WO2023142685A1 - 电池绝缘故障的预警方法、装置、系统和计算机设备 - Google Patents

电池绝缘故障的预警方法、装置、系统和计算机设备 Download PDF

Info

Publication number
WO2023142685A1
WO2023142685A1 PCT/CN2022/136584 CN2022136584W WO2023142685A1 WO 2023142685 A1 WO2023142685 A1 WO 2023142685A1 CN 2022136584 W CN2022136584 W CN 2022136584W WO 2023142685 A1 WO2023142685 A1 WO 2023142685A1
Authority
WO
WIPO (PCT)
Prior art keywords
insulation
early warning
battery
stable
feature
Prior art date
Application number
PCT/CN2022/136584
Other languages
English (en)
French (fr)
Inventor
李智周
赵薇
Original Assignee
宁德时代新能源科技股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 宁德时代新能源科技股份有限公司 filed Critical 宁德时代新能源科技股份有限公司
Publication of WO2023142685A1 publication Critical patent/WO2023142685A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
    • G01R31/392—Determining battery ageing or deterioration, e.g. state of health

Definitions

  • the present application relates to the field of battery technology, in particular to a battery insulation fault early warning method, device, system, computer equipment, storage medium and computer program product.
  • the traditional insulation detection method needs to redesign the battery circuit, which increases the cost of insulation warning.
  • the traditional insulation detection method is only applicable to circuits that have already experienced insulation faults, and cannot effectively avoid insulation risks.
  • the traditional battery insulation fault detection method has a high cost and cannot effectively avoid the danger caused by the degradation of the battery insulation performance.
  • a battery insulation fault early warning method device, system, computer equipment, storage medium and computer program product are provided.
  • a method for early warning of battery insulation failure comprising:
  • the insulation stability working condition includes the working condition when the fluctuation range of the insulation value of the battery is within a preset range
  • Feature extraction is performed on the historical data of the battery according to the stable insulation condition to obtain the stable insulation feature;
  • the stable insulation feature includes the characteristics of the battery used to predict the insulation failure of the battery under the stable insulation condition;
  • the characteristic early warning threshold is determined according to the stable insulation characteristics, and/or, the insulation early warning model is generated according to the stable insulation characteristics; the characteristic early warning threshold and/or the insulation early warning model are used for early warning of the insulation failure of the battery.
  • the above-mentioned early warning method for battery insulation failure by determining the stable insulation working conditions, and obtaining the characteristics related to the insulation failure under the stable working conditions of the battery insulation value as the stable insulation characteristics, and further determining the characteristic early warning threshold and/or based on the stable insulation characteristics Generate an insulation early warning model, so that the corresponding insulation fault early warning can be carried out based on the characteristic early warning threshold and/or the insulation early warning model. Since the insulation value of the battery is relatively stable under stable insulation conditions, there will be no large fluctuations. Based on the insulation The insulation value and related parameters under stable working conditions are used to determine the characteristic early warning threshold and/or generate an insulation early warning model, and the characteristic early warning threshold that can reflect insulation faults and/or an insulation early warning model with high recognition accuracy can be obtained. Redesigning the battery circuit can realize the early warning of battery insulation failure, and reduce the cost of insulation early warning. Moreover, the insulation failure of the battery can be predicted in advance, effectively avoiding the danger caused by the decline of battery insulation performance.
  • determining the feature warning threshold according to the stable insulation feature includes:
  • the determination of the characteristic early warning threshold by means of the correlation coefficient consumes less computing resources and is more efficient, reducing the cost of insulation fault early warning.
  • determining the feature warning threshold according to the stable insulation feature includes:
  • determining the characteristic early warning threshold by means of the average distance consumes less computing resources and is more efficient, reducing the cost of insulation fault early warning.
  • determining the feature warning threshold according to the stable insulation feature includes:
  • the determination of the characteristic early warning threshold through mathematical processing consumes less computing resources and is more efficient, reducing the cost of insulation fault early warning.
  • determining a threshold corresponding to a key insulation feature as a feature warning threshold includes:
  • the threshold corresponding to the key insulation characteristics is determined as the characteristic early warning threshold.
  • determining a threshold corresponding to a key insulation feature as a feature warning threshold includes:
  • the value corresponding to the segmentation point is used as the feature early warning threshold.
  • the feature early warning threshold is determined by means of a single-layer decision tree, which can accurately and objectively determine the feature early warning threshold.
  • the insulation early warning model is generated according to the stable insulation characteristics, including:
  • each similar feature early warning model is fused to obtain an insulation early warning model.
  • an insulation early warning model for accurately identifying insulation fault risks is obtained through model training, and early warning is performed based on the insulation early warning model, which improves the accuracy of insulation fault early warning.
  • dividing key insulation features to obtain at least one feature set includes:
  • the key insulation features are divided to obtain at least one feature set.
  • the preset classification conditions include insulation failure mechanism, insulation stable working condition or feature extraction cycle.
  • key insulation features can be efficiently and accurately divided and multiple feature sets can be obtained through classification conditions including insulation failure mechanism, insulation stability working condition, or feature extraction cycle.
  • the stable insulation working conditions include the thermal management working conditions with charging turned on, the thermal management working conditions with discharging turned on, the high current discharge working conditions, the high current charging working conditions, the high state of charge discharging working conditions, and the charging gun insertion working conditions. At least one of the working conditions, the charging gun pulling out working conditions, and the working conditions in the open circuit voltage platform area.
  • the charge-on thermal management condition includes a condition in which the temperature of the charging section is greater than the first temperature
  • the discharge-on thermal management condition includes a condition in which the temperature of the discharge section is greater than the second temperature
  • the large current discharge condition includes a condition in which the discharge current is greater than the first current
  • the high-current charging condition includes a working condition in which the charging current is greater than the second current
  • the high state of charge discharge condition includes the condition that the state of charge is between the first state of charge and the second state of charge during discharge;
  • the working condition of charging gun insertion includes the working conditions of the first few frames of the charging section
  • the working condition of pulling out the charging gun includes the working conditions of the first few frames of the non-charging section;
  • the operating conditions in the open circuit voltage plateau region include the operating conditions in which the battery voltage changes slowly.
  • An early warning device for battery insulation failure comprising:
  • the data module is configured to determine the insulation stability working condition, and obtain the historical data of the battery; the insulation stability working condition includes the working condition when the fluctuation range of the insulation value of the battery is within a preset range;
  • the feature module is configured to perform feature extraction on the historical data of the battery according to the stable insulation working condition to obtain the stable insulation feature;
  • the stable insulation feature includes the characteristics of the battery used to predict the battery insulation failure under the stable insulation working condition;
  • the early warning module is configured to determine the characteristic early warning threshold according to the stable insulation characteristics, and/or generate an insulation early warning model according to the stable insulation characteristics; the characteristic early warning threshold and/or the insulation early warning model are used for early warning of the insulation failure of the battery.
  • the above-mentioned early warning device for battery insulation fault determines the stable insulation working condition, and obtains the characteristics related to the insulation failure under the stable insulation value of the battery as the stable insulation characteristic, and then further determines the characteristic early warning threshold and/or based on the stable insulation characteristic. Generate an insulation early warning model, so that the corresponding insulation fault early warning can be carried out based on the characteristic early warning threshold and/or the insulation early warning model. Since the insulation value of the battery is relatively stable under stable insulation conditions, there will be no large fluctuations. Based on the insulation The insulation value and related parameters under stable working conditions are used to determine the characteristic early warning threshold and/or generate an insulation early warning model, and the characteristic early warning threshold that can reflect insulation faults and/or an insulation early warning model with high recognition accuracy can be obtained. Redesigning the battery circuit can realize the early warning of battery insulation failure, and reduce the cost of insulation early warning. Moreover, the insulation failure of the battery can be predicted in advance, effectively avoiding the danger caused by the decline of battery insulation performance.
  • a battery insulation fault early warning system comprising: a battery management terminal and an insulation fault early warning background;
  • the battery management terminal is configured to collect the battery data of the vehicle battery, and send the collected battery data to the insulation fault early warning background for the insulation fault early warning background to extract historical battery data from the battery data;
  • the insulation fault early warning background is configured to extract the characteristics of the historical data of the battery according to the determined insulation stability working conditions to obtain stable insulation characteristics; the insulation stability working conditions include the working conditions when the fluctuation range of the insulation value of the battery is within a preset range
  • the stable insulation characteristics include the characteristics of the battery used to predict the insulation failure of the battery under stable insulation conditions; the characteristic early warning threshold is determined according to the stable insulation characteristics, and/or, the insulation early warning model is generated according to the stable insulation characteristics; the characteristic early warning threshold and/or The insulation early warning model is used for early warning of battery insulation failure.
  • the battery management terminal can send the collected battery data to the insulation fault early warning background for the insulation fault early warning background to extract historical battery data from the battery data.
  • the insulation fault early warning background can determine the insulation stability working condition, And obtain the features related to the insulation fault under the condition of stable insulation value of the battery as the stable insulation feature, and then further determine the feature warning threshold and/or generate the insulation warning model based on the stable insulation feature, so that the feature warning threshold and/or insulation
  • the early warning model carries out the corresponding insulation fault early warning. Since the insulation value of the battery is relatively stable under the stable insulation condition, there will not be large fluctuations. Based on the insulation value and related parameters under the stable insulation condition, the characteristic early warning is determined.
  • Threshold and/or generate an insulation early warning model the characteristic early warning threshold that can reflect the insulation fault and/or an insulation early warning model with high recognition accuracy can be obtained, so that the early warning of the battery insulation fault can be realized without redesigning the battery circuit, The cost of insulation early warning is reduced, and the insulation failure of the battery can be predicted in advance, effectively avoiding the danger caused by the degradation of the insulation performance of the battery.
  • a computer device comprising a memory and one or more processors, wherein computer readable instructions are stored in the memory, and when executed by the one or more processors, the computer readable instructions cause the one or more processors to perform the following steps: determine Insulation stable working conditions, and obtaining battery historical data; Insulation stable working conditions include working conditions when the fluctuation range of the insulation value of the battery is within a preset range;
  • Feature extraction is performed on the historical data of the battery according to the stable insulation condition to obtain the stable insulation feature;
  • the stable insulation feature includes the characteristics of the battery used to predict the insulation failure of the battery under the stable insulation condition;
  • the characteristic early warning threshold according to the stable insulation characteristics, and/or generate an insulation early warning model according to the stable insulation characteristics; the characteristic early warning threshold and/or the insulation early warning model are used for early warning of the insulation failure of the battery.
  • One or more computer storage media storing computer readable instructions that, when executed by one or more processors, cause the one or more processors to perform the following steps:
  • the insulation stability working condition includes the working condition when the fluctuation range of the insulation value of the battery is within a preset range
  • Feature extraction is performed on the historical data of the battery according to the stable insulation condition to obtain the stable insulation feature;
  • the stable insulation feature includes the characteristics of the battery used to predict the insulation failure of the battery under the stable insulation condition;
  • the characteristic early warning threshold according to the stable insulation characteristics, and/or generate an insulation early warning model according to the stable insulation characteristics; the characteristic early warning threshold and/or the insulation early warning model are used for early warning of the insulation failure of the battery.
  • a computer program product comprising a computer program which, when executed by one or more processors, causes the one or more processors to perform the following steps:
  • the insulation stability working condition includes the working condition when the fluctuation range of the insulation value of the battery is within a preset range
  • Feature extraction is performed on the historical data of the battery according to the stable insulation condition to obtain the stable insulation feature;
  • the stable insulation feature includes the characteristics of the battery used to predict the insulation failure of the battery under the stable insulation condition;
  • the characteristic early warning threshold according to the stable insulation characteristics, and/or generate an insulation early warning model according to the stable insulation characteristics; the characteristic early warning threshold and/or the insulation early warning model are used for early warning of the insulation failure of the battery.
  • the above-mentioned computer equipment, computer-readable storage medium and computer program product by determining the insulation stable working conditions, and obtaining the characteristics related to the insulation fault under the working conditions of the stable insulation value of the battery as the stable insulation characteristics, and further determining based on the stable insulation characteristics
  • the characteristic early warning threshold and/or generate the insulation early warning model so that the corresponding insulation fault early warning can be carried out based on the characteristic early warning threshold and/or the insulation early warning model. Since the insulation value of the battery is relatively stable under the condition of stable insulation, there will be no large fluctuations.
