Battery abnormal event detection method and device, computer equipment and readable storage medium
By screening historical abnormal fluctuations and building thresholds, combined with sliding window technology, efficient and accurate detection of battery abnormal events is achieved, and the problems of poor real-time and accuracy in the existing technology are solved, ensuring the safety of the battery.
Patent Information
- Application Number
- CN202411219687.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-02
- Publication Date
- 2024-11-22
AI Technical Summary
The existing battery abnormality detection methods have problems such as poor real-time and poor accuracy, and cannot detect battery abnormal events efficiently and accurately. Especially in new energy vehicles, the monitoring system cannot identify the abnormal battery position and mode in a timely and accurate manner, resulting in serious safety failures.
By obtaining the historical time series data of the battery, filtering out historical abnormal fluctuations, constructing an abnormal fluctuation set, determining the dual-standard threshold combination and average amplitude ratio threshold, combining sliding window translation technology, real-time detection of battery parameters and rapid identification of abnormal events.
It realizes efficient and accurate battery abnormal event detection, avoids the need to build large-scale equivalent circuit models and intelligent algorithm models, improves the real-time and accuracy of detection, and ensures the safety of the battery.
Smart Images

Figure CN119001464A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of battery management technology, and more specifically, to a battery abnormal event detection method, device, computer equipment and readable storage medium. Background Art
[0002] With the continuous growth of the number of new energy vehicles, the demand for power batteries has also increased. At the same time, it is also accompanied by the uncertainty of the power battery manufacturing process level and the vehicle operating conditions. That is, the power battery will inevitably have too high or too low voltage, too large or too small current, too high or too low temperature during use, which will lead to related faults. Especially when the battery abnormality occurs during vehicle operation, the monitoring system is often unable to accurately identify the abnormal battery position and abnormal mode in time, resulting in serious safety failures of the battery, endangering personal and property safety. Therefore, the battery management system (BMS) has become an essential part of electric vehicles to provide guarantees for vehicle performance and safety. At present, the abnormality detection of new energy vehicle batteries is mainly divided into abnormality detection based on equivalent circuit models, abnormality detection based on statistical methods, and abnormality detection based on data-driven intelligent algorithm models. However, it is found in the study that since the above methods need to be based on a large amount of statistical data to build a large-scale equivalent circuit model or intelligent algorithm model, the above detection methods all have the problems of poor real-time performance and poor accuracy, making it impossible to detect battery abnormal events efficiently and accurately. Summary of the invention
[0003] In view of this, an object of the present invention is to provide a battery abnormal event detection method, device, computer equipment and readable storage medium to efficiently and accurately detect battery abnormal events.
[0004] In a first aspect, an embodiment of the present application provides a method for detecting abnormal battery events, the method comprising:
[0005] Before performing battery abnormal event detection, for each detection channel, a historical time series set of the battery to be detected under the detection channel is obtained, a historical abnormal fluctuation moment set in the historical time series set is screened out, and an abnormal fluctuation set is constructed based on the historical abnormal fluctuation moment set, wherein the historical time series set is composed of battery parameters under the detection channel at several different historical moments;
[0006] Determine a dual-standard threshold combination and an average amplitude ratio threshold under the detection channel based on a set of historical abnormal fluctuations under the detection channel;
[0007] When performing battery abnormality event detection, obtaining battery parameters of the battery to be detected at each current moment in the detection channel;
[0008] Based on the battery parameters of the battery to be detected in the detection channel at each current moment, the abnormal event of the battery to be detected is determined according to the first sliding hop count, the dual standard threshold combination in the detection channel and the average amplitude ratio threshold.
[0009] Optionally, determining an average amplitude ratio threshold value under the detection channel based on a set of historical abnormal fluctuations under the detection channel includes:
[0010] Perform sliding window translation on the historical abnormal fluctuation set under the detection channel with a second sliding jump number to determine the short-time window average amplitude and the long-time window average amplitude of each historical abnormal fluctuation set;
[0011] The average amplitude ratio threshold is determined based on the short-time window average amplitude and the long-time window average amplitude.
[0012] Optionally, determining a dual-standard threshold combination under the detection channel based on a set of historical abnormal fluctuations under the detection channel includes:
[0013] Set a number of extended time windows, and optimize the historical abnormal fluctuation set under the detection channel based on each extended time window;
[0014] The dual-standard threshold combination is determined based on the amplitude of the optimized historical abnormal fluctuation set.
