Battery voltage difference anomaly detection method and related equipment

Abstract

The invention provides a battery voltage difference anomaly detection method and related equipment, and the method comprises the steps: obtaining the voltage data of a battery pack of each vehicle, and determining the single voltage difference data from the voltage data of the battery pack; performing positive sequence sorting on the single pressure difference data of the same vehicle to obtain first pressure difference distribution, and calculating a single pressure difference lower edge of the vehicle according to a first type of quantile of the first pressure difference distribution; determining 50 quantiles of the first differential pressure distribution, and performing positive sequence sorting on the 50 quantiles of the same battery model to obtain second differential pressure distribution; calculating a group differential pressure upper edge corresponding to the vehicle according to a second type quantile of the second differential pressure distribution; and comparing the single-vehicle lower edge of the vehicle with the group differential pressure upper edge corresponding to the vehicle, and determining that the battery differential pressure of the vehicle is abnormal under the condition that the single-vehicle differential pressure lower edge is smaller than the group differential pressure upper edge. According to the method, the battery state can be predicted in real time according to the to-be-evaluated data of the vehicle, and the abnormal condition of the battery state is processed in time.

Description

Technical Field

[0001] The present application relates to the technical field of automobile batteries, and in particular to a method for detecting abnormal battery voltage difference and related equipment. Background Art

[0002] With the widespread use of electric vehicles, the safety requirements for vehicles are getting higher and higher, especially in terms of the safety performance of the vehicle's power battery. The safety requirements of users are constantly increasing. Usually, the pressure difference of the vehicle's power battery is monitored to help users promptly discover abnormal problems in the power cell voltage data, and by analyzing the abnormal pressure difference data, the power battery can be repaired in a timely manner.

[0003] In the related art, the pressure difference detection method for power batteries is relatively simple, and the abnormal condition of the battery can only be evaluated by obtaining the pressure difference data of the power battery in the charging state. The pressure difference data obtained in this way is seriously delayed, resulting in a relatively lagging feedback processing of the battery. At the same time, the existing technology is also unable to predict the abnormal condition of the power battery and generate solutions for possible abnormal conditions. Summary of the invention

[0004] In view of this, the present application provides a battery voltage abnormality detection method and related equipment, which can accurately detect abnormal conditions of the battery.

[0005] A first aspect of an embodiment of the present application provides a method for detecting abnormal battery pressure difference, including: obtaining battery pack voltage data of each vehicle, and determining single cell pressure difference data from the battery pack voltage data; sorting the single cell pressure difference data of the same vehicle in positive order to obtain a first pressure difference distribution; determining a first-class quantile of the first pressure difference distribution, and calculating the lower edge of the single-vehicle pressure difference of the vehicle based on the first-class quantile; determining the 50th quantile of the first pressure difference distribution, and sorting the 50th quantile of the same battery model in positive order to obtain a second pressure difference distribution; determining a second-class quantile of the second pressure difference distribution, and calculating the upper edge of the group pressure difference corresponding to the vehicle based on the second-class quantile; comparing the single-vehicle lower edge of the vehicle with the corresponding group pressure difference upper edge of the vehicle, and when the single-vehicle pressure difference lower edge is smaller than the group pressure difference upper edge, determining that the battery pressure difference of the vehicle is abnormal.

[0006] Compared with the related art, the embodiments of the present application have at least the following advantages: By receiving the battery pack voltage data of multiple vehicles, the voltage data of the vehicle is compared with the voltage data of the battery of the same model to accurately detect the pressure difference of the vehicle battery. Among them, the battery pack is composed of multiple battery cells. The voltage value of the battery cell with the lowest voltage in the entire vehicle can be detected through the lower edge of the single-vehicle pressure difference, so as to find the battery cell with over-discharge risk. When calculating the lower edge of the single-vehicle pressure difference, the first pressure difference distribution is used for calculation, so that the lower edge of the single-vehicle pressure difference can be accurately calculated while retaining the consistency of the vehicle battery cells. Extracting the 50th percentile from the first pressure difference distribution to generate the second pressure difference distribution is to reduce the interference of the battery pressure difference of the vehicle with excessive pressure difference deviation in the same model battery, determine the single cell pressure difference data of the same model battery within the same baseline range, and the upper edge of the group pressure difference calculated by the second pressure difference distribution is the maximum battery pressure difference of the battery cell within the baseline range of the same model vehicle. When the lower edge of the single-vehicle pressure difference is less than the upper edge of the group pressure difference, it means that the battery pack pressure difference is large, indicating that there is an abnormality in the battery cell and the battery has the risk of overcharge or over-discharge.

[0007] Optionally, determining the single cell pressure difference data from the battery pack voltage data includes: determining the single cell voltage data from the battery pack voltage data, the single cell voltage data is the battery pack cell voltage; determining the maximum single cell voltage data and the minimum single cell voltage data in the single cell voltage data, and determining the single cell pressure difference data of the vehicle according to the maximum single cell voltage data and the minimum single cell voltage data.