  • the characteristic early warning threshold that can reflect insulation faults and/or insulation with high recognition accuracy can be obtained Early warning model, so that the early warning of battery insulation failure can be realized without redesigning the battery circuit, the cost of insulation early warning is reduced, and the insulation failure of the battery can be predicted in advance, effectively avoiding the failure caused by the decline of battery insulation performance. Danger.
  • Fig. 1 is the application environment diagram of the early warning method of battery insulation failure in some embodiments
  • Fig. 2 is a schematic flowchart of a battery insulation fault early warning method in some embodiments
  • Fig. 3a is a schematic diagram of insulation fault early warning based on characteristic early warning thresholds in some embodiments
  • Fig. 3b is a schematic diagram of insulation fault early warning based on an insulation early warning model in some embodiments
  • Fig. 4 is a structural block diagram of a battery insulation fault early warning device in some embodiments.
  • Fig. 5 is a structural block diagram of a battery insulation fault early warning system in some embodiments.
  • Figure 6 is a diagram of the internal structure of a computer device in some embodiments.
  • the early warning method for battery insulation failure can be applied to the application environment shown in FIG. 1 .
  • the battery management terminal 102 communicates with the insulation fault early warning background 104 through the network.
  • the data storage system can store the data required by the insulation fault early warning background 104 to execute the battery insulation fault early warning method.
  • the data storage system can be integrated on the server 104, or placed on the cloud or other network servers.
  • the battery management terminal 102 may specifically be a terminal equipped with a battery management system (Battery Management System, BMS). For example, it can be an on-board battery management terminal deployed on a new energy vehicle equipped with a battery management system.
  • BMS Battery Management System
  • the battery management system is mainly used to intelligently manage and maintain each battery unit, prevent the battery from overcharging and overdischarging, prolong the service life of the battery, and monitor the state of the battery.
  • the insulation fault early warning background 104 can be implemented by an independent server or a server cluster composed of multiple servers.
  • a battery insulation fault early warning method is provided.
  • the application of the method to the insulation fault early warning background 104 in FIG. 1 is used as an example for illustration, including the following steps:
  • Step S202 determining the stable insulation condition, and acquiring battery history data.
  • the above-mentioned stable insulation working condition includes the working condition when the fluctuation range of the insulation value of the battery is within a preset range.
  • the insulation value fluctuates within the range of 0-60K ⁇ under certain working conditions.
  • the insulation value fluctuates within the preset range, that is, the insulation value of the battery tends to be stable.
  • the insulation failure early warning background 104 can combine historical data and insulation failure mechanism (for example, high humidity in the box leads to insulation failure, insulation failure causes insulation failure) to determine the working condition that the battery insulation value tends to be stable, as the above-mentioned insulation stable working condition .
  • the insulation stability working conditions may include charging-on thermal management working conditions, discharging-on thermal management working conditions, high-current discharging working conditions, high-current charging working conditions, and high state of charge (State of Charge, SOC) discharging.
  • the charging start thermal management working conditions may include that the temperature of the charging section is greater than the first A working condition of a temperature T1 ;
  • the discharge opening thermal management working condition may include a working condition in which the temperature of the discharge section is greater than the second temperature T2 ;
  • a large current discharge working condition may include a working condition in which the discharge current is greater than the first current I1 ;
  • the high current The charging condition can include a working condition in which the charging current is greater than the second current I 2 ;
  • the high state of charge discharge working condition can include a working condition in which the state of charge is between the first state of charge and the second state of charge when discharging;
  • charging The working condition of the gun insertion can include the working conditions of the first several frames (for example, the first n frames, n>1) of the charging section;
  • the charging gun pull-out working condition includes the working conditions of the first few frames of the
  • T 1 and T 2 can be different temperature values, also can be the same temperature value;
  • I 1 and I 2 can be different current values, also can be the same current value;
  • the aforementioned battery history data may include data generated when the battery is in operation. For example, the temperature of the battery when it is charged, the current when it is discharged, etc.
  • the battery management terminal 102 can collect battery data for the battery, and upload the battery data to the data storage system.
  • the insulation fault early warning background 104 can extract required data from the battery data stored in the data storage system as the above-mentioned battery history data.
  • step S204 feature extraction is performed on the historical data of the battery according to the stable insulation condition to obtain the stable insulation feature; the stable insulation feature includes the feature used to predict the battery insulation failure under the stable insulation condition of the battery.
  • the insulation fault early warning background 104 can extract features that can be used to predict insulation faults under stable insulation conditions from the battery history data, as the above-mentioned stable insulation features.
  • the historical data of the battery includes multiple working conditions and their corresponding insulation value, voltage and other data, determine the insulation stable working conditions in multiple working conditions, and then further determine the voltage change rate according to the data corresponding to the stable insulation working conditions , insulation value change rate, voltage drop and other stable insulation characteristics.
  • the insulation value of the battery is relatively stable, and there will be no large fluctuations.
  • the insulation fault warning based on the insulation value under stable insulation conditions can improve the accuracy of early warning.
  • the stable insulation features may include insulation value average, insulation value quantile, insulation value extremum, insulation value standard deviation, voltage change rate, insulation value change rate, voltage difference, temperature difference, current, abnormal frame number at least one of the Those skilled in the art can understand that in practical applications, other parameters or mathematical statistics methods can also be used to obtain stable insulation features. limit.
  • Step 206 determining a characteristic early warning threshold according to the stable insulation characteristics, and/or generating an insulation early warning model according to the stable insulation characteristics; the characteristic early warning threshold and/or the insulation early warning model are used for early warning of the insulation failure of the battery.
  • the insulation fault early warning background 104 can determine the characteristic early warning threshold based on the stable insulation characteristics. Therefore, in the real-time insulation fault early warning, the insulation fault early warning can be realized based on the comparison result of the real-time battery data and the characteristic early warning threshold. For example, a threshold corresponding to a stable insulation feature can be set according to historical data (the value of a certain stable insulation feature when an insulation fault occurs), as the aforementioned feature warning threshold.
  • the insulation fault early warning background 104 can also train an insulation early warning model based on stable insulation characteristics, so that during real-time insulation fault early warning, the real-time data of the battery can be input into the insulation early warning model, and according to the output results of the insulation early warning model Carry out insulation fault early warning.
  • the battery insulation fault early warning method by determining the insulation stable working condition, and obtaining the characteristics related to the insulation fault under the stable insulation value of the battery as the stable insulation characteristic, and further determining the characteristic early warning threshold and/or based on the stable insulation characteristic Or generate an insulation early warning model, so that the corresponding insulation fault early warning can be carried out based on the characteristic early warning threshold and/or the insulation early warning model. Since the insulation value of the battery is relatively stable under the condition of stable insulation, there will be no large fluctuations.
  • the insulation value and related parameters under stable insulation conditions are used to determine the characteristic early warning threshold and/or generate an insulation early warning model, and the characteristic early warning threshold that can reflect insulation faults and/or an insulation early warning model with high identification accuracy can be obtained, thus,
  • the early warning of battery insulation failure can be realized without redesigning the battery circuit, the cost of insulation early warning is reduced, and the insulation failure of the battery can be predicted in advance, effectively avoiding the danger caused by the decline of the battery insulation performance.
  • determining the feature warning threshold according to the stable insulation feature includes:
  • the key insulation features are selected from the stable insulation features through the correlation coefficient and/or the mean distance; the threshold corresponding to the key insulation features is determined as the feature early warning threshold.
  • the insulation fault early warning background 104 can select the characteristics that can fully reflect the insulation fault from various stable insulation characteristics as the above-mentioned key insulation characteristics through processing methods such as correlation coefficient and mean distance.
  • the average insulation value before the battery insulation failure is 10000 ⁇ when the thermal management is turned on during charging, and the average insulation value without insulation failure is 16000 ⁇ ; while the average insulation value before the insulation failure is It is 15000 ⁇ , and the average insulation value without insulation fault is 17000 ⁇ .
  • the difference between the mean value of the insulation value before and after the fault under the condition of charging with thermal management turned on and the condition of charging with thermal management not turned on is 6000 ⁇ and 2000 ⁇ respectively, indicating that the characteristics of the average value of the insulation value under the condition of charging with thermal management turned on are more correlated with insulation faults. Therefore, according to the average distance, the stable insulation characteristics of the charging-on thermal management condition are selected as the key insulation characteristics.
  • the candidate characteristic early warning threshold is selected in a cycle from 16000 ⁇ downwards for testing until the test The result satisfies the false alarm rate and the early warning accuracy rate until the preset value is reached.
  • a single-layer decision tree can be used to find the segmentation point with the largest change in the Gini coefficient, and the value corresponding to the segmentation point can be used as the early warning threshold of the feature.
  • Fig. 3a is a schematic diagram of insulation fault pre-warning based on characteristic pre-warning thresholds in some embodiments.
  • the insulation failure early warning background can obtain a large number of stable insulation characteristics based on the insulation failure mechanism such as high humidity in the box, and then filter out key insulation characteristics through mathematical statistics such as mean distance and correlation coefficient. And determine the corresponding feature early warning threshold. Therefore, when performing early warning, the insulation fault early warning background can perform insulation fault early warning according to the comparison result of the real-time battery data and the characteristic early warning threshold.
  • the determination of the characteristic early warning threshold through mathematical processing consumes less computing resources and is more efficient, reducing the cost of insulation fault early warning.
  • the insulation early warning model is generated according to the stable insulation characteristics, including:
  • the insulation fault early warning background 104 can select the characteristics that can fully reflect the insulation fault from various stable insulation characteristics as the above-mentioned key insulation characteristics through processing methods such as correlation coefficient and mean distance.
  • the key insulation features can be divided into multiple feature sets, and the decision tree and support vector machines (support vector machines, SVM) can be used for each feature set. ) or other machine learning model training, thus obtaining the corresponding model of each feature set, as the above-mentioned similar feature early warning model.
  • SVM support vector machines
  • multiple similar feature models can be fused based on the output results of the same feature early warning model to obtain the final insulation early warning model.
  • Those skilled in the art can design a fusion method according to needs, for example, fusion based on a voting mechanism, linear weighted fusion, and the like.
  • the feature set of the charging section can include the key insulation features of the charging start thermal management condition, the high current charging condition, and the charging gun insertion condition
  • the feature set of the discharging section can include the discharge Turn on the key insulation characteristics of thermal management conditions, high current discharge conditions, and charging gun pull-out conditions.
  • Two feature sets are used to train the decision tree model respectively, and the similar feature early warning model_charging section and the similar feature early warning model_discharging section are obtained. Finally, the similar feature early warning model_charging section and the similar feature early warning model_discharging section are fused.
  • Generate the final insulation warning model for example, use the "or" judgment logic to output the final warning result for the output results of the two models of the same kind of feature warning model_charging section and the same kind of feature warning model_discharging section.
  • Fig. 3b is a schematic diagram of an embodiment of an insulation fault early warning based on an insulation early warning model.
  • the insulation fault early warning background can obtain a large number of stable insulation features based on the insulation fault mechanism such as high humidity in the box leading to insulation, etc., and then classify according to the insulation fault mechanism, insulation stable working conditions, and feature extraction cycle
  • the key insulation features are divided into multiple feature sets, and model training is performed on each feature set to obtain the corresponding early warning models of the feature sets.
  • the various early warning models are fused to obtain the final insulation early warning model. Therefore, when performing early warning, the insulation fault early warning background can input the real-time data of the battery into the insulation early warning model, and perform insulation fault early warning according to the output result of the insulation early warning model.
  • an insulation early warning model that accurately identifies the risk of insulation fault is obtained through model training, and the early warning is performed based on the insulation early warning model, which improves the accuracy of the insulation fault early warning.
  • steps in the flow charts involved in the above embodiments are shown sequentially as indicated by the arrows, these steps are not necessarily executed sequentially in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least some of the steps in the flow charts involved in the above-mentioned embodiments may include multiple steps or stages, and these steps or stages are not necessarily executed at the same time, but may be performed at different times For execution, the execution order of these steps or stages is not necessarily performed sequentially, but may be executed in turn or alternately with other steps or at least a part of steps or stages in other steps.
  • an embodiment of the present application further provides a battery insulation fault early warning device for implementing the above-mentioned battery insulation fault early warning method.
  • the solution to the problem provided by this device is similar to the implementation described in the above method, so the specific limitations in one or more embodiments of the battery insulation fault early warning device provided below can be referred to above for the battery insulation fault early warning The limitation of the method will not be repeated here.
  • a battery insulation fault early warning device 400 including: a data module 402, a feature module 404 and an early warning module 406, wherein:
