Voltage anomaly detection method, detection device, detection system and storage medium

Abstract

The invention discloses a voltage anomaly detection method, a voltage anomaly detection device, a voltage anomaly detection system and a computer readable storage medium. The voltage anomaly detection method comprises the following steps: acquiring a plurality of voltage curves of a plurality of sample batteries which change along with time in the charging and discharging process, wherein the plurality of sample batteries comprise normal batteries and abnormal batteries; acquiring a time characteristic range when the abnormal battery is abnormal; constructing an unsupervised anomaly detection model according to the plurality of voltage curves and the time characteristic ranges; and performing voltage anomaly detection on the to-be-detected battery based on the unsupervised anomaly detection model. According to the voltage anomaly detection method, the voltage anomaly detection device, the voltage anomaly detection system and the computer readable storage medium provided by the embodiment of the invention, the unsupervised anomaly detection model is constructed according to the plurality of voltage curves and the time characteristic range so as to perform voltage anomaly detection on the to-be-detected battery; real-time accurate detection of voltage abnormity in the whole charging and discharging process of the multi-step multi-gradient current of the to-be-detected battery can be realized.

Description

Technical Field

[0001] The present application relates to the technical field of voltage anomaly detection, and in particular to a voltage anomaly detection method, a voltage anomaly detection device, a voltage anomaly detection system and a computer-readable storage medium. Background Art

[0002] As the core component for storing electric energy in new energy vehicles, batteries have broad application prospects in the field of new energy vehicles. Their stability and safety are important factors affecting the promotion of new energy vehicles. How to effectively detect voltage anomalies is crucial to improving the stability and safety of batteries. Summary of the invention

[0003] Embodiments of the present application provide a voltage anomaly detection method, a voltage anomaly detection device, a voltage anomaly detection system, and a computer-readable storage medium.

[0004] The voltage anomaly detection method according to the embodiment of the present application includes:

[0005] Acquire a plurality of voltage curves of voltage changes over time of a plurality of sample batteries during a charge and discharge process, wherein the plurality of sample batteries include normal batteries and abnormal batteries;

[0006] Obtaining a time characteristic range when the abnormal battery is abnormal;

[0007] constructing an unsupervised anomaly detection model according to the plurality of voltage curves and the time feature range;

[0008] Voltage anomaly detection is performed on the battery to be detected based on the unsupervised anomaly detection model.

[0009] In some embodiments, obtaining the time characteristic range of the abnormal battery when it is abnormal includes:

[0010] Acquire an abnormal voltage curve corresponding to the abnormal battery from the multiple voltage curves;

[0011] A time characteristic analysis is performed on the abnormal voltage curve to obtain the time characteristic range.

[0012] In some embodiments, constructing an unsupervised anomaly detection model according to the multiple voltage curves and the time feature range includes:

[0013] Acquire, according to the multiple voltage curves, multiple standard score curves showing changes of the standard scores of the multiple sample batteries over time during the charging and discharging process;

[0014] The unsupervised anomaly detection model is constructed according to the multiple standard score curves and the time feature range.

[0015] In some embodiments, the performing voltage anomaly detection on the battery to be detected based on the unsupervised anomaly detection model includes:

[0016] determining an anomaly score threshold based on the unsupervised anomaly detection model;

[0017] It is determined whether the battery to be detected has a voltage abnormality according to the abnormal score threshold.

[0018] In some embodiments, determining an anomaly score threshold based on the unsupervised anomaly detection model includes:

[0019] Set the initial score threshold;

[0020] Performing model training on the unsupervised anomaly detection model according to the multiple standard score curves and the initial score threshold;

[0021] The anomaly score threshold is determined by the trained unsupervised anomaly detection model.

[0022] In some embodiments, judging whether the battery to be detected has a voltage abnormality according to the abnormal score threshold includes:

[0023] Obtaining a voltage curve to be detected, in which the voltage of the battery to be detected changes with time during the charging and discharging process;

[0024] Acquire a standard score curve to be detected according to the voltage curve to be detected, in which the standard score of the battery to be detected changes with time during the charging and discharging process;

[0025] Whether the voltage of the battery to be detected is abnormal is determined according to the relationship between the standard score curve to be detected and the abnormal score threshold.

[0026] In some embodiments, judging whether the voltage of the battery to be detected is abnormal according to the relationship between the standard score curve to be detected and the abnormal score threshold value includes:

[0027] When the standard score in the to-be-detected standard score curve is greater than the abnormal score threshold, obtaining a current time feature corresponding to the standard score greater than the abnormal score threshold;

[0028] When the current time feature is within the time feature range, it is determined that the voltage of the battery to be detected is abnormal.