[0015] Optionally, the determining of the abnormal event of the battery to be detected based on the battery parameters of the battery to be detected in the detection channel at each current moment according to the first sliding hop count, the dual standard threshold combination in the detection channel and the average amplitude ratio threshold includes:
[0016] Performing sliding window translation on the battery parameters of the battery to be detected at each current moment in the detection channel with the first sliding hop number;
[0017] When performing sliding window translation, the ratio of long and short time windows is calculated, the ratio of long and short time windows is compared with the average amplitude ratio threshold to obtain a first comparison result, and the window size when performing sliding window translation is compared with the dual standard threshold combination to obtain a second comparison result;
[0018] An abnormal event of the battery to be detected is determined based on the first comparison result and the second comparison result.
[0019] In a second aspect, an embodiment of the present application provides a battery abnormal event detection device, the device comprising:
[0020] A historical abnormal fluctuation set determination module is used to obtain, for each detection channel, a historical time series set of the battery to be detected under the detection channel before performing battery abnormal event detection, filter out a historical abnormal fluctuation moment set in the historical time series set, and construct an abnormal fluctuation set based on the historical abnormal fluctuation moment set, wherein the historical time series set is composed of battery parameters under the detection channel at several different historical moments;
[0021] A detection standard determination module, used to determine a dual-standard threshold combination and an average amplitude ratio threshold under the detection channel based on a historical abnormal fluctuation set under the detection channel;
[0022] A battery parameter acquisition module, used to acquire the battery parameters of the battery to be detected in the detection channel at each current moment when performing battery abnormality event detection;
[0023] The abnormal event detection module is used to determine the abnormal event of the battery to be detected based on the battery parameters of the battery to be detected in the detection channel at each current moment, according to the first sliding hop count, the dual standard threshold combination in the detection channel and the average amplitude ratio threshold.
[0024] Optionally, determining an average amplitude ratio threshold value under the detection channel based on a set of historical abnormal fluctuations under the detection channel includes:
[0025] Perform sliding window translation on the historical abnormal fluctuation set under the detection channel with a second sliding jump number to determine the short-time window average amplitude and the long-time window average amplitude of each historical abnormal fluctuation set;
[0026] The average amplitude ratio threshold is determined based on the short-time window average amplitude and the long-time window average amplitude.
[0027] Optionally, determining a dual-standard threshold combination under the detection channel based on a set of historical abnormal fluctuations under the detection channel includes:
[0028] Set a number of extended time windows, and optimize the historical abnormal fluctuation set under the detection channel based on each extended time window;
[0029] The dual-standard threshold combination is determined based on the amplitude of the optimized historical abnormal fluctuation set.
[0030] Optionally, the determining of the abnormal event of the battery to be detected based on the battery parameters of the battery to be detected in the detection channel at each current moment according to the first sliding hop count, the dual standard threshold combination in the detection channel and the average amplitude ratio threshold includes:
[0031] Performing sliding window translation on the battery parameters of the battery to be detected at each current moment in the detection channel with the first sliding hop number;
[0032] When performing sliding window translation, the ratio of long and short time windows is calculated, the ratio of long and short time windows is compared with the average amplitude ratio threshold to obtain a first comparison result, and the window size when performing sliding window translation is compared with the dual standard threshold combination to obtain a second comparison result;
[0033] An abnormal event of the battery to be detected is determined based on the first comparison result and the second comparison result.
[0034] In a third aspect, an embodiment of the present application provides a computer device, comprising: a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the computer device is running, the processor and the memory communicate through the bus, and when the machine-readable instructions are executed by the processor, the steps of the battery abnormal event detection method described in any optional implementation manner of the first aspect are performed.
[0035] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the battery abnormal event detection method described in any optional implementation manner in the above-mentioned first aspect are executed.
[0036] The technical solution provided by this application includes but is not limited to the following beneficial effects:
[0037] This application extracts effective time series data (voltage, current, SOC, etc.) of the battery to be detected, sets the basic parameters of the long and short time window method, sets the threshold of the long and short time window method and the double standard threshold group through historical experience values to preliminarily screen the abnormal fluctuation time series set within the time period, and determines the double standard threshold group and the average amplitude ratio threshold value based on the existing small sample time series abnormal fluctuation set, which can provide an objective quantitative detection standard for battery abnormal event detection. Then, based on the battery parameters of the battery to be detected at each current moment under the detection channel, the abnormal event of the battery to be detected is determined according to the first sliding jump number, the double standard threshold combination under the detection channel and the average amplitude ratio threshold value, which can realize the detection of battery abnormal events without the need to build a large-scale equivalent circuit model, an intelligent algorithm model, and perform a large amount of data Egyptian statistics, thereby realizing efficient and accurate battery abnormal event detection.
[0038] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0039] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.
