training LSTM model, calling the model in the production environment, and optimizing the model based on the results.

[0057] Optionally, the LSTM model can be obtained through machine learning based on the running characteristic data of the target vehicle in a stationary state collected in historical periods, and the running characteristic data of the target vehicle in a non-stationary state. In other words, the running characteristic data of the target vehicle in a stationary state collected in historical periods, and the running characteristic data of the target vehicle in a non-stationary state can be used as training data, and the training data can be preprocessed, and the LSTM model can be obtained through machine learning based on the preprocessed training data. The LSTM model can be implemented in the form of microservices. Specifically, the preprocessing of the training data requires selecting the data of the stationary vehicle and the data of the non-stationary state before and after the stationary vehicle identified by real-time calculation, performing feature selection, retaining the characteristic values ​​(i.e., characteristic data) such as the voltage array, temperature array, mileage, vehicle speed, and charging and discharging current, and calculating the rate of change of the voltage array and adding it to the characteristic value, normalizing these characteristic values ​​as the input of the model, taking the stationary time as the output, selecting a part of the non-stationary vehicle data to add to the training data, and the stationary time of this part of the vehicle is 0; using these data as input and output to train the LSTM model, by adjusting the model parameters such as the model layer, the number of units in each layer, the model learning rate, the batch size, and the time step, to ensure that the performance indicators of the trained LSTM model are qualified and the model results are relatively accurate, and the generalization ability of the model can be verified by the validation set data to check whether it is underfitting or overfitting; after training a suitable model, the model can be called in the production environment to ensure the success of the model call, and the implementation method packages the model call method into a microservice cluster. When the target vehicle is judged to be stationary in real-time calculation, a request is issued to call the microservice to make an accuracy judgment on the stationary state. If the result returned by the microservice call shows that the target vehicle does not meet the stationary condition, the stationary state count will be reset to zero in the real-time calculation, and the stationary state judgment will be made again to prevent the real-time calculation from making an error in the stationary state judgment due to the inability to process the reissued data.

[0058] In an optional embodiment, the method also includes: when the target vehicle is in a stationary state, obtaining multiple frames of second feature data of the target vehicle collected in a first time period according to a preset timing task, wherein the multiple frames of second feature data correspond to different sampling moments, and each frame of feature data is used to indicate the operating characteristics of the target vehicle at the corresponding moment, and the first time period is the time period after the current time period; based on the multiple frames of second feature data, a long short-term memory network model is used to obtain a second predicted state of the target vehicle and a second predicted stationary time, wherein the second predicted state is used to indicate whether the target vehicle is in a stationary state during the first time period; when the second predicted state is a stationary state, and the difference between the cumulative stationary time and the second predicted stationary time is less than a preset difference threshold, the cumulative stationary time is corrected to obtain a corrected cumulative stationary time.

[0059] Optionally, the long short-term memory network model is called every certain period of time to determine the static time of a vehicle through the static time accumulation in real-time calculation, and the static time is corrected to ensure the accuracy of the static time. This prevents the real-time processing thread from being stuck when requesting to call the model because the model calls the Application Programming Interface (API) offline or reports an error and does not return results for a long time. This can also ensure the availability of the model calling service.

[0060] Optionally, when the second predicted state is a non-stationary state, it may also be determined that the target vehicle ends the stationary state.

[0061] Optionally, the preset timed task can be two hours, that is, the microservice is called once every two hours to verify the static state and cumulative static time of the vehicle through the LSTM model.

[0062] Optionally, when the LSTM model is implemented in the form of a microservice, the static time of a vehicle can be determined by accumulating the static time in real-time calculation, and the microservice can be called every two hours to correct the static time to ensure the accuracy of the static time. The purpose of using a microservice cluster is to prevent the real-time computing thread from being stuck when requesting to call the model because the model call API is disconnected or reports an error, and the result is not returned for a long time, which can also ensure the availability of the model call service. After a period of operation, the static results of the real-time calculation and model calculation can be verified and counted, and the verified and corrected data can be used to continuously optimize the LSTM model, which can effectively improve the accuracy of the model results and ensure that the model can meet the current vehicle conditions.

