Fault diagnosis method, device, medium, and equipment based on charging data

Abstract

A fault diagnosis method, device, medium, and equipment based on charging data are provided. [Solution] Charging data is collected during the charging process, static features and time series features are extracted from the charging data, the static features are input into a support vector machine model to obtain a first failure probability, the time series features are input into a long short-term memory network to obtain a second failure probability, and weighted addition is performed based on the first failure probability and the second failure probability to obtain an overall reliability; that is, charging data is collected in real time during the charging process, static features and time series features are extracted from it, and the support vector machine model and the long short-term memory network are used to identify the static features and time series features respectively to obtain the first failure probability and the second failure probability, and the first failure probability and the second failure probability are combined to obtain the reliability of charging failure, thereby improving the accuracy of fault diagnosis.

Description

[Technical Field]

[0001] The present invention relates to the field of charging monitoring technology, and more particularly to a fault diagnosis method, device, medium and equipment based on charging data. [Background technology]

[0002] Electric vehicles have been rapidly gaining popularity in recent years due to their environmentally friendly, clean, and energy-efficient features. Accordingly, electric vehicle charging facilities, as an important component in this trend, are rapidly developing. The number and frequency of use of charging piles, the primary charging devices for electric vehicles, are also rapidly increasing. However, various factors can cause charging failures during the charging process, leading to charging abnormalities and further severely impacting charging safety. Therefore, there is a need for fault diagnosis during the charging process of electric vehicles. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] China Patent Publication Publication CN116819328A [Non-patent literature]

[0004] [Non-Patent Document 1] "Sensor Drift Compensation Based on the Improved LSTM and SVM Multi-Class Ensemble Learning Models," Xia Zhao et al., Sensors, pp. 1-25 Summary of the Invention [Problem to be solved by the invention]

[0005] Conventional charging fault diagnosis techniques rely primarily on static parameters, such as charging current and charging voltage, acquired by sensors or controllers during the charging process to determine whether or not the battery is hot. However, excessive charging current during actual charging can cause rapid temperature rises and lead to safety hazards. Furthermore, it is difficult to predict faults in advance using static parameters alone. Therefore, there is a need for a method that can quickly and accurately predict charging faults. [Means for solving the problem]

[0006] In order to solve the above technical problems, the present invention is proposed. The embodiments of the present invention provide a method, a device, a medium and an apparatus for diagnosing a fault based on charging data.

[0007] According to one aspect of the present invention, a fault diagnosis method based on charging data is provided. The fault diagnosis method includes: collecting charging data during a charging process; extracting static features and time series features from the charging data; inputting the static features into a support vector machine model to calculate a first failure probability; inputting the time series features into a long short-term memory network to calculate a second failure probability; and performing weighted addition based on the first failure probability and the second failure probability to obtain an overall reliability, wherein the collection frequency of the charging data is set according to the state of charge of a rechargeable battery; the overall reliability indicates the reliability that a fault exists in the current state of charge; the weight of the first failure probability is positively correlated with the accuracy rate of the support vector machine model; and the weight of the second failure probability is positively correlated with the accuracy rate of the long short-term memory network.

[0008] Another aspect of the present invention provides a fault diagnosis device based on charging data, the fault diagnosis device comprising: a charging data collection module that collects charging data during a charging process; a data feature extraction module that extracts static features and time series features from the charging data; a first probability calculation module that inputs the static features into a support vector machine model to calculate a first failure probability; a second probability calculation module that inputs the time series features into a long short-term memory network to calculate a second failure probability; and a reliability calculation module that performs weighted addition based on the first failure probability and the second failure probability to calculate an overall reliability, wherein the collection frequency of the charging data is related to the state of charge of a rechargeable battery, the overall reliability indicates the reliability that a fault exists in the current state of charge, and the weight of the first failure probability is positively correlated with the accuracy rate of the support vector machine model and the weight of the second failure probability is positively correlated with the accuracy rate of the long short-term memory network.