  • the data module 402 is configured to determine the insulation stability working condition, and obtain battery historical data; the insulation stability working condition includes the working condition when the fluctuation range of the insulation value of the battery is within a preset range;
  • the feature module 404 is configured to perform feature extraction on the historical data of the battery according to the stable insulation working condition to obtain the stable insulation feature;
  • the stable insulation feature includes the feature used to predict the battery insulation failure of the battery under the stable insulation working condition;
  • the early warning module 406 is configured to determine a characteristic early warning threshold according to the stable insulation characteristics, and/or generate an insulation early warning model according to the stable insulation characteristics; the characteristic early warning threshold and/or the insulation early warning model are used for early warning of the insulation failure of the battery.
  • the battery insulation fault early warning device by determining the insulation stable working condition, and obtaining the characteristics related to the insulation fault under the stable insulation value of the battery as the stable insulation characteristic, and further determining the characteristic early warning threshold and/or based on the stable insulation characteristic Or generate an insulation early warning model, so that the corresponding insulation fault early warning can be carried out based on the characteristic early warning threshold and/or the insulation early warning model. Since the insulation value of the battery is relatively stable under the condition of stable insulation, there will be no large fluctuations.
  • the insulation value and related parameters under stable insulation conditions are used to determine the characteristic early warning threshold and/or generate an insulation early warning model, and the characteristic early warning threshold that can reflect insulation faults and/or an insulation early warning model with high identification accuracy can be obtained, thus,
  • the early warning of battery insulation failure can be realized without redesigning the battery circuit, the cost of insulation early warning is reduced, and the insulation failure of the battery can be predicted in advance, effectively avoiding the danger caused by the decline of the battery insulation performance.
  • the early warning module 406 is further configured to: select a key insulation feature from stable insulation features through the correlation coefficient and/or mean distance; determine a threshold corresponding to the key insulation feature as the feature early warning threshold.
  • the early warning module 406 is further configured to: perform periodic statistics on key insulation features to obtain statistical results; based on the statistical results, determine thresholds corresponding to key insulation features as feature early warning thresholds.
  • the early warning module 406 is further configured to: use a single-layer decision tree to find the segmentation point where the Gini coefficient of the key insulation feature changes the most; use the value corresponding to the segmentation point as the feature early warning threshold.
  • the early warning module 406 is further configured to: select key insulation features from stable insulation features; divide key insulation features to obtain at least one feature set; perform model training based on each feature set, and obtain the similar features corresponding to each feature set.
  • Feature early warning model According to the output results of each similar feature early warning model, each similar feature early warning model is fused to obtain an insulation early warning model.
  • the early warning module 406 is further configured to: classify key insulation features according to preset classification conditions to obtain at least one feature set, where the preset classification conditions include insulation failure mechanism, insulation stability working condition or feature extraction period .
  • the stable insulation working conditions include the thermal management working conditions with charging turned on, the thermal management working conditions with discharging turned on, the high current discharge working conditions, the high current charging working conditions, the high state of charge discharging working conditions, and the charging gun insertion working conditions. At least one of the working conditions, the charging gun pulling out working conditions, and the working conditions in the open circuit voltage platform area.
  • the charge-on thermal management condition includes a condition in which the temperature of the charging section is greater than the first temperature; the discharge-on thermal management condition includes a condition in which the temperature of the discharge section is greater than the second temperature; the high-current discharge condition includes a discharge current The working condition greater than the first current; the high current charging condition includes the charging current greater than the second current working condition; the high state of charge discharge condition includes the state of charge when discharging is between the first state of charge and the second state of charge
  • the charging gun insertion working condition includes the working conditions of the first few frames of the charging section; the charging gun pull-out working condition includes the working conditions of the first few frames of the non-charging section; the working condition of the open circuit voltage platform area includes the slow change of the battery voltage working conditions.
  • the stable insulation features include insulation value mean, insulation value quantile, insulation value extremum, insulation value standard deviation, voltage change rate, insulation value change rate, voltage difference, temperature difference, current, abnormal frame number at least one of the
  • Each module in the above-mentioned battery insulation fault early warning device can be fully or partially realized by software, hardware and a combination thereof.
  • the above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, and can also be stored in the memory of the computer device in the form of software, so that the processor can invoke and execute the corresponding operations of the above-mentioned modules.
  • an embodiment of the present application further provides a battery insulation fault early warning system for implementing the above-mentioned battery insulation fault early warning method.
  • the solution to the problem provided by the system is similar to the implementation described in the above method, so the specific limitations in one or more embodiments of the battery insulation fault early warning system provided below can be referred to above for the battery insulation fault early warning The limitation of the method will not be repeated here.
  • a battery insulation fault early warning system 500 including: a battery management terminal 502 and an insulation fault early warning background 504, wherein:
  • the battery management terminal 502 is configured to collect battery data of the vehicle battery, and send the collected battery data to the insulation fault early warning background for the insulation fault early warning background to extract historical battery data from the battery data;
  • the insulation fault early warning background 504 is configured to perform feature extraction on the historical data of the battery according to the determined insulation stability working conditions to obtain stable insulation characteristics; the insulation stability working conditions include working conditions when the fluctuation range of the insulation value of the battery is within a preset range
  • the stable insulation characteristics include the characteristics of the battery used to predict the insulation failure of the battery under the stable insulation condition; the characteristic early warning threshold is determined according to the stable insulation characteristics, and/or, the insulation early warning model is generated according to the stable insulation characteristics; the characteristic early warning threshold and/or Or the insulation early warning model is used for early warning of battery insulation failure.
  • a computer device is provided.
  • the computer device may be a server, and its internal structure may be as shown in FIG. 6 .
  • the computer device includes a processor, memory and a network interface connected by a system bus. Wherein, the processor of the computer device is used to provide calculation and control capabilities.
  • the memory of the computer device includes a non-volatile storage medium and an internal memory.
  • the non-volatile storage medium stores an operating system, computer programs and databases.
  • the internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium.
  • the computer device's database is used to store battery data.
  • the network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a battery insulation fault warning method is realized.
  • FIG. 6 is only a block diagram of a part of the structure related to the solution of this application, and does not constitute a limitation on the computer equipment to which the solution of this application is applied.
  • the specific computer equipment can be More or fewer components than shown in the figures may be included, or some components may be combined, or have a different arrangement of components.
  • a computer device including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the following steps when executing the computer program: determining insulation stability conditions, and acquiring battery history data; Insulation stable working conditions include working conditions when the fluctuation range of the battery’s insulation value is within a preset range; feature extraction is performed on the historical data of the battery according to the insulation stable working conditions to obtain stable insulation characteristics; stable insulation characteristics include the battery in the insulation stable working conditions The characteristics used to predict the battery insulation fault under the following characteristics; determine the characteristic early warning threshold according to the stable insulation characteristics, and/or generate an insulation early warning model according to the stable insulation characteristics; the characteristic early warning threshold and/or the insulation early warning model are used to predict the insulation fault of the battery early warning.
  • the processor when the processor executes the computer program, the following steps are also implemented: selecting key insulation features from the stable insulation features through the correlation coefficient and/or mean distance; determining the threshold corresponding to the key insulation feature as the feature warning threshold.
  • the processor when the processor executes the computer program, the following steps are also implemented: selecting key insulation features from stable insulation features; dividing key insulation features to obtain at least one feature set; performing model training based on each feature set to obtain each feature set Corresponding early warning models of the same kind of characteristics; according to the output results of the early warning models of the same kind of characteristics, the early warning models of the same kind of characteristics are fused to obtain the insulation early warning model.
  • the stable insulation working conditions include the thermal management working conditions with charging turned on, the thermal management working conditions with discharging turned on, the high current discharge working conditions, the high current charging working conditions, the high state of charge discharging working conditions, and the charging gun insertion working conditions. At least one of the working conditions, the charging gun pulling out working conditions, and the working conditions in the open circuit voltage platform area.
  • the charge-on thermal management condition includes a condition in which the temperature of the charging section is greater than the first temperature; the discharge-on thermal management condition includes a condition in which the temperature of the discharge section is greater than the second temperature; the high-current discharge condition includes a discharge current The working condition greater than the first current; the high current charging condition includes the charging current greater than the second current working condition; the high state of charge discharge condition includes the state of charge when discharging is between the first state of charge and the second state of charge
  • the charging gun insertion working condition includes the working conditions of the first few frames of the charging section; the charging gun pull-out working condition includes the working conditions of the first few frames of the non-charging section; the working condition of the open circuit voltage platform area includes the slow change of the battery voltage working conditions.
  • a computer-readable storage medium on which a computer program is stored.
  • the following steps are implemented: determining the insulation stability working condition, and obtaining battery historical data;
  • the condition includes the working condition when the fluctuation range of the insulation value of the battery is within the preset range;
  • the feature extraction of the historical data of the battery is carried out according to the stable insulation condition, and the stable insulation feature is obtained;
  • the stable insulation feature includes the usage of the battery under the stable insulation condition
  • the following steps are also implemented: selecting key insulation features from stable insulation features through the correlation coefficient and/or mean distance; determining the threshold corresponding to the key insulation feature as a feature warning threshold .
  • the following steps are also implemented: selecting key insulation features from stable insulation features; dividing key insulation features to obtain at least one feature set; performing model training based on each feature set to obtain each feature Collect the corresponding early warning models of the same kind of characteristics; according to the output results of the early warning models of the same kind of characteristics, the early warning models of the same kind of characteristics are fused to obtain the insulation early warning model.
  • the stable insulation working conditions include the thermal management working conditions with charging turned on, the thermal management working conditions with discharging turned on, the high current discharge working conditions, the high current charging working conditions, the high state of charge discharging working conditions, and the charging gun insertion working conditions. At least one of the working conditions, the charging gun pulling out working conditions, and the working conditions in the open circuit voltage platform area.
  • the charge-on thermal management condition includes a condition in which the temperature of the charging section is greater than the first temperature; the discharge-on thermal management condition includes a condition in which the temperature of the discharge section is greater than the second temperature; the high-current discharge condition includes a discharge current The working condition greater than the first current; the high current charging condition includes the charging current greater than the second current working condition; the high state of charge discharge condition includes the state of charge when discharging is between the first state of charge and the second state of charge
  • the charging gun insertion working condition includes the working conditions of the first few frames of the charging section; the charging gun pull-out working condition includes the working conditions of the first few frames of the non-charging section; the working condition of the open circuit voltage platform area includes the slow change of the battery voltage working conditions.
  • user information including but not limited to user equipment information, user personal information, etc.
  • data including but not limited to data used for analysis, stored data, displayed data, etc.
  • the computer-readable instructions can be stored in a non-volatile computer
  • the computer-readable instructions may include the processes of the embodiments of the above-mentioned methods when executed.
  • any reference to storage, database or other media used in the various embodiments provided in the present application may include at least one of non-volatile and volatile storage.
  • Non-volatile memory can include read-only memory (Read-Only Memory, ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive variable memory (ReRAM), magnetic variable memory (Magnetoresistive Random Access Memory, MRAM), Ferroelectric Random Access Memory (FRAM), Phase Change Memory (Phase Change Memory, PCM), graphene memory, etc.
  • the volatile memory may include random access memory (Random Access Memory, RAM) or external cache memory, etc.
  • RAM Random Access Memory
  • RAM Random Access Memory
  • RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).
  • the databases involved in the various embodiments provided in this application may include at least one of a relational database and a non-relational database.
  • the non-relational database may include a blockchain-based distributed database, etc., but is not limited thereto.
  • the processors involved in the various embodiments provided by this application can be general-purpose processors, central processing units, graphics processors, digital signal processors, programmable logic devices, data processing logic devices based on quantum computing, etc., and are not limited to this.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Charge And Discharge Circuits For Batteries Or The Like (AREA)
  • Secondary Cells (AREA)