[0008] Optionally, the first type of quantiles includes: the 25th quantile of the first pressure differential distribution and the 75th quantile of the first pressure differential distribution; calculating the lower edge of the vehicle's single-vehicle pressure differential based on the first type of quantiles, including: based on a first difference between the 75th quantile and the 25th quantile; obtaining a first coefficient, calculating a first product of the first coefficient and the first difference; calculating the lower edge of the vehicle's single-vehicle pressure differential based on the difference between the 25th quantile data and the first product.

[0009] Optionally, the second type of quantiles includes: the 25th quantile of the second pressure differential distribution and the 75th quantile of the second pressure differential distribution; calculating the upper edge of the group pressure differential corresponding to the vehicle based on the second type of quantiles, including: determining a second difference between the 75th quantile and the 25th quantile; obtaining a second coefficient, calculating a second product of the second coefficient and the second difference; calculating the upper edge of the group pressure differential based on the sum of the second product and the 75th quantile.

[0010] Optionally, after determining that there is an abnormality in the battery pressure difference of the vehicle, the method includes: determining an abnormal vehicle with an abnormal battery pressure difference, obtaining battery operation data and driving condition data corresponding to the abnormal vehicle; and generating abnormal sample data based on the battery operation data and driving condition data.

[0011] Optionally, after generating the abnormal sample data, the method includes: generating an abnormal battery status time series and a status label based on the abnormal sample data; inputting the abnormal battery status time series into a preset prediction model to obtain a prediction result output by the preset prediction model; and updating the model parameters of the preset prediction model based on the difference between the prediction result and the status label to obtain a target prediction model.

[0012] Optionally, after obtaining the target prediction model, it includes: receiving the battery operation data to be tested and the driving condition data to be tested of the vehicle to be tested; inputting the battery operation data to be tested and the driving condition data to be tested into the target prediction model to obtain the battery state prediction result of the vehicle to be tested.

[0013] In a second aspect, an embodiment of the present application further provides a battery voltage abnormality detection device, comprising: An acquisition unit, used to acquire battery pack voltage data of each vehicle, and determine single cell pressure difference data from the battery pack voltage data; A first calculation unit is used to sort the single pressure difference data of the same vehicle in positive order to obtain a first pressure difference distribution; A second calculation unit is used to determine a first type quantile of the first pressure difference distribution, and calculate a single vehicle pressure difference lower edge of the vehicle according to the first type quantile; A third calculation unit is used to determine the 50th percentile of the first pressure difference distribution, and sort the 50th percentiles of the same battery model in positive order to obtain a second pressure difference distribution; A fourth calculation unit is used to determine a second type quantile of the second pressure difference distribution, and calculate an upper edge of the group pressure difference corresponding to the vehicle according to the second type quantile; The detection unit is used to compare the lower edge of the vehicle with the upper edge of the group pressure difference corresponding to the vehicle. When the lower edge of the single vehicle pressure difference is smaller than the upper edge of the group pressure difference, it is determined that the battery pressure difference of the vehicle is abnormal.

[0014] In a third aspect, an embodiment of the present application further provides an electronic device, the electronic device comprising a processor and a memory, the memory being used to store instructions, and the processor being used to call the instructions in the memory so that the electronic device executes the battery voltage abnormality detection method as in the first aspect.

[0015] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, which stores computer instructions. When the computer instructions are executed on an electronic device, the electronic device executes the battery voltage abnormality detection method as in the first aspect.

[0016] The technical effects obtained by the above-mentioned second, third and fourth aspects are similar to the technical effects obtained by the corresponding technical means in the first aspect, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A flowchart of the steps of a battery voltage abnormality detection method provided in one embodiment of the present application.

[0018] Figure 2 A flowchart of the steps of a method for detecting abnormal battery voltage difference provided in yet another embodiment of the present application.

[0019] Figure 3 A flowchart of the steps of a method for detecting abnormal battery voltage difference provided in yet another embodiment of the present application.

[0020] Figure 4 A flowchart of the steps of a method for detecting abnormal battery voltage difference provided in yet another embodiment of the present application.

[0021] Figure 5 A schematic diagram of the application environment of the target prediction model provided in one embodiment of the present application.

[0022] Figure 6 A functional module diagram of a battery voltage abnormality detection device provided in one embodiment of the present application.

[0023] Figure 7 A schematic diagram of the structure of an electronic device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0024] In order to more clearly understand the above-mentioned purposes, features and advantages of the present application, the present application is described in detail below in conjunction with the accompanying drawings and specific implementation methods. It should be noted that the implementation methods of the present application and the features in the implementation methods can be combined with each other without conflict.