[0063] Step S104, obtaining a total discharge rate of a vehicle battery of a target vehicle within a target period of time;

[0064] Optionally, the total discharge rate can be used to directly reflect the average rate of battery energy loss during the target period. By calculating the total discharge rate, the battery's energy consumption can be quantified, so that it can be combined with the actual SOC capacity of the vehicle battery at rest to help determine the remaining discharge time of the vehicle battery, that is, to predict the time required for the battery to drop from the current SOC to the empty state.

[0065] Optionally, the target period may be the time one month before the current moment, and the corresponding total discharge rate may be the monthly total discharge rate.

[0066] In an optional embodiment, the total discharge rate of the vehicle battery of the target vehicle within the target time period is obtained, including: obtaining the current discharge rate of the vehicle battery within the target time period; obtaining the self-discharge rate of the vehicle battery within the target time period; and obtaining the total discharge rate based on the current discharge rate and the self-discharge rate.

[0067] Optionally, the total discharge rate is composed of two parts: the current discharge rate and the self-discharge rate. The current discharge rate can reflect the energy consumption of the target vehicle due to the power consumption of the on-board electronic equipment when it is stationary, while the self-discharge rate is the energy naturally lost by the chemical reaction inside the battery when the battery is not in use. By combining the total discharge rate with the current discharge rate and the self-discharge rate, the energy loss rate of the battery in a stationary state can be comprehensively evaluated. When calculating the total discharge rate, considering both the current discharge rate and the self-discharge rate can more accurately predict the time required for the battery to drop from the current SOC to the empty state, that is, the remaining discharge time. It should be noted that the current discharge rate and self-discharge rate of different vehicles may vary due to factors such as battery type, vehicle configuration and usage conditions. By calculating the total discharge rate of a specific vehicle, a personalized risk assessment can be provided for each vehicle, which helps to develop a more accurate early warning strategy. Based on the current discharge rate and self-discharge rate of the vehicle, the early warning system can determine when the remaining energy of the vehicle battery may not be sufficient to support its energy needs in a stationary state, thereby triggering an early warning in time to avoid battery feeding.

[0068] Optionally, after determining that the vehicle is in a stationary state, the current SOC of the target vehicle can be further detected. If the current SOC is less than a predetermined value, the total discharge rate is calculated. For example, the data of vehicles with an SOC of less than 30% can be used in the subsequent stationary power feeding warning algorithm. For vehicles with an SOC of less than 30%, it is necessary to analyze the total discharge rate of the vehicle based on historical data. The monthly total discharge rate = current discharge rate + self-discharge rate. The current discharge rate is used to estimate the monthly current discharge rate based on the current power consumption of the last two hours of stationary state. The current power consumption of the last two hours of stationary state is calculated in a similar integral manner. The calculation formula is as follows:

[0069] Current consumption = I1*T+I2*T+I3*T+…+I n *T

[0070] Among them, T is the collection interval, I is the current at each collection point, and the small current consumption for two hours is calculated by summing the power consumption in each collection period. Because the collection point interval is short, generally collected once every 10 seconds, the current consumption calculated by this method is close to the actual current consumption. The monthly current discharge rate is:

[0071]

[0072] The formula assumes that each month is 30 days long. The self-discharge rate of the vehicle battery can be calculated by calling the self-discharge rate calculation result calculated offline, which mainly uses the historical voltage drop during static state, and calculates the corresponding SOC value through the voltage before and after static state. With the current discharge rate and self-discharge rate of the static vehicle, the monthly total discharge rate can be obtained by summing them.

[0073] Step S106, determining the actual SOC capacity of the vehicle battery in a static state;

[0074] Optionally, the actual SOC capacity can more accurately reflect the current charge and discharge state of the battery. Since the SOC value calculated by the BMS may be affected by a variety of factors, such as battery aging, temperature changes, etc., it may deviate from the actual SOC. These deviations can be eliminated by determining the actual SOC. The actual SOC capacity is a key parameter for calculating the remaining discharge time and evaluating the results of the recharging risk identification. A lower SOC means that the battery has less remaining energy. Combined with the total discharge rate, it can more accurately predict when the battery may be recharged due to energy exhaustion when it is stationary. Recharging risk assessment based on the actual SOC capacity can help optimize the early warning strategy. For example, for vehicles with lower SOC, the system can set a stricter early warning threshold to ensure that the owner or the car manufacturer can be informed in time when the vehicle battery is close to recharging so that necessary precautions can be taken, such as charging in advance or checking the battery status. Considering that the actual SOC of different vehicle batteries may vary due to factors such as aging, usage habits, etc., determining the actual SOC capacity helps provide personalized risk identification results. This means that the early warning information obtained by each vehicle based on its battery condition will be more accurate, thereby avoiding unnecessary alarms or omissions.