Abstract

一种电池绝缘故障的预警方法，包括:确定绝缘稳定工况以及获取电池历史数据（S202）；根据绝缘稳定工况对电池历史数据进行特征提取，得到稳定绝缘特征（S204）；根据稳定绝缘特征确定特征预警阈值，和/或，根据稳定绝缘特征生成绝缘预警模型（S206）。采用本方法能够有效避免由于电池绝缘性能下降而导致的危险。一种电池绝缘故障的预警装置、系统和计算机设备。

Description

电池绝缘故障的预警方法、装置、系统和计算机设备
相关申请的交叉引用
本申请引用于2022年1月29日递交的名称为“电池绝缘故障的预警方法、装置、系统和计算机设备”的第202210110117X号中国专利申请，其通过引用被全部并入本申请。
技术领域
本申请涉及电池技术领域，特别是涉及一种电池绝缘故障的预警方法、装置、系统、计算机设备、存储介质和计算机程序产品。
背景技术
随着新能源电动汽车的发展，动力电池逐步取代传统车辆的油箱，成为新能源等电动汽车的关键零部件之一。然而，动力电池的绝缘介质可能会因为环境潮湿、使用不当等原因而导致老化，影响绝缘性能，可能由此而产生漏电等危险。因此，有必要对电池的绝缘性能进行检测。
然而，传统的绝缘检测方法需要对电池电路进行重新设计，增加了绝缘预警的成本，而且，传统的绝缘检测方法只适用于已经出现绝缘故障的电路，无法有效避免绝缘风险。
因此，传统的电池绝缘故障检测方法存在着成本较高且无法有效避免因电池绝缘性能下降而导致的危险。
发明内容
根据本申请公开的各种实施例，提供一种电池绝缘故障的预警方法、装置、系统、计算机设备、存储介质和计算机程序产品。
一种电池绝缘故障的预警方法，包括：
确定绝缘稳定工况，以及，获取电池历史数据；绝缘稳定工况包括电池的绝缘值的波动幅度处于预设范围内时的工况；
根据绝缘稳定工况对电池历史数据进行特征提取，得到稳定绝缘特征；稳定绝缘特征包括电池在绝缘稳定工况下的用于预测电池绝缘故障的特征；及
根据稳定绝缘特征确定特征预警阈值，和/或，根据稳定绝缘特征生成绝缘预警模型；特 征预警阈值和/或绝缘预警模型用于对电池的绝缘故障进行预警。
上述电池绝缘故障的预警方法，通过确定绝缘稳定工况，并获取电池绝缘值稳定的工况下的与绝缘故障相关的特征作为稳定绝缘特征，再进一步基于稳定绝缘特征确定特征预警阈值和/或生成绝缘预警模型，从而可以基于特征预警阈值和/或绝缘预警模型进行相应的绝缘故障预警，由于在绝缘稳定工况下电池的绝缘值较为稳定，不会出现波动幅度较大的变化，基于绝缘稳定工况下的绝缘值和相关的参数来确定特征预警阈值和/或生成绝缘预警模型，可以得到能够反映绝缘故障的特征预警阈值和/或识别准确率较高的绝缘预警模型，从而，无须对电池电路进行重新设计即可实现电池绝缘故障的预警，进行绝缘预警的成本降低，而且，可以提前预测到电池的绝缘故障，有效避免了由于电池绝缘性能下降而导致的危险。
在一些实施例中，根据稳定绝缘特征确定特征预警阈值，包括：
通过相关性系数在稳定绝缘特征中选取关键绝缘特征；及
确定关键绝缘特征所对应的阈值，作为特征预警阈值。
上述实施例中，通过相关性系数的方式确定特征预警阈值所耗费计算资源较少且效率较高，降低了绝缘故障预警的成本。
在一些实施例中，根据稳定绝缘特征确定特征预警阈值，包括：
通过均值距离在稳定绝缘特征中选取关键绝缘特征；及
确定关键绝缘特征所对应的阈值，作为特征预警阈值。
上述实施例中，通过均值距离的方式确定特征预警阈值所耗费计算资源较少且效率较高，降低了绝缘故障预警的成本。
在一些实施例中，根据稳定绝缘特征确定特征预警阈值，包括：
通过相关性系数和均值距离在稳定绝缘特征中选取关键绝缘特征；及
确定关键绝缘特征所对应的阈值，作为特征预警阈值。
上述实施例中，通过数学处理的方式确定特征预警阈值所耗费计算资源较少且效率较高，降低了绝缘故障预警的成本。
在一些实施例中，确定关键绝缘特征所对应的阈值，作为特征预警阈值包括：
对关键绝缘特征进行周期性统计，得到统计结果；及
基于统计结果，确定关键绝缘特征对应的阈值，作为特征预警阈值。
上述实施例中，通过对关键绝缘特征进行周期性统计，进而基于统计结果确定特征预警阈值，能够减少计算资源且效率较高，降低了绝缘故障预警的成本。
在一些实施例中，确定关键绝缘特征所对应的阈值，作为特征预警阈值包括：
通过单层决策树查找关键绝缘特征的基尼系数变化最大的分割点；及
以分割点对应的数值作为特征预警阈值。
上述实施例中，通过单层决策树的方式确定特征预警阈值，能够准确客观地确定特征预警阈值。
在一些实施例中，根据稳定绝缘特征生成绝缘预警模型，包括：
在稳定绝缘特征中选取关键绝缘特征；
划分关键绝缘特征，得到至少一个特征集合；
基于各个特征集合进行模型训练，得到各个特征集合对应的同类特征预警模型；及
根据各个同类特征预警模型的输出结果，对各个同类特征预警模型进行融合，得到绝缘预警模型。
上述实施例中，通过模型训练的方式得到准确识别绝缘故障风险的绝缘预警模型，基于该绝缘预警模型进行预警，提升了绝缘故障预警的准确率。
在一些实施例中，划分关键绝缘特征，得到至少一个特征集合包括：
根据预设的分类条件，划分关键绝缘特征，得到至少一个特征集合，预设的分类条件包括绝缘故障机理、绝缘稳定工况或特征提取周期。
上述实施例中，通过包括绝缘故障机理、绝缘稳定工况或特征提取周期在内的分类条件，能够高效且准确的划分关键绝缘特征，得到多个特征集合。
在一些实施例中，绝缘稳定工况，包括充电开启热管理工况、放电开启热管理工况、大电流放电工况、大电流充电工况、高荷电状态放电工况、充电枪插入工况、充电枪拔出工况、开路电压平台区工况中的至少一种。
上述实施例中，通过确定诸多种类到的绝缘稳定工况，能够实现在不同的绝缘稳定工况下获取稳定绝缘特征，实现绝缘故障的预警。
在一些实施例中，充电开启热管理工况包括充电段温度大于第一温度的工况；
放电开启热管理工况包括放电段温度大于第二温度的工况；
大电流放电工况包括放电电流大于第一电流的工况；
大电流充电工况包括充电电流大于第二电流的工况；
高荷电状态放电工况包括放电时荷电状态在第一荷电状态和第二荷电状态之间的工况；
充电枪插入工况包括充电段的前若干帧的工况；
充电枪拔出工况包括非充电段的前若干帧的工况；及
开路电压平台区工况包括电池电压变化缓慢的工况。
一种电池绝缘故障的预警装置，包括：
数据模块，被配置为确定绝缘稳定工况，以及，获取电池历史数据；绝缘稳定工况包括电池的绝缘值的波动幅度处于预设范围内时的工况；
特征模块，被配置为根据绝缘稳定工况对电池历史数据进行特征提取，得到稳定绝缘特征；稳定绝缘特征包括电池在绝缘稳定工况下的用于预测电池绝缘故障的特征；及
预警模块，被配置为根据稳定绝缘特征确定特征预警阈值，和/或，根据稳定绝缘特征生成绝缘预警模型；特征预警阈值和/或绝缘预警模型用于对电池的绝缘故障进行预警。
上述电池绝缘故障的预警装置，通过确定绝缘稳定工况，并获取电池绝缘值稳定的工况下的与绝缘故障相关的特征作为稳定绝缘特征，再进一步基于稳定绝缘特征确定特征预警阈值和/或生成绝缘预警模型，从而可以基于特征预警阈值和/或绝缘预警模型进行相应的绝缘故障预警，由于在绝缘稳定工况下电池的绝缘值较为稳定，不会出现波动幅度较大的变化，基于绝缘稳定工况下的绝缘值和相关的参数来确定特征预警阈值和/或生成绝缘预警模型，可以得到能够反映绝缘故障的特征预警阈值和/或识别准确率较高的绝缘预警模型，从而，无须对电池电路进行重新设计即可实现电池绝缘故障的预警，进行绝缘预警的成本降低，而且，可以提前预测到电池的绝缘故障，有效避免了由于电池绝缘性能下降而导致的危险。
一种电池绝缘故障预警系统，包括：电池管理端和绝缘故障预警后台；
电池管理端，被配置为采集车辆电池的电池数据，并将采集的电池数据发送至绝缘故障预警后台，供绝缘故障预警后台从电池数据中提取历史电池数据；
绝缘故障预警后台，被配置为根据所确定的绝缘稳定工况对电池历史数据进行特征提取，得到稳定绝缘特征；绝缘稳定工况包括电池的绝缘值的波动幅度处于预设范围内时的工况；稳定绝缘特征包括电池在绝缘稳定工况下的用于预测电池绝缘故障的特征；根据稳定绝缘特征确定特征预警阈值，和/或，根据稳定绝缘特征生成绝缘预警模型；特征预警阈值和/或绝缘预警模型用于对电池的绝缘故障进行预警。
上述电池绝缘故障的预警系统，电池管理端可将采集的电池数据发送至绝缘故障预警后台，供绝缘故障预警后台从电池数据中提取历史电池数据，绝缘故障预警后台可通过确定绝缘稳定工况，并获取电池绝缘值稳定的工况下的与绝缘故障相关的特征作为稳定绝缘特征，再进一步基于稳定绝缘特征确定特征预警阈值和/或生成绝缘预警模型，从而可以基于特征预警阈值和/或绝缘预警模型进行相应的绝缘故障预警，由于在绝缘稳定工况 下电池的绝缘值较为稳定，不会出现波动幅度较大的变化，基于绝缘稳定工况下的绝缘值和相关的参数来确定特征预警阈值和/或生成绝缘预警模型，可以得到能够反映绝缘故障的特征预警阈值和/或识别准确率较高的绝缘预警模型，从而，无须对电池电路进行重新设计即可实现电池绝缘故障的预警，进行绝缘预警的成本降低，而且，可以提前预测到电池的绝缘故障，有效避免了由于电池绝缘性能下降而导致的危险。
一种计算机设备，包括存储器及一个或多个处理器，存储器中储存有计算机可读指令，计算机可读指令被一个或多个处理器执行时，使得一个或多个处理器执行以下步骤：确定绝缘稳定工况，以及，获取电池历史数据；绝缘稳定工况包括电池的绝缘值的波动幅度处于预设范围内时的工况；
根据绝缘稳定工况对电池历史数据进行特征提取，得到稳定绝缘特征；稳定绝缘特征包括电池在绝缘稳定工况下的用于预测电池绝缘故障的特征；及
根据稳定绝缘特征确定特征预警阈值，和/或，根据稳定绝缘特征生成绝缘预警模型；特征预警阈值和/或绝缘预警模型用于对电池的绝缘故障进行预警。
一个或多个存储有计算机可读指令的计算机存储介质，计算机可读指令被一个或多个处理器执行时，使得一个或多个处理器执行以下步骤：
确定绝缘稳定工况，以及，获取电池历史数据；绝缘稳定工况包括电池的绝缘值的波动幅度处于预设范围内时的工况；
根据绝缘稳定工况对电池历史数据进行特征提取，得到稳定绝缘特征；稳定绝缘特征包括电池在绝缘稳定工况下的用于预测电池绝缘故障的特征；及
根据稳定绝缘特征确定特征预警阈值，和/或，根据稳定绝缘特征生成绝缘预警模型；特征预警阈值和/或绝缘预警模型用于对电池的绝缘故障进行预警。
一种计算机程序产品，包括计算机程序，该计算机程序被一个或多个处理器执行时，使得一个或多个处理器执行以下步骤：
确定绝缘稳定工况，以及，获取电池历史数据；绝缘稳定工况包括电池的绝缘值的波动幅度处于预设范围内时的工况；
根据绝缘稳定工况对电池历史数据进行特征提取，得到稳定绝缘特征；稳定绝缘特征包括电池在绝缘稳定工况下的用于预测电池绝缘故障的特征；及
根据稳定绝缘特征确定特征预警阈值，和/或，根据稳定绝缘特征生成绝缘预警模型；特征预警阈值和/或绝缘预警模型用于对电池的绝缘故障进行预警。
上述计算机设备、计算机可读存储介质和计算机程序产品，通过确定绝缘稳定工况， 并获取电池绝缘值稳定的工况下的与绝缘故障相关的特征作为稳定绝缘特征，再进一步基于稳定绝缘特征确定特征预警阈值和/或生成绝缘预警模型，从而可以基于特征预警阈值和/或绝缘预警模型进行相应的绝缘故障预警，由于在绝缘稳定工况下电池的绝缘值较为稳定，不会出现波动幅度较大的变化，基于绝缘稳定工况下的绝缘值和相关的参数来确定特征预警阈值和/或生成绝缘预警模型，可以得到能够反映绝缘故障的特征预警阈值和/或识别准确率较高的绝缘预警模型，从而，无须对电池电路进行重新设计即可实现电池绝缘故障的预警，进行绝缘预警的成本降低，而且，可以提前预测到电池的绝缘故障，有效避免了由于电池绝缘性能下降而导致的危险。
本申请的一个或多个实施例的细节在下面的附图和描述中提出。本申请的其它特征和优点将从说明书、附图以及权利要求书变得明显。
附图说明
为了更清楚地说明本申请实施例中的技术方案，下面将对实施例中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其它的附图。
图1为一些实施例中电池绝缘故障的预警方法的应用环境图；
图2为一些实施例中电池绝缘故障预警方法的流程示意图；