[0029] The voltage anomaly detection device according to the embodiment of the present application includes:

[0030] A first acquisition module, used to acquire a plurality of voltage curves of voltage changes over time of a plurality of sample batteries during a charge and discharge process, wherein the plurality of sample batteries include normal batteries and abnormal batteries;

[0031] A second acquisition module is used to acquire a time characteristic range when the abnormal battery is abnormal;

[0032] A construction module, configured to construct an unsupervised anomaly detection model according to the plurality of voltage curves and the time feature range;

[0033] The detection module is used to perform voltage anomaly detection on the battery to be detected based on the unsupervised anomaly detection model.

[0034] The voltage anomaly detection system of the embodiment of the present application includes one or more processors and a memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, the voltage anomaly detection method of any of the above embodiments is implemented.

[0035] The computer-readable storage medium of the embodiment of the present application stores a computer program thereon, and when the program is executed by a processor, the voltage anomaly detection method of any of the above-mentioned embodiments is implemented.

[0036] The voltage anomaly detection method, voltage anomaly detection device, voltage anomaly detection system and computer-readable storage medium of the embodiments of the present application construct an unsupervised anomaly detection model based on multiple voltage curves and time feature ranges to perform voltage anomaly detection on the battery to be detected, and can realize real-time and accurate detection of voltage anomalies in the entire charging and discharging process of the battery to be detected with multiple steps and multiple gradient currents.

[0037] Additional aspects and advantages of the embodiments of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the embodiments of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which:

[0039] Figure 1 is a flowchart of a voltage anomaly detection method according to certain embodiments of the present application;

[0040] Figure 2 is a flowchart of a voltage anomaly detection method according to certain embodiments of the present application;

[0041] Figure 3 is a schematic diagram of multiple voltage curves of multiple batteries in certain embodiments of the present application showing voltage changes over time during the charging process;

[0042] Figure 4 It is a schematic diagram of multiple voltage curves of multiple batteries in certain embodiments of the present application, showing voltage changes over time during the charging and discharging process;

[0043] Figure 5 is a flowchart of a voltage anomaly detection method according to certain embodiments of the present application;

[0044] Figure 6 is a flowchart of a voltage anomaly detection method according to certain embodiments of the present application;

[0045] Figure 7 is a schematic diagram of multiple standard score curves showing changes of standard scores over time during charging of multiple batteries in certain embodiments of the present application;

[0046] Figure 8 is a schematic diagram of multiple standard score curves showing changes of standard scores over time during the charge and discharge process of multiple batteries in certain embodiments of the present application;

[0047] Fig. 9 is a flowchart of a voltage anomaly detection method according to certain embodiments of the present application;

[0048] Fig.10 is a flowchart of a voltage anomaly detection method according to certain embodiments of the present application;

[0049] Fig.11 is a flowchart of a voltage anomaly detection method according to certain embodiments of the present application;

[0050] Fig.12 is a flowchart of a voltage anomaly detection method according to certain embodiments of the present application;

[0051] Fig.13 is a module schematic diagram of a voltage anomaly detection device according to certain embodiments of the present application;

[0052] Fig.14 is a module schematic diagram of a voltage anomaly detection system in certain embodiments of the present application;

[0053] Fig.15 It is a schematic diagram of the connection status between a computer-readable storage medium and a processor in certain embodiments of the present application.

[0054] Description of reference numerals:

[0055] Voltage anomaly detection device 100 , first acquisition module 10 , second acquisition module 20 , construction module 30 , detection module 40 , voltage anomaly detection system 200 , processor 210 , memory 220 , computer readable storage medium 300 , computer program 310 , processor 320 . DETAILED DESCRIPTION

[0056] The following is a further description of the embodiments of the present application in conjunction with the accompanying drawings. The same or similar reference numerals in the accompanying drawings represent the same or similar elements or elements with the same or similar functions from beginning to end. In addition, the embodiments of the present application described below in conjunction with the accompanying drawings are exemplary and are only used to explain the embodiments of the present application, and cannot be understood as limiting the present application.

[0057] See also Figure 1 and Figure 2 The present application embodiment provides a voltage anomaly detection method. The voltage anomaly detection method includes:

[0058] 010: obtaining multiple voltage curves of multiple sample batteries in which voltage changes with time during the charging and discharging process, where the multiple sample batteries include normal batteries and abnormal batteries;

[0059] 020: Get the time characteristic range of abnormal battery abnormality;

[0060] 030: Construct an unsupervised anomaly detection model based on multiple voltage curves and time feature ranges;

[0061] 040: Voltage anomaly detection of the battery to be tested based on the unsupervised anomaly detection model.