[0075] In an optional embodiment, determining the actual SOC capacity of a vehicle battery in a stationary state includes: obtaining a voltage array of the vehicle battery collected in a stationary state, wherein the voltage array includes voltage values ​​of a plurality of single cells included in the vehicle battery in a stationary state; identifying a maximum voltage value and a minimum voltage value in the voltage array; determining a maximum SOC capacity corresponding to the maximum voltage value and a minimum SOC capacity corresponding to the minimum voltage value based on a corresponding relationship between voltage values ​​and SOC capacities; and determining the actual SOC capacity based on the maximum SOC capacity and the minimum SOC capacity.

[0076] Optionally, the SOC evaluation in the related art usually relies on the data provided by the BMS, and the BMS may give an inaccurate SOC value due to factors such as algorithm errors, battery aging or temperature effects. By directly collecting the battery voltage array in a static state, identifying the maximum voltage value and the minimum voltage value therein to determine the maximum SOC capacity and the minimum SOC capacity, and then calculating the actual SOC capacity, the above method can provide an SOC value that is closer to the actual state of the battery. Using the maximum and minimum voltage values ​​in the voltage array to calculate the SOC capacity can avoid overall SOC evaluation errors caused by abnormal data of a single cell. When the vehicle is at rest, the battery voltage value is obtained in real time and the SOC capacity is calculated immediately. The above method supports real-time monitoring of the battery status, thereby quickly identifying power feeding risks.

[0077] Optionally, for the SOC data transmitted from the BMS in the battery, the BMS calculates the power consumed by the discharge current by integrating the discharge current, subtracts the power consumed by the discharge current from the initial power, and obtains the current remaining power. The current remaining power / total battery capacity is used to obtain the battery SOC, but this calculation is not accurate. This embodiment uses a new method to calculate the true SOC of the battery. SOC min is a voltage array {V0,V1,V2,V3,…,V n The minimum voltage V min By querying the preset state of charge-voltage (SOC-OCV) correspondence table, the voltage V of the minimum voltage cell is obtained. min Corresponding SOC capacity SOC min Similarly, SOC max is a voltage array {V0,V1,V2,V3,…,V n}The maximum voltage V max By querying the SOC-OCV table, the voltage V of the maximum voltage cell is obtained. max Corresponding SOC capacity SOC max , because the actual SOC should take into account the situation of the entire battery system. If the SOC of the minimum voltage cell is significantly different from the average SOC of all other battery cells, when the capacity consumption of the minimum voltage cell is 0, even if other cells still have remaining power, they cannot be released for the entire battery system. Similarly, when the entire battery system is charging, when the highest voltage cell is charged to 100% SOC, other cells cannot be charged even if they have not reached full capacity. After considering the above reasons, the following calculation of the actual SOC capacity SOC is obtained. real The calculation formula is:

[0078]

[0079] The SOC calculated by this formula can be used as the SOC of the entire battery system, and it can also prevent the calculated SOC from being falsely high.

[0080] Step S108, determining the remaining discharge time of the vehicle battery based on the total discharge rate and the actual SOC capacity;

[0081] Optionally, the calculation of the remaining discharge time combines the energy consumption rate (total discharge rate) of the battery in a static state and the current energy state (actual SOC capacity), and can estimate the time required for the battery to drop to the feeding state without charging. This is of great value for preventing battery feeding, ensuring battery safety and extending its life. The determination of the remaining discharge time can be used for early warning classification and intelligent notification. When the calculated remaining discharge time is lower than the preset threshold, different levels of early warning can be sent to the owner or the car manufacturer in time according to the cumulative static time and aging of the vehicle to avoid the vehicle being unable to start or other more serious battery damage caused by battery feeding. The batteries of different vehicles may differ in total discharge rate and actual SOC capacity due to different models, years of use, maintenance conditions, etc. In the above way, the battery feeding risk of each vehicle can be personalized and evaluated to ensure that the early warning information is accurate and timely for each target vehicle.