[0009] According to another aspect of the present invention, there is provided a computer-readable storage medium having a computer program stored therein, the computer program being used to cause any one of the above methods to be performed.

[0010] According to another aspect of the present invention, there is provided an electronic device comprising a processor and a memory storing instructions executable by the processor, the processor being configured to perform any of the methods described above. [Effects of the Invention]

[0011] The present invention provides a fault diagnosis method, device, medium, and equipment based on charging data, which collects charging data during a charging process, sets the frequency of collecting the charging data according to the state of charge of a rechargeable battery, extracts static features and time series features from the charging data, inputs the static features into a support vector machine model to obtain a first failure probability, inputs the time series features into a long short-term memory network to obtain a second failure probability, and performs weighted addition based on the first failure probability and the second failure probability to calculate an overall reliability, which can accurately evaluate the failure occurrence rate at the current state of charge. The weight of the first failure probability is positively correlated with the accuracy rate of the support vector machine model, and the weight of the second failure probability is positively correlated with the accuracy rate of the long short-term memory network. That is, charging data is collected in real time during the charging process, static features and time series features are extracted from the charging data, and the static features and time series features are respectively identified using a support vector machine model and a long short-term memory network to calculate the first fault probability and the second fault probability. The first fault probability and the second fault probability are then combined to obtain the reliability of the charging fault, thereby improving the accuracy of the fault diagnosis. [Brief explanation of the drawings]

[0012] The above and other objects, features, and advantages of the present invention will become more apparent by describing the embodiments of the present invention in more detail with reference to the drawings. The drawings are used to provide a further understanding of the embodiments of the present invention, constitute a part of the specification, and are used to interpret the present invention together with the embodiments of the present invention, but do not limit the present invention. In the drawings, the same reference numerals generally indicate the same or corresponding parts or steps. [Figure 1] 4 is a flowchart of a fault diagnosis method based on charging data according to an exemplary embodiment of the present invention. [Figure 2] 1 is a schematic diagram illustrating the configuration of a fault diagnosis device based on charge data according to an exemplary embodiment of the present invention; [Figure 3] 1 is a block diagram of an electronic device according to an exemplary embodiment of the present invention; DETAILED DESCRIPTION OF THE INVENTION

[0013] Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the drawings. Those skilled in the art will understand that the described embodiments are only some of the embodiments of the present invention, but are not all of the embodiments of the present invention, and the present invention is not limited to the exemplary embodiments described herein.

[0014] 1 is a flowchart of a fault diagnosis method based on charging data according to an exemplary embodiment of the present invention. As shown in FIG. 1, the fault diagnosis method based on charging data of this embodiment includes the following steps: In step 110, charging data is collected during the charging process.

[0015] The charging data collection frequency is set according to the charging state of the rechargeable battery. In the present invention, the charging process of the rechargeable battery is monitored in real time by collecting charging data during the charging process of the rechargeable battery. The charging data collection frequency is related to the state of charge (SOC) of the rechargeable battery. Specifically, when the SOC of the rechargeable battery is less than 20%, the collection frequency is 1 Hz, when the SOC of the rechargeable battery is 20% to 80%, the collection frequency is 10 Hz, and when the SOC of the rechargeable battery is more than 80%, the collection frequency is 1 Hz. That is, when the SOC of the rechargeable battery is too low or too high (when the charging current is usually small), charging data is collected at a low collection frequency, but when the SOC is in the range of 20% to 80% (when the charging current is usually large), charging data is collected at a high frequency, thereby improving the monitoring effect.

[0016] In step 120, static and time series features are extracted from the charging data.

[0017] In the present invention, after collecting charging data, static features (e.g., voltage average value, temperature extreme value, etc.) and time series features (e.g., voltage fluctuation curve, temperature change trend, etc.) are extracted from the charging data, and based on these features, it is determined whether a fault exists in the charging process.

[0018] In step 130, the extracted static features are input into a support vector machine model to calculate the first failure probability.