[0062] The voltage anomaly detection method of the implementation mode of the present application constructs an unsupervised anomaly detection model based on multiple voltage curves and time feature ranges to perform voltage anomaly detection on the battery to be detected, and can realize real-time and accurate detection of voltage anomalies in the entire charging and discharging process of the battery to be detected with multiple steps and multiple gradient currents.

[0063] Specifically, multiple sample batteries refer to a group of batteries used to collect voltages at various time points during the charging and discharging process to construct an unsupervised anomaly detection model. Multiple sample batteries can be produced by the production line at the same time to reduce variable interference. Multiple sample batteries include normal batteries and abnormal batteries to ensure the randomness and comprehensiveness of the samples, so that the constructed unsupervised anomaly detection model is more accurate. Among them, abnormal batteries can also be used to obtain the time characteristic range of the abnormality. The battery to be tested refers to a battery to be determined whether the voltage is abnormal. The battery to be tested may be a newly produced battery from the production line, or a battery that has been used for a period of time, or other batteries that may have abnormal conditions to be detected. In the example of the present application, multiple sample batteries and batteries to be tested may all be lithium-ion batteries.

[0064] In 010, multiple voltage curves of the voltage changes over time of multiple sample batteries during the charging and discharging process are obtained. The sample batteries are placed in trays, and multiple sample batteries are placed in one tray, and flow through the corresponding charging and discharging processes in sequence. During the charging and discharging process, the voltages of multiple sample batteries in the same tray at each time point are collected in real time to obtain multiple voltage curves of the voltage changes over time of multiple sample batteries during the charging and discharging process. Figure 3 and Figure 4 As shown, each sample battery corresponds to a voltage curve. The horizontal axis represents time by the number of times the sample battery voltage is collected (the voltage is collected once at a predetermined interval, and there is a corresponding relationship between the collection time and the number of collections), and the vertical axis represents the voltage corresponding to the collection time.

[0065] Figure 3 and Figure 4 These are multi-step voltage curves of multiple sample batteries. Figure 3 These are multiple voltage curves showing the voltage changes over time during the charging process of multiple sample batteries. Figure 4 These are multiple voltage curves showing the voltage changes over time for multiple sample batteries during the charge and discharge process. Figure 3 The charging voltage range of the sample battery is 3.35V-3.5V, and the corresponding voltage is collected every 1s. Figure 4 The charge and discharge voltage range of the sample battery is 2.0V-4.0V, and the corresponding voltage is collected every 1s.

[0066] In the embodiment of the present application, the charging and discharging process may include multiple charging and discharging steps to comprehensively detect the performance of the sample battery or the battery to be tested to avoid the outflow of unqualified batteries. The process of the charging and discharging process may include one or more shelving steps, gradient charging steps or discharging steps. The gradient refers to the different magnitudes of the battery current. For example, Figure 3 The multi-step sample battery charging voltage curve includes a rest step and a charging step. The curve is relatively flat between 0-30 times, indicating that the sample battery is in a rest state; the curve rises between 30-700 times, indicating that the sample battery is in a charging state. For another example, Figure 4 The multi-step sample battery charge and discharge voltage curve includes a rest step, a charging step and a discharging step. The curve is relatively flat between 0-10 times, indicating that the sample battery is in a rest state; the curve rises rapidly between 10-20 times, indicating that the sample battery is in a charging state; the curve drops gently between 20-125 times, indicating that the sample battery is in a rest state; the curve drops rapidly between 125-130 times, indicating that the sample battery is in a discharging state; the curve is relatively flat between 130-1750 times, indicating that the sample battery is in a rest state; the curve drops rapidly between 1750-2375 times, indicating that the sample battery is in a discharging state.

[0067] In 020, the time characteristic range of the abnormal battery when it is abnormal is obtained. As mentioned above, the multiple sample batteries include normal batteries and abnormal batteries, and the abnormal battery can be used to obtain the time characteristic range when it is abnormal. It can be understood that during the entire charging and discharging process, the abnormal battery may be abnormal at some time or all of the time. In the embodiment of the present application, the time characteristic range represents the time characteristic range when the abnormal battery is abnormal, represented by r. Figure 3 In the example, the time feature range is 107≤r≤590; Figure 4 In the example of , the time feature range is 472≤r≤6484. Of course, the above specific values ​​are only exemplary. In other examples, the time feature range may also be other values, which are not limited here.