图3a为一些实施例中基于特征预警阈值进行绝缘故障预警的示意图；
图3b为一些实施例的基于绝缘预警模型进行绝缘故障预警的示意图；
图4为一些实施例中电池绝缘故障预警装置的结构框图；
图5为一些实施例中电池绝缘故障预警系统的结构框图；
图6为一些实施例中计算机设备的内部结构图。
具体实施方式
为了使本申请的技术方案及优点更加清楚明白，以下结合附图及实施例，对本申请进行进一步详细说明。应当理解，此处描述的具体实施例仅仅用以解释本申请，并不用于限定本申请。
除非另有定义，本文所使用的所有的技术和科学术语与属于本申请的技术领域的技术人员通常理解的含义相同；本文中所使用的术语只是为了描述具体的实施例的目的，不 是旨在于限制本申请；本申请的说明书和权利要求书及上述附图说明中的术语“包括”和“具有”以及它们的任何变形，意图在于覆盖不排他的包含。
在本申请实施例的描述中，技术术语“第一”“第二”等仅用于区别不同对象，而不能理解为指示或暗示相对重要性或者隐含指明所指示的技术特征的数量、特定顺序或主次关系。在本申请实施例的描述中，“多个”的含义是两个以上，除非另有明确具体的限定。同时，在本说明书中使用的术语“和/或”包括相关所列项目的任何及所有组合。
本申请实施例提供的电池绝缘故障的预警方法，可以应用于如图1所示的应用环境中。其中，电池管理端102通过网络与绝缘故障预警后台104进行通信。数据存储系统可以存储绝缘故障预警后台104执行电池绝缘故障预警方法所需的数据。数据存储系统可以集成在服务器104上，也可以放在云上或其他网络服务器上。电池管理端102可以具体为搭载有电池管理系统(Battery Management System,BMS)的终端。例如，可以是部署在新能源汽车上搭载电池管理系统的车载电池管理端。电池管理系统主要用于智能化管理及维护各个电池单元，防止电池出现过充电和过放电，延长电池的使用寿命，监控电池的状态。绝缘故障预警后台104可以用独立的服务器或者是多个服务器组成的服务器集群来实现。
在一些实施例中，如图2所示，提供了一种电池绝缘故障预警方法，以该方法应用于图1中的绝缘故障预警后台104为例进行说明，包括以下步骤：
步骤S202，确定绝缘稳定工况，以及，获取电池历史数据。
上述绝缘稳定工况包括电池的绝缘值的波动幅度处于预设范围内时的工况。例如，在某个工况下绝缘值在0-60KΩ的幅度内波动。当绝缘值在预设的范围内时波动，即电池的绝缘值趋于稳定。绝缘故障预警后台104可以结合历史数据和绝缘故障机理(例如，箱体内湿度高导致绝缘故障、绝缘胶缺陷导致绝缘故障)来确定电池绝缘值趋于稳定的工况，作为上述的绝缘稳定工况。
在一些实施例中，绝缘稳定工况可以包括充电开启热管理工况、放电开启热管理工况、大电流放电工况、大电流充电工况、高荷电状态(State of Charge，SOC)放电工况、充电枪插入工况、充电枪拔出工况、开路电压(Open Circuit Voltage，OCV)平台区工况中的至少一种；其中，充电开启热管理工况可以包括充电段温度大于第一温度T 1的工况；放电开启热管理工况可以包括放电段温度大于第二温度T 2的工况；大电流放电工况可以包括放电电流大于第一电流I 1的工况；大电流充电工况可以包括充电电流大于第二电流I 2的工况；高荷电状态放电工况可以包括放电时荷电状态在第一荷电状态和第二荷电状态之间的工况；充电枪插入工况可以包括充电段的前若干帧(例如，前n帧，n＞1)的工况； 充电枪拔出工况包括非充电段的前若干帧的工况；开路电压平台区工况包括电池电压变化缓慢的工况。其中，T 1和T 2可以为不同的温度值，也可以为相同的温度值；同样地，I 1和I 2可以为不同的电流值，也可以为相同的电流值；本领域技术人员可以根据实际情况而设定。本领域技术人员可以理解的是，在实际应用中，还可以根据其他的绝缘值稳定的工况获取稳定绝缘特征，上述实施例仅用作绝缘稳定工况的具体示例，并非对具体的绝缘稳定工况进行限制。
上述的电池历史数据，可以包括电池运作时所产生的数据。例如，电池在充电状态下的温度、放电时的电流等。此外，电池管理端102可以针对电池采集电池数据，并将该电池数据上发至数据存储系统。绝缘故障预警后台104可以从数据存储系统所存储的电池数据中提取所需的数据作为上述的电池历史数据。
步骤S204，根据绝缘稳定工况对电池历史数据进行特征提取，得到稳定绝缘特征；稳定绝缘特征包括电池在绝缘稳定工况下用于预测电池绝缘故障的特征。
绝缘故障预警后台104可以从电池历史数据中提取出绝缘稳定工况下的能够用于对绝缘故障进行预测的特征，作为上述的稳定绝缘特征。例如，电池历史数据中包括有多个工况及其对应的绝缘值、电压等数据，确定多个工况中的绝缘稳定工况，然后根据绝缘稳定工况所对应的数据进一步确定电压变化率、绝缘值变化率、压差等稳定绝缘特征。在绝缘稳定工况下，电池的绝缘值较为稳定，不会出现波动幅度较大的变化，基于绝缘稳定工况下的绝缘值进行绝缘故障预警，可以提升预警准确率。在一些实施例中，稳定绝缘特征可以包括绝缘值均值、绝缘值分位数、绝缘值极值、绝缘值标准差、电压变化率、绝缘值变化率、压差、温差、电流、异常帧数中的至少一种。本领域技术人员可以理解的是，在实际应用中，还可以采用其他参数或数学统计方式获取到稳定绝缘特征，上述实施例仅用作稳定绝缘特征的具体示例，并非对具体的稳定绝缘特征进行限制。
步骤206，根据稳定绝缘特征确定特征预警阈值，和/或，根据稳定绝缘特征生成绝缘预警模型；特征预警阈值和/或绝缘预警模型用于对电池的绝缘故障进行预警。
绝缘故障预警后台104可以基于稳定绝缘特征来确定特征预警阈值，从而，在进行实时的绝缘故障预警中，可以基于电池实时数据与特征预警阈值的比较结果来实现绝缘故障预警。例如，可以根据历史数据(出现绝缘故障时某个稳定绝缘特征的数值)来设定稳定绝缘特征对应的阈值，作为上述的特征预警阈值。
此外，绝缘故障预警后台104也可以基于稳定绝缘特征来训练出绝缘预警模型，从而，在进行实时的绝缘故障预警中，可以将电池实时数据输入至绝缘预警模型，根据绝缘 预警模型的输出结果来进行绝缘故障预警。
上述的电池绝缘故障预警方法中，通过确定绝缘稳定工况，并获取电池绝缘值稳定的工况下的与绝缘故障相关的特征作为稳定绝缘特征，再进一步基于稳定绝缘特征确定特征预警阈值和/或生成绝缘预警模型，从而可以基于特征预警阈值和/或绝缘预警模型进行相应的绝缘故障预警，由于在绝缘稳定工况下电池的绝缘值较为稳定，不会出现波动幅度较大的变化，基于绝缘稳定工况下的绝缘值和相关的参数来确定特征预警阈值和/或生成绝缘预警模型，可以得到能够反映绝缘故障的特征预警阈值和/或识别准确率较高的绝缘预警模型，从而，无须对电池电路进行重新设计即可实现电池绝缘故障的预警，进行绝缘预警的成本降低，而且，可以提前预测到电池的绝缘故障，有效避免了由于电池绝缘性能下降而导致的危险。
在一些实施例中，根据稳定绝缘特征确定特征预警阈值，包括：
通过相关性系数和/或均值距离在稳定绝缘特征中选取关键绝缘特征；确定关键绝缘特征所对应的阈值，作为特征预警阈值。
绝缘故障预警后台104可以通过相关性系数、均值距离等的处理方式，从多种稳定绝缘特征中，筛选出可以充分反映绝缘故障的特征，作为上述的关键绝缘特征。
以在充电开启热管理工况和充电未开启热管理工况的稳定绝缘特征中选取关键绝缘特征为例说明，首先，可以针对充电开启热管理工况，确定一定时间内(例如，一天，一个充电循环或一小时)的绝缘值并求均值，获得充电开启热管理工况和充电未开启热管理工况的绝缘值均值，作为上述的稳定绝缘特征。例如，在充电开启热管理工况下电池绝缘故障前的绝缘值均值为10000Ω，无绝缘故障的绝缘值均值为16000Ω；而在充电未开启热管理工况下，电池绝缘故障前的绝缘值均值为15000Ω，无绝缘故障的绝缘值均值为17000Ω。
充电开启热管理工况和充电未开启热管理工况的故障前后的缘值均值的差值分别为6000Ω和2000Ω，说明充电开启热管理工况下绝缘值均值特征与绝缘故障相关度更大，因此，根据均值距离选取出充电开启热管理工况的稳定绝缘特征作为关键绝缘特征。
通过对关键绝缘特征进行周期性统计(例如，以1天或30天为周期统计大电流放电工况下绝缘值25分位数)，并基于统计结果确定各个关键绝缘特征所对应的阈值，作为特征预警阈值。在一种确定特征预警阈值的具体示例中，假设在充电开启热管理工况下电池无绝缘故障时的绝缘值均值为16000Ω，则从16000Ω向下循环选取候选的特征预警阈值进行测试，直至测试结果满足误报率和预警准确率达到预设值为止。在另一种确定特征 预警阈值的具体示例中，可以通过单层决策树查找基尼系数变化最大的分割点，并以该分割点对应的数值作为特征预警阈值。
图3a为一些实施例中基于特征预警阈值进行绝缘故障预警的示意图。如图所示，绝缘故障预警后台可以基于如箱体内湿度较高导致绝缘等的绝缘故障机理得到大量的稳定绝缘特征，然后通过均值距离、相关性系数等数学统计方式，筛选出关键绝缘特征，并确定对应的特征预警阈值。从而，在进行预警时，绝缘故障预警后台可以根据电池实时数据与特征预警阈值的比较结果进行绝缘故障预警。
上述的绝缘故障预警方法中，通过数学处理的方式确定特征预警阈值所耗费计算资源较少且效率较高，降低了绝缘故障预警的成本。
在一些实施例中，根据稳定绝缘特征生成绝缘预警模型，包括：
在稳定绝缘特征中选取关键绝缘特征；划分关键绝缘特征，得到至少一个特征集合；基于各个特征集合进行模型训练，得到各个特征集合对应的同类特征预警模型；根据各个同类特征预警模型的输出结果，对各个同类特征预警模型进行融合，得到绝缘预警模型。
绝缘故障预警后台104可以通过相关性系数、均值距离等的处理方式，从多种稳定绝缘特征中，筛选出可以充分反映绝缘故障的特征，作为上述的关键绝缘特征。
然后，可以根据绝缘故障机理、绝缘稳定工况、特征提取的周期等分类条件，将关键绝缘特征划分成多个特征集合，并对各个特征集合进行决策树、支持向量机(support vector machines，SVM)或其他机器学习模型训练，由此得到各个特征集合所对应的模型，作为上述的同类特征预警模型。
最后，可以基于同类特征预警模型的输出结果对多个同类特征模型进行融合，以得到最终的绝缘预警模型。本领域技术人员可以根据需要设计融合的方式，例如可以是基于投票机制的融合、线性加权融合等。
例如，按照充电段和放电段划分特征，充电段的特征集合可以包括有充电开启热管理工况、大电流充电工况、充电枪插入工况的关键绝缘特征，放电段的特征集合可以包括放电开启热管理工况、大电流放电工况、充电枪拔出工况的关键绝缘特征。采用两个特征集合分别对决策树模型进行训练，得到同类特征预警模型_充电段和同类特征预警模型_放电段，最后对同类特征预警模型_充电段和同类特征预警模型_放电段进行融合，生成最终的绝缘预警模型，例如，对同类特征预警模型_充电段和同类特征预警模型_放电段该两个模型的输出结果采用“或”判断逻辑来输出最终的预警结果。
图3b为一个实施例的基于绝缘预警模型进行绝缘故障预警的示意图。如图所示， 绝缘故障预警后台可以基于如箱体内湿度较高导致绝缘等的绝缘故障机理得到大量的稳定绝缘特征，然后根据绝缘故障机理、绝缘稳定工况、特征提取的周期等分类条件，将关键绝缘特征划分成多个特征集合，并对各个特征集合分别进行模型训练，得到特征集合各自对应的预警模型，最后对各个预警模型进行融合，得到最终的绝缘预警模型。从而，在进行预警时，绝缘故障预警后台可以将电池实时数据输入至绝缘预警模型，根据绝缘预警模型的输出结果进行绝缘故障预警。
上述的绝缘故障预警方法中，通过模型训练的方式得到准确识别绝缘故障风险的绝缘预警模型，基于该绝缘预警模型进行预警，提升了绝缘故障预警的准确率。
本领域技术人员可以理解的是，虽然如上所述的各实施例所涉及的流程图中的各个步骤按照箭头的指示依次显示，但是这些步骤并不是必然按照箭头指示的顺序依次执行。除非本文中有明确的说明，这些步骤的执行并没有严格的顺序限制，这些步骤可以以其它的顺序执行。而且，如上所述的各实施例所涉及的流程图中的至少一部分步骤可以包括多个步骤或者多个阶段，这些步骤或者阶段并不必然是在同一时刻执行完成，而是可以在不同的时刻执行，这些步骤或者阶段的执行顺序也不必然是依次进行，而是可以与其它步骤或者其它步骤中的步骤或者阶段的至少一部分轮流或者交替地执行。