[0019] In the present invention, the support vector machine model is trained by a plurality of sample data, and the first failure probability can be obtained by inputting the extracted static features.

[0020] In step 140, the extracted time series features are input into a long short-term memory network (LSTM) to calculate the second failure probability.

[0021] In the present invention, the long short-term memory network is also trained by a plurality of sample data, and the second failure probability is obtained by inputting the time series features.

[0022] In step 150, a weighted sum is performed based on the first failure probability and the second failure probability to calculate an overall reliability.

[0023] Here, the overall reliability is an index indicating the reliability of the existence of a fault in the current charging state. The weight of the first fault probability is positively correlated with the accuracy rate of the support vector machine model, and the weight of the second fault probability is positively correlated with the accuracy rate of the long short-term memory network. According to the present invention, the probability of a fault occurring in the charging process is determined using the support vector machine model and the long short-term memory network, respectively, and the first fault probability and the second fault probability are weighted and added to obtain a final fault probability (i.e., the overall reliability), thereby utilizing the advantages of the two models in a complementary manner and improving the accuracy of fault determination.

[0024] The fault diagnosis method based on charging data provided by the present invention collects charging data during the charging process, sets the charging data collection frequency according to the charging state of the rechargeable battery, extracts static features and time series features from the charging data, inputs the static features into a support vector machine model to calculate a first fault probability, inputs the time series features into a long short-term memory network to calculate a second fault probability, and performs weighted addition based on the first and second fault probabilities to calculate an overall reliability, where the overall reliability indicates the reliability of a fault occurring in the current charging state, the weight of the first fault probability is positively correlated with the accuracy rate of the support vector machine model, and the weight of the second fault probability is positively correlated with the accuracy rate of the long short-term memory network. That is, charging data is collected in real time during the charging process, and static features and time series features are extracted therefrom. The static features and time series features are respectively identified using the support vector machine model and the long short-term memory network to calculate the first and second fault probabilities, and the first and second fault probabilities are combined to obtain the reliability of charging faults, thereby improving the accuracy of fault diagnosis.

[0025] In one embodiment, the specific implementation of step 120 may be as follows: extract a voltage-temperature coupling coefficient and a charging state of health index from the charging data, where the voltage-temperature coupling coefficient indicates the dynamic relationship between the cell voltage and temperature of the charging battery, and the charging state of health index indicates the similarity between the current charging curve and the historical reference charging curve.

[0026] By extracting the voltage-temperature coupling coefficient and the charging health index, the present invention obtains an indicator for determining whether a fault exists in the charging process. The voltage-temperature coupling coefficient indicates the dynamic correlation between cell voltage and temperature, and the state-of-charge index represents the similarity between the current charge curve and the historical reference charge curve.

[0027] In one embodiment, the specific implementation of step 120 may be as follows: The voltage-temperature coupling coefficient is calculated using the following formula (1). JPEG0007828506000002.jpg47160The calculation formula for the charging health index is expressed as follows: JPEG0007828506000003.jpg51157

[0028] The present invention calculates the voltage-temperature coupling coefficient and the state-of-charge index using the above formulas. Here, the value range of the voltage-temperature coupling coefficient is [-1, 1]. The voltage-temperature coupling coefficient is used to identify the risk of thermal runaway in a rechargeable battery, and the state-of-charge index is an index for evaluating the state of health of a rechargeable battery. If the state-of-charge index is continuously below a set threshold, it indicates a deterioration in the capacity of the rechargeable battery or an increase in its internal resistance. In such cases, further diagnosis is required to determine whether a fault exists in the charging process.

[0029] In one embodiment, the above-described fault diagnosis method based on charging data may further include determining that a fault exists in the current charging state if a difference between the voltage-temperature coupling coefficient at the current time and the voltage-temperature coupling coefficient at the previous time exceeds a predetermined difference threshold.