[0040] Figure 1 A flow chart of a battery abnormal event detection method provided by the first embodiment of the present invention is shown;
[0041] Figure 2 A flow chart of a method for determining a detection standard provided in the first embodiment of the present invention is shown;
[0042] Figure 3 A flow chart of a method for determining a dual-standard threshold combination provided by Embodiment 1 of the present invention is shown;
[0043] Figure 4 A schematic diagram of an abnormal fluctuation time series segment provided by the first embodiment of the present invention is shown;
[0044] Figure 5 A flowchart of a specific abnormal event determination method provided by the first embodiment of the present invention is shown;
[0045] Figure 6 A schematic diagram showing the structure of a battery abnormal event detection device provided by Embodiment 2 of the present invention is shown;
[0046] Figure 7 A schematic diagram of the structure of a computer device provided in Embodiment 3 of the present invention is shown. DETAILED DESCRIPTION
[0047] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present invention.
[0048] Embodiment 1
[0049] To facilitate understanding of this application, Figure 1The flowchart of a battery abnormal event detection method provided in the first embodiment of the present invention is shown to describe the contents of the first embodiment of the present application in detail.
[0050] See also Figure 1 As shown, Figure 1 A flowchart of a battery abnormal event detection method provided in Embodiment 1 of the present invention is shown, wherein the method comprises steps S101 to S104:
[0051] S101: Before performing battery abnormal event detection, for each detection channel, obtain a historical time series set of the battery to be detected under the detection channel, filter out a historical abnormal fluctuation moment set in the historical time series set, and construct an abnormal fluctuation set based on the historical abnormal fluctuation moment set, wherein the historical time series set is composed of battery parameters under the detection channel at several different historical moments.
[0052] Specifically, a time series is a curve that changes over time with equal-interval sampling, and is mostly expressed as two-dimensional data. It can contain multiple frequencies, which can reflect the state of the monitored object and facilitate researchers or users to grasp the dynamics of the monitored object in real time. Time series often do not appear alone. Users deploy multiple sensors or monitoring devices. Each sensor or monitoring device is usually called a channel. These channels set the working time according to user needs to form a time series set. When monitoring certain special objects, such as low-frequency objects, high-frequency sampling is required to capture these low-frequency objects. Therefore, a large number of high-dimensional and massive time series sets formed by high-frequency sampling are generated. For ordinary users, there is a limit to computer performance, and the two are in conflict. An anomaly detection method that can balance accuracy and real-time performance is needed to accurately extract the abnormal fluctuation set of the time series while achieving real-time monitoring.
[0053] The purpose of abnormal moment detection is to extract the final abnormal fluctuation set, which is critical for the subsequent analysis of the monitored objects, especially for low-frequency objects. If a more accurate abnormal moment detection method is used at the beginning, when the number of channels is large, real-time monitoring may not be achieved, which is very unfavorable for emergency response. For high-frequency objects or objects with slow state changes, although the time series frequency is low and a more accurate abnormal moment detection method can be used, the real-time performance of the monitored objects under this condition is not strong, and there is no need for real-time monitoring.
[0054] Furthermore, for multiple channels distributed in different time periods, a set of historical time series with a finite length and containing abnormal fluctuation sets is obtained. Assume that their frequencies are all F, and the time interval between every two data points (battery voltage, current or SOC charge) is 1 / F. Determine the set of historical abnormal fluctuation moments of channel p according to the long and short time window method represents the i-th channel p1 Historical abnormal fluctuation moments, i 1 =1,2,…,k, where k is the total number of historical abnormal fluctuation moments of the pth channel, and the number of each two is determined based on empirical values. When the interval time difference Δ is satisfied Delete invalid Forming an effective set of historical abnormal fluctuation moments
[0055] use Construct an effective abnormal fluctuation set, The corresponding time series is taken from the front on the x-axis Seconds, take the latter seconds, combined to form an abnormal fluctuation set TS i2 ={ts 1 ,ts 2 ,...,ts m}, where ts p ={ts p (1),ts p (2),...,ts p (n)} represents the time series of the pth channel, with a length of n data ts p (n) represents the n data in the time series of the pth channel. Here, the time series of each channel belongs to the same moment in time, that is, Time(ts p (1))=Time(ts p+1 (1)), and |TS i2 |=|TS i2+1 |,i 2 ∈[1,|TS i2 |]; Multiple TS i2 Constitute multiple valid abnormal fluctuation sets TS r×m×n ={TS 1 ,...,TS r}, j≠i 2 ,i 2 and j is the serial number of the abnormal fluctuation set, the i-th 2 The jth abnormal fluctuation set and the jth abnormal fluctuation set are not the same abnormal fluctuation set, TS i2 is the number of abnormal fluctuation sets, r is the number of abnormal fluctuation sets, m is the number of channels of each abnormal fluctuation set, each channel is numbered with p, and a and b are preset parameters.