[0025] In the following description, many specific details are set forth to facilitate a full understanding of the present application. The described implementations are only part of the implementations of the present application, rather than all the implementations.

[0026] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein in the specification of this application are only for the purpose of describing specific embodiments and are not intended to limit this application.

[0027] It should be further noted that, in this article, the terms "comprises", "includes" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or device including the element.

[0028] In this application, "at least one" means one or more, and "more" means two or more than two. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone, where A and B can be singular or plural. The terms "first", "second", "third", "fourth", etc. (if any) in the specification, claims and drawings of this application are used to distinguish similar objects, rather than to describe a specific order or sequence.

[0029] In the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0030] To facilitate understanding, some illustrations of concepts related to the embodiments of the present application are given by way of example for reference.

[0031] The battery of a vehicle is usually a battery module, which is composed of multiple single cells. A single cell is an independent battery unit, which usually includes a battery cell, a positive electrode, a negative electrode, an electrolyte and a separator. It is commonly found in consumer electronics, power tools and electric vehicles. Single cells usually have a small capacity and voltage and are the basic components of a battery pack.

[0032] The vehicle also includes: Battery Monitoring and Management System (BMS), thermal management system, and electrical system to regulate the operation of the battery. The BMS system monitors and manages battery parameters such as voltage, current, and temperature. The thermal management system is used to control the dimensions of the battery so that the battery operates within the indicated temperature range.

[0033] The TSP (Telematics Service Provider) platform is one of the core components of the Internet of Vehicles. It is responsible for providing telematics services and connecting multiple parties such as vehicle manufacturers, application / content service providers / battery cell operators and call centers. Its main functions include: vehicle monitoring, navigation positioning, remote upgrades, etc. In terms of battery management, the TSP platform can monitor the battery status in real time, including parameters such as voltage / current / temperature. When a battery abnormality is detected, it can record and report the problem so that maintenance personnel can take timely measures.

[0034] The present application provides a battery voltage difference abnormality detection method and related equipment, which can accurately detect abnormal conditions of the battery. Figure 1 , Figure 1 The flowchart of an embodiment of the method for detecting abnormal battery voltage difference of the present invention comprises the following steps: Step 110 , obtaining battery pack voltage data of each vehicle, and determining single cell pressure difference data from the battery pack voltage data.

[0035] After the vehicle is started, the cloud receives the battery pack voltage data of each vehicle, and determines the single cell voltage data from the battery pack voltage data, where the single cell voltage data is the battery pack cell voltage; determines the maximum single cell voltage data and the minimum single cell voltage data in the single cell voltage data, and determines the vehicle's single cell pressure difference data based on the maximum single cell voltage data and the minimum single cell voltage data.

[0036] Step 120 , sorting the individual pressure difference data of the same vehicle in positive order to obtain a first pressure difference distribution.

[0037] The pressure difference distribution can reflect the abnormal mode and abnormal trend of the battery. According to the first pressure difference distribution of the single cell voltage data of the same vehicle, the abnormal single cell pressure difference data can be detected. The balancing current is dynamically adjusted according to the abnormal single cell pressure difference data to ensure the consistency of the battery and prolong the battery life.

[0038] Step 130, determining a first type quantile of the first pressure difference distribution, and calculating a lower edge of the single-vehicle pressure difference of the vehicle according to the first type quantile.

[0039] The single vehicle pressure drop edge refers to the lowest value of the cell voltage in a single vehicle battery pack. The weakest cell in the battery pack can be found through the single vehicle pressure drop edge. Optionally, the first type of quantiles includes: the 25th percentile of the first pressure difference distribution and the 75th percentile of the first pressure difference distribution; calculating the single vehicle pressure difference lower edge of the vehicle according to the first type of quantiles includes: according to the first difference between the 75th percentile and the 25th percentile; obtaining the first coefficient, calculating the first product of the first coefficient and the first difference; calculating the single vehicle pressure difference lower edge according to the difference between the 25th percentile data and the first product.

[0040] Specifically, the first coefficient is 1.5, and the calculation formula of the lower edge value V1 of the single-vehicle pressure difference is formula (1), including: V1=Q25-1.5ⅹ(Q75-Q25) formula (1); Among them, Q25 is the 25th quantile of the first category mentioned above, and Q75 is the 75th quantile of the first category.

[0041] Step 140, determining the 50th percentile of the first pressure difference distribution, and sorting the 50th percentiles of the same battery model in positive order to obtain a second pressure difference distribution.

[0042] After receiving the first differential pressure distribution of each vehicle, the cloud extracts the 50th percentile from the first differential pressure distribution to generate the second differential pressure distribution. The 50th percentile is selected to reduce the error of the battery differential pressure of vehicles with excessive differential pressure deviation of the same battery signal, so as to determine the single differential pressure data of the same model battery within the same baseline range. The upper edge of the group differential pressure calculated by the second differential pressure distribution can be the maximum battery differential pressure of the battery cells within the baseline range of the same model vehicle.