[0030] The present invention can extract the voltage-temperature coupling coefficient and then determine whether there is a risk of thermal runaway in the rechargeable battery based on changes in the coefficient. Specifically, the difference between the voltage-temperature coupling coefficient at the current time and the voltage-temperature coupling coefficient at the previous time is calculated, and if this difference exceeds a predetermined difference threshold (i.e., if the voltage-temperature coupling coefficient increases suddenly), it is determined that abnormal heat is occurring inside the rechargeable battery, and an immediate alarm is issued to avoid a safety accident.

[0031] In one embodiment, the specific implementation of the above step 150 may be as follows: The calculation formula for the overall reliability is expressed by the following formula (3). JPEG0007828506000004.jpg59156

[0032] The present invention integrates the diagnosis results of the support vector machine model and the long short-term memory network by calculating the overall reliability using the above formula to obtain a final fault diagnosis result, i.e., the final fault probability. A higher overall reliability indicates a higher probability of a fault existing in the charging process. The present invention further improves the accuracy of fault diagnosis by determining the weights of the two models based on the accuracy rates of the support vector machine model and the long short-term memory network (obtained from the validation data set).

[0033] After calculating the overall reliability, the present invention can execute an operation based on a warning measure corresponding to a preset failure probability. For example, the warning mechanism of the present invention can be designed into five stages: if the overall reliability is less than 30%, no operation is triggered; if the overall reliability is 30% to 50%, a notification is triggered; if the overall reliability is 50% to 70%, current-limited charging is triggered; if the overall reliability is 70% to 90%, power-reduced charging is triggered; and if the overall reliability is over 90%, a power-off operation is triggered.

[0034] In one embodiment, the specific implementation manner of the above step 110 may be as follows: During the charging process, the charging battery data collected by the charging battery and the charging pile data collected by the charging pile are obtained, and these data are matched to generate charging data.

[0035] This invention uses both the rechargeable battery (e.g., BMS) and the charging pile to simultaneously collect rechargeable battery data and charging pile data during the charging process. It also incorporates a data consistency check mechanism, which cross-verifies the rechargeable battery data and charging pile data to ensure they are within acceptable ranges. It also cross-checks the rechargeable battery data and charging pile data to complement missing data and remove outliers and noise, thereby improving data reliability. If a discrepancy is found between the rechargeable battery data and charging pile data during the verification process, it can trigger a data recollection or an alarm notification.

[0036] In one embodiment, the specific implementation of the above step 110 may be as follows: If the inter-cell voltage difference of the rechargeable battery exceeds a preset voltage threshold, increase the frequency of collecting charging data.

[0037] The present invention collects cell voltages of a rechargeable battery to obtain cell state information. If the voltage difference between cells exceeds a preset voltage threshold, it indicates a large difference in the state of charge between the cells. In this case, the frequency of collecting charging data can be increased (for example, up to 100 Hz), thereby improving charging safety.

[0038] 2 is a schematic diagram of a charging data-based fault diagnosis device according to an exemplary embodiment of the present invention. As shown in FIG. 2, the charging data-based fault diagnosis device 20 includes a charging data collection module 21 that collects charging data during a charging process, a data feature extraction module 22 that extracts static and time-series features from the charging data, a first probability calculation module 23 that inputs the static features into a support vector machine model to calculate a first failure probability, a second probability calculation module 24 that inputs the time-series features into a long short-term memory network to calculate a second failure probability, and a reliability calculation module 25 that performs weighted summation based on the first and second failure probabilities to calculate an overall reliability. The charging data collection frequency is set according to the state of charge of the rechargeable battery. The overall reliability indicates the reliability of a fault occurring at the current state of charge. The weight of the first failure probability is positively correlated with the accuracy rate of the support vector machine model, and the weight of the second failure probability is positively correlated with the accuracy rate of the long short-term memory network.