[0056] S102: Determine a dual-standard threshold combination and an average amplitude ratio threshold under the detection channel based on a set of historical abnormal fluctuations under the detection channel.
[0057] Specifically, when the signal reaches the sensor or device, it is mainly divided into large amplitude events and small amplitude events. Large amplitude events are more likely to be a large change in the state of the monitored object itself, which requires special attention from the user, while small amplitude events are more likely to be the precursor accumulation information of large amplitude events. Users need to analyze the possible large amplitude events and the nature of the monitored object itself based on these accumulated information, and make timely adjustments to the monitored object. For example, when sensors or monitoring devices from several different sources jointly monitor an object, the abnormal moments of the time series on different devices can jointly analyze the state of the monitored object, and the possible large amplitude events can be discovered in advance. For another example, there is a lot of noise near some monitored objects. This type of data is likely to be a small amplitude event, prompting the user to transfer the sensor or monitoring device to achieve higher monitoring benefits. Therefore, it is necessary to distinguish between large and small amplitude events with dual standard thresholds. Therefore, based on the existing small sample time series abnormal fluctuation set, a dual standard threshold group is determined to extract the basic characteristics of the effective time series to prevent the lack of initial parameters during the cold start of the system.
[0058] S103: when performing battery abnormality event detection, obtaining battery parameters of the battery to be detected in the detection channel at each current moment.
[0059] Specifically, the battery parameters of the battery to be detected at each current moment in the detection channel are obtained through each sensor device, including but not limited to voltage, current, SOC, etc.
[0060] S104: Based on the battery parameters of the battery to be detected in the detection channel at each current moment, determine the abnormal event of the battery to be detected according to the first sliding hop count, the dual standard threshold combination in the detection channel and the average amplitude ratio threshold.
[0061] Specifically, the appropriate first sliding jump number is determined according to the real-time requirements, the abnormal fluctuation moment is corrected, the judgment conditions are tightened, the position of the abnormal moment is corrected, and the accuracy of abnormal moment detection and fluctuation set interception is improved. At the same time, the double standard threshold combination and the average amplitude ratio threshold under the detection channel are used as the basis for event detection, and the abnormal events of the battery to be detected are detected based on the battery parameters under the detection channel, and an abnormal fluctuation set is established. An abnormal fluctuation set based on the existing monitoring data in a certain field will be established for subsequent in-depth analysis of the monitored object or prediction of abnormal behavior of the monitored object.
[0062] In an alternative embodiment, see Figure 2 As shown, Figure 2 A flow chart of a detection standard determination method provided in the first embodiment of the present invention is shown, wherein the average amplitude ratio threshold value under the detection channel is determined based on the historical abnormal fluctuation set under the detection channel, including steps S201 to S202:
[0063] S201: Perform sliding window translation on the historical abnormal fluctuation set under the detection channel with a second sliding jump number to determine the short-time window average amplitude and the long-time window average amplitude of each historical abnormal fluctuation set.
[0064] S202: Determine the average amplitude ratio threshold based on the short-time window average amplitude and the long-time window average amplitude.
[0065] Specifically, by translating the sliding window Win, the second sliding hop number is hop 2 , calculate TS i2 The average amplitudes of all short and long time windows are and u represents the number of short-term windows sliding on the time series, and v represents the number of long-term windows sliding on the time series. The third quartile IQR is used. 3 As the average amplitude ratio threshold in the short and long time windows As shown in the following formula:
[0066]
[0067] Use the mean of the ratio thresholds of all abnormal fluctuation sets As the average amplitude ratio threshold in the global short and long time windows, The subscript i 3 Indicates the i 3 The ratio of short-time windows to long-time windows, r is the total number of abnormal fluctuation sets.
[0068] In an alternative embodiment, see Figure 3 As shown, Figure 3 A flowchart of a dual-standard threshold combination determination method provided in Embodiment 1 of the present invention is shown, wherein a dual-standard threshold combination under the detection channel is determined based on a historical abnormal fluctuation set under the detection channel, including steps S301 to S302:
[0069] S301: setting a number of extended time windows, and optimizing the historical abnormal fluctuation set under the detection channel based on each extended time window.
[0070] S302: Determine the dual-standard threshold combination based on the amplitude of the optimized historical abnormal fluctuation set.
[0071] Specifically, extended time windows of different sizes are set around the abnormal position of the time series, and a double standard threshold group is determined according to the amplitude of the time series.
[0072] Determine three extended windows, namely the front extended window Win be, rear extension window Win af , rear thin extension window Win afr They monitor the background noise, small fluctuation range and large fluctuation range respectively, with lengths of |Win be |=F / 10、|Win af |=F / 10、|Win afr |=F×1.2.