[0068] In 030, an unsupervised anomaly detection model is constructed based on multiple voltage curves and time feature ranges. An unsupervised anomaly detection model is a model that extracts intrinsic features of data and performs feature analysis to distinguish whether the data is normal or abnormal in the absence of labeled data. The unsupervised anomaly detection model has the advantages of no need for labeled data, powerful pattern discovery capabilities, high flexibility, and strong scalability.

[0069] In 040, voltage anomaly detection is performed on the battery to be detected based on the unsupervised anomaly detection model. In this way, it can be determined whether the battery to be detected has voltage anomaly.

[0070] In the related art, the lithium battery industry usually uses manual visual methods to detect abnormal voltage curves of batteries to be tested, or detects the voltage at specific time points (for example, setting two voltage detection points for some charging and discharging steps). The coverage of the detection points is very low, resulting in the inability to identify abnormal batteries in advance. In the related art, voltage abnormality detection of batteries to be tested is a post-detection, which cannot detect abnormal behavior during the charging and discharging process of the production line battery in real time, and is not universal for different charging and discharging conditions. It needs to be recalibrated and is not suitable for charging and discharging processes with multiple steps and unstable actual ideal voltage curves.

[0071] In the implementation mode of the present application, multiple sample batteries are used as research objects, and an unsupervised anomaly detection model is constructed to perform voltage anomaly detection on the battery to be detected. The unsupervised anomaly detection model is not limited by the battery charging and discharging conditions, and can detect the entire charging and discharging process of the battery to be detected in real time and is not limited by the ideal curve of the battery voltage. It helps to timely discover and resolve potential safety hazards, and is suitable for scenarios such as new energy vehicles that require real-time detection of battery status. In addition, with the continuous development of battery technology and the emergence of new battery types, the unsupervised voltage anomaly detection model can be applied to batteries with different raw materials and different processes. It can adapt to new battery types by updating the data set, and there is no need to rebuild the battery anomaly detection model.

[0072] See also Figure 5 In some embodiments, obtaining the time characteristic range (i.e., 020) when the abnormal battery is abnormal includes:

[0073] 021: Obtain an abnormal voltage curve corresponding to an abnormal battery from multiple voltage curves;

[0074] 022: Perform time characteristic analysis on the abnormal voltage curve to obtain the time characteristic range.

[0075] Specifically, the plurality of voltage curves may include a normal voltage curve corresponding to a normal battery and an abnormal voltage curve corresponding to an abnormal battery. Figure 3 As shown, the normal voltage curve is a plurality of relatively dense curves at the top, and the abnormal voltage curve is a curve at the bottom. The abnormal voltage curve is far away from the normal voltage curve and exists out of the group, so the abnormal voltage curve can be identified from the plurality of voltage curves.

[0076] Abnormal batteries cause abnormal battery system energy due to factors such as battery moisture changes, which makes the corresponding abnormal voltage curve usually show obvious abnormalities within a certain time range. Based on the process mechanism analysis of the time characteristics of the abnormal voltage curve, the time characteristic range r, T of the abnormal battery can be obtained. a ≤r≤T b This time feature range can be used to build an unsupervised anomaly detection model and as a criterion for subsequent determination of whether the battery to be tested has voltage anomaly detection.

[0077] See also Figure 6 In some embodiments, an unsupervised anomaly detection model (i.e., 030) is constructed based on multiple voltage curves and time feature ranges, including:

[0078] 031: Obtain multiple standard score curves of the standard scores of multiple sample batteries changing with time during the charging and discharging process according to multiple voltage curves;

[0079] 032: Build an unsupervised anomaly detection model based on multiple standard score curves and time feature ranges.

[0080] Specifically, multiple voltage curves can be converted to obtain multiple standard score curves of the standard scores of multiple sample batteries changing with time during the charging and discharging process. The specific calculation formula is as follows:

[0081] Z_Score=(s-np.mean(s)) / np.std(s)

[0082] Among them, Z_Score is the standard score of each sample battery at the current time point, s is the voltage of each sample battery at the current time point, np.mean(s) is the average voltage of multiple sample batteries at the current time point, and np.std(s) is the standard deviation of the voltage of multiple sample batteries at the current time point.

[0083] A standard score curve can be obtained according to the standard score of each sample battery at each time point, and multiple standard score curves can be obtained according to the standard scores of multiple sample batteries at each time point, such as Figure 7 and Figure 8 As shown. Among them, Figure 7 for Figure 3 Multiple voltage curves in the graph correspond to multiple standard score curves. Figure 8 for Figure 4 Multiple standard score curves corresponding to multiple voltage curves in . An unsupervised anomaly detection model can be constructed based on multiple standard score curves and time feature ranges.