[0039] The charging data-based fault diagnosis device provided by the present invention uses a charging data collection module 21 to collect charging data during the charging process, where the frequency of collecting charging data is set according to the charging state of the rechargeable battery. A data feature extraction module 22 extracts static features and time series features from the charging data. A first probability calculation module 23 inputs the static features into a support vector machine model to calculate a first failure probability. A second probability calculation module 24 inputs the time series features into a long short-term memory network to calculate a second failure probability. A reliability calculation module 25 performs weighted addition based on the first failure probability and second failure probability to calculate an overall reliability. Here, the overall reliability indicates the reliability of a fault in the current charging state, the weight of the first fault probability has a positive correlation with the accuracy rate of the support vector machine model, and the weight of the second fault probability has a positive correlation with the accuracy rate of the long short-term memory network. That is, according to the present invention, charging data is collected in real time during the charging process, and its static features and time series features are extracted and analyzed. The static features and time series features are respectively identified using the support vector machine model and the long short-term memory network, thereby obtaining and synthesizing the first fault probability and the second fault probability, and the first fault probability and the second fault probability, thereby improving the accuracy of fault diagnosis.

[0040] In one embodiment, the data feature extraction module 22 may be further configured to: extract a voltage-temperature coupling coefficient and a charging state-of-health index from the charging data, where the voltage-temperature coupling coefficient indicates the dynamic relationship between the cell voltage and temperature of the charging battery, and the charging state-of-health index indicates the similarity between the current charging curve and a historical reference charging curve.

[0041] In one embodiment, the data feature extraction module 22 may be further configured as follows: The voltage-temperature coupling coefficient is calculated using the following formula (1). JPEG0007828506000005.jpg149159

[0042] In one embodiment, the charging data-based fault diagnosis device 20 may be further configured as follows: if the difference between the voltage-temperature coupling coefficient at the current time and the voltage-temperature coupling coefficient at the previous time exceeds a preset difference threshold, determine that a fault exists in the current charging state.

[0043] In one embodiment, the reliability calculation module 25 may be further configured as follows: The calculation formula for the overall reliability is expressed by the following formula (3). JPEG0007828506000006.jpg55154

[0044] In one embodiment, the charging data collection module 21 may be further configured as follows: acquire charging battery data collected by the charging battery and charging pile data collected by the charging pile during the charging process, and match these data to generate charging data.

[0045] In one embodiment, the charging data collection module 21 may be further configured as follows: When the voltage difference between cells of the rechargeable battery exceeds a preset voltage threshold, increase the frequency of collecting charging data.

[0046] An electronic device according to an embodiment of the present invention will be described below with reference to Fig. 3. The electronic device may be either one or both of the first and second devices, or a standalone device independent of them, and the standalone device can communicate with the first and second devices and receive input signals collected therefrom.

[0047] FIG. 3 is a block diagram of an electronic device according to an embodiment of the present invention.

[0048] As shown in FIG. 3, the electronic device 10 includes one or more processors 11 and a memory 12.

[0049] The processor 11 may be a central processing unit (CPU) or other type of processing device having data processing and / or instruction execution capabilities, and may further control other components within the electronic device 10 to perform desired functions.

[0050] The memory 12 can store one or more computer program products, which can include various types of computer-readable storage media, such as volatile memory and / or nonvolatile memory. The volatile memory can include, for example, random access memory (RAM) and / or cache memory. The non-volatile memory can include, for example, read-only memory (ROM), a hard disk, or flash memory. One or more computer program instructions can be stored in the computer-readable storage medium, and the processor 11 executes the program instructions to implement the methods according to the above-described embodiments of the present invention and / or other desired functions. The computer-readable storage medium can also store various contents, such as an input signal, a signal component, or a noise component.

[0051] In one embodiment, electronic device 10 may further include input devices 13 and output devices 14, with these components connected to one another via a bus system and / or other connection mechanism (not shown).

[0052] If the electronic device is a stand-alone device, the input device 13 may be a communication network connector for receiving input signals collected from the first device and the second device.

[0053] The input device 13 may further include, for example, a keyboard, a mouse, and the like.

[0054] The output device 14 can output various information to the outside, including the determined distance information, direction information, etc. The output device 14 can include, for example, a display, a speaker, a printer, a communication network, and a remote output device connected thereto.