[0043] Step 150, determining the second type quantile of the second pressure difference distribution, and calculating the upper edge of the group pressure difference corresponding to the vehicle according to the second type quantile.

[0044] Optionally, the second type of quantiles includes: the 25th quantile of the second pressure differential distribution and the 75th quantile of the second pressure differential distribution; calculating the upper edge of the group pressure differential corresponding to the vehicle based on the second type of quantiles, including: determining a second difference between the 75th quantile and the 25th quantile; obtaining a second coefficient, calculating a second product of the second coefficient and the second difference; calculating the upper edge of the group pressure differential based on the sum of the second product and the 75th quantile. The group pressure difference refers to the difference in electricity between different battery packs in a group of vehicles or battery packs. Usually, the voltage of each battery pack is compared with the average battery difference between the battery packs. In this embodiment, the 50th percentile of the first category of quantiles is used to form the group pressure difference distribution between battery packs, and then the 25th and 75th percentiles of the group pressure difference distribution are obtained to calculate the group pressure difference distribution. The group pressure difference distribution can reflect the consistent performance of the fleet or battery pack. By analyzing the group pressure difference, it is possible to identify which battery packs are performing abnormally, so as to carry out targeted maintenance and optimization.

[0045] The upper edge of the group pressure difference refers to the highest value of the cell voltage in the battery pack in a group of vehicles. The upper edge of the group pressure difference is helpful to find the best new energy cell in the fleet and understand its contribution to the overall performance. Calculating the lower edge of the group pressure difference requires calculating the second type of quantile of the group pressure difference distribution.

[0046] Specifically, the second coefficient is 1.5, and the second category quantiles include: 25th percentile pressure difference and 75th percentile pressure difference. The formula for calculating the upper edge V0 of the group pressure difference is formula (2), including: V0=Q75_total+1.5ⅹ(Q75_total-Q25_total) Formula (2); Among them, Q25_total is the 25th quantile of the second category, and Q75_total is the 75th quantile of the second category.

[0047] Step 160, comparing the lower edge of the vehicle and the upper edge of the group pressure difference corresponding to the vehicle, when the lower edge of the single vehicle pressure difference is smaller than the upper edge of the group pressure difference, it is determined that the battery pressure difference of the vehicle is abnormal.

[0048] The upper edge of the group pressure difference calculated by the second pressure difference distribution is the maximum battery pressure difference of the battery cells within the baseline range of the same model of vehicles. When the lower edge of the single vehicle pressure difference is less than the upper edge of the group pressure difference, it means that the battery pack pressure difference is large, indicating that the battery cell is abnormal and there is a risk of overcharge or over-discharge.

[0049] This embodiment mainly evaluates the single cell voltage data through the preset battery evaluation standard. The preset battery evaluation standard mainly compares the lower edge of the single vehicle pressure difference with the upper edge of the group pressure difference of the vehicle battery to determine whether the battery is abnormal. When the difference between the lower edge of the single vehicle pressure difference and the upper edge of the group pressure difference of the vehicle battery is too large, it will cause the performance of the battery pack to decline, affecting the vehicle's cruising range and battery life. In this embodiment, if the lower edge value of the single vehicle pressure difference is greater than the lower edge value of the group pressure difference, that is, (V1>V0), it is marked as abnormal pressure difference data.

[0050] It should be noted that the preset battery evaluation standards can be stored in the cloud. After receiving the lower edge of the single vehicle and the upper edge of the group pressure difference, the cloud directly detects whether the vehicle has any abnormalities and sends the results to the vehicle end. The preset battery evaluation standards can also be stored in the vehicle end. The cloud sends the lower edge of the single vehicle and the upper edge of the group pressure difference to the vehicle end, and the vehicle end makes a judgment based on the preset battery evaluation standards.

[0051] In one embodiment, if Figure 2 After determining that the battery voltage difference of the vehicle is abnormal, step 160 includes: Step 171, determining that there is an abnormal vehicle with an abnormal battery pressure difference, and obtaining battery operation data and driving condition data corresponding to the abnormal vehicle.

[0052] Step 172, generating abnormal sample data according to the battery operation data and the driving condition data.

[0053] In one embodiment, the battery operation data of the vehicle includes: one or more of voltage, current, ambient temperature, SOC, and SOH, and the driving condition data of the vehicle includes: one or more of vehicle identification, user driving habits, and vehicle driving environment parameters.