[0082] Optionally, the calculated total discharge rate of the battery when it is at rest and the actual SOC capacity can be used to calculate how long the battery can be left without charging before the SOC reaches 0 under the current battery conditions. If the battery continues to be left at SOC 0, there is a risk of power feeding, resulting in an irreversible decrease in the total battery capacity. To prevent this from happening, this embodiment calculates and predicts the time it takes for the battery to be discharged to empty under the current conditions, and issues an early warning. The calculation formula is:

[0083]

[0084] Among them, because the monthly total discharge rate is used, the T 剩余 The remaining discharge months are the number of remaining discharge days. To get the remaining discharge days, you can multiply it by 30, because when calculating the monthly total discharge rate, it is assumed that there are 30 days in each month.

[0085] Step S110: determining a power feeding risk identification result of the vehicle battery based on the remaining discharge time and the accumulated rest time.

[0086] Optionally, the risk of power feeding can be graded by comparing the remaining discharge time with the cumulative rest time. For example, if the remaining discharge time accounts for a low proportion of the cumulative rest time, it means that the battery is about to run out of energy and the risk of power feeding is high; when the proportion is high, the risk is relatively low. Based on the analysis of the remaining discharge time and the cumulative rest time, the time point of battery power feeding can be predicted in advance, so that early warning information can be sent to the owner or the car manufacturer at a reasonable time before the battery is fed, leaving enough reaction time to take preventive measures, such as charging in advance or adjusting the vehicle's rest state, to avoid damage caused by battery power feeding. Different vehicles may be in different resting states and SOC levels. Therefore, the combined analysis of the remaining discharge time and the cumulative rest time can provide a personalized risk assessment for each vehicle. The power feeding risk identification results provide important decision support for car owners and car manufacturers.

[0087] In an optional embodiment, a power feeding risk identification result of a vehicle battery is determined based on a remaining discharge time and a cumulative standing time, including: determining a target time range to which the remaining discharge time belongs from a plurality of time ranges, wherein the plurality of time ranges are determined based on different proportions of the cumulative standing time; determining a target power feeding risk level corresponding to the target time range from a plurality of power feeding risk levels, wherein the plurality of power feeding risk levels correspond one-to-one to the plurality of time ranges; and determining a power feeding risk identification result based on the target power feeding risk level.

[0088] Optionally, by converting the relationship between the remaining discharge time and the accumulated static time into different risk levels, a quantitative representation of the risk is achieved. This makes the risk warning not limited to a simple "yes" or "no", but can give the degree of risk according to the specific value. Dividing multiple time ranges and corresponding risk levels can realize a multi-level warning mechanism. This means that different countermeasures can be taken for different risk levels, thereby improving the flexibility and pertinence of the warning. By providing a target power feeding risk level, the owner or the car manufacturer can quickly understand the health status and potential power feeding risks of the current vehicle battery based on the warning results, so as to facilitate the formulation of corresponding battery management or maintenance plans. By introducing the proportional relationship between the remaining discharge time and the accumulated static time and converting it into a power feeding risk level, not only can the risk be quantified and the warning be multi-level, but also the intelligence and reliability of the warning system can be enhanced, providing strong support for the efficient management and maintenance of new energy vehicle batteries.

[0089] Optionally, for appropriate thresholds for alarms, the multiple power feeding risk levels can be three levels of alarm levels: level one is the lowest risk alarm, indicating that the vehicle has a risk of static power feeding, but it is not a big deal for the time being and can be paid a little attention; level two is a medium risk alarm, indicating that the vehicle has a greater risk of static power feeding and needs to pay attention to the vehicle condition. If the situation is true, manual intervention can be performed in advance to prevent the battery condition from further deteriorating; level three is the highest risk alarm, indicating that the vehicle has a serious risk of static power feeding or battery power feeding has already occurred, and timely manual intervention is now required, such as timely charging or sending the vehicle to the after-sales service department for timely repair to prevent malicious incidents caused by further deterioration of the battery condition. Different thresholds can be set for the three alarm levels. The threshold algorithm mainly considers the cumulative static time and the actual SOC capacity. If the actual SOC of the vehicle is lower than 3%, a third-level static power feed alarm will be issued as long as the cumulative static time is judged to be greater than 2 hours; on this basis, the first-level static power feed alarm needs to meet the requirement that the remaining discharge time is less than 30% of the vehicle's static time, the second-level static power feed alarm needs to meet the requirement that the remaining discharge time is less than 20% of the vehicle's static time, and the third-level static power feed alarm needs to meet the requirement that the remaining discharge time is less than 15% of the vehicle's static time. The possibility of static time in the remaining discharge time is inferred by the proportion of the remaining discharge time to the static time. In this way, graded alarms can not only ensure the authenticity of the remaining static time, but also meet the condition that the less battery power of the static vehicle, the greater the risk.