[0055] Of course, for simplicity, Figure 3 shows only some of the components of electronic device 10 that are relevant to the present invention, and omits components such as buses, input / output interfaces, etc. Furthermore, electronic device 10 may optionally include other suitable components depending on the specific application.

[0056] In addition to the methods and apparatus described above, embodiments of the present invention may also be a computer program product including computer program instructions that, when executed by a processor, cause the processor to perform the method steps according to various embodiments of the present invention as described in the "Exemplary Methods" section herein above.

[0057] The computer program product may be written in any combination of one or more programming languages ​​to create program code for performing operations according to embodiments of the present invention, including object-oriented programming languages ​​such as Java, C++, and the like, as well as traditional procedural programming languages ​​such as "C" or similar. The program code may run entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0058] It should be noted that an embodiment of the present invention may also be a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, cause the processor to perform the method steps according to various embodiments of the present invention as described herein above in the "Exemplary Methods" section.

[0059] The computer-readable storage medium may be composed of any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may include, for example, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (non-exhaustive list) of readable storage media include an electrical connection having one or more conductors, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0060] Although the basic principles of the present invention have been described above with reference to specific embodiments, the benefits, advantages, effects, etc. mentioned in the present invention are merely examples and not limitations. These benefits, advantages, effects, etc. should not be considered essential to each embodiment of the present invention. Furthermore, the specific details disclosed above are merely for illustrative purposes and to facilitate understanding, and are not limiting. The details do not limit the scope of the present invention that must be realized by adopting the specific details.

[0061] Block diagrams of devices, apparatus, instruments, and systems according to the present invention are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner depicted in the block diagrams. As one skilled in the art will recognize, these devices, apparatus, instruments, and systems may be connected, arranged, or configured in any manner. As one skilled in the art will recognize, these devices, apparatus, instruments, and systems may be connected, arranged, or configured in any manner. Terms such as "comprising," "including," and "having" are open-ended terms and mean, and can be used interchangeably, "including but not limited to." As used herein, the terms "or" and "and" refer to, and can be used interchangeably with, the term "and / or," unless the context clearly indicates otherwise. As used herein, the term "for example," "for example," or "for example" refers to, and can be used interchangeably with, the phrase "such as, but not limited to."

[0062] It should be noted that in the device, apparatus and method of the present invention, each component or each step can be disassembled and / or reassembled, and such disassembly and / or reassembly should be considered as an equivalent solution of the present invention.

[0063] The previous description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present disclosure. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of the present invention. Thus, the present invention is not intended to be limited to the aspects shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0064] The foregoing description has been presented for purposes of illustration and description. Furthermore, this description is not intended to limit embodiments of the invention to the form disclosed herein. While several exemplary aspects and embodiments have been described above, those skilled in the art will recognize certain variations, modifications, variations, additions, and subcombinations thereof.

Claims

1. 1. A method for fault diagnosis based on charging data executed by a processor, comprising: collecting charging data during the charging process; extracting static features and time series features from the charging data; inputting the static features into a support vector machine model to calculate a first failure probability; inputting the time series features into a long short-term memory network to calculate a second failure probability; and calculating an overall reliability by performing weighted addition based on the first failure probability and the second failure probability, The frequency of collecting the charging data is related to the charging state of the rechargeable battery, the overall reliability indicates the reliability that a fault exists in the current charging state, the weight of the first fault probability is positively correlated with the accuracy rate of the support vector machine model, and the weight of the second fault probability is positively correlated with the accuracy rate of the long short-term memory network; extracting static features and time series features from the charging data includes extracting a voltage-temperature coupling coefficient and a state-of-charge index from the charging data, the voltage-temperature coupling coefficient indicating a dynamic relationship between cell voltage and temperature of the rechargeable battery, and the state-of-charge index indicating a similarity between a current charging curve and a historical reference charging curve; The voltage-temperature coupling coefficient is calculated using the following equation (1): Here, C ｖｔ denotes the voltage-temperature coupling coefficient, and ΔV ｉ is the voltage change at the i-th sampling point, ΔT ｉ is the temperature change at the i-th sampling point, The calculation formula for the state of charge index is expressed by the following formula (2): where CHI is the state of charge index and D current is the current charging curve, D baseline indicates the historical reference charge curve, and DTW(D current , D baseline ) indicates the similarity distance between the current charging curve and the historical reference charging curve, and MaxLength indicates the maximum length of the curves.