[0054] Voltage includes: single cell voltage and total battery pack voltage. Current includes: charging current and discharging current. Temperature includes: temperature of each battery cell in the battery pack. Capacity: remaining capacity and total capacity of the battery. The battery's state of charge SOC (State of Charge) is used to indicate the percentage of the current power in the total power. The battery's state of health SOH (State of Health) indicates the percentage of the battery's current performance equivalent to a new battery. Battery operating data also includes: battery internal resistance, number of cycles (number of charge and discharge cycles of the battery), etc. The vehicle is identified by the vehicle identification number VIN. User driving habits include: user switching habits, driving habits, etc. Vehicle driving environment parameters include the road the vehicle is driving on, temperature, etc.

[0055] Step 172 is used to train a preset model after generating abnormal sample data to obtain a target prediction model, so that when the battery operation data to be tested and the driving condition data to be tested of the vehicle to be tested are received, the battery operation data to be tested and the driving condition data to be tested of the vehicle to be tested are input into the target prediction model to accurately predict the abnormal condition of the battery.

[0056] In one embodiment, if Figure 3 As shown, after generating abnormal sample data, it includes: Step 181, generating an abnormal battery status time series and status label according to the abnormal sample data.

[0057] Step 182, inputting the abnormal battery status time series into a preset prediction model to obtain a prediction result output by the preset prediction model.

[0058] Step 183, based on the difference between the prediction result and the state label, the model parameters of the preset prediction model are updated to obtain the target prediction model.

[0059] In this embodiment, the abnormal sample data is divided into training data and training results. The battery features of the training data are extracted, and the abnormal battery state time series is formed according to the battery features of the training data. The abnormal battery state time series and the abnormal state label are used as sample inputs of the preset prediction model.

[0060] The training process of the preset prediction model includes: inputting the abnormal sample data into the preset prediction model, obtaining the output result of the preset prediction model and the predicted value corresponding to the output result, calculating the loss function according to the difference between the predicted value and the actual value of the training result, and adjusting the parameters of the preset prediction model when the loss function does not meet the preset conditions, and re-inputting the output result into the adjusted preset prediction model to obtain a new prediction result. Repeat the above input and output process, and calculate the loss function of the output prediction value until the loss function converges, and the updated preset prediction model is considered to be the target prediction model. Since the abnormal sample data is obtained through the Internet of Vehicles, the abnormal sample data comes from different vehicles, so the source of the abnormal sample data is wide and diverse. A rich sample set can improve the robustness of the model when training the preset model.

[0061] In a specific embodiment, the abnormal data related to the vehicles with abnormal pressure difference in the past three days are extracted to generate abnormal battery status time series, feature extraction is performed, and standardized processing is performed, and the three-day abnormal data is divided into a training set and a validation set. The training set is used to train the model using a long short-term memory network (LSTM), and back propagation and hyperparameter tuning methods are used to find the optimal network result and training parameter combination to obtain the optimal model. The validation set is used to verify the model of the training number to avoid overfitting.

[0062] Since voltage difference is usually the result of various factors acting on the battery during long-term use, in order to avoid gradient vanishing or gradient explosion caused by the passage of time, the LSTM model is used for training. Compared with the traditional RNN (recurrent neural network), LSTM introduces three gates (input gate, forget gate, output gate) and a cell state. These mechanisms enable LSTM to better handle long-term dependencies in sequences.

[0063] In this embodiment, a target prediction model is trained offline and then deployed on the cloud or the vehicle side, or on both the cloud and the vehicle side, to predict candidate battery states.

[0064] In one embodiment, if Figure 4 As shown, after obtaining the target prediction model, it includes: Step 191, receiving the battery operation data and driving condition data of the vehicle to be tested.

[0065] Step 192, input the battery operation data to be tested and the driving condition data to be tested into the target prediction model to obtain the battery state prediction result of the vehicle to be tested.

[0066] In this implementation, the real-time battery signal and real-time driving signal generated by the vehicle during driving are obtained. The real-time battery signal is used to measure the battery operation data to be tested of the vehicle, and the real-time driving signal is used to measure the driving condition data to be tested of the vehicle. The target prediction model can predict the potential battery problems that may occur during driving of the vehicle based on the real-time battery signal and the real-time driving signal, so as to generate an early warning plan for the abnormal state of the battery in advance. By discovering potential battery problems in advance, this embodiment can take preventive measures to reduce traffic safety hazards of the vehicle before the vehicle completely fails.

[0067] In one embodiment, when the battery state prediction result is an abnormal prediction result, the abnormal prediction result is verified.

[0068] When the battery status prediction result is an abnormal prediction result, the abnormal prediction result is verified, including: obtaining the abnormal quantile of the abnormal prediction result; when the abnormal quantile reaches a preset quantile threshold, performing abnormal verification on the abnormal prediction result based on a preset battery evaluation standard.