[0084] In the implementation mode of the present application, multiple voltage curves are first converted into multiple standard score curves, and then an unsupervised anomaly detection model is constructed based on the multiple standard score curves and the time feature range. In this way, by converting to the standard score curve, the differences in the amplitude and change trend of the voltage curve due to factors such as battery type, usage conditions, and aging degree can be eliminated, so that the data of different sample batteries can be compared and analyzed on the same scale, and noise interference can be reduced, so as to better capture the inherent laws of the data and improve the generalization ability of the unsupervised anomaly detection model. In addition, in the standard score curve, since the outliers are far away from the mean, the outliers can be more easily identified, thereby improving the detection accuracy of the unsupervised anomaly detection model.

[0085] See also Fig. 9 In some embodiments, voltage anomaly detection (i.e., 040) is performed on the battery to be detected based on an unsupervised anomaly detection model, including:

[0086] 041: Determine anomaly score threshold based on unsupervised anomaly detection model;

[0087] 042: Determine whether the battery to be tested has voltage abnormality according to the abnormality score threshold.

[0088] In the implementation manner of the present application, an abnormality score threshold is determined based on an unsupervised abnormality detection model. When voltage abnormality detection is performed on a battery to be detected, it can be determined whether there is a voltage abnormality in the battery to be detected based on the abnormality score threshold.

[0089] Among them, the anomaly score threshold determined based on the unsupervised anomaly detection model is fixed and does not require retraining the model for a new batch of batteries to be tested on the production line. According to the anomaly score threshold, it is possible to determine whether the voltage of the battery to be tested is abnormal, and effectively distinguish between normal batteries and abnormal batteries in the batteries to be tested.

[0090] See also Fig.10 In some embodiments, determining an anomaly score threshold (i.e., 0.41) based on an unsupervised anomaly detection model includes:

[0091] 0411: Set the initial score threshold;

[0092] 0412: Model training for unsupervised anomaly detection models based on multiple standard score curves and initial score thresholds;

[0093] 0413: Determine anomaly score threshold using a trained unsupervised anomaly detection model.

[0094] In the implementation mode of the present application, the unsupervised anomaly detection model is trained by using multiple standard score curves and score thresholds. First, the initial score threshold of the unsupervised anomaly detection model is set manually, and multiple standard score curves are input into the unsupervised anomaly detection model. The unsupervised anomaly detection model is used to determine whether the sample battery corresponding to the standard score curve is a normal battery or an abnormal battery. The process of the unsupervised anomaly detection model determining whether the sample battery is a normal battery or an abnormal battery is as follows: when the standard score in the standard score curve is greater than the initial score threshold, and the time feature corresponding to the standard score is within the time feature range, the sample battery is determined to be an abnormal battery; otherwise, the sample battery is determined to be a normal battery.

[0095] When the unsupervised anomaly detection model can accurately determine whether the sample battery is a normal battery or an abnormal battery, it indicates that the unsupervised anomaly detection model training is completed, and the current score threshold is reasonable, which can be used as the anomaly score threshold for subsequent determination of whether the battery to be detected has voltage anomalies. If the unsupervised anomaly detection model cannot accurately determine whether the sample battery is a normal battery or an abnormal battery, adjust the initial score threshold of the unsupervised anomaly detection model and retrain the above model.

[0096] See also Fig.11 In some embodiments, judging whether the battery to be tested has a voltage abnormality (i.e., 042) according to the abnormal score threshold includes:

[0097] 0421: Obtain a voltage curve of the battery to be tested, which shows the voltage changing with time during the charging and discharging process;

[0098] 0422: Obtain a standard score curve to be detected in which the standard score of the battery to be detected changes with time during the charging and discharging process according to the voltage curve to be detected;

[0099] 0423: Determine whether the voltage of the battery to be tested is abnormal based on the relationship between the standard score curve to be tested and the abnormal score threshold.

[0100] It should be pointed out that in the implementation manner of the present application, the number of batteries to be detected can also be multiple, and the corresponding number of voltage curves to be detected and standard score curves to be detected is also multiple. The explanations of "obtaining multiple voltage curves of multiple sample batteries changing with time during the charging and discharging process" in 010 and "obtaining multiple standard score curves of multiple sample batteries changing with time during the charging and discharging process based on multiple voltage curves" in 031 in the aforementioned implementation manner are also applicable to the corresponding schemes in 0421 and 0422 of the implementation manner of the present application. The only difference is that the sample batteries are replaced by the batteries to be detected, which will not be repeated here. Afterwards, based on the relationship between the standard score curve to be detected and the abnormal score threshold, it is determined whether the voltage of the battery to be detected is abnormal. Among them, Figure 3 and Figure 4 It can also be used as a schematic diagram of the voltage curve of the battery to be tested; Figure 7 and Figure 8 It can also be used as a schematic diagram of the standard score curve of the battery to be tested.