2. The fault diagnosis method based on charging data further includes:

2. The fault diagnosis method based on charging data according to claim 1, further comprising: determining that a fault exists in the current charging state when a difference between the voltage-temperature coupling coefficient at the current time and the voltage-temperature coupling coefficient at the previous time exceeds a predetermined difference threshold.

3. The step of performing weighted addition based on the first failure probability and the second failure probability to calculate an overall reliability includes: The calculation of the overall reliability is represented by the following formula (3): where m(A) represents the overall reliability and ω ＳＶＭ is the weight of the first failure probability, ω ＬＳＴＭ is the weight of the second failure probability, SVM ｐｒｏｂ is the first failure probability, LSTM ｐｒｏｂ denotes the second failure probability, Ａｃｃ SVM is the accuracy rate of the support vector machine model, Ａｃｃ 2. The fault diagnosis method based on charging data according to claim 1, wherein LSTM indicates the accuracy rate of a long short-term memory network.

4. The step of collecting charging data during the charging process includes: acquiring charging battery data collected by the charging battery and charging pile data collected by the charging pile during a charging process; 2. The fault diagnosis method based on charge data according to claim 1, further comprising: obtaining the charge data by matching based on the charge battery data and the charge pile data.

5. The step of collecting charging data during the charging process includes:

2. The method for fault diagnosis based on charging data according to claim 1, further comprising: increasing the frequency of collecting the charging data when a voltage difference between cells of the rechargeable battery exceeds a preset voltage threshold.

6. A fault diagnosis device based on charging data, a charging data collection module that collects charging data during the charging process; a data feature extraction module that extracts static features and time series features from the charging data; a first probability calculation module that inputs the static features into a support vector machine model to calculate a first failure probability; a second probability calculation module that inputs the time series features into a long short-term memory network to calculate a second failure probability; a reliability calculation module that performs weighted addition based on the first failure probability and the second failure probability to obtain an overall reliability; The frequency of collecting the charging data is set according to the charging state of the rechargeable battery; the overall reliability indicates the reliability that a fault exists in the current charging state, the weight of the first fault probability is positively correlated with the accuracy rate of the support vector machine model, and the weight of the second fault probability is positively correlated with the accuracy rate of the long short-term memory network; the data feature extraction module is further configured to extract a voltage-temperature coupling coefficient and a state-of-charge index in the charging data; The voltage-temperature coupling coefficient indicates a dynamic relationship between the cell voltage and temperature of the rechargeable battery, and the state-of-charge index indicates a similarity between a current charging curve and a historical reference charging curve; The voltage-temperature coupling coefficient is calculated using the following equation (1): Here, C ｖｔ denotes the voltage-temperature coupling coefficient, and ΔV ｉ is the voltage change at the i-th sampling point, ΔT ｉ is the temperature change at the i-th sampling point, The calculation formula for the state of charge index is expressed by the following formula (2): where CHI is the state of charge index and D current is the current charging curve, D baseline indicates the historical reference charge curve, and DTW(D current , D baseline ) indicates the similarity distance between the current charging curve and the historical reference charging curve, and MaxLength indicates the maximum length of the curves.

7. A computer-readable storage medium, comprising: A computer-readable storage medium having a computer program stored therein, the computer program being used to execute the method according to any one of claims 1 to 5.

8. An electronic device, a processor; a memory that stores instructions executable by the processor; 6. An electronic device, characterized in that the processor is used to execute the method according to any one of claims 1 to 5.

Images