[0069] Combination Figure 5 The application environment diagram of verifying the abnormal prediction results is explained in detail. Figure 5 In the example, the cloud is 510 and the vehicle is 520. A target prediction model is deployed on each of the cloud 510 and the vehicle 520. After the cloud 510 completes the prediction of the battery information to be evaluated, it obtains the abnormal result, verifies the abnormal result, generates an indication S1 for the verified abnormal result, and sends part of the S1 indication to the vehicle 520 for processing. The target prediction model deployed on the vehicle 520 is used as a supplement to the cloud 510. After the vehicle 520 obtains the abnormal result, S2 sends the abnormal result to the cloud 510. The cloud 510 verifies the abnormal result and sends the abnormal indication generated by the verification result to the vehicle 520.

[0070] In this embodiment, the pressure difference abnormality prediction is performed on the cloud data or CAN bus data in real time. The target prediction model is divided into two deployment modes, one is cloud deployment and the other is vehicle-side deployment. For the target prediction model deployed on the cloud: deploy the trained target prediction model on a cloud computing platform, such as a computing cloud center; then predict the pressure difference abnormality based on the real-time data on the cloud. The target preset model deployed on the vehicle side is to directly deploy the target prediction model on the vehicle computer, generally as a supplementary solution to the target prediction model deployed on the cloud. The target preset model on the vehicle side directly receives the battery operation data and driving condition data based on the CAN bus data to predict battery abnormalities.

[0071] In this embodiment, the target prediction model is deployed in the cloud, abnormal results in the prediction results are obtained, and the abnormal results are compressed and verified, including: scoring the abnormal results; when the score of the abnormal result reaches a preset quantile threshold, the abnormal result is compressed and verified.

[0072] In one embodiment, after generating the abnormal prediction result, the method further includes determining the abnormal level of the abnormal prediction result and generating an abnormal indication corresponding to the abnormal level.

[0073] Among them, the abnormality level includes: a first abnormality level and a second abnormality level, the risk of the first abnormality level is lower than the risk of the second abnormality level, and the abnormal indication includes: a first indication and a second indication, judging the abnormality level of the abnormal result after verification, and generating abnormal indications corresponding to the abnormality levels, including: for the first abnormality level, generating a first indication, the first indication is used to instruct the battery control system to adjust itself; for the second abnormality level, generating a second indication, the second indication is used to limit the battery function and alert the user to handle the battery.

[0074] The first instruction is used to instruct the battery control system to adjust the battery pack itself, and to contact the owner to go to the store for battery maintenance by pushing information, etc. The first instruction does not limit the function of the battery.

[0075] Adjustments to the battery pack include: voltage balancing. The BMS system monitors the voltage of each battery cell and adjusts the voltage telecommunications through active or passive balancing circuits to ensure that the voltage of all batteries remains consistent and reduce the voltage difference. Temperature management: The BMS detects the temperature of each battery cell in the battery pack and adjusts the temperature through heat dissipation methods such as air cooling, liquid cooling, or phase change materials to prevent overheating or overcooling. Charge and discharge control: The BMS system adjusts the charge and discharge rate according to the state of the battery to avoid overcharging or over-discharging to delay battery life. Fault detection and isolation: The BMS system can detect faulty batteries in the battery pack and isolate them to prevent the fault from spreading to other batteries to ensure the safety of the battery pack. Record battery operating data, including: voltage, temperature, current, etc., analyze data, predict and discover potential problems in advance, and perform preventive maintenance.

[0076] For the second indication, the power is linearly limited through the power battery management system, and the feedback is sent to the TSP platform at the same time. The platform promptly contacts the customer for processing. The second indication limits the output power of the battery, which now specifically includes: Limiting the maximum discharge current: By limiting the maximum discharge current of the battery, the battery is prevented from overheating or other safety problems under high load. Reduce the charging power to prevent the battery from overheating or overcharging, thereby reducing safety hazards. Dynamically adjust the power output, such as by adjusting the temperature, voltage, and internal resistance, to dynamically adjust the power output of the battery to ensure that the battery is operating within a safe range. Furthermore, a warning signal is issued to limit the power output or cut off the power supply to prevent further damage or danger.

[0077] In one embodiment, the target prediction model can be stored in the cloud or on the vehicle side. The battery status prediction is performed with cloud prediction as the main method and vehicle side prediction as the auxiliary method. When the vehicle computer signal is good, the cloud prediction is used as the standard. When the vehicle computer signal is not good, the vehicle side prediction is used as the standard.

[0078] Cloud prediction: The cloud receives vehicle signals in real time, processes the proposal segment signals in real time, extracts the required data features, and then inputs the data into the model. The model returns the prediction results and the pressure difference abnormality score; if the pressure difference abnormality score reaches the threshold, the vehicle will be labeled with data; then the background will bring the vehicle's pressure difference abnormality-related data into the pressure difference abnormality calculation formula for verification in real time; if there is a risk of pressure difference abnormality, different treatment plans will be adopted based on the risk level. If the level is low, the battery control system will adjust the battery pack itself, and at the same time, the owner will be prompted to come to the store for battery maintenance through methods such as pushing information; if the risk level is high, the power will be linearly limited through the power battery management system, and feedback will be given to the TSP platform at the same time, and the customer will be contacted in time through the platform to come to the station for processing.