[0073] See also Figure 4 As shown, Figure 4 A schematic diagram of an abnormal fluctuation time series segment provided by the first embodiment of the present invention is shown, wherein, it includes three extended windows, namely, a front extended window, a rear coarse extended window, and a rear fine extended window. The long window, the short window, and each window and their names are marked, and the direction of new data entry is on the right.
[0074] Determine the dual standard threshold set T h and T l The methods include: According to TS i2 The amplitude in the abnormal fluctuation set is sorted, the distribution of the amplitude is observed, and after eliminating the noise and error fluctuation sets, a set of low standard threshold values T is determined using the abnormal fluctuation time position of the fluctuation set with the smallest amplitude l =(T l1 ,T l2 ), and then determine the high standard threshold group T according to the amplitude distribution h =(T h1 ,T h2 ). l1 It's Win af The low standard (small amplitude) threshold, T l2 It's Win afr The low standard (small amplitude) threshold, T h1 It's Win af The high standard (large amplitude) threshold, T h2 It's Win afe High standard (large amplitude) threshold.
[0075] In an alternative embodiment, see Figure 5 As shown, Figure 5 A flowchart of a specific abnormal event determination method provided in the first embodiment of the present invention is shown, wherein the abnormal event of the battery to be detected is determined based on the battery parameters of the battery to be detected in the detection channel at each current moment, according to the first sliding hop count, the dual standard threshold combination in the detection channel and the average amplitude ratio threshold, including steps S501 to S503:
[0076] S501: Perform sliding window translation on the battery parameters of the battery to be detected in the detection channel at each current moment with the first sliding hop number.
[0077] Specifically, before executing step S401, the first sliding hop count is determined by the following method;
[0078] Determine the first sliding hop number hop 1 (pcs), the sliding hop number after adjusting the accuracy is hop′, which depends on the real-time requirements of the calculation (h<0.1ms, m<50ms, l<1s), the number of channels ch (pcs), the sampling frequency F (K represents 10 3 , M represents 10 6 ), comprehensive consideration, only the computer performance (hardware parameter) requirements and hop count parameter setting reference under different data characteristics under the real-time requirement of l<1s are listed, as shown below. The following table shows the minimum configuration of computer performance under different data characteristics:
[0079]
[0080] As shown in the above table, the first column of data characteristics includes the number of channels and sampling frequency. These two indicators have the greatest impact on computer hardware. Unless there is special data in a special field, the size of the space occupied by each set of data and the data format have little effect on performance.
[0081] Other parameters are as follows: (1) High and low standard threshold groups T l ,T h , (2) sliding window hop count hop (pcs), (3) number of channels ch (pcs), (4) sampling frequency F (pcs), (5) front extension window be, (6) rear coarse extension window af, (7) rear fine extension window afr and its size, (8) average amplitude ratio threshold β in short and long time windows, (9) two and The time difference Δ between the intervals of the last data window and the time range t of the last data window. (10) The time range t from the latest time and the range t′ of the previous data window are determined based on experience. (2), (5), (6), (7), (9), (10) are fixed in advance. (3) and (4) need to be determined in steps 1 and 2. (1) and (8) need to be determined in steps 1 and 2. The amount of data to be processed in the window is ch×F×t. When the real-time requirement is <50ms, the processing time of each window is hop′ is the sliding window hop number after adjusting the accuracy, that is, the new sliding window hop number determined under the maximum accuracy of the current server or personal computer performance, which is equivalent to the actual value, and hop 1 It is a theoretical value.
[0082] For applications with high real-time requirements, due to the large hop setting, each slide hop data will cause abnormal fluctuations at the same time. The maximum detection accuracy is hop. Therefore, when improving the accuracy afterwards, adjust the window size after the accuracy to hop′=1, re-execute the long and short time window method, and use the same set of high and low standard threshold groups to determine the new abnormal fluctuation time.
[0083] Fine-tune the Detect the extreme points on the current time series. The maximum / minimum point that appears first is the final one with a small window w extending to its left and right. hop , slide left and right respectively,
[0084]
[0085] S502: When performing sliding window translation, calculate the ratio of long and short time windows, compare the ratio of long and short time windows with the average amplitude ratio threshold to obtain a first comparison result, and compare the window size when performing sliding window translation with the dual standard threshold combination to obtain a second comparison result.
[0086] S503: Determine an abnormal event of the battery to be detected based on the first comparison result and the second comparison result.