基于同样的发明构思，本申请实施例还提供了一种用于实现上述所涉及的电池绝缘故障预警方法的电池绝缘故障预警装置。该装置所提供的解决问题的实现方案与上述方法中所记载的实现方案相似，故下面所提供的一个或多个电池绝缘故障预警装置实施例中的具体限定可以参见上文中对于电池绝缘故障预警方法的限定，在此不再赘述。
在一些实施例中，如图4所示，提供了一种电池绝缘故障预警装置400，包括：数据模块402、特征模块404和预警模块406，其中：
数据模块402，被配置为确定绝缘稳定工况，以及，获取电池历史数据；绝缘稳定工况包括电池的绝缘值的波动幅度处于预设范围内时的工况；
特征模块404，被配置为根据绝缘稳定工况对电池历史数据进行特征提取，得到稳定绝缘特征；稳定绝缘特征包括电池在绝缘稳定工况下的用于预测电池绝缘故障的特征；
预警模块406，被配置为根据稳定绝缘特征确定特征预警阈值，和/或，根据稳定绝缘特征生成绝缘预警模型；特征预警阈值和/或绝缘预警模型用于对电池的绝缘故障进行预警。
上述的电池绝缘故障预警装置中，通过确定绝缘稳定工况，并获取电池绝缘值稳定的工况下的与绝缘故障相关的特征作为稳定绝缘特征，再进一步基于稳定绝缘特征确定特征预警阈值和/或生成绝缘预警模型，从而可以基于特征预警阈值和/或绝缘预警模型进行 相应的绝缘故障预警，由于在绝缘稳定工况下电池的绝缘值较为稳定，不会出现波动幅度较大的变化，基于绝缘稳定工况下的绝缘值和相关的参数来确定特征预警阈值和/或生成绝缘预警模型，可以得到能够反映绝缘故障的特征预警阈值和/或识别准确率较高的绝缘预警模型，从而，无须对电池电路进行重新设计即可实现电池绝缘故障的预警，进行绝缘预警的成本降低，而且，可以提前预测到电池的绝缘故障，有效避免了由于电池绝缘性能下降而导致的危险。
在一些实施例中，预警模块406还被配置为：通过相关性系数和/或均值距离在稳定绝缘特征中选取关键绝缘特征；确定关键绝缘特征所对应的阈值，作为特征预警阈值。
在一些实施例中，预警模块406还被配置为：对关键绝缘特征进行周期性统计，得到统计结果；基于统计结果，确定关键绝缘特征对应的阈值，作为特征预警阈值。
在一些实施例中，预警模块406还被配置为：通过单层决策树查找关键绝缘特征的基尼系数变化最大的分割点；以分割点对应的数值作为所述特征预警阈值。
在一些实施例中，预警模块406还被配置为：在稳定绝缘特征中选取关键绝缘特征；划分关键绝缘特征，得到至少一个特征集合；基于各个特征集合进行模型训练，得到各个特征集合对应的同类特征预警模型；根据各个同类特征预警模型的输出结果，对各个同类特征预警模型进行融合，得到绝缘预警模型。
在一些实施例中，预警模块406还被配置为：根据预设的分类条件，划分关键绝缘特征，得到至少一个特征集合，预设的分类条件包括绝缘故障机理、绝缘稳定工况或特征提取周期。
在一些实施例中，绝缘稳定工况，包括充电开启热管理工况、放电开启热管理工况、大电流放电工况、大电流充电工况、高荷电状态放电工况、充电枪插入工况、充电枪拔出工况、开路电压平台区工况中的至少一种。
在一些实施例中，充电开启热管理工况包括充电段温度大于第一温度的工况；放电开启热管理工况包括放电段温度大于第二温度的工况；大电流放电工况包括放电电流大于第一电流的工况；大电流充电工况包括充电电流大于第二电流的工况；高荷电状态放电工况包括放电时荷电状态在第一荷电状态和第二荷电状态之间的工况；充电枪插入工况包括充电段的前若干帧的工况；充电枪拔出工况包括非充电段的前若干帧的工况；开路电压平台区工况包括电池电压变化缓慢的工况。
在一些实施例中，稳定绝缘特征，包括绝缘值均值、绝缘值分位数、绝缘值极值、绝缘值标准差、电压变化率、绝缘值变化率、压差、温差、电流、异常帧数中的至少一种。
上述电池绝缘故障预警装置中的各个模块可全部或部分通过软件、硬件及其组合来实现。上述各模块可以硬件形式内嵌于或独立于计算机设备中的处理器中，也可以以软件形式存储于计算机设备中的存储器中，以便于处理器调用执行以上各个模块对应的操作。
基于同样的发明构思，本申请实施例还提供了一种用于实现上述所涉及的电池绝缘故障预警方法的电池绝缘故障预警系统。该系统所提供的解决问题的实现方案与上述方法中所记载的实现方案相似，故下面所提供的一个或多个电池绝缘故障预警系统实施例中的具体限定可以参见上文中对于电池绝缘故障预警方法的限定，在此不再赘述。
在一些实施例中，如图5所示，提供了一种电池绝缘故障预警系统500，包括：电池管理端502和绝缘故障预警后台504，其中：
电池管理端502，被配置为采集车辆电池的电池数据，并将采集的电池数据发送至绝缘故障预警后台，供绝缘故障预警后台从电池数据中提取历史电池数据；
绝缘故障预警后台504，被配置为根据所确定的绝缘稳定工况对电池历史数据进行特征提取，得到稳定绝缘特征；绝缘稳定工况包括电池的绝缘值的波动幅度处于预设范围内时的工况；稳定绝缘特征包括电池在绝缘稳定工况下的用于预测电池绝缘故障的特征；根据稳定绝缘特征确定特征预警阈值，和/或，根据稳定绝缘特征生成绝缘预警模型；特征预警阈值和/或绝缘预警模型用于对电池的绝缘故障进行预警。
在一些实施例中，提供了一种计算机设备，该计算机设备可以是服务器，其内部结构图可以如图6所示。该计算机设备包括通过系统总线连接的处理器、存储器和网络接口。其中，该计算机设备的处理器用于提供计算和控制能力。该计算机设备的存储器包括非易失性存储介质和内存储器。该非易失性存储介质存储有操作系统、计算机程序和数据库。该内存储器为非易失性存储介质中的操作系统和计算机程序的运行提供环境。该计算机设备的数据库用于存储电池数据。该计算机设备的网络接口用于与外部的终端通过网络连接通信。该计算机程序被处理器执行时以实现一种电池绝缘故障预警方法。
本领域技术人员可以理解，图6中示出的结构，仅仅是与本申请方案相关的部分结构的框图，并不构成对本申请方案所应用于其上的计算机设备的限定，具体的计算机设备可以包括比图中所示更多或更少的部件，或者组合某些部件，或者具有不同的部件布置。
在一些实施例中，提供了一种计算机设备，包括存储器和处理器，存储器中存储有计算机程序，该处理器执行计算机程序时实现以下步骤：确定绝缘稳定工况，以及，获取电池历史数据；绝缘稳定工况包括电池的绝缘值的波动幅度处于预设范围内时的工况；根据绝缘稳定工况对电池历史数据进行特征提取，得到稳定绝缘特征；稳定绝缘特征包括电 池在绝缘稳定工况下的用于预测电池绝缘故障的特征；根据稳定绝缘特征确定特征预警阈值，和/或，根据稳定绝缘特征生成绝缘预警模型；特征预警阈值和/或绝缘预警模型用于对电池的绝缘故障进行预警。
在一些实施例中，处理器执行计算机程序时还实现以下步骤：通过相关性系数和/或均值距离在稳定绝缘特征中选取关键绝缘特征；确定关键绝缘特征所对应的阈值，作为特征预警阈值。
在一些实施例中，处理器执行计算机程序时还实现以下步骤：在稳定绝缘特征中选取关键绝缘特征；划分关键绝缘特征，得到至少一个特征集合；基于各个特征集合进行模型训练，得到各个特征集合对应的同类特征预警模型；根据各个同类特征预警模型的输出结果，对各个同类特征预警模型进行融合，得到绝缘预警模型。
在一些实施例中，绝缘稳定工况，包括充电开启热管理工况、放电开启热管理工况、大电流放电工况、大电流充电工况、高荷电状态放电工况、充电枪插入工况、充电枪拔出工况、开路电压平台区工况中的至少一种。
在一些实施例中，充电开启热管理工况包括充电段温度大于第一温度的工况；放电开启热管理工况包括放电段温度大于第二温度的工况；大电流放电工况包括放电电流大于第一电流的工况；大电流充电工况包括充电电流大于第二电流的工况；高荷电状态放电工况包括放电时荷电状态在第一荷电状态和第二荷电状态之间的工况；充电枪插入工况包括充电段的前若干帧的工况；充电枪拔出工况包括非充电段的前若干帧的工况；开路电压平台区工况包括电池电压变化缓慢的工况。
在一些实施例中，提供了一种计算机可读存储介质，其上存储有计算机程序，计算机程序被处理器执行时实现以下步骤：确定绝缘稳定工况，以及，获取电池历史数据；绝缘稳定工况包括电池的绝缘值的波动幅度处于预设范围内时的工况；根据绝缘稳定工况对电池历史数据进行特征提取，得到稳定绝缘特征；稳定绝缘特征包括电池在绝缘稳定工况下的用于电池预测绝缘故障的特征；根据稳定绝缘特征确定特征预警阈值，和/或，根据稳定绝缘特征生成绝缘预警模型；特征预警阈值和/或绝缘预警模型用于对电池的绝缘故障进行预警。
在一些实施例中，计算机程序被处理器执行时还实现以下步骤：通过相关性系数和/或均值距离在稳定绝缘特征中选取关键绝缘特征；确定关键绝缘特征所对应的阈值，作为特征预警阈值。
在一些实施例中，计算机程序被处理器执行时还实现以下步骤：在稳定绝缘特征中 选取关键绝缘特征；划分关键绝缘特征，得到至少一个特征集合；基于各个特征集合进行模型训练，得到各个特征集合对应的同类特征预警模型；根据各个同类特征预警模型的输出结果，对各个同类特征预警模型进行融合，得到绝缘预警模型。
在一些实施例中，绝缘稳定工况，包括充电开启热管理工况、放电开启热管理工况、大电流放电工况、大电流充电工况、高荷电状态放电工况、充电枪插入工况、充电枪拔出工况、开路电压平台区工况中的至少一种。
在一些实施例中，充电开启热管理工况包括充电段温度大于第一温度的工况；放电开启热管理工况包括放电段温度大于第二温度的工况；大电流放电工况包括放电电流大于第一电流的工况；大电流充电工况包括充电电流大于第二电流的工况；高荷电状态放电工况包括放电时荷电状态在第一荷电状态和第二荷电状态之间的工况；充电枪插入工况包括充电段的前若干帧的工况；充电枪拔出工况包括非充电段的前若干帧的工况；开路电压平台区工况包括电池电压变化缓慢的工况。
需要说明的是，本申请所涉及的用户信息(包括但不限于用户设备信息、用户个人信息等)和数据(包括但不限于用于分析的数据、存储的数据、展示的数据等)，均为经用户授权或者经过各方充分授权的信息和数据。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程，是可以通过计算机可读指令来指令相关的硬件来完成，所述的计算机可读指令可存储于一非易失性计算机可读取存储介质中，该计算机可读指令在执行时，可包括如上述各方法的实施例的流程。其中，本申请所提供的各实施例中所使用的对存储器、数据库或其它介质的任何引用，均可包括非易失性和易失性存储器中的至少一种。非易失性存储器可包括只读存储器(Read-Only Memory，ROM)、磁带、软盘、闪存、光存储器、高密度嵌入式非易失性存储器、阻变存储器(ReRAM)、磁变存储器(Magnetoresistive Random Access Memory，MRAM)、铁电存储器(Ferroelectric Random Access Memory，FRAM)、相变存储器(Phase Change Memory，PCM)、石墨烯存储器等。易失性存储器可包括随机存取存储器(Random Access Memory，RAM)或外部高速缓冲存储器等。作为说明而非局限，RAM可以是多种形式，比如静态随机存取存储器(Static Random Access Memory，SRAM)或动态随机存取存储器(Dynamic Random Access Memory，DRAM)等。本申请所提供的各实施例中所涉及的数据库可包括关系型数据库和非关系型数据库中至少一种。非关系型数据库可包括基于区块链的分布式数据库等，不限于此。本申请所提供的各实施例中所涉及的处理器可为通用处理器、中央处理器、图形处理器、数字信号处理器、可编程逻辑器、基于量子 计算的数据处理逻辑器等，不限于此。
以上实施例的各技术特征可以进行任意的组合，为使描述简洁，未对上述实施例中的各个技术特征所有可能的组合都进行描述，然而，只要这些技术特征的组合不存在矛盾，都应当认为是本说明书记载的范围。
以上所述实施例仅表达了本申请的几种实施方式，其描述较为具体和详细，但并不能因此而理解为对发明专利范围的限制。应当指出的是，对于本领域的普通技术人员来说，在不脱离本申请构思的前提下，还可以做出若干变形和改进，这些都属于本申请的保护范围。因此，本申请专利的保护范围应以所附权利要求为准。