[0090] Optionally, after outputting the static power feed alarm level of the static vehicle, the warning result can be sent to the car owner or car manufacturer through SMS, email or warning platform. For vehicles on the battery warning big data platform, the static power feed alarm result can be directly saved in the corresponding database, allowing the battery warning big data platform to read the alarm result by itself; for other car manufacturers that have their own battery big data management platform, this solution will send the static power feed alarm vehicle, alarm level and related alarm details to the car manufacturer through the Kafka server. Through the above methods, it is ensured that the car owner can know the warning risk of the vehicle battery in time, and take corresponding measures in time according to the alarm type and alarm level to ensure that the battery condition of the vehicle will not deteriorate further, thereby increasing the service life of the vehicle battery. Optimizing the use of the battery can also greatly improve the car owner's car experience.

[0091] Through the above steps S102 to S110, the remaining discharge time and the accumulated stationary time of the vehicle battery can be obtained in real time, and the purpose of dynamically identifying the feeding risk of the vehicle battery can be achieved, thereby achieving the technical effect of improving the efficiency and accuracy of identifying the feeding risk of the vehicle battery when the vehicle is stationary, and then solving the technical problem of low accuracy and efficiency of feeding risk prediction in the related art of identifying the feeding risk of the vehicle battery when the vehicle is stationary.

[0092] Based on the above embodiments and optional embodiments, the present invention proposes an optional implementation mode: Figure 2 is a flow chart of an optional vehicle battery feeding risk identification method according to an embodiment of the present invention, such as Figure 2 As shown, the method includes:

[0093] S1, receives the battery voltage, current, temperature and other data of the target vehicle from the Kafka server in real time for data cleaning and regularization;

[0094] S2, using the real-time computing framework to determine the static state and accumulate the static time of the vehicle real-time data;

[0095] S3, verifying the static time of the vehicle using the LSTM model based on offline calculation, such as verifying that the target vehicle is in a static state;

[0096] S4, detect whether the current SOC capacity of the vehicle (i.e., the actual SOC capacity) is less than 30%. If so, calculate the total discharge rate of the vehicle at rest by the voltage change trend, the vehicle's historical self-discharge rate, and the current data; otherwise, do not output an alarm;

[0097] S5, predict the remaining discharge time from discharge to empty power according to the actual SOC capacity of the current battery of the stationary vehicle and the current total discharge rate, and give a stationary power supply warning to the vehicle according to the aging condition of the vehicle. Among them, three levels of alarm levels can be set, and different thresholds are set for the three alarm levels. Specifically:

[0098] If the actual SOC of the vehicle is lower than 3%, the third-level static power supply alarm will be issued as long as the cumulative static time is judged to be greater than 2 hours; on this basis, the first-level static power supply alarm needs to meet the remaining discharge time less than 30% of the vehicle's static time, the second-level static power supply alarm needs to meet the remaining discharge time less than 20% of the vehicle's static time, and the third-level static power supply alarm needs to meet the remaining discharge time less than 15% of the vehicle's static time;

[0099] S6, sending the warning result to the car owner or car manufacturer via SMS, email or warning platform.

[0100] When it is detected that the target vehicle is in a stationary state, the cumulative stationary time of the target vehicle is obtained; the total discharge rate of the vehicle battery of the target vehicle within the target time period is obtained; the actual SOC capacity of the vehicle battery in the stationary state is determined; based on the total discharge rate and the actual SOC capacity, the remaining discharge time of the vehicle battery is determined; based on the remaining discharge time and the cumulative stationary time, the feeding risk identification result of the vehicle battery is determined.