[0079] Vehicle-side prediction: Vehicle-side prediction is generally used as a supplement to cloud-side prediction. It uses edge computing to extract the characteristics of the vehicle CAN bus data or LIN bus data, and then uses the target prediction model deployed on the vehicle to predict the pressure difference abnormality. At the same time, the data with higher pressure difference abnormality scores are marked and then passed to the cloud for verification. The cloud adopts different processing strategies based on the verification results. At the same time, the vehicle is regularly upgraded through OTA multi-model.

[0080] This application can improve the sample richness of abnormal battery data through the vehicle battery operation data uploaded by multiple vehicles. Calculate the single cell voltage data according to the vehicle battery operation data; obtain the preset battery evaluation standard, evaluate the single cell voltage data based on the preset battery evaluation standard, and obtain the abnormal single cell voltage data, so as to find accurate abnormal battery information. Secondly, by training the preset prediction model according to the vehicle battery operation data corresponding to the abnormal single cell voltage data, a target prediction model is obtained, thereby obtaining a target prediction model that can predict abnormal batteries according to battery characteristics. According to the target prediction model, the real-time evaluation of the battery status of the vehicle can be completed according to the battery operation data of the vehicle to be evaluated received in real time, so as to generate filing measures for the impending battery abnormality, so as to improve the safety of the vehicle battery during driving.

[0081] like Figure 6 As shown, the present application provides a battery voltage difference abnormality detection device 60, comprising: an acquisition unit 610, a first calculation unit 620, a second calculation unit 630, a third calculation unit 640, a fourth calculation unit 650 and a detection unit 660. Among them: The acquisition unit 610 is used to acquire the battery pack voltage data of each vehicle and determine the single cell pressure difference data from the battery pack voltage data. The first calculation unit 620 is used to sort the single pressure difference data of the same vehicle in positive order to obtain a first pressure difference distribution.

[0082] The second calculation unit 630 is used to determine a first type quantile of the first pressure difference distribution, and calculate a lower edge of the single-vehicle pressure difference of the vehicle according to the first type quantile.

[0083] The third calculation unit 640 is used to determine the 50th percentile of the first pressure difference distribution, and sort the 50th percentiles of the same battery model in positive order to obtain a second pressure difference distribution.

[0084] The fourth calculation unit 650 is used to determine the second type quantile of the second pressure difference distribution, and calculate the upper edge of the group pressure difference corresponding to the vehicle according to the second type quantile.

[0085] The detection unit 660 is used to compare the lower edge of the single vehicle with the upper edge of the group pressure difference corresponding to the vehicle, and determine that the battery pressure difference of the vehicle is abnormal when the lower edge of the single vehicle pressure difference is smaller than the upper edge of the group pressure difference.

[0086] Please refer to Figure 7 , Figure 7 This is a schematic diagram of an embodiment of an electronic device of the present application.

[0087] The electronic device 100 includes a memory 20, a processor 30, and a computer program 40 stored in the memory 20 and executable on the processor 30. When the processor 30 executes the computer program 40, the steps in the above-mentioned battery voltage difference abnormality detection method embodiment are implemented, for example Figure 1 Steps 110 to 160 are shown.

[0088] Exemplarily, the computer program 40 can also be divided into one or more modules / units, one or more modules / units are stored in the memory 20 and executed by the processor 30. One or more modules / units can be a series of computer program instruction segments that can complete specific functions, and the instruction segments are used to describe the execution process of the computer program 40 in the electronic device 100. For example, it can be divided into the acquisition unit 610, the first calculation unit 620, the second calculation unit 630, the third calculation unit 640, the fourth calculation unit 650 and the detection unit 660 shown.

[0089] Those skilled in the art will appreciate that the schematic diagram is merely an example of the electronic device 100 and does not constitute a limitation on the electronic device 100 . The electronic device 100 may include more or fewer components than shown in the diagram, or a combination of certain components, or different components. For example, the electronic device 100 may also include input and output devices, network access devices, buses, etc.

[0090] The processor 30 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor, a single-chip microcomputer, or the processor 30 may also be any conventional processor, etc.

[0091] The memory 20 can be used to store the computer program 40 and / or the module / unit. The processor 30 realizes various functions of the electronic device 100 by running or executing the computer program and / or the module / unit stored in the memory 20 and calling the data stored in the memory 20. The memory 20 can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data (such as audio data) created according to the use of the electronic device 100, etc. In addition, the memory 20 can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other non-volatile solid-state storage devices.