Claims (15)

  1. 一种电池绝缘故障的预警方法，包括：
    确定绝缘稳定工况，以及，获取电池历史数据；所述绝缘稳定工况包括电池的绝缘值的波动幅度处于预设范围内时的工况；
    根据所述绝缘稳定工况对所述电池历史数据进行特征提取，得到稳定绝缘特征；所述稳定绝缘特征包括电池在所述绝缘稳定工况下的用于预测所述电池绝缘故障的特征；及
    根据所述稳定绝缘特征确定特征预警阈值，和/或，根据所述稳定绝缘特征生成绝缘预警模型；所述特征预警阈值和/或所述绝缘预警模型用于对所述电池的绝缘故障进行预警。
  2. 根据权利要求1所述的电池绝缘故障的预警方法，其中，所述根据所述稳定绝缘特征确定特征预警阈值，包括：
    通过相关性系数在所述稳定绝缘特征中选取关键绝缘特征；及
    确定所述关键绝缘特征所对应的阈值，作为所述特征预警阈值。
  3. 根据权利要求1所述的电池绝缘故障的预警方法，其中，所述根据所述稳定绝缘特征确定特征预警阈值，包括：
    通过均值距离在所述稳定绝缘特征中选取关键绝缘特征；及
    确定所述关键绝缘特征所对应的阈值，作为所述特征预警阈值。
  4. 根据权利要求1所述的电池绝缘故障的预警方法，其中，所述根据所述稳定绝缘特征确定特征预警阈值，包括：
    通过相关性系数和均值距离在所述稳定绝缘特征中选取关键绝缘特征；及
    确定所述关键绝缘特征所对应的阈值，作为所述特征预警阈值。
  5. 根据权利要求1至4任意一项所述的电池绝缘故障的预警方法，其中，所述确定所述关键绝缘特征所对应的阈值，作为所述特征预警阈值包括：
    对所述关键绝缘特征进行周期性统计，得到统计结果；及
    基于所述统计结果，确定所述关键绝缘特征对应的阈值，作为所述特征预警阈值。
  6. 根据权利要求1至4任意一项所述的电池绝缘故障的预警方法，其中，所述确定所述关键绝缘特征所对应的阈值，作为所述特征预警阈值包括：
    通过单层决策树查找所述关键绝缘特征的基尼系数变化最大的分割点；及
    以所述分割点对应的数值作为所述特征预警阈值。
  7. 根据权利要求1至4任意一项所述的电池绝缘故障的预警方法，其中，所述根据 所述稳定绝缘特征生成绝缘预警模型，包括：
    在所述稳定绝缘特征中选取关键绝缘特征；
    划分所述关键绝缘特征，得到至少一个特征集合；
    基于各个所述特征集合进行模型训练，得到各个所述特征集合对应的同类特征预警模型；及
    根据各个所述同类特征预警模型的输出结果，对各个所述同类特征预警模型进行融合，得到所述绝缘预警模型。
  8. 根据权利要求7所述的电池绝缘故障的预警方法，其中，所述划分所述关键绝缘特征，得到至少一个特征集合包括：
    根据预设的分类条件，划分所述关键绝缘特征，得到至少一个特征集合，所述预设的分类条件包括绝缘故障机理、绝缘稳定工况或特征提取周期。
  9. 根据权利要求7所述的电池绝缘故障的预警方法，其中，所述绝缘稳定工况，包括充电开启热管理工况、放电开启热管理工况、大电流放电工况、大电流充电工况、高荷电状态放电工况、充电枪插入工况、充电枪拔出工况、开路电压平台区工况中的至少一种。
  10. 根据权利要求9所述的电池绝缘故障的预警方法，其中，
    所述充电开启热管理工况包括充电段温度大于第一温度的工况；
    所述放电开启热管理工况包括放电段温度大于第二温度的工况；
    所述大电流放电工况包括放电电流大于第一电流的工况；
    所述大电流充电工况包括充电电流大于第二电流的工况；
    所述高荷电状态放电工况包括放电时荷电状态在第一荷电状态和第二荷电状态之间的工况；
    所述充电枪插入工况包括充电段的前若干帧的工况；
    所述充电枪拔出工况包括非充电段的前若干帧的工况；及
    所述开路电压平台区工况包括电池电压变化缓慢的工况。
  11. 一种电池绝缘故障的预警装置，包括：
    数据模块，被配置为确定绝缘稳定工况，以及，获取电池历史数据；所述绝缘稳定工况包括电池的绝缘值的波动幅度处于预设范围内时的工况；
    特征模块，被配置为根据所述绝缘稳定工况对所述电池历史数据进行特征提取，得到稳定绝缘特征；所述稳定绝缘特征包括电池在所述绝缘稳定工况下的用于预测所述电池绝缘故障的特征；及
    预警模块，被配置为根据所述稳定绝缘特征确定特征预警阈值，和/或，根据所述稳定绝缘特征生成绝缘预警模型；所述特征预警阈值和/或所述绝缘预警模型用于对电池的绝缘故障进行预警。
  12. 一种电池绝缘故障预警系统，包括：电池管理端和绝缘故障预警后台；
    所述电池管理端，被配置为采集车辆电池的电池数据，并将采集的所述电池数据发送至所述绝缘故障预警后台，供所述绝缘故障预警后台从所述电池数据中提取历史电池数据；
    所述绝缘故障预警后台，被配置为根据所确定的绝缘稳定工况对所述电池历史数据进行特征提取，得到稳定绝缘特征；所述绝缘稳定工况包括电池的绝缘值的波动幅度处于预设范围内时的工况；所述稳定绝缘特征包括电池在所述绝缘稳定工况下的用于预测所述电池绝缘故障的特征；根据所述稳定绝缘特征确定特征预警阈值，和/或，根据所述稳定绝缘特征生成绝缘预警模型；所述特征预警阈值和/或所述绝缘预警模型用于对所述电池的绝缘故障进行预警。
  13. 一种计算机设备，包括存储器及一个或多个处理器，所述存储器中储存有计算机可读指令，所述计算机可读指令被所述一个或多个处理器执行时，使得所述一个或多个处理器执行以下步骤：
    确定绝缘稳定工况，以及，获取电池历史数据；所述绝缘稳定工况包括电池的绝缘值的波动幅度处于预设范围内时的工况；
    根据所述绝缘稳定工况对所述电池历史数据进行特征提取，得到稳定绝缘特征；所述稳定绝缘特征包括电池在所述绝缘稳定工况下的用于预测所述电池绝缘故障的特征；及
    根据所述稳定绝缘特征确定特征预警阈值，和/或，根据所述稳定绝缘特征生成绝缘预警模型；所述特征预警阈值和/或所述绝缘预警模型用于对所述电池的绝缘故障进行预警。
  14. 一个或多个存储有计算机可读指令的计算机存储介质，所述计算机可读指令被一个或多个处理器执行时，使得所述一个或多个处理器执行以下步骤：
    确定绝缘稳定工况，以及，获取电池历史数据；所述绝缘稳定工况包括电池的绝缘值的波动幅度处于预设范围内时的工况；
    根据所述绝缘稳定工况对所述电池历史数据进行特征提取，得到稳定绝缘特征；所述稳定绝缘特征包括电池在所述绝缘稳定工况下的用于预测所述电池绝缘故障的特征；及
    根据所述稳定绝缘特征确定特征预警阈值，和/或，根据所述稳定绝缘特征生成绝缘预警模型；所述特征预警阈值和/或所述绝缘预警模型用于对所述电池的绝缘故障进行预 警。
  15. 一种计算机程序产品，包括计算机程序，该计算机程序被一个或多个处理器执行时，使得所述一个或多个处理器执行以下步骤：
    确定绝缘稳定工况，以及，获取电池历史数据；所述绝缘稳定工况包括电池的绝缘值的波动幅度处于预设范围内时的工况；
    根据所述绝缘稳定工况对所述电池历史数据进行特征提取，得到稳定绝缘特征；所述稳定绝缘特征包括电池在所述绝缘稳定工况下的用于预测所述电池绝缘故障的特征；及
    根据所述稳定绝缘特征确定特征预警阈值，和/或，根据所述稳定绝缘特征生成绝缘预警模型；所述特征预警阈值和/或所述绝缘预警模型用于对所述电池的绝缘故障进行预警。
PCT/CN2022/136584 2022-01-29 2022-12-05 电池绝缘故障的预警方法、装置、系统和计算机设备 WO2023142685A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210110117.XA CN115825790B (zh) 2022-01-29 2022-01-29 电池绝缘故障的预警方法、装置、系统和计算机设备
CN202210110117.X 2022-01-29