[0101] See also Fig.12 In some embodiments, judging whether the voltage of the battery to be detected is abnormal (i.e., 0423) according to the relationship between the standard score curve to be detected and the abnormal score threshold value includes:

[0102] 04231: When the standard score in the standard score curve to be detected is greater than the abnormal score threshold, the current time feature corresponding to the standard score greater than the abnormal score threshold is obtained;

[0103] 04232: When the current time feature is within the time feature range, it is determined that the voltage of the battery to be detected is abnormal.

[0104] In the implementation mode of the present application, when the standard score in the standard score curve to be detected is greater than the abnormal score threshold, the current time feature corresponding to the standard score greater than the abnormal score threshold is obtained to compare with the time feature range to determine whether the battery to be detected has a voltage abnormality. If the current time feature is not within the time feature range r, it is determined that the battery to be detected does not have a voltage abnormality and is a normal battery. If the current time feature is within the time feature range r, it is determined that the battery to be detected has a voltage abnormality and is an abnormal battery, and the abnormal battery barcode is output at the same time. Among them, the abnormal battery barcode refers to a coding identifier used to identify a battery product, which is used to obtain the corresponding specific abnormal battery so that the abnormal battery can be processed later, such as shielding and isolating the abnormal battery. In addition, when the standard score in the standard score curve to be detected is greater than the abnormal score threshold, the abnormal voltage value corresponding to the standard score greater than the abnormal score threshold can also be obtained to subsequently evaluate the degree of abnormality and determine the corresponding processing method.

[0105] by Figure 7 and Figure 8 For example, Figure 7 The following is a schematic diagram of multiple standard score curves showing the change of standard scores over time during the charging process of multiple batteries to be tested. The time characteristic range of abnormal batteries in the figure is 107≤r≤590. The voltage of one of the multiple batteries to be tested is 3.409V at the 588th time (e.g. Figure 3 As shown in FIG. 1 ), after analysis by the constructed unsupervised anomaly detection model, the standard score of the battery to be detected is greater than the anomaly score threshold of 5V (as shown in FIG. Figure 7 As shown), and the time feature T588 corresponding to the standard score is within the time feature range 107≤r≤590, it is determined that the battery to be tested has a voltage abnormality and is an abnormal battery. Figure 8 The following is a schematic diagram of multiple standard score curves showing the change of standard scores over time during the charge and discharge process of multiple batteries to be tested. The time characteristic range of abnormal batteries in the figure is 472≤r≤6484. The voltage of one of the multiple batteries to be tested is 3.127V at the 1053rd time (e.g. Figure 4 As shown in FIG. 1 ), after analysis by the constructed unsupervised anomaly detection model, the standard score of the battery to be detected is greater than the anomaly score threshold of 5V (as shown in FIG. Figure 8 As shown), and the time feature T1053 corresponding to the standard score is within the time feature range 472≤r≤6484, it is determined that the battery to be tested has no voltage abnormality and is a normal battery.

[0106] See also Fig.13, the embodiment of the present application also provides a voltage anomaly detection device 100. The voltage anomaly detection device 100 includes a first acquisition module 10, a second acquisition module 20, a construction module 30 and a detection module 40. The first acquisition module 10 can be used to obtain multiple voltage curves of voltage changes over time during the charging and discharging process of multiple sample batteries, and the multiple sample batteries include normal batteries and abnormal batteries. The second acquisition module 20 can be used to obtain the time characteristic range when the abnormal battery is abnormal. The construction module 30 can be used to construct an unsupervised anomaly detection model based on multiple voltage curves and time characteristic ranges. The detection module 40 can be used to perform voltage anomaly detection on the battery to be detected based on the unsupervised anomaly detection model.

[0107] In some embodiments, the second acquisition module 20 is specifically used to: acquire an abnormal voltage curve corresponding to the abnormal battery from a plurality of voltage curves; and perform time characteristic analysis on the abnormal voltage curve to obtain a time characteristic range.

[0108] In some embodiments, the construction module 30 is specifically used to: obtain multiple standard score curves of the standard scores of multiple sample batteries changing with time during the charging and discharging process according to multiple voltage curves; and construct an unsupervised anomaly detection model according to the multiple standard score curves and the time feature range.