[0092] If the module / unit integrated in the electronic device 100 is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

Claims

1. A method for detecting abnormal battery voltage difference, characterized in that: include: Acquire battery pack voltage data of each vehicle, and determine single cell pressure difference data from the battery pack voltage data; Sorting the single pressure difference data of the same vehicle in positive order to obtain a first pressure difference distribution; Determine a first type quantile of the first pressure difference distribution, and calculate a lower edge of the single-vehicle pressure difference of the vehicle according to the first type quantile; Determine the 50th percentile of the first pressure difference distribution, and sort the 50th percentiles of the same battery model in positive order to obtain a second pressure difference distribution; Determine a second type quantile of the second pressure difference distribution, and calculate an upper edge of the group pressure difference corresponding to the vehicle according to the second type quantile; The single vehicle lower edge of the vehicle is compared with the group pressure difference upper edge corresponding to the vehicle. When the single vehicle pressure difference lower edge is smaller than the group pressure difference upper edge, it is determined that the battery pressure difference of the vehicle is abnormal.

2. The method according to claim 1, characterized in that: The determining of single cell voltage difference data from the battery pack voltage data includes: Determine single cell voltage data from the battery pack voltage data, the single cell voltage data being the battery pack cell voltage; The maximum cell voltage data and the minimum cell voltage data in the cell voltage data are determined, and the cell pressure difference data of the vehicle is determined according to the maximum cell voltage data and the minimum cell voltage data.

3. The method according to claim 1, characterized in that The first type of quantiles includes: the 25th quantile of the first pressure difference distribution and the 75th quantile of the first pressure difference distribution; The calculating the lower edge of the vehicle pressure difference of the vehicle according to the first type of quantiles includes: According to a first difference between the 75th percentile and the 25th percentile; Obtaining a first coefficient, and calculating a first product of the first coefficient and the first difference; The lower edge of the bicycle pressure difference is calculated based on the difference between the 25th percentile data and the first product.

4. The method according to claim 1, characterized in that: The second type of quantiles include: the 25th quantile of the second pressure difference distribution and the 75th quantile of the second pressure difference distribution; The calculating the upper edge of the group pressure difference corresponding to the vehicle according to the second type quantile includes: determining a second difference between the 75th percentile and the 25th percentile; Obtaining a second coefficient, and calculating a second product of the second coefficient and the second difference; The upper edge of the population pressure difference is calculated based on the sum of the second product and the 75th quantile.

5. The method according to claim 1, characterized in that After determining that the battery voltage difference of the vehicle is abnormal, the method includes: Determine that there is an abnormal vehicle with abnormal battery pressure difference, and obtain battery operation data and driving condition data corresponding to the abnormal vehicle; Abnormal sample data is generated according to the battery operation data and the driving condition data.

6. The method according to claim 5, characterized in that After generating the abnormal sample data, the method includes: Generate an abnormal battery status time series and a status label according to the abnormal sample data; Inputting the abnormal battery status time series into a preset prediction model to obtain a prediction result output by the preset prediction model; According to the difference between the prediction result and the state label, the model parameters of the preset prediction model are updated to obtain the target prediction model.

7. The method according to claim 6, characterized in that After obtaining the target prediction model, the method includes: Receiving the battery operation data and driving condition data of the vehicle to be tested; The battery operation data to be detected and the vehicle operating condition data to be detected are input into the target prediction model to obtain a battery state prediction result of the vehicle to be detected.

8. A battery voltage abnormality detection device, characterized in that: include: An acquisition unit, used to acquire battery pack voltage data of each vehicle, and determine single cell pressure difference data from the battery pack voltage data; A first calculation unit, used for sorting the single pressure difference data of the same vehicle in positive order to obtain a first pressure difference distribution; a second calculation unit, configured to determine a first type quantile of the first pressure difference distribution, and calculate a lower edge of the single-vehicle pressure difference of the vehicle according to the first type quantile; a third calculation unit, configured to determine the 50th percentile of the first pressure difference distribution, and sort the 50th percentiles of the same battery model in positive order to obtain a second pressure difference distribution; A fourth calculation unit, configured to determine a second type quantile of the second pressure difference distribution, and calculate an upper edge of the group pressure difference corresponding to the vehicle according to the second type quantile; The detection unit is used to compare the lower edge of the single vehicle with the upper edge of the group pressure difference corresponding to the vehicle, and determine that the battery pressure difference of the vehicle is abnormal when the lower edge of the single vehicle pressure difference is smaller than the upper edge of the group pressure difference.

9. An electronic device, comprising a processor and a memory, characterized in that: The memory is used to store instructions, and the processor is used to call the instructions in the memory, so that the electronic device executes the battery voltage abnormality detection method according to any one of claims 1 to claim 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and when the computer instructions are executed on an electronic device, the electronic device executes the battery voltage abnormality detection method according to any one of claims 1 to 7.

Images