[0087] Specifically, when data arrives, the window size is set to |Win be +Win afr |, according to hop 1 Slide, first calculate the ratio of long and short time windows Is it satisfied? If satisfied, then judge No, if not satisfied, continue to judge the next
[0088] judge No, if satisfied, then judge If not satisfied, continue to judge the next
[0089] judge No, if satisfied, then judge End to The mean value of >T h2 If not satisfied, continue to judge the next
[0090] judge End to The mean value of >T h2 No, if satisfied, then judge If not satisfied, Perform 2x denoising.
[0091] judge No, if satisfied, then judge If not satisfied, Perform 2x denoising.
[0092] judge No, if satisfied, then Perform 4 times denoising. If not satisfied, Perform 2x denoising.
[0093] Then judge No, if it is satisfied, then the abnormal fluctuation set is judged to be a large amplitude event, and the program ends. If it is not satisfied, then Perform 2x denoising.
[0094] Final judgment No, if it is satisfied, then the abnormal fluctuation set is judged as a small amplitude event, and the program ends. If it is not satisfied, continue to judge the next
[0095] When it is determined to be a large / small amplitude event, the method in S101 is used to intercept the abnormal fluctuation time before and after seconds, execute the abnormal fluctuation time correction method, each abnormal fluctuation TS detected i2 , each constitutes an abnormal fluctuation set of a channel, and all channels are combined into an abnormal fluctuation set TS.
[0096] Large amplitude events are more likely to be a significant change in the state of the monitored object itself, which requires special attention from the user, while small amplitude events are more likely to be the accumulated information of the precursors of large amplitude events. Users need to analyze the possible large amplitude events and the nature of the monitored object itself based on this accumulated information, and make timely adjustments to the monitored object. For example, when sensors or monitoring devices from several different sources jointly monitor an object, the abnormal moments of the time series on different devices can be jointly analyzed to analyze the state of the monitored object, and possible large amplitude events can be discovered in advance. For another example, there is a lot of noise near some monitored objects. This type of data is likely to be a small amplitude event, prompting users to transfer sensors or monitoring devices to achieve higher monitoring benefits.
[0097] In practical applications, the battery extreme difference curve of a battery charging and discharging process can be selected, where the voltage extreme difference curve is selected as the detection of the abnormal fluctuation set, and the SOC curve is selected as a comparison. The relationship between the two can show the battery's high load condition. The battery extreme difference curve contains 210,000 data, with a sampling frequency of 100Hz, or 2100 seconds. Assuming that the dual standard threshold combination and the average amplitude ratio threshold have been determined through a small sample set of the abnormal fluctuation set, when improving the accuracy afterwards, adjust hop′=1, re-execute the long and short time window method, and use the same set of high and low standard threshold groups to determine the new abnormal fluctuation time. According to the sampling frequency, the sliding window hop detection is determined to obtain 7 detected abnormal moments After correction, high-precision abnormal time is obtained The time series data set extracted by the present invention can obtain 7 abnormal fluctuation sets TS 1~7 .
[0098] This application pre-sets parameters for battery data such as voltage, current, SOC, etc., and performs real-time abnormal moment detection on the existing high-dimensional, massive multi-channel time series data set. By tightening and loosening (i.e., "tightening" and "relaxing") the judgment conditions and using dual standard thresholds, the detection accuracy of abnormal moments is improved while meeting the real-time requirements, thereby improving the accuracy of picking up abnormal fluctuation sets. In the data closed loop of the entire monitoring object, the user information is collected and analyzed by the big data system (i.e., the vehicle system operation data), which is ultimately used to monitor battery abnormalities in real time and improve user experience.
[0099] Embodiment 2
[0100] See also Figure 6 As shown, Figure 6 A schematic diagram of the structure of a battery abnormal event detection device provided by Embodiment 2 of the present invention is shown, wherein the device comprises:
[0101] The historical abnormal fluctuation set determination module 601 is used to obtain, for each detection channel, a historical time series set of the battery to be detected under the detection channel before performing battery abnormal event detection, filter out a historical abnormal fluctuation moment set in the historical time series set, and construct an abnormal fluctuation set based on the historical abnormal fluctuation moment set, wherein the historical time series set is composed of battery parameters under the detection channel at several different historical moments;
[0102] A detection standard determination module 602 is used to determine a dual standard threshold combination and an average amplitude ratio threshold under the detection channel based on a historical abnormal fluctuation set under the detection channel;
[0103] The battery parameter acquisition module 603 is used to acquire the battery parameters of the battery to be detected in the detection channel at each current moment when performing battery abnormality event detection;
[0104] The abnormal event detection module 604 is used to determine the abnormal event of the battery to be detected based on the battery parameters of the battery to be detected in the detection channel at each current moment, according to the first sliding hop count, the dual standard threshold combination in the detection channel and the average amplitude ratio threshold.