[0101] In this embodiment, a vehicle battery feeding risk identification device is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the terms "module" and "device" can implement a combination of software and / or hardware for a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0102] According to an embodiment of the present invention, there is also provided an apparatus embodiment for implementing the above-mentioned vehicle battery feeding risk identification method. Figure 3 is a schematic diagram of the structure of a vehicle battery feeding risk identification device according to an embodiment of the present invention. Figure 3 As shown, the above-mentioned vehicle battery feeding risk identification device includes: a timing module 200, an acquisition module 202, a capacity module 204, a determination module 206, and an identification module 208, wherein:

[0103] The timing module 200 is used to obtain the cumulative static time of the target vehicle in the static state when it is detected that the target vehicle is in the static state;

[0104] The acquisition module 202 is connected to the timing module 200 and is used to acquire the total discharge rate of the vehicle battery of the target vehicle within the target period;

[0105] A capacity module 204, connected to the acquisition module 202, for determining an actual SOC capacity of the vehicle battery in a static state;

[0106] A determination module 206 , connected to the capacity module 204 , for determining a remaining discharge time of the vehicle battery based on the total discharge rate and the actual SOC capacity;

[0107] The identification module 208 is connected to the determination module 206 and is used to determine the power feeding risk identification result of the vehicle battery based on the remaining discharge time and the accumulated static time.

[0108] It should be noted that the above modules can be implemented by software or hardware. For example, for the latter, it can be implemented in the following ways: the above modules can be located in the same processor; or the above modules can be located in different processors in any combination.

[0109] It should be noted that the timing module 200, the acquisition module 202, the capacity module 204, the determination module 206, and the identification module 208 correspond to steps S102 to S110 in the embodiment, and the examples and application scenarios implemented by the modules and the corresponding steps are the same, but are not limited to the contents disclosed in the above embodiment. It should be noted that the modules as part of the device can be run in a computer terminal.

[0110] It should be noted that the optional or preferred implementation of this embodiment can refer to the relevant description in the embodiment, which will not be repeated here.

[0111] The above-mentioned vehicle battery feeding risk identification device may also include a processor and a memory. The above-mentioned timing module 200, acquisition module 202, capacity module 204, determination module 206, identification module 208, etc. are all stored in the memory as program modules, and the processor executes the above-mentioned program modules stored in the memory to realize corresponding functions.

[0112] The processor includes a kernel, which retrieves the corresponding program module from the memory. The kernel may be one or more. The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. The memory includes at least one memory chip.

[0113] According to an embodiment of the present application, an embodiment of a non-volatile storage medium is also provided. Optionally, in this embodiment, the non-volatile storage medium includes a stored program, wherein when the program is executed, the device where the non-volatile storage medium is located is controlled to execute any of the above-mentioned vehicle battery feeding risk identification methods.

[0114] Optionally, in this embodiment, the non-volatile storage medium may be located in any computer terminal in a computer terminal group in a computer network, or in any mobile terminal in a mobile terminal group, and the non-volatile storage medium includes a stored program.

[0115] Optionally, when the program is running, the device where the non-volatile storage medium is located is controlled to perform the following functions: when it is detected that the target vehicle is in a stationary state, obtaining the cumulative stationary time of the target vehicle in the stationary state; obtaining the total discharge rate of the vehicle battery of the target vehicle during the target time period; determining the actual SOC capacity of the vehicle battery in the stationary state; based on the total discharge rate and the actual SOC capacity, determining the remaining discharge time of the vehicle battery; based on the remaining discharge time and the cumulative stationary time, determining the feeding risk identification result of the vehicle battery.

[0116] According to an embodiment of the present application, an embodiment of a processor is also provided. Optionally, in this embodiment, the processor is used to run a program, wherein when the program is run, any one of the vehicle battery feeding risk identification methods is executed.

[0117] According to an embodiment of the present application, an embodiment of a computer program product is also provided. Optionally, in this embodiment, the computer program product includes a computer program, and when the computer program is executed by a processor, the computer program implements any of the steps of the vehicle battery feeding risk identification method.

[0118] Optionally, the above-mentioned computer program product, when executed on a data processing device, is suitable for executing a program initialized with the following method steps: when it is detected that the target vehicle is in a stationary state, obtaining the cumulative stationary time of the target vehicle in the stationary state; obtaining the total discharge rate of the vehicle battery of the target vehicle within the target time period; determining the actual SOC capacity of the vehicle battery in the stationary state; based on the total discharge rate and the actual SOC capacity, determining the remaining discharge time of the vehicle battery; based on the remaining discharge time and the cumulative stationary time, determining the power feeding risk identification result of the vehicle battery.