[0109] In some embodiments, the detection module 40 is specifically used to: determine an abnormality score threshold based on an unsupervised abnormality detection model; and determine whether the battery to be detected has a voltage abnormality according to the abnormality score threshold.

[0110] In some embodiments, the detection module 40 is specifically used to: set an initial score threshold; perform model training on an unsupervised anomaly detection model according to a plurality of standard score curves and the initial score threshold; and determine the anomaly score threshold through the trained unsupervised anomaly detection model.

[0111] In some embodiments, the detection module 40 is specifically used to: obtain a voltage curve to be detected in which the voltage of the battery to be detected changes with time during the charging and discharging process; obtain a standard score curve to be detected in which the standard score of the battery to be detected changes with time during the charging and discharging process according to the voltage curve to be detected; and determine whether the voltage of the battery to be detected is abnormal based on the relationship between the standard score curve to be detected and the abnormal score threshold.

[0112] In some embodiments, the detection module 40 is specifically used to: when the standard score in the standard score curve to be detected is greater than the abnormal score threshold, obtain the current time feature corresponding to the standard score greater than the abnormal score threshold; when the current time feature is within the time feature range, determine that the voltage of the battery to be detected is abnormal.

[0113] It should be pointed out that the explanation of the voltage anomaly detection method in the aforementioned embodiment is also applicable to the voltage anomaly detection device 100 in the embodiment of the present application, and will not be elaborated here.

[0114] See also Fig.14 The embodiment of the present application further provides a voltage anomaly detection system 200. The voltage anomaly detection system 200 includes one or more processors 210 and a memory 220. The memory 220 stores a computer program, and when the computer program is executed by the processor 210, the voltage anomaly detection method of any of the above embodiments is implemented.

[0115] For example, when the computer program is executed by the processor 210, the following voltage anomaly detection method is implemented:

[0116] 010: obtaining multiple voltage curves of multiple sample batteries in which voltage changes with time during the charging and discharging process, where the multiple sample batteries include normal batteries and abnormal batteries;

[0117] 020: Get the time characteristic range of abnormal battery abnormality;

[0118] 030: Construct an unsupervised anomaly detection model based on multiple voltage curves and time feature ranges;

[0119] 040: Voltage anomaly detection of the battery to be tested based on the unsupervised anomaly detection model.

[0120] For another example, when the computer program is executed by the processor 210, the following voltage anomaly detection method is implemented:

[0121] 021: Obtain an abnormal voltage curve corresponding to an abnormal battery from multiple voltage curves;

[0122] 022: Perform time characteristic analysis on the abnormal voltage curve to obtain the time characteristic range.

[0123] It should be noted that the explanation of the voltage anomaly detection method in the aforementioned embodiment is also applicable to the voltage anomaly detection system 200 in the embodiment of the present application, and will not be elaborated here.

[0124] See also Fig.15 The present application also provides a computer-readable storage medium 300 on which a computer program 310 is stored. When the program is executed by the processor 320, the voltage anomaly detection method of any of the above embodiments is implemented.

[0125] For example, when the program is executed by the processor 320, the following voltage anomaly detection method is implemented:

[0126] 010: obtaining multiple voltage curves of multiple sample batteries in which voltage changes with time during the charging and discharging process, where the multiple sample batteries include normal batteries and abnormal batteries;

[0127] 020: Get the time characteristic range of abnormal battery abnormality;

[0128] 030: Construct an unsupervised anomaly detection model based on multiple voltage curves and time feature ranges;

[0129] 040: Voltage anomaly detection of the battery to be tested based on the unsupervised anomaly detection model.

[0130] For another example, when the program is executed by the processor 320, the following voltage anomaly detection method is implemented:

[0131] For another example, when the computer program 310 is executed by the processor 320, the following voltage anomaly detection method is implemented:

[0132] 021: Obtain an abnormal voltage curve corresponding to an abnormal battery from multiple voltage curves;

[0133] 022: Perform time characteristic analysis on the abnormal voltage curve to obtain the time characteristic range.

[0134] It should be noted that the explanation of the voltage anomaly detection method in the aforementioned embodiment is also applicable to the computer-readable storage medium 300 in the embodiment of the present application, and will not be elaborated here.

[0135] In summary, the voltage anomaly detection method, voltage anomaly detection device 100, voltage anomaly detection system 200 and computer-readable storage medium 300 of the embodiments of the present application construct an unsupervised anomaly detection model based on multiple voltage curves and time feature ranges to perform voltage anomaly detection on the battery to be detected, and can realize real-time and accurate detection of voltage anomalies in the entire charging and discharging process of the battery to be detected with multiple steps and multiple gradient currents.