[0105] In an optional embodiment, determining the average amplitude ratio threshold value under the detection channel based on the historical abnormal fluctuation set under the detection channel includes:
[0106] Perform sliding window translation on the historical abnormal fluctuation set under the detection channel with a second sliding jump number to determine the short-time window average amplitude and the long-time window average amplitude of each historical abnormal fluctuation set;
[0107] The average amplitude ratio threshold is determined based on the short-time window average amplitude and the long-time window average amplitude.
[0108] In an optional embodiment, determining a dual-standard threshold combination under the detection channel based on a set of historical abnormal fluctuations under the detection channel includes:
[0109] Set a number of extended time windows, and optimize the historical abnormal fluctuation set under the detection channel based on each extended time window;
[0110] The dual-standard threshold combination is determined based on the amplitude of the optimized historical abnormal fluctuation set.
[0111] In an optional embodiment, the abnormal event of the battery to be detected is determined based on the battery parameters of the battery to be detected in the detection channel at each current moment, according to the first sliding hop count, the dual standard threshold combination in the detection channel and the average amplitude ratio threshold, including:
[0112] Performing sliding window translation on the battery parameters of the battery to be detected at each current moment in the detection channel with the first sliding hop number;
[0113] When performing sliding window translation, the ratio of long and short time windows is calculated, the ratio of long and short time windows is compared with the average amplitude ratio threshold to obtain a first comparison result, and the window size when performing sliding window translation is compared with the dual standard threshold combination to obtain a second comparison result;
[0114] An abnormal event of the battery to be detected is determined based on the first comparison result and the second comparison result.
[0115] Embodiment 3
[0116] Based on the same application concept, see Figure 7 As shown, Figure 7 FIG. 4 shows a schematic diagram of the structure of a computer device provided by Embodiment 3 of the present invention, wherein: Figure 7 As shown, a computer device 700 provided in Embodiment 3 of the present application includes:
[0117] A processor 701, a memory 702 and a bus 703, wherein the memory 702 stores machine-readable instructions executable by the processor 701. When the computer device 700 is running, the processor 701 communicates with the memory 702 via the bus 703. When the processor 701 is running, the machine-readable instructions execute the steps of the battery abnormal event detection method shown in the above-mentioned embodiment 1.
[0118] Embodiment 4
[0119] Based on the same application concept, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the battery abnormal event detection method described in any one of the above embodiments are executed.
[0120] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the system and device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.
[0121] The computer program product for detecting abnormal battery events provided in the embodiments of the present invention includes a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the method described in the previous method embodiment. The specific implementation can be found in the method embodiment and will not be repeated here.
[0122] The battery abnormal event detection device provided in the embodiment of the present invention can be specific hardware on the device or software or firmware installed on the device. The implementation principle and technical effects of the device provided in the embodiment of the present invention are the same as those of the aforementioned method embodiment. For the sake of brief description, the parts not mentioned in the device embodiment can refer to the corresponding contents in the aforementioned method embodiment. Technical personnel in the relevant field can clearly understand that for the convenience and simplicity of description, the specific working processes of the systems, devices and units described above can all refer to the corresponding processes in the aforementioned method embodiment, and will not be repeated here.
[0123] In the embodiments provided by the present invention, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of devices or units can be electrical, mechanical or other forms.
[0124] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0125] In addition, each functional unit in the embodiment provided by the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
[0126] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.
[0127] It should be noted that similar numbers and letters represent similar items in the following figures. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. In addition, the terms "first", "second", "third", etc. are only used to distinguish the description and are not to be understood as indicating or implying relative importance.
[0128] Finally, it should be noted that the above-described embodiments are only specific implementations of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The protection scope of the present invention is not limited thereto. Although the present invention is described in detail with reference to the above-described embodiments, those skilled in the art should understand that any person skilled in the art can still modify the technical solutions described in the above-described embodiments within the technical scope disclosed by the present invention, or can easily think of changes, or perform equivalent replacements on some of the technical features thereof; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention. They should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims
1. A battery abnormal event detection method, characterized in that: The method comprises: Before performing battery abnormal event detection, for each detection channel, a historical time series set of the battery to be detected under the detection channel is obtained, a historical abnormal fluctuation moment set in the historical time series set is screened out, and an abnormal fluctuation set is constructed based on the historical abnormal fluctuation moment set, wherein the historical time series set is composed of battery parameters under the detection channel at several different historical moments; Determine a dual-standard threshold combination and an average amplitude ratio threshold under the detection channel based on a set of historical abnormal fluctuations under the detection channel; When performing battery abnormality event detection, obtaining battery parameters of the battery to be detected at each current moment in the detection channel; Based on the battery parameters of the battery to be detected in the detection channel at each current moment, the abnormal event of the battery to be detected is determined according to the first sliding hop count, the dual standard threshold combination in the detection channel and the average amplitude ratio threshold.