[0119] An embodiment of the present invention provides an electronic device, which includes a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, the following steps are implemented: when a target vehicle is detected to be in a stationary state, the cumulative stationary time of the target vehicle is obtained; the total discharge rate of a vehicle battery of the target vehicle within a target period of time is obtained; the actual SOC capacity of the vehicle battery in the stationary state is determined; based on the total discharge rate and the actual SOC capacity, the remaining discharge time of the vehicle battery is determined; based on the remaining discharge time and the cumulative stationary time, the feeding risk identification result of the vehicle battery is determined.

[0120] The above sequence of the embodiments of the present invention is for description only and does not represent the superiority or inferiority of the embodiments.

[0121] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0122] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the above modules can be a logical function division. There may be other division methods in actual implementation, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, modules or indirect coupling or communication connection of modules, which can be electrical or other forms.

[0123] The modules described above as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, that is, they may be located in one place or distributed on multiple modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment.

[0124] In addition, each functional module in each embodiment of the present invention may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of software functional modules.

[0125] If the above-mentioned integrated module is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable non-volatile storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a non-volatile storage medium, including a number of instructions to enable a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the various embodiments of the present invention. The aforementioned non-volatile storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.

[0126] The above are only preferred embodiments of the present invention. It should be pointed out that, for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A vehicle battery feeding risk identification method, characterized in that: include: When it is detected that the target vehicle is in a stationary state, obtaining a cumulative stationary time length of the target vehicle in the stationary state; Obtaining a total discharge rate of a vehicle battery of the target vehicle within a target period of time; determining an actual SOC capacity of the vehicle battery in the stationary state; determining a remaining discharge time of the vehicle battery based on the total discharge rate and the actual SOC capacity; Based on the remaining discharge time and the accumulated rest time, a power feeding risk identification result of the vehicle battery is determined.

2. The method according to claim 1, characterized in that In the case where the target vehicle is detected to be in a stationary state, before obtaining the accumulated stationary time length of the target vehicle in the stationary state, the method further includes: Acquire multiple frames of operating data of the target vehicle collected in the current period, wherein the multiple frames of operating data correspond to different sampling moments, and each frame of operating data at least includes: a voltage array, a charging current, a discharging current, a vehicle mileage, and a vehicle speed of the target vehicle at a corresponding moment, wherein the voltage array includes voltage values ​​corresponding to a plurality of single cells in the vehicle battery; Detecting whether the multiple frames of operating data meet a predetermined operating condition; In the case where, among the multiple frames of operating data, the number of operating data frames that meet the predetermined operating condition is greater than a predetermined number of consecutive frames, it is determined that the target vehicle is in the stationary state.

3. The method according to claim 2, characterized in that The detecting whether the multiple frames of operation data meet a predetermined operation condition includes: Whether any frame of operation data in the multiple frames of operation data satisfies the predetermined operation condition is detected in the following manner: Based on the voltage data of the target vehicle at any time, detect whether the voltage change rate of the vehicle battery is less than a preset change rate threshold, whether the target vehicle has a charging current at any time, whether the discharge current of the target vehicle at any time is less than a preset current, whether the vehicle mileage of the target vehicle at any time changes, and whether the vehicle speed of the target vehicle at any time is less than a preset speed, and obtain a detection result, wherein the voltage change rate includes the overall voltage change rate of the vehicle battery and the single cell voltage change rate of each single cell in the vehicle battery; Determining, according to the detection result, whether any frame of operation data meets the predetermined operation condition; Whether the multiple frames of operating data meet the predetermined operating condition is detected by detecting whether any frame of operating data meets the predetermined operating condition.

4. The method according to claim 3, characterized in that The step of determining, according to the detection result, whether any frame of operation data satisfies the predetermined operation condition comprises: When the voltage change rate is less than the preset change rate threshold, and the target vehicle has no charging current at any time, and the discharge current is less than the preset current, and the vehicle mileage does not change, and the vehicle speed is less than the preset speed, it is determined that any frame of operating data meets the predetermined operating condition; or When the voltage change rate is greater than or equal to the preset change rate threshold, or the target vehicle has a charging current at any time, or the discharging current is greater than or equal to the preset current, or the vehicle mileage changes, or the vehicle speed is greater than or equal to the preset speed, it is determined that any frame of operating data does not meet the predetermined operating condition.

