Insulation fault detection method, device and equipment of power battery and medium

Abstract

The invention provides an insulation fault detection method and device of a power battery, equipment and a medium. The method comprises the steps of obtaining multiple pieces of state time sequence data of the power battery; wherein the multiple pieces of state time sequence data comprise insulation resistance value time sequence data, collected by an insulation monitoring unit, of a battery module in the power battery and sensor time sequence data, collected by each preset sensor, of the power battery; performing weighted data fusion on the plurality of pieces of state time sequence data, and determining an insulation health score of the power battery; and comparing the insulation health score with a plurality of preset score thresholds, and determining an insulation fault detection result of the power battery. Through the method and the device, the degradation trend of the insulation performance can be recognized earlier, the sensitivity and the accuracy of fault detection are improved, safety accidents caused by lagging alarm of insulation faults are effectively avoided, and the operation safety and the reliability of a power battery system are remarkably improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of battery management, in particular to an insulation fault detection method and device for a power battery, equipment and a medium. BACKGROUND

[0002] With the rapid development of the new energy vehicle industry, as a core component, the high-voltage safety problem of the power battery is increasingly prominent. In the high-voltage system of an electric vehicle, the insulation performance is directly related to the safe operation of the vehicle. In the prior art, the power battery system usually adopts a battery management system (BMS) for real-time monitoring, and insulation detection is one of the important functions to ensure high-voltage safety.

[0003] The insulation detection scheme commonly used in the industry at present is mainly based on an active or passive balancing method, which measures the insulation resistance value of the total positive electrode and the total negative electrode of the battery to the vehicle body to determine the insulation state. When the insulation resistance is detected to be lower than the preset safety threshold, the system will trigger an alarm mechanism to prompt the existence of an insulation fault risk. This traditional detection method can only issue an alarm after the insulation resistance value drops below the safety threshold, and cannot capture the slow degradation process of the insulation performance. This lag cannot timely discover potential insulation hazards, and cannot provide early warning for possible insulation faults. When the insulation resistance value is detected to be abnormal, it has often developed to a stage that may cause serious accidents such as electric leakage, short circuit or fire, and there is a significant safety hazard. SUMMARY

[0004] Therefore, the purpose of the present application is to provide an insulation fault detection method, device and equipment for a power battery, which can identify the degradation trend of the insulation performance earlier, improve the sensitivity and accuracy of fault detection, effectively avoid safety accidents caused by lagging alarm of insulation faults, and significantly improve the operation safety and reliability of the power battery system.

[0005] In a first aspect, an insulation fault detection method for a power battery is provided, which comprises: obtaining a plurality of state time series data of the power battery; wherein the plurality of state time series data comprises insulation resistance value time series data of a battery module in the power battery collected by an insulation monitoring unit, and sensor time series data of the power battery collected by each preset sensor; performing weighted data fusion on the plurality of state time series data to determine an insulation health score of the power battery; comparing the insulation health score with a plurality of preset score thresholds to determine an insulation fault detection result of the power battery.

[0006] Further, the weighting data fusion on the plurality of state time series data determines an insulation health score of the power battery, comprising: Time axis synchronization processing is performed on the plurality of state time series data to obtain a plurality of target time series data; For each target time series data, filtering and denoising and normalization processing are performed on the target time series data to obtain corresponding sensor target data; The weight corresponding to each preset sensor and the sensor target data are weighted and fused to obtain the insulation health score.

[0007] Further, after the plurality of state time series data of the power battery is obtained, the insulation fault detection method further comprises: The plurality of state time series data is simultaneously input into a pre-trained insulation fault prediction model to obtain an insulation fault prediction result output by the insulation fault prediction model.

[0008] Further, the insulation fault prediction model is trained by the following steps: Obtaining training samples; wherein the training samples comprise a plurality of state time series sample data and insulation fault sample labels; The training samples are input into an insulation fault original prediction model to obtain insulation fault prediction labels; A loss function is determined according to the insulation fault sample labels and the insulation fault prediction labels, and the insulation fault original prediction model is iteratively trained based on the loss function until a preset training completion condition is met, thereby obtaining the trained insulation fault prediction model.