2. The method according to claim 1, characterized in that Based on the historical abnormal fluctuation set under the detection channel, the average amplitude ratio threshold under the detection channel is determined, including: Perform sliding window translation on the historical abnormal fluctuation set under the detection channel with a second sliding jump number to determine the short-time window average amplitude and the long-time window average amplitude of each historical abnormal fluctuation set; The average amplitude ratio threshold is determined based on the short-time window average amplitude and the long-time window average amplitude.
3. The method according to claim 1, characterized in that Based on the historical abnormal fluctuation set under the detection channel, a dual-standard threshold combination under the detection channel is determined, including: Set a number of extended time windows, and optimize the historical abnormal fluctuation set under the detection channel based on each extended time window; The dual-standard threshold combination is determined based on the amplitude of the optimized historical abnormal fluctuation set.
4. The method according to claim 1, characterized in that: The method of determining the abnormal event of the battery to be detected based on the battery parameters of the battery to be detected in the detection channel at each current moment and according to the first sliding hop count, the dual standard threshold combination in the detection channel and the average amplitude ratio threshold includes: Performing sliding window translation on the battery parameters of the battery to be detected at each current moment in the detection channel with the first sliding hop number; When performing sliding window translation, the ratio of long and short time windows is calculated, the ratio of long and short time windows is compared with the average amplitude ratio threshold to obtain a first comparison result, and the window size when performing sliding window translation is compared with the dual standard threshold combination to obtain a second comparison result; An abnormal event of the battery to be detected is determined based on the first comparison result and the second comparison result.
5. A battery abnormal event detection device, characterized in that: The device comprises: A historical abnormal fluctuation set determination module is used to obtain, for each detection channel, a historical time series set of the battery to be detected under the detection channel before performing battery abnormal event detection, filter out a historical abnormal fluctuation moment set in the historical time series set, and construct an abnormal fluctuation set based on the historical abnormal fluctuation moment set, wherein the historical time series set is composed of battery parameters under the detection channel at several different historical moments; A detection standard determination module, used to determine a dual-standard threshold combination and an average amplitude ratio threshold under the detection channel based on a historical abnormal fluctuation set under the detection channel; A battery parameter acquisition module, used to acquire the battery parameters of the battery to be detected in the detection channel at each current moment when performing battery abnormality event detection; The abnormal event detection module is used to determine the abnormal event of the battery to be detected based on the battery parameters of the battery to be detected in the detection channel at each current moment, according to the first sliding hop count, the dual standard threshold combination in the detection channel and the average amplitude ratio threshold.
6. The device according to claim 5, characterized in that Based on the historical abnormal fluctuation set under the detection channel, the average amplitude ratio threshold under the detection channel is determined, including: Perform sliding window translation on the historical abnormal fluctuation set under the detection channel with a second sliding jump number to determine the short-time window average amplitude and the long-time window average amplitude of each historical abnormal fluctuation set; The average amplitude ratio threshold is determined based on the short-time window average amplitude and the long-time window average amplitude.
7. The device according to claim 5, characterized in that Based on the historical abnormal fluctuation set under the detection channel, a dual-standard threshold combination under the detection channel is determined, including: Set a number of extended time windows, and optimize the historical abnormal fluctuation set under the detection channel based on each extended time window; The dual-standard threshold combination is determined based on the amplitude of the optimized historical abnormal fluctuation set.
8. The device according to claim 5, characterized in that The method of determining the abnormal event of the battery to be detected based on the battery parameters of the battery to be detected in the detection channel at each current moment and according to the first sliding hop count, the dual standard threshold combination in the detection channel and the average amplitude ratio threshold includes: Performing sliding window translation on the battery parameters of the battery to be detected at each current moment in the detection channel with the first sliding hop number; When performing sliding window translation, the ratio of long and short time windows is calculated, the ratio of long and short time windows is compared with the average amplitude ratio threshold to obtain a first comparison result, and the window size when performing sliding window translation is compared with the dual standard threshold combination to obtain a second comparison result; An abnormal event of the battery to be detected is determined based on the first comparison result and the second comparison result.
9. A computer device, characterized in that: include: A processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the computer device is running, the processor and the memory communicate via the bus, and when the machine-readable instructions are executed by the processor, the steps of the battery abnormal event detection method as described in any one of claims 1 to 4 are performed.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the battery abnormal event detection method as described in any one of claims 1 to 4 are executed.
Citation Information
Cited By
Battery abnormal data clustering method and clustering device
CN120724189A
Abnormality detection method and device, equipment and storage medium
CN121978554A