5. The method according to claim 2, characterized in that: When the number of frames of operation data satisfying the predetermined operation condition among the multiple frames of operation data is greater than a predetermined number of consecutive frames, determining that the target vehicle is in the stationary state includes: When the number of frames of operation data satisfying the predetermined operation condition in the multiple frames of operation data is greater than the predetermined number of consecutive frames, obtaining the target time corresponding to the first frame of operation data satisfying the predetermined operation condition in the multiple frames of operation data and the end time of the current time period; Determining an initial static time based on the target time and the end time; Acquire multiple frames of first feature data of the target vehicle collected during the current period, wherein the multiple frames of first feature data correspond to different sampling moments, and each frame of feature data is used to indicate the running characteristics of the target vehicle at the corresponding moment; Based on the multi-frame first feature data, a long short-term memory network model is used to obtain a first predicted state of the target vehicle and a first predicted stationary duration, wherein the first predicted state is used to indicate whether the target vehicle is in a stationary state within a period corresponding to the initial stationary duration; When the first predicted state is the stationary state and a difference between the initial stationary time length and the first predicted stationary time length is less than a preset difference threshold, it is determined that the target vehicle is in the stationary state.

6. The method according to claim 1, characterized in that The obtaining of the total discharge rate of the vehicle battery of the target vehicle within the target period includes: Obtaining a current discharge rate of the vehicle battery during the target period; Obtaining a self-discharge rate of the vehicle battery during the target period; The total discharge rate is obtained based on the current discharge rate and the self-discharge rate.

7. The method according to claim 1, characterized in that Determining the actual SOC capacity of the vehicle battery in the stationary state includes: Acquire a voltage array of the vehicle battery collected in the static state, wherein the voltage array includes voltage values ​​of a plurality of single cells included in the vehicle battery in the static state; Identifying a maximum voltage value and a minimum voltage value in the voltage array; Based on the corresponding relationship between the voltage value and the SOC capacity, determining the maximum SOC capacity corresponding to the maximum voltage value and the minimum SOC capacity corresponding to the minimum voltage value; Based on the maximum SOC capacity and the minimum SOC capacity, the actual SOC capacity is determined.

8. The method according to any one of claims 1 to 7, characterized in that The determining, based on the remaining discharge time and the accumulated static time, a result of identifying the feeding risk of the vehicle battery includes: Determine a target duration range to which the remaining discharge duration belongs from a plurality of duration ranges, wherein the plurality of duration ranges are determined based on different proportions of the accumulated resting duration; Determining a target power feeding risk level corresponding to the target duration range from a plurality of power feeding risk levels, wherein the plurality of power feeding risk levels correspond one-to-one to the plurality of duration ranges; Based on the target power feeding risk level, the power feeding risk identification result is determined.

9. The method according to any one of claims 1 to 7, characterized in that: The method further comprises: When the target vehicle is in the stationary state, multiple frames of second feature data of the target vehicle collected in the first time period are obtained according to a preset timing task, wherein the multiple frames of second feature data correspond to different sampling moments, and each frame of feature data is used to indicate the running characteristics of the target vehicle at the corresponding moment, and the first time period is the time period after the current time period; based on the multiple frames of second feature data, a long short-term memory network model is used to obtain a second predicted state of the target vehicle and a second predicted stationary duration, wherein the second predicted state is used to indicate whether the target vehicle is in a stationary state during the first time period; When the second predicted state is the static state and the difference between the accumulated static time and the second predicted static time is less than a preset difference threshold, the accumulated static time is corrected to obtain a corrected accumulated static time.

10. A vehicle battery feeding risk identification device, characterized in that: include: A timing module, when detecting that the target vehicle is in a stationary state, obtains the accumulated stationary time length of the target vehicle in the stationary state; An acquisition module, for acquiring a total discharge rate of a vehicle battery of the target vehicle within a target period of time; A capacity module, determining an actual SOC capacity of the vehicle battery in the static state; a discharge module, determining a remaining discharge time of the vehicle battery based on the total discharge rate and the actual SOC capacity; An identification module determines a feeding risk identification result of the vehicle battery based on the remaining discharge time and the accumulated static time.

Images