[0009] Further, after the plurality of state time series data of the power battery is obtained, the insulation fault detection method further comprises: For each battery module, a current resistance value of the battery module is determined based on the insulation resistance value time series data of the battery module; When it is detected that the current resistance value is lower than a first resistance threshold, first insulation fault warning information of the battery module is generated; When it is detected that the current resistance value is lower than a second resistance threshold, second insulation fault warning information of the battery module is generated, and a high-voltage cutoff instruction is sent to a battery management module; wherein the second resistance threshold is lower than the first resistance threshold.

[0010] In a second aspect, the embodiments of the present application further provide an insulation fault detection device of a power battery, comprising: The data acquisition module is configured to acquire multiple state time series data of the power battery, wherein the multiple state time series data include insulation resistance value time series data of a battery module in the power battery collected by an insulation monitoring unit and sensor time series data of the power battery collected by each preset sensor. The insulation health score calculation module is configured to perform weighted data fusion on the multiple state time series data to determine an insulation health score of the power battery. The fault detection result determination module is configured to compare the insulation health score with multiple preset score thresholds to determine an insulation fault detection result of the power battery.

[0011] Further, when performing the weighted data fusion on the multiple state time series data to determine the insulation health score of the power battery, the insulation health score calculation module is further configured to: perform time axis synchronization processing on the multiple state time series data to obtain multiple target time series data; perform filtering and denoising and normalization processing on each target time series data to obtain corresponding sensor target data; perform weighted fusion calculation on the weight corresponding to each preset sensor and the sensor target data to obtain the insulation health score.

[0012] Further, the insulation fault detection device further includes a model prediction module, which is configured to: input the multiple state time series data into a pre-trained insulation fault prediction model to obtain an insulation fault prediction result output by the insulation fault prediction model.

[0013] In a third aspect, an electronic device is provided, which includes a processor, a memory and a bus. The memory stores machine readable instructions executable by the processor. When the electronic device is running, the processor and the memory communicate through the bus. The machine readable instructions are executed by the processor to perform the steps of the insulation fault detection method of the power battery as described above.

[0014] In a fourth aspect, a computer readable storage medium is provided, which stores a computer program. When the computer program is run by a processor, the steps of the insulation fault detection method of the power battery as described above are performed.

[0015] The application provides a kind of insulation fault detection method, device, equipment and medium of power battery, first, obtain the multiple state time series data of power battery;Wherein, the multiple state time series data include the insulation resistance value time series data of battery module in the power battery collected by insulation monitoring unit, and the sensor time series data of the power battery collected by each preset sensor;Then, multiple state time series data are weighted data fusion, determine the insulation health score of the power battery;Finally, the insulation health score is compared with multiple preset score thresholds, and the insulation fault detection result of the power battery is determined.

[0016] The application obtains multiple source state time series data of power battery, and determines insulation health score by weighted data fusion, and comprehensively judges by combining preset score threshold, realizes multi-dimensional evaluation of the insulation state of power battery.Compared with the traditional method of only relying on single insulation resistance threshold alarm, the application can identify the degradation trend of insulation performance earlier, improve the sensitivity and accuracy of fault detection, effectively avoid safety accidents caused by insulation fault lag alarm, significantly improve the operation safety and reliability of power battery system.

[0017] In order to make the above-mentioned purposes, features and advantages of the application more obvious and easy to understand, the following preferred embodiments are described in detail below, and the accompanying drawings are described as follows. BRIEF DESCRIPTION OF DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the application, the following will briefly introduce the drawings needed to be used in the embodiments, it should be understood that the following drawings only show some embodiments of the application, therefore should not be regarded as the limitation to the scope, for ordinary skilled in the art, without paying creative labor, other related drawings can also be obtained according to these drawings.

[0019] Figure 1 The flow chart of the insulation fault detection method of the power battery provided by the embodiments of the application is shown in the figure. Figure 2 The structure schematic diagram of the insulation fault detection device of the power battery provided by the embodiments of the application is shown in the figure. Figure 3 The structure schematic diagram of the insulation fault detection device of the power battery provided by the embodiments of the application is shown in the figure. Figure 4 The structure schematic diagram of the electronic device provided by the embodiments of the application is shown in the figure. DETAILED DESCRIPTION

[0020] To make the objectives, technical solutions and advantages of the embodiments of the present application clearer, the following will be combined with the accompanying drawings in the embodiments of the present application to make a clear and complete description of the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application and not all embodiments. The components of the embodiments of the present application described and shown in the accompanying drawings can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. Based on the embodiments of the present application, every other embodiment obtained by a person skilled in the art without creative work belongs to the scope of protection of the present application.

[0021] Firstly, the application scenarios applicable to the present application are introduced. The present application can be applied to the technical field of battery management.

[0022] With the rapid development of the new energy vehicle industry, the high-voltage safety problem of the power battery as a core component is increasingly prominent. In the high-voltage system of an electric vehicle, the advantages and disadvantages of the insulation performance are directly related to the safe operation of the vehicle. In the prior art, the power battery system usually adopts a battery management system (BMS) for real-time monitoring, and insulation detection is one of the important functions to ensure high-voltage safety.

[0023] It is found through research that the insulation detection scheme commonly used in the industry at present is mainly based on an active or passive balancing method, which judges the insulation state by measuring the insulation resistance value of the total positive electrode and the total negative electrode of the battery to the vehicle body. When the insulation resistance is detected to be lower than the preset safety threshold, the system will trigger an alarm mechanism to prompt the existence of insulation failure risk. This traditional detection method can only issue an alarm after the insulation resistance value drops below the safety threshold, and cannot capture the slow degradation process of the insulation performance. This lag cannot timely discover potential insulation hazards, and cannot give early warning of possible insulation failures. When the insulation resistance value is detected to be abnormal, it has often developed to a stage that may cause serious accidents such as electric leakage, short circuit and fire, and there is a significant safety hazard.

[0024] Based on this, the embodiments of the present application provide an insulation fault detection method for a power battery, which can identify the degradation trend of the insulation performance earlier, improve the sensitivity and accuracy of fault detection, effectively avoid safety accidents caused by lagging alarm of insulation failure, and significantly improve the operation safety and reliability of the power battery system.

[0025] Please refer to Figure 1 , Figure 1 The flowchart of the insulation fault detection method for a power battery provided by the embodiments of the present application is shown in FIG. 1. As shown in FIG. 1, the insulation fault detection method provided by the embodiments of the present application comprises the following steps. Figure 1 ​S101, acquire multiple state time series data of the power battery.

[0026] For the above step S101, in the specific implementation, multiple state time series data of the power battery are acquired. According to the embodiments provided in the present application, the multiple state time series data include insulation resistance value time series data of the battery module in the power battery collected by the insulation monitoring unit, and sensor time series data of the power battery collected by each preset sensor. Specifically, on the basis of the total insulation resistance measured by the conventional BMS master control unit, a distributed insulation detection subunit is deployed inside or near each battery module for real-time or periodic monitoring of the local insulation resistance value of each module. The preset sensor includes an air pressure sensor, a humidity sensor, a gas sensor and a temperature sensor. The air pressure sensor is installed in the electrolyte storage area and the battery module gap position to monitor the internal air pressure change of the battery pack. Leakage can cause air pressure to drop or humidity to fluctuate abnormally. The humidity sensor is deployed in the key areas inside the battery pack, such as the module gap and the high-voltage connector, to monitor the humidity value of the environment inside the pack. The gas sensor is installed in the module gap or the top space area of the cell to monitor specific gases generated due to insulation failure and local discharge, such as HF. The temperature sensor is uniformly distributed between the battery modules to monitor the temperature of the cell and the module. Insulation failure is often accompanied by local overheating. The air pressure sensor collects air pressure time series data of the power battery, the humidity sensor collects humidity time series data of the power battery, the gas sensor collects concentration time series data of the characteristic gas inside the power battery, and the temperature sensor collects temperature time series data of the power battery. The above-mentioned preset sensor continuously collects data at a certain collection frequency to form a multi-dimensional time series.

[0027] S102, weighting data fusion is performed on the multiple state time series data to determine the insulation health score of the power battery.

[0028] For the above step S102, in the specific implementation, the multiple source heterogeneous state time series data are fused and processed to comprehensively evaluate the overall insulation health level of the battery system. The score not only reflects whether the current insulation resistance is abnormal, but also considers potential factors that may cause insulation degradation, thereby achieving more comprehensive health status evaluation.

[0029] Specifically, for the above step S102, the weighting data fusion is performed on the multiple state time series data to determine the insulation health score of the power battery, including: Step 1021, time axis synchronization processing is performed on the multiple state time series data to obtain multiple target time series data.

[0030] For the above step 1021, in specific implementation, due to the differences in data acquisition frequency and communication delay of different sensors (such as the IMU reporting insulation resistance once per second, and the temperature sensor updating once every 2 seconds), time alignment needs to be performed on all time series data. By interpolation method (such as linear interpolation, spline interpolation) or time stamp-based resampling strategy, the same time reference (such as 1 data point per second) is unified to generate the synchronized target time series data set.

[0031] Specifically, first, due to the different sampling frequencies of different sensors, the weighted data fusion algorithm needs to unify the time axis of each sensor by linear interpolation, and the average sampling time is calculated using the sampling time of each sensor, and the formula is:

[0032] Among them, T(res), T(Pre), T(wet), T(gas), and T(temperature) represent the sampling time of the corresponding sensor.

[0033] The T(mean) and linear interpolation method are used to unify the time axis of each sensor. Taking the temperature sensor as an example, the linear interpolation formula is:

[0034] Among them, represents the data value of the average sampling time of the temperature sensor, , needs to satisfy ≤ ≤ .

[0035] Step 1022, for each target time series data, the target time series data is filtered and denoised and normalized to obtain the corresponding sensor target data.

[0036] For the above step 1022, in specific implementation, after the time axis synchronization processing, for each synchronized target time series data, the target time series data is filtered and denoised and normalized to obtain the corresponding sensor target data. Specifically, after time axis alignment, the sensor data needs to be denoised by using sliding average filtering, and the sliding average filtering formula is:

[0037] Among them, represents the above-mentioned T(mean), N represents the window size, generally taking a value of 3, and t is the gradient sampling time with T(mean) as the sampling frequency.

[0038] Since the data units for pressure, humidity, and temperature from different sensors are different, the data from different sensors need to be normalized to the same scale using Min-Max normalization. The normalization formula is as follows:

[0039] Where x represents the normalized data, i.e., the sensor target data, and the output range is... , , Represents the minimum and maximum values ​​of the sensor data. , The sliding window statistical method is used for calculation. For example, 100 data points from the pressure sensor are selected as a fixed-length sliding window. The pressure sensor data is: [102.1, 101.8, 103.0, ..., 105.2]. At this time... =101.8, =105.2.

[0040] Step 1023: Perform weighted fusion calculation on the weight corresponding to each preset sensor and the target data of the sensor to obtain the insulation health score.

[0041] Regarding step 1023 above, in specific implementation, firstly, the weight corresponding to each preset sensor is determined. Then, a weighted fusion calculation is performed on the weight of each preset sensor and the target data to obtain the insulation health score. Specifically, for each sensor, the weight ratio of each sensor's data is calculated using the inverse variance weighting method to complete the weighted data fusion. The formula for the inverse variance weighting method is:

[0042] in, This represents the weighting coefficients of each sensor, and w(resistance) + w(pressure) + w(humidity) + w(air) + w(temperature) = 1. The standard deviation of the data from each sensor is represented by the following formula:

[0043] Then, using the weights of each sensor and normalized sensor target data The weighted fusion calculation of the comprehensive insulation health score s is performed. The weighted fusion calculation formula is as follows:

[0044] Thus, according to steps 1021-1023 above, time axis synchronization, filtering and denoising, and normalization are introduced during the multi-source data fusion process to ensure the consistency of sensor data with different sampling frequencies and units in terms of time and numerical scale. Furthermore, by assigning corresponding weights to the target data of each sensor for weighted fusion, the data of key sensors have a more significant impact on the final insulation health score, improving the robustness and accuracy of the insulation health status assessment.

[0045] S103, compare the insulation health score with multiple preset scoring thresholds to determine the insulation fault detection result of the power battery.

[0046] Regarding step S103 above, in specific implementation, after the insulation health score of the power battery is calculated, the insulation health score is compared with multiple preset scoring thresholds to determine the insulation fault detection result of the power battery. As an example, S... l S m and S h The threshold can be calculated based on the distribution of historical data. If S < S l If S is normal, it indicates that the airtightness is normal; continue monitoring. l ≤S<S m If S m ≤S<S h If S h ≤ S If the insulation fault detection result is a serious warning, then the result will be as follows.

[0047] As an optional embodiment, after acquiring multiple state timing data of the power battery, the insulation fault detection method provided in this application further includes: Multiple state time series data are simultaneously input into a pre-trained insulation fault prediction model to obtain the insulation fault prediction result output by the insulation fault prediction model.

[0048] In implementing the above steps, the prediction model is a deep learning model (such as LSTM, GRU, Transformer, or a CNN-LSTM hybrid structure) capable of capturing the dynamic evolution patterns in time-series data. The input is historical time-series data for N periods (e.g., data snippets from the past 5 minutes), and the output is a binary or multi-class classification result: for example, whether an insulation fault exists or the type of fault, such as positive electrode leakage, negative electrode leakage, or overall moisture absorption.

[0049] As an optional embodiment, the insulation fault prediction model is trained through the following steps: A: Obtain training samples.

[0050] For the above step A, in specific implementation, training samples are obtained. The training samples are derived from real vehicle operation data, laboratory accelerated aging experiments and simulation platform simulation data. Each training sample includes: an input part: a plurality of state time series sample data, each sensor data in a time window with a length of T. The state time series sample data herein is the same as the content contained in the state time series data in step S101, and will not be repeated here. A label part: an insulation fault sample label.

[0051] B: inputting the training sample into the insulation fault original prediction model to obtain an insulation fault prediction label.

[0052] C: determining a loss function according to the insulation fault sample label and the insulation fault prediction label, and iteratively training the insulation fault original prediction model based on the loss function until a preset training completion condition is met, to obtain the trained insulation fault prediction model.

[0053] The preset training completion condition herein can be that the number of iterations reaches a preset number, or the loss function reaches a preset threshold, which is not limited in the present application.

[0054] For the above steps B-C, in specific implementation, the training sample is input into the insulation fault original prediction model to obtain the insulation fault prediction label output by the insulation fault original prediction model. Then the insulation fault sample label and the insulation fault prediction label are compared to determine the loss function, and the insulation fault original prediction model is iteratively trained based on the loss function until a preset training completion condition is met, to obtain the trained insulation fault prediction model.

[0055] As an optional embodiment, after obtaining the plurality of state time series data of the power battery, the insulation fault detection method provided by the present application further includes: I: for each battery module, determining a current resistance value of the battery module based on the insulation resistance value time series data of the battery module.

[0056] II: when it is detected that the current resistance value is lower than a first resistance threshold, generating first insulation fault warning information of the battery module.

[0057] III: when it is detected that the current resistance value is lower than a second resistance threshold, generating second insulation fault warning information of the battery module, and sending a high-voltage cutoff instruction to a battery management module.

[0058] Wherein, the second resistance threshold is lower than the first resistance threshold.

[0059] For the above steps I-step III, in the specific implementation, for each battery module, the current resistance value of the battery module is determined based on the insulation resistance value time series data of the battery module. When the current resistance value is detected to be lower than the first resistance threshold, the first insulation fault warning information of the battery module is generated, and the first insulation fault warning information indicates that the battery module is suspected to have insulation fault. When the current resistance value is detected to be lower than the second resistance threshold, the second insulation fault warning information of the battery module is generated, and the high-voltage cut-off instruction is sent to the battery management module.

[0060] In this way, according to the above steps I-step III, the insulation resistance value of each battery module is independently monitored, and two-level warning mechanisms are set, the first warning is used to prompt maintenance, and the second warning triggers high-voltage cut-off, realizing a fine and hierarchical fault response strategy. When the resistance value is lower than the lower second resistance threshold, the high-voltage cut-off instruction is immediately sent, which can quickly isolate the high-voltage loop before serious leakage or short circuit occurs, improving the flexibility and reliability of the high-voltage safety management of the whole vehicle.

[0061] The insulation fault detection method of the power battery provided by the embodiment of the application first acquires a plurality of state time series data of the power battery; wherein the plurality of state time series data includes insulation resistance value time series data of a battery module in the power battery collected by an insulation monitoring unit, and sensor time series data of the power battery collected by each preset sensor; then, the plurality of state time series data is subjected to weighted data fusion to determine an insulation health score of the power battery; finally, the insulation health score is compared with a plurality of preset score thresholds to determine an insulation fault detection result of the power battery.

[0062] The application acquires a plurality of source state time series data of the power battery, and subjects the plurality of source state time series data to weighted data fusion to determine an insulation health score, and then comprehensively judges in combination with a preset score threshold, thereby realizing multi-dimensional evaluation of the insulation state of the power battery. Compared with the traditional method of only relying on a single insulation resistance threshold alarm, the application can earlier identify the degradation trend of the insulation performance, improve the sensitivity and accuracy of fault detection, effectively avoid safety accidents caused by late alarm of insulation fault, and significantly improve the operation safety and reliability of the power battery system.

[0063] Please refer to Figure 2 、 Figure 3 , Figure 2 Fig. 1 is a structural schematic diagram of an insulation fault detection device of a power battery provided by an embodiment of the application, Figure 3 Fig. 2 is a structural schematic diagram of an insulation fault detection device of a power battery provided by an embodiment of the application. As shown in Figure 2 Fig. 1, the insulation fault detection device 200 includes: The data acquisition module 201 is configured to acquire a plurality of state time series data of the power battery; wherein the plurality of state time series data comprises insulation resistance value time series data of a battery module in the power battery collected by an insulation monitoring unit, and sensor time series data of the power battery collected by each preset sensor. The insulation health score calculation module 202 is configured to perform weighted data fusion on the plurality of state time series data to determine an insulation health score of the power battery. The fault detection result determination module 203 is configured to compare the insulation health score with a plurality of preset score thresholds to determine an insulation fault detection result of the power battery.

[0064] Further, when the insulation health score calculation module 202 is configured to perform weighted data fusion on the plurality of state time series data to determine the insulation health score of the power battery, the insulation health score calculation module 202 is further configured to: perform time axis synchronization processing on the plurality of state time series data to obtain a plurality of target time series data; for each target time series data, perform filtering and denoising and normalization processing on the target time series data to obtain corresponding sensor target data; perform weighted fusion calculation on the weight corresponding to each preset sensor and the sensor target data to obtain the insulation health score.

[0065] Please refer to Figure 3 The insulation fault detection device 200 further comprises a model prediction module 204, which is configured to: input the plurality of state time series data into a pre-trained insulation fault prediction model to obtain an insulation fault prediction result output by the insulation fault prediction model.

[0066] Further, the model prediction module 204 is further configured to train the insulation fault prediction model by the following steps: acquire training samples; wherein the training samples comprise a plurality of state time series sample data and insulation fault sample labels; input the training samples into an insulation fault original prediction model to obtain insulation fault prediction labels; determine a loss function according to the insulation fault sample labels and the insulation fault prediction labels, and iteratively train the insulation fault original prediction model based on the loss function until a preset training completion condition is met, to obtain the trained insulation fault prediction model.

[0067] Please refer to Figure 3The insulation fault detection apparatus 200 further comprises a battery module detection module 205, after the plurality of state time sequence data of the power battery is acquired, the battery module detection module 205 is used for: For each battery module, the current resistance value of the battery module is determined based on the insulation resistance value time sequence data of the battery module; When the current resistance value is detected to be lower than a first resistance threshold, first insulation fault warning information of the battery module is generated; When the current resistance value is detected to be lower than a second resistance threshold, second insulation fault warning information of the battery module is generated, and a high-voltage cut-off instruction is sent to the battery management module; wherein the second resistance threshold is lower than the first resistance threshold.

[0068] Please refer to Figure 4 , Figure 4 A structural schematic diagram of an electronic device provided by an embodiment of the present application is shown in FIG. 4. As shown in FIG. 4, the electronic device 400 comprises a processor 410, a memory 420 and a bus 430. Figure 4

[0069] The memory 420 stores machine readable instructions executable by the processor 410. When the electronic device 400 is running, the processor 410 and the memory 420 communicate through the bus 430. The machine readable instructions, when executed by the processor 410, can perform the steps of the insulation fault detection method of the power battery in the method embodiment shown in the above Figure 1 The specific implementation can be referred to the method embodiment, and will not be described here.

[0070] The present application further provides a computer readable storage medium, the computer readable storage medium stores a computer program, the computer program is run by the processor, and the computer program can perform the steps of the insulation fault detection method of the power battery in the method embodiment shown in the above Figure 1 The specific implementation can be referred to the method embodiment, and will not be described here.

[0071] Those skilled in the art can clearly understand that, for the convenience and brevity of the description, the specific working process of the above-described system, device and unit can refer to the corresponding process in the foregoing method embodiment, and will not be described here.

[0072] ​In several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented in other manners. The described device embodiments are merely schematic, for example, the division of the units is only a logical function division, and there can be another division manner in actual implementation, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections between different units, or the among different units, can be indirect couplings or communication connections through some interfaces, devices or units, and can be in electrical, mechanical or other forms.

[0073] The units described as separated components can or can not be physically separated, and the components displayed as units can or can not be physical units, i.e., can be located in one place, or can be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purposes of the embodiments of the present application.

[0074] In addition, each functional unit in the embodiments of the present application can be integrated in one processing unit, or each unit can exist physically as a separate unit, or two or more units can be integrated in one unit.

[0075] If the functions are implemented in the form of software function units and sold or used as independent products, they can be stored in a non-volatile computer readable storage medium executable by a processor. Based on this understanding, the technical solutions of the present application or the parts of the technical solutions that make essential contributions to the prior art can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0076] Finally, it should be noted that the above-described embodiments are merely specific embodiments of the present application, which are used to illustrate the technical solutions of the present application, but not to limit the same. The protection scope of the present application is not limited thereto. Although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those skilled in the art that any skilled person in the art can still modify or easily think of changes to the technical solutions recorded in the foregoing embodiments, or make equivalent replacements to some of the technical features, within the technical scope disclosed by the present application. The modifications, changes or replacements do not make the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for detecting insulation faults of a power battery, characterized in that, The insulation fault detection method comprises: Obtaining a plurality of state time series data of the power battery; wherein the plurality of state time series data comprises insulation resistance value time series data of a battery module in the power battery collected by an insulation monitoring unit, and sensor time series data of the power battery collected by each preset sensor; Performing weighted data fusion on the plurality of state time series data to determine an insulation health score of the power battery; Comparing the insulation health score with a plurality of preset score thresholds to determine an insulation fault detection result of the power battery.

2. The insulation fault detection method according to claim 1, characterized by, The insulation health score of the power battery is determined by performing weighted data fusion on the plurality of state time series data, comprising: Performing time axis synchronization processing on the plurality of state time series data to obtain a plurality of target time series data; For each target time series data, performing filtering and denoising and normalization processing on the target time series data to obtain corresponding sensor target data; Performing weighted fusion calculation on the weight corresponding to each preset sensor and the sensor target data to obtain the insulation health score.

3. The insulation fault detection method according to claim 1, characterized by, After the plurality of state time series data of the power battery is obtained, the insulation fault detection method further comprises: Inputting the plurality of state time series data into a pre-trained insulation fault prediction model to obtain an insulation fault prediction result output by the insulation fault prediction model.

4. The insulation fault detection method according to claim 3, characterized by, The insulation fault prediction model is trained by the following steps: Obtaining training samples; wherein the training samples comprise a plurality of state time series sample data and insulation fault sample labels; Inputting the training samples into an insulation fault original prediction model to obtain insulation fault prediction labels; Determining a loss function according to the insulation fault sample labels and the insulation fault prediction labels, and iteratively training the insulation fault original prediction model based on the loss function until a preset training completion condition is met, thereby obtaining the trained insulation fault prediction model.

5. The insulation fault detection method according to claim 1, characterized by, After the plurality of state time series data of the power battery is obtained, the insulation fault detection method further comprises: For each battery module, determining a current resistance value of the battery module based on the insulation resistance value time series data of the battery module; When it is detected that the current resistance value is lower than a first resistance threshold, first insulation fault warning information of the battery module is generated; When it is detected that the current resistance value is lower than a second resistance threshold, second insulation fault warning information of the battery module is generated, and a high-voltage cutoff instruction is sent to a battery management module; wherein the second resistance threshold is lower than the first resistance threshold.

6. An insulation fault detection device for a power cell, characterized by The insulation fault detection device comprises: A data acquisition module for obtaining a plurality of state time series data of the power battery; wherein the plurality of state time series data comprises insulation resistance value time series data of a battery module in the power battery collected by an insulation monitoring unit, and sensor time series data of the power battery collected by each preset sensor; An insulation health score calculation module for performing weighted data fusion on the plurality of state time series data to determine an insulation health score of the power battery; The fault detection result determination module is configured to compare the insulation health score with a plurality of preset score thresholds, and determine an insulation fault detection result of the power battery.

7. The insulation fault detection apparatus according to claim 6, characterized by When the insulation health score calculation module is configured to perform weighted data fusion on the plurality of state time series data to determine the insulation health score of the power battery, the insulation health score calculation module is further configured to: synchronize the plurality of state time series data in a time axis to obtain a plurality of target time series data; for each target time series data, perform filtering and denoising and normalization processing on the target time series data to obtain corresponding sensor target data; perform weighted fusion calculation on the weight corresponding to each preset sensor and the sensor target data to obtain the insulation health score.

8. The insulation fault detection apparatus according to claim 6, characterized by, The insulation fault detection device further comprises a model prediction module, after the plurality of state time series data of the power battery are obtained, the model prediction module is configured to: input the plurality of state time series data into a pre-trained insulation fault prediction model to obtain an insulation fault prediction result output by the insulation fault prediction model.

9. An electronic device, comprising: comprise: a processor, a memory and a bus, the memory stores machine readable instructions executable by the processor, when the electronic device is running, the processor and the memory communicate through the bus, the machine readable instructions are executed by the processor to perform the steps of the insulation fault detection method of the power battery as claimed in any one of claims 1 to 5.

10. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, the computer program is executed by the processor to perform the steps of the insulation fault detection method of the power battery as claimed in any one of claims 1 to 5.

Images