[0136] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0137] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0138] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable storage medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purposes of this specification, a computer-readable storage medium can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples (non-exhaustive list) of computer-readable storage media include the following: an electrical connection with one or more wires (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable storage medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways if necessary, and then stored in a computer memory.

[0139] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0140] A person of ordinary skill in the art will appreciate that all or part of the steps carried by the method for implementing the above-mentioned embodiment can be completed by instructing the relevant hardware through a program, and the program can be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment. In addition, each functional unit in each embodiment of the present application can be integrated into a processing module, or each unit can exist physically alone, or two or more units can be integrated into one module. The above-mentioned integrated module can be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. The above-mentioned storage medium can be a read-only memory, a disk or an optical disk, etc.

[0141] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application. The scope of the present application is defined by the claims and their equivalents.

Claims

1. A voltage anomaly detection method, characterized in that: include: Acquire a plurality of voltage curves of voltage changes over time of a plurality of sample batteries during a charge and discharge process, wherein the plurality of sample batteries include normal batteries and abnormal batteries; Obtaining a time characteristic range when the abnormal battery is abnormal; constructing an unsupervised anomaly detection model according to the plurality of voltage curves and the time feature range; Voltage anomaly detection is performed on the battery to be detected based on the unsupervised anomaly detection model.

2. The voltage anomaly detection method according to claim 1, characterized in that: Obtaining the time characteristic range of the abnormal battery when it is abnormal, including: Acquire an abnormal voltage curve corresponding to the abnormal battery from the multiple voltage curves; A time characteristic analysis is performed on the abnormal voltage curve to obtain the time characteristic range.

3. The voltage anomaly detection method according to claim 1, characterized in that: The constructing an unsupervised anomaly detection model according to the multiple voltage curves and the time feature range includes: Acquire, according to the multiple voltage curves, multiple standard score curves showing changes of the standard scores of the multiple sample batteries over time during the charging and discharging process; The unsupervised anomaly detection model is constructed according to the multiple standard score curves and the time feature range.

4. The voltage anomaly detection method according to claim 3, characterized in that: The performing voltage anomaly detection on the battery to be detected based on the unsupervised anomaly detection model includes: determining an anomaly score threshold based on the unsupervised anomaly detection model; It is determined whether the battery to be detected has a voltage abnormality according to the abnormal score threshold.

5. The voltage anomaly detection method according to claim 4, characterized in that: The determining of anomaly score threshold based on the unsupervised anomaly detection model comprises: Set the initial score threshold; Performing model training on the unsupervised anomaly detection model according to the multiple standard score curves and the initial score threshold; The anomaly score threshold is determined by the trained unsupervised anomaly detection model.

6. The voltage anomaly detection method according to claim 4, characterized in that: The determining, according to the abnormal score threshold, whether the battery to be detected has a voltage abnormality includes: Obtaining a voltage curve to be detected, in which the voltage of the battery to be detected changes with time during the charging and discharging process; Acquire a standard score curve to be detected according to the voltage curve to be detected, in which the standard score of the battery to be detected changes with time during the charging and discharging process; Whether the voltage of the battery to be detected is abnormal is determined according to the relationship between the standard score curve to be detected and the abnormal score threshold.

7. The voltage anomaly detection method according to claim 6, characterized in that: The step of judging whether the voltage of the battery to be detected is abnormal according to the relationship between the standard score curve to be detected and the abnormal score threshold comprises: When the standard score in the to-be-detected standard score curve is greater than the abnormal score threshold, obtaining a current time feature corresponding to the standard score greater than the abnormal score threshold; When the current time feature is within the time feature range, it is determined that the voltage of the battery to be detected is abnormal.

8. A voltage anomaly detection device, characterized in that: include: A first acquisition module, used to acquire a plurality of voltage curves of voltage changes over time of a plurality of sample batteries during a charge and discharge process, wherein the plurality of sample batteries include normal batteries and abnormal batteries; A second acquisition module is used to acquire a time characteristic range when the abnormal battery is abnormal; A construction module, configured to construct an unsupervised anomaly detection model according to the plurality of voltage curves and the time feature range; The detection module is used to perform voltage anomaly detection on the battery to be detected based on the unsupervised anomaly detection model.

9. A voltage anomaly detection system, characterized in that: The voltage anomaly detection system includes one or more processors and a memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, the voltage anomaly detection method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the voltage anomaly detection method according to any one of claims 1 to 7 is implemented.

Images