Publications (1)

Publication Number Publication Date
WO2023142685A1 true WO2023142685A1 (zh) 2023-08-03

Family

ID=85522405

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/136584 WO2023142685A1 (zh) 2022-01-29 2022-12-05 电池绝缘故障的预警方法、装置、系统和计算机设备

Country Status (2)

Country Link
CN (1) CN115825790B (zh)
WO (1) WO2023142685A1 (zh)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106654405A (zh) * 2015-11-02 2017-05-10 三星电子株式会社 电池管理方法和设备
US20180143257A1 (en) * 2016-11-21 2018-05-24 Battelle Energy Alliance, Llc Systems and methods for estimation and prediction of battery health and performance
CN111308376A (zh) * 2020-02-24 2020-06-19 上海蔚来汽车有限公司 动力电池绝缘监测方法、系统以及装置
CN111381170A (zh) * 2020-05-15 2020-07-07 上海工程技术大学 基于大数据的电动汽车电池包健康状态预测方法及系统
CN111426955A (zh) * 2020-04-23 2020-07-17 华南理工大学 一种锂离子电池故障诊断方法
CN112444748A (zh) * 2020-10-12 2021-03-05 武汉蔚来能源有限公司 电池异常检测方法、装置、电子设备及存储介质

Family Cites Families (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102385381A (zh) * 2011-09-30 2012-03-21 中联重科股份有限公司 一种运行设备的故障预警提示系统
CN102495373A (zh) * 2011-11-30 2012-06-13 重庆长安汽车股份有限公司 一种动力电池绝缘检测系统及检测方法
CN103605056B (zh) * 2013-11-20 2017-01-04 奇瑞新能源汽车技术有限公司 一种高压直流绝缘监测装置及电动汽车
CN105277787B (zh) * 2015-09-30 2018-09-11 上海凌翼动力科技有限公司 电动汽车绝缘电阻故障预测方法及系统
CN105334434B (zh) * 2015-10-19 2019-03-15 金龙联合汽车工业（苏州）有限公司 一种电动车辆绝缘监测和报警方法
CN206479582U (zh) * 2017-02-28 2017-09-08 科力远混合动力技术有限公司 一种混合动力汽车绝缘检测电路
CN108091947B (zh) * 2017-12-18 2024-02-02 清华大学 电动车用动力电池组安全防控系统
CN110534822B (zh) * 2019-07-19 2023-06-02 贵州热缆技术有限责任公司 一种动力电池故障预警系统
CN111653840B (zh) * 2020-06-08 2021-07-09 中国第一汽车股份有限公司 一种预警方法、装置、设备及存储介质
CN113567870A (zh) * 2021-07-07 2021-10-29 浙江绿色慧联有限公司 基于边缘计算的电池安全预警系统及方法
CN113850979B (zh) * 2021-09-24 2023-09-15 重庆云宸新能源科技有限公司 一种基于ai算法的电池远程预警系统

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106654405A (zh) * 2015-11-02 2017-05-10 三星电子株式会社 电池管理方法和设备
US20180143257A1 (en) * 2016-11-21 2018-05-24 Battelle Energy Alliance, Llc Systems and methods for estimation and prediction of battery health and performance
CN111308376A (zh) * 2020-02-24 2020-06-19 上海蔚来汽车有限公司 动力电池绝缘监测方法、系统以及装置
CN111426955A (zh) * 2020-04-23 2020-07-17 华南理工大学 一种锂离子电池故障诊断方法
CN111381170A (zh) * 2020-05-15 2020-07-07 上海工程技术大学 基于大数据的电动汽车电池包健康状态预测方法及系统
CN112444748A (zh) * 2020-10-12 2021-03-05 武汉蔚来能源有限公司 电池异常检测方法、装置、电子设备及存储介质

Also Published As

Publication number Publication date
CN115825790A (zh) 2023-03-21
CN115825790B (zh) 2023-10-17

Similar Documents

Publication Publication Date Title
Richardson et al. Battery health prediction under generalized conditions using a Gaussian process transition model
US9255973B2 (en) System and method for estimating long term characteristics of battery
US8412658B2 (en) System and method for estimating long term characteristics of battery
Qiu et al. A novel entropy-based fault diagnosis and inconsistency evaluation approach for lithium-ion battery energy storage systems
Yang et al. State of health assessment of lithium-ion batteries based on deep Gaussian process regression considering heterogeneous features
US20240348064A1 (en) Battery Equalization Method and System
US20220373609A1 (en) State Value for Rechargeable Batteries
US20200217896A1 (en) Device and method for evaluating energy storage device and evaluation system
CN115805810B (zh) 电池故障预测方法、装置、设备、存储介质和程序产品
CN113083739B (zh) 电芯分选方法、装置以及计算机设备
CN113589174A (zh) 一种锂离子动力电池系统电芯一致性预测方法及装置
JP7326237B2 (ja) 複数の電池に関する判定装置、蓄電システム、判定方法及び判定プログラム
CN112084459A (zh) 电池充放电循环寿命的预测方法、装置、电子终端、及存储介质
WO2022267979A1 (zh) 动力电池容量预测方法、装置及设备
WO2023052910A1 (en) System and method for estimating state of health and remaining useful life of a battery
CN116482540A (zh) 电池电压不一致性的分析预测方法、装置及系统
CN114970734A (zh) 异常电池确定方法、装置、计算机设备和存储介质
Kharlamova et al. Evaluating machine-learning-based methods for modeling a digital twin of battery systems providing frequency regulation
Jiang et al. State of Health Estimation for Second-Life Lithium-Ion Batteries in Energy Storage System With Partial Charging-Discharging Workloads
Nunes et al. An online unscented Kalman filter remaining useful life prediction method applied to second-life lithium-ion batteries
WO2023142685A1 (zh) 电池绝缘故障的预警方法、装置、系统和计算机设备
CN115856641A (zh) 一种电池剩余充电时间预测方法、装置及电子设备
Li et al. Evaluation method for consistency of lithium-ion battery packs in electric vehicles based on the Mahalanobis-Taguchi system
She et al. Accurate State of Health Estimation of Battery System based on Multi-stage Constant Current Charging and Behavior Analysis in Real-world Electric Vehicles
TWI852365B (zh) 資訊處理裝置、資訊處理方法、電腦可讀取記錄媒體以及資訊處理系統

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22923476

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE