Battery arc discharge anomaly detection method and device and storage medium

Abstract

The invention relates to the technical field of batteries, and discloses a battery arc discharge anomaly detection method and device and a storage medium, the method comprises the steps that a to-be-detected battery is sampled to obtain a group of sampling data, the sampling data comprises sampling data of n sampling points, and the sampling data of each sampling point comprises voltage and current; converting the sampling data into first frequency domain data and determining a first high-frequency component according to the first frequency domain data; determining first time domain data according to the sampling data, the first time domain data including at least one of a voltage instantaneous maximum variation amplitude, a current instantaneous maximum variation amplitude, a maximum instantaneous voltage change rate, a maximum instantaneous current change rate, a voltage standard deviation and a current standard deviation; performing dimension reduction fusion on a multi-dimensional matrix formed by the first high-frequency component and the first time domain data to obtain a single fusion index value; and determining whether the to-be-detected battery has arc discharge abnormity or not according to the fusion index value. According to the invention, whether arc discharge abnormity occurs in the battery can be accurately detected.

Description

Technical Field

[0001] Embodiments of the present application relate to the technical field of batteries, and particularly to a method, device, and storage medium for detecting abnormal arcing of batteries. Background Art

[0002] With the rapid development of renewable energy, battery systems, as energy storage systems that can be repeatedly charged and discharged, show great application potential in the energy field. However, during the operation of battery systems, abnormal arcing of batteries may occur due to loose connectors, poor contact, or degradation or failure of battery insulation performance in the battery system. After abnormal arcing of the battery occurs, if the abnormal arcing cannot be detected in a timely and accurate manner for abnormal handling, it may cause serious damage to the battery. Currently, there is a lack of a detection method that can accurately detect whether abnormal arcing occurs in the battery. Summary of the Invention

[0003] In view of the above problems, embodiments of the present application provide a method, device, and storage medium for detecting abnormal arcing of batteries, which are used to solve the problem in the prior art that it is impossible to accurately detect whether abnormal arcing occurs in the battery.

[0004] According to one aspect of the embodiments of the present application, a method for detecting abnormal arcing of batteries is provided. The method includes: sampling a battery to be detected to obtain a set of sampling data, where the sampling data includes sampling data of n sampling points, and the sampling data of each sampling point includes voltage and current, n is a positive integer, and n>1; converting the sampling data into first frequency-domain data, and determining a first high-frequency component according to the first frequency-domain data, where the first high-frequency component includes a voltage high-frequency component and a current high-frequency component; determining first time-domain data according to the sampling data, where the first time-domain data includes at least one of the instantaneous maximum voltage change amplitude, instantaneous maximum current change amplitude, maximum instantaneous voltage change rate, maximum instantaneous current change rate, voltage standard deviation, and current standard deviation; performing dimensionality reduction and fusion on the multi-dimensional matrix composed of the first high-frequency component and the first time-domain data to obtain a single fusion index value; and determining whether the battery to be detected has abnormal arcing according to the fusion index value.

[0005] In an optional manner, the performing dimensionality reduction and fusion on the multi-dimensional matrix composed of the first high-frequency component and the first time-domain data to obtain a single fusion index value includes: performing dimensionality reduction and fusion on the multi-dimensional matrix through the formula where is the fusion index value, is a preset matrix, is the multi-dimensional matrix, is a preset mean matrix, and and Determined according to N groups of sample data collected during battery arcing, each group of the sample data includes sampling data of n sampling points, and N is a positive integer.

[0006] In an alternative manner, the sampling frequency of the sampling data and the sample data is the same.

[0007] In an alternative manner, the preset matrix and the preset mean matrix are determined through the following steps: Collect the N groups of sample data during battery arcing; convert each group of the sample data into second frequency-domain data, and determine the second high-frequency components according to the second frequency-domain data, where the second high-frequency components include a voltage high-frequency component and a current high-frequency component; determine the second time-domain data according to each group of the sample data respectively, where the data types of the data included in the second time-domain data are the same as those of the data included in the first time-domain data, and the second time-domain data includes data of M data types; form an N×(M + 2) matrix with the second high-frequency components and the second time-domain data, and the data types of the data in the same column of the N×(M + 2) matrix are the same; determine the average value of each column of data in the N×(M + 2) matrix respectively to obtain (M + 2) average values corresponding to (M + 2) data types; determine the matrix composed of the (M + 2) average values as the preset mean matrix ; determine the difference between each data in the N×(M + 2) matrix and the average value corresponding to the data type of the data to obtain an N×(M + 2) de-meaned matrix ; determine the covariance matrix through the formula ; determine the eigenvalues of the covariance matrix and the eigenvectors corresponding to the eigenvalues; determine the eigenvector corresponding to the largest eigenvalue of the covariance matrix as the matrix ; determine the transpose matrix of the matrix W as the preset matrix W ; .

[0008] In an alternative manner, the determination of whether the battery to be detected has an arcing anomaly according to the fusion index value includes: if the fusion index value is greater than a first preset threshold, it is determined that the battery to be detected has an arcing anomaly.

[0009] In an alternative approach, before determining whether the battery under test has an arcing anomaly based on the fusion index value, the method further includes: repeatedly performing the step of sampling the battery under test to obtain a set of sampling data until the step of performing dimensionality reduction fusion on the multi-dimensional matrix composed of the first high-frequency component and the first time-domain data to obtain a single fusion index value for multiple times, thereby obtaining multiple fusion index values; determining whether the battery under test has an arcing anomaly based on the fusion index value includes: if there are k consecutive fusion index values that are all greater than a first preset threshold, it is determined that the battery under test has an arcing anomaly, where k is a positive integer and k≥2.

[0010] In an alternative approach, after determining that the battery under test has an arcing anomaly if there are k consecutive fusion index values that are all greater than a first preset threshold, the method further includes: repeatedly performing the step of sampling the battery under test to obtain a set of sampling data until the step of performing dimensionality reduction fusion on the multi-dimensional matrix composed of the first high-frequency component and the first time-domain data to obtain a single fusion index value for multiple times, thereby obtaining multiple fusion index values; if there are s consecutive fusion index values that are all less than a second preset threshold, it is determined that the arcing anomaly of the battery under test is lifted, where the second preset threshold is less than or equal to the first preset threshold, and s is a positive integer and s≥2.

[0011] In an alternative approach, determining the first time-domain data based on the sampling data includes: using the formula to determine the instantaneous maximum voltage change amplitude , where and are the voltages at two adjacent sampling points in the sampling data; using the formula to determine the instantaneous maximum current change amplitude , where and are the currents at two adjacent sampling points in the sampling data; using the formula to determine the instantaneous maximum voltage change rate , is the sampling time interval between two adjacent sampling points in the sampling data; using the formula to determine the instantaneous maximum current change rate ; using the formula to determine the voltage standard deviation , where is the voltage at the i-th sampling point in the sampling data, is the average value of all voltages in the sampling data; using the formula to determine the current standard deviation , where is the current of the i-th sampling point in the sampling data, is the average value of all currents in the sampling data.

[0012] According to another aspect of the embodiments of the present application, a battery arcing anomaly detection device is provided, including a memory, a processor, and a computer program stored on the memory. The processor executes the computer program to implement the battery arcing anomaly detection method as described above.

[0013] According to still another aspect of the embodiments of the present application, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the battery arcing anomaly detection method as described above is implemented.

[0014] Since the state of the battery is affected not only by voltage and current but also by other factors, in the embodiments of the present application, by simultaneously combining the time-domain data and frequency-domain data of the battery to determine whether the battery to be detected has an arcing anomaly, the signal anomaly when the battery has an arcing anomaly can be effectively captured, thereby improving the accuracy of the detection result. Moreover, in the embodiments of the present application, by simultaneously combining the time-domain data and frequency-domain data of the battery to determine whether the battery to be detected has an arcing anomaly, compared with the method of determining whether the battery to be detected has an arcing anomaly solely by frequency-domain data or time-domain data, some working conditions with large signal disturbances but no arcing can be filtered out, reducing the false alarm rate and enhancing the robustness of the system in a complex electrical environment. Furthermore, in the embodiments of the present application, there is no need to additionally set up devices such as infrared cameras, high-frame-rate cameras, and sensors in the battery system, which not only applies to battery systems with limited space but also reduces costs, and can also support on-site real-time detection of whether the battery has an arcing anomaly, thus providing guarantee for the safe operation of the battery system.

[0015] The above description is only an overview of the technical solutions of the embodiments of the present application. In order to be able to understand the technical means of the embodiments of the present application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features, and advantages of the embodiments of the present application more obvious and understandable, the following specifically gives the specific implementation manners of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The drawings are only used to illustrate the embodiments and are not considered as a limitation to the present application. And throughout the drawings, the same reference numerals are used to represent the same components. In the drawings: Figure 1 shows a schematic flow chart of the battery arcing anomaly detection method provided by the embodiments of the present application; Figure 2 shows a preset matrix provided by the embodiments of the present application and a schematic flow chart of the determination method of the preset mean matrix ; Figure 3 The structural schematic diagram of the battery arcing abnormality detection device provided by the embodiment of the present application is shown. Specific embodiments

[0017] The exemplary embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present application are shown in the drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments set forth herein.

[0018] New energy batteries are increasingly widely used in life and industries. For example, new energy vehicles equipped with batteries have been widely used. In addition, batteries are also increasingly used in the energy storage field and so on.

[0019] During the operation of the battery system, the battery may experience arcing abnormalities due to loose connectors, poor contact, or a decrease or even failure in the battery insulation performance in the battery system. "Arcing" refers to an abnormal discharge phenomenon that occurs inside or outside the battery, manifested as an electric arc generated when current passes through air or other non-conductive media. This electric arc is usually caused by poor electrical connection, insulation damage, short circuit, accidental contact between metal components, or other electrical faults. Arcing abnormalities in the battery system are extremely dangerous because they may cause fires, explosions, or damage to battery components. The high temperature generated by the electric arc can quickly ignite the flammable materials inside the battery, such as electrolyte or diaphragm, resulting in a fire. In addition, the electric arc may also cause physical damage to the electrical components of the battery system, further exacerbating the fault. Therefore, during the operation of the battery system, it is necessary to detect in a timely and accurate manner whether the battery has an arcing abnormality, so that when the battery has an arcing abnormality, the abnormality can be processed in a timely manner, thereby reducing losses.

[0020] As introduced above, when the battery has an arcing abnormality, it is manifested as an electric arc generated when current passes through air or other non-conductive media. Therefore, an infrared camera device or a high-frame-rate camera device can be set in the area where an electric arc may appear in the battery system to perform image recognition and analysis on this area through the set device to detect whether an electric arc appears, so as to detect whether the battery has an arcing abnormality. However, since both the infrared camera device and the high-frame-rate camera device have a certain volume, they are not suitable for battery systems with limited space, and the infrared camera device and the high-frame-rate camera device are expensive. If the infrared camera device or the high-frame-rate camera device is used to detect whether the battery has an arcing abnormality, the detection cost is relatively high.

[0021] When an arc-drawing anomaly occurs in the battery, the voltage and current at the load end connected to the battery will change. Therefore, it is possible to detect whether an arc-drawing anomaly has occurred in the battery by checking whether there are significant fluctuations in the voltage and current at the load end. However, this detection method has a weak ability to identify high-frequency and small-amplitude arc-drawing anomalies and is easily affected by interference factors such as load disturbances and system switching, resulting in false alarms and missed detections.

[0022] To detect whether an arc-drawing anomaly has occurred in the battery, a signal sensor can also be set in the battery system to detect the sound or light signal in the battery system to detect whether there is an electric arc in the battery system, thereby detecting whether an arc-drawing anomaly has occurred in the battery. However, this detection method relies on sensors and has a high system complexity and is not suitable for closed or strongly electromagnetically interfered environments.

[0023] When an arc-drawing anomaly occurs in the battery, abnormal frequency components usually appear in the frequency-domain signals corresponding to the voltage and current signals of the battery. These abnormal frequency components reflect the high-frequency signals in the arc-discharge process. However, due to the limited generalization ability of spectral features for different working conditions and the difficulty in covering all arc-drawing scenarios, if the occurrence of arc-drawing is detected only based on frequency-domain data, the accuracy is relatively low. Moreover, for arc-drawing anomalies with a relatively smooth change in spectral energy distribution, since the change in frequency-domain data is not significant enough, detection failures may occur.

[0024] Based on the above considerations, without additionally setting devices such as infrared cameras, high-frame-rate cameras, and sensors in the battery system, in order to improve the accuracy of detecting whether an arc-drawing anomaly has occurred in the battery, this application proposes a method for detecting battery arc-drawing anomalies. By sampling the voltage and current of the battery to be detected, sampling data including multiple voltages and multiple currents is obtained. The sampling data is converted into frequency-domain data. The voltage high-frequency component and the current high-frequency component are determined from the frequency-domain data, and the time-domain data is determined based on the sampling data, where the time-domain data includes at least one of the instantaneous maximum voltage change amplitude, the instantaneous maximum current change amplitude, the maximum instantaneous voltage change rate, the maximum instantaneous current change rate, the voltage standard deviation, and the current standard deviation. Then, the multi-dimensional matrix composed of the voltage high-frequency component, the current high-frequency component, and the time-domain data is dimension-reduced and fused to obtain a single fusion index value. Finally, based on the fusion index value, it can be determined whether an arc-drawing anomaly has occurred in the battery. In this application, by simultaneously combining frequency-domain and time-domain data to detect whether an arc-drawing anomaly has occurred in the battery, rather than solely analyzing frequency-domain data or time-domain data, the accuracy of the detection result is improved.

[0025] Figure 1The flowchart of the battery arc discharge anomaly detection method provided by the embodiments of the present application is shown. This method is executed by a terminal device, which can be a device including one or more processors, such as an industrial and commercial energy storage cabinet. The processor may be a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application, which is not limited herein. One or more processors included in the device may be of the same type of processor, such as one or more CPUs; or may be of different types of processors, such as one or more CPUs and one or more ASICs, which is not limited herein. As Figure 1 shown, the method includes the following steps: Step 110: Sample the battery to be detected to obtain a set of sampled data.

[0026] Among them, the battery to be detected may be a high-voltage DC battery. In this step, specifically, the voltage and current of the battery to be detected are collected at a preset sampling frequency (for example, 50 KHz) to obtain a set of sampled data. This set of sampled data includes the sampled data of n sampling points, and the sampled data of each sampling point includes voltage and current. n is a positive integer, and n>1.

[0027] In this step, the sampled data is used to determine whether the battery to be detected has an arc discharge anomaly. Therefore, the more the amount of the sampled data obtained in this step, the higher the accuracy of the result obtained by determining whether the battery to be detected has an arc discharge anomaly based on the sampled data. However, it can be understood that if the amount of sampled data is larger, the corresponding time consumed for processing the sampled data is longer, which will affect the efficiency of detecting whether the battery to be detected has an arc discharge anomaly. Therefore, in this step, n can be set as needed. For example, n is 256, 512, or 1024, etc. In the embodiments of the present application, in order to balance the accuracy of the detection result of determining whether the battery to be detected has an arc discharge anomaly and the detection efficiency, preferably, n is 1024.

[0028] Step 120: Convert the sampled data into first frequency domain data, and determine the first high-frequency component according to the first frequency domain data.

[0029] Among them, the n voltage and n current data sampled in step 110 are all time domain data. In this step, the sampled data is converted to obtain the first frequency domain data, where the first frequency domain data includes voltage frequency domain data and current frequency domain data. Specifically, the Fourier transform, fast Fourier transform, or wavelet transform is performed on the n voltage data in the time domain to obtain the voltage frequency domain data. Similarly, the Fourier transform, fast Fourier transform, or wavelet transform is performed on the n current data in the time domain to obtain the current frequency domain data.

[0030] As introduced above, when an abnormal arcing occurs in the battery, abnormal frequency components usually appear in the frequency-domain signals corresponding to the voltage and current signals of the battery. These abnormal frequency components reflect the high-frequency signals during the arc discharge process. Therefore, in this step, a first high-frequency component is determined based on the first frequency-domain data, so as to subsequently determine whether an abnormal arcing has occurred in the battery to be detected based on the first high-frequency component, where the first high-frequency component includes a voltage high-frequency component and a current high-frequency component.

[0031] Specifically, the sum of the amplitudes with frequencies greater than 3 KHz in the voltage frequency-domain data is determined, and this sum is determined as the voltage high-frequency component. Similarly, the sum of the amplitudes with frequencies greater than 3 KHz in the current frequency-domain data is determined, and this sum is determined as the current high-frequency component.

[0032] Step 130: Determine the first time-domain data based on the sampling data.

[0033] Among them, the first time-domain data includes at least one of the instantaneous maximum voltage change amplitude, the instantaneous maximum current change amplitude, the maximum instantaneous voltage change rate, the maximum instantaneous current change rate, the voltage standard deviation, and the current standard deviation.

[0034] Among them, the instantaneous maximum voltage change amplitude can be determined by the formula where and are the voltages of two adjacent sampling points in the sampling data. Specifically, after determining the absolute values of the voltage differences between all adjacent two sampling points among the n sampling points, the maximum absolute value among all the absolute values of the voltage differences is determined as the instantaneous maximum voltage change amplitude . For example, if n is 3, first determine the absolute value of the voltage difference between the first sampling point and the second sampling point, and determine the absolute value of the voltage difference between the second sampling point and the third sampling point, and then determine the maximum absolute value among these two absolute values of the voltage differences as the instantaneous maximum voltage change amplitude . .

[0035] The instantaneous maximum current change amplitude can be determined by the formula where and are the currents of two adjacent sampling points in the sampling data. Among them, the specific determination method of the instantaneous maximum current change amplitude is similar to the specific determination method of the instantaneous maximum voltage change amplitude . Therefore, the specific determination method of the instantaneous maximum current change amplitude can refer to the specific determination method of the above-mentioned instantaneous maximum voltage change amplitude , and will not be elaborated here. .

[0036] The instantaneous maximum rate of change of voltage can be determined by the formula where , is the sampling time interval between two adjacent sampling points in the sampling data. If the sampling data is collected at the first sampling rate , then .

[0037] The instantaneous maximum rate of change of current can be determined by the formula where .

[0038] The standard deviation of voltage can be determined by the formula where , is the voltage at the i-th sampling point in the sampling data is the average value of all voltages in the sampling data, and i is a positive integer

[0039] The standard deviation of current can be determined by the formula where , is the current at the i-th sampling point in the sampling data is the average value of all currents in the sampling data

[0040] In this step, the instantaneous maximum change amplitude of voltage , the instantaneous maximum change amplitude of current , the maximum instantaneous rate of change of voltage , the maximum instantaneous rate of change of current , the standard deviation of voltage and the standard deviation of current can be determined in the above manner, and at least one of them is used as the first time-domain data. By determining the time-domain data, the instantaneous dynamic characteristics when the battery has an arc-drawing anomaly can be captured

[0041] It should be noted that since the first time-domain data is used to determine whether the battery to be detected has an arc-drawing anomaly in the subsequent process, the more data types included in the first time-domain data, the higher the accuracy of the detection result obtained by determining whether the battery to be detected has an arc-drawing anomaly based on the first time-domain data. Therefore, in the embodiments of the present application, preferably, the first time-domain data includes the instantaneous maximum change amplitude of voltage , the instantaneous maximum change amplitude of current , the maximum instantaneous rate of change of voltage , the maximum instantaneous rate of change of current , the standard deviation of voltage and the standard deviation of current .

[0042] Step 140: Perform dimensionality reduction and fusion on the multi-dimensional matrix composed of the first high-frequency component and the first time-domain data to obtain a single fusion index value.

[0043] Specifically, the dimensionality reduction and fusion of the multi-dimensional matrix can be performed through the formula . Among them, is the fusion index value, is the preset matrix, is the multi-dimensional matrix composed of the first high-frequency component and the first time-domain data, is the preset mean matrix. The matrix includes the preset means corresponding to each data type in the matrix , is each data in the matrix minus the preset mean corresponding to the data type of the data. If the matrix includes m time-domain data, then the dimension of the matrix is 1×(m + 2), the matrix is a 1×(m + 2) matrix, and the matrix is a (m + 2)×1 matrix. and can be set as needed, and the determination methods of and will be specifically introduced below.

[0044] To improve the efficiency of determining whether the battery under test has an arcing anomaly, in Step 140, weights corresponding to the voltage high-frequency component, current high-frequency component, and time-domain data of each data type can also be set, and then each data in the multi-dimensional matrix composed of the high-frequency component and time-domain data is weighted and summed with the weight corresponding to the data type of each data. The obtained sum value is the single fusion index value, thereby improving the efficiency of obtaining the fusion index value and also improving the efficiency of determining whether the battery under test has an arcing anomaly.

[0045] Step 150: Determine whether the battery under test has an arcing anomaly according to the fusion index value.

[0046] Among them, the threshold can be set to determine whether the battery under test has an arcing anomaly. Specifically, if the fusion index value determined in Step 140 is greater than the first preset threshold, it is determined that the battery under test has an arcing anomaly. The first preset threshold can be determined as needed and is not limited here.

[0047] Since the state of the battery is affected not only by voltage and current, but also by other factors, in the embodiments of the present application, by simultaneously combining the time-domain data and frequency-domain data of the battery to determine whether the battery to be detected has an arc-drawing anomaly, the signal anomaly when the battery has an arc-drawing anomaly can be effectively captured, thereby improving the accuracy of the detection result. Moreover, in the embodiments of the present application, by simultaneously combining the time-domain data and frequency-domain data of the battery to determine whether the battery to be detected has an arc-drawing anomaly, compared with the method of determining whether the battery to be detected has an arc-drawing anomaly solely through frequency-domain data or time-domain data, some working conditions with large signal disturbances but no arc-drawing can be filtered out, reducing the false alarm rate and enhancing the robustness of the system in a complex electrical environment. Furthermore, in the embodiments of the present application, there is no need to additionally set up devices such as infrared cameras, high-frame-rate cameras, and sensors in the battery system, which not only applies to battery systems with limited detection space, but also reduces costs, and can also support on-site real-time detection of whether the battery has an arc-drawing anomaly, thereby providing guarantee for the safe operation of the battery system.

[0048] Generally speaking, in the embodiments of the present application, by collecting signals such as voltage and current in real time during the operation of the battery to be detected, extracting its spectral features (such as energy distribution in specific frequency bands, peak frequency changes, etc.), and combining the first time-domain data features to construct a high-dimensional feature vector. Subsequently, mathematical tools such as principal component analysis (PCA) are used to perform dimensionality reduction processing on the feature vector, and the multi-dimensional features are fused into a single feature quantity to achieve efficient and accurate identification of the possible arc-drawing state of the battery to be detected. This method has the advantages of strong real-time performance, good anti-noise performance, and suitability for implementation in embedded systems, and can be widely applied to the safety protection of DC systems in fields such as electric vehicles and energy storage power stations.

[0049] Moreover, compared with traditional arc-drawing detection methods (such as those based on high-frequency hardware filtering, fixed threshold judgment, or single physical quantity monitoring), the advantage of the embodiments of the present application lies in stronger fusion analysis of multi-source information and feature space mapping ability. By constructing a multi-dimensional feature vector and performing normalization and standardization processing, the problem of misjudgment caused by the lack of obvious change in a single signal is effectively avoided; at the same time, by means of data dimensionality reduction and feature extraction algorithms such as PCA, the operation amount can be greatly reduced while ensuring high-sensitivity identification, adapting to the resource-limited MCU platform. In addition, the present application also has good scalability and can be adapted to battery systems with different power levels and different structural forms. By combining model training with empirical thresholds, the intelligence and robustness of arc-drawing judgment can be further improved. Overall, the embodiments of the present application provide an efficient, embeddable, and intelligent DC arc-drawing detection method, providing a solid guarantee for the safe operation of the battery system.

[0050] To reduce losses, in the embodiments of the present application, after determining that an arc discharge anomaly occurs in the battery, an emergency protection strategy is immediately executed. For example, the battery output loop is quickly cut off to prevent continuous arc discharge from causing thermal damage and safety hazards. At the same time, an alarm signal is sent to the system or the user, and key electrical characteristic data is recorded for subsequent analysis and model optimization.

[0051] Since the arc discharge anomaly in the battery lasts for a period of time, in order to avoid misjudgment, in the embodiments of the present application, steps 110 to 140 Figure 1 are repeatedly executed multiple times, and correspondingly, multiple fusion index values are obtained. And step 150 includes: if there are k consecutive fusion index values that are all greater than a first preset threshold, it is determined that an arc discharge anomaly occurs in the battery to be detected, where k is a positive integer and k≥2, and k can be set as needed, for example, set to 2, 3, or 5, etc.

[0052] As introduced above, when the battery does not have an arc discharge anomaly but is in a working condition with large signal disturbances, it will also affect the voltage and current of the battery. In the embodiments of the present application, by repeatedly executing steps 110 to 140 multiple times, if k consecutive fusion index values all indicate that an arc discharge anomaly occurs in the battery to be detected, then fluctuations in the battery voltage and current caused by factors such as signal disturbances can be excluded, so that it can be accurately determined that an arc discharge anomaly occurs in the battery to be detected.

[0053] When the above embodiments are specifically executed, a counting parameter p for the number of consecutive fusion index values greater than the first preset threshold can be set. The initial value of p is 0, and the counting threshold is k. After obtaining a fusion index value each time step 140 is executed, it is determined whether the fusion index value is greater than the first preset threshold. If the judgment result is yes, then p = p + 1. Subsequently, continue to execute the next round of steps 110 to 140. If the fusion index value obtained in this round is still greater than the first preset threshold, then p = p + 1 (that is, continue to increment the p value by 1), otherwise reset p to the initial value 0. In this way, when p = k, it is determined that an arc discharge anomaly occurs in the battery to be detected.

[0054] After detecting that an arc discharge anomaly occurs in the battery, in order to determine whether the arc discharge anomaly is eliminated, in the embodiments of the present application, on the basis of the embodiments Figure 1 provided, after determining that an arc discharge anomaly occurs in the battery, the method further includes: repeatedly executing steps 110 to 140 multiple times, and correspondingly, multiple fusion index values are obtained. If there are s consecutive fusion index values that are all less than a second preset threshold, it is determined that the arc discharge anomaly of the battery to be detected is eliminated, where the second preset threshold is less than or equal to the first preset threshold, s is a positive integer, and s≥2, and the second preset threshold can be set as needed.

[0055] In the embodiment of the present application, the arc strike anomaly is determined to be lifted only when s consecutive fusion index values are all less than the second preset threshold, which avoids misjudgment and thus improves the accuracy of the result of whether the arc strike anomaly is lifted finally obtained.

[0056] In the specific implementation of the above embodiment, a quantity counting parameter q for counting the number of consecutive fusion index values greater than the first preset threshold can be set. The initial value of q is 0 and the counting threshold is s. After detecting that the battery has an arc strike anomaly, in the subsequent battery sampling and data processing process, each time a fusion index value is obtained after executing step 140, it is determined whether the fusion index value is less than the second preset threshold. If the determination result is yes, then q = q + 1. Subsequently, continue to execute the next round of steps 110 to 140. If the fusion index value obtained in this round is still less than the second preset threshold, then q = q + 1 (that is, continue to increment the q value by 1), otherwise reset q to the initial value 0. In this way, when q = s, it is determined that the arc strike anomaly of the battery to be detected is lifted.

[0057] In order to enable the battery to resume normal operation, in the embodiment of the present application, after determining that the arc strike anomaly of the battery is lifted, the power supply capacity of the battery is restored, and the arc strike recovery event is reported to the remote monitoring platform for monitoring the battery through the communication interface, so as to realize the real-time safety supervision and early warning response of the battery system, and comprehensively improve the intelligence, safety and operation stability of the system.

[0058] In order to further improve the accuracy of determining whether the battery to be detected has an arc strike anomaly, Figure 2 shows the preset matrix provided by the embodiment of the present application and the preset mean matrix The flow schematic diagram of the determination method. As Figure 2 shown, the preset matrix and the preset mean matrix can be determined through the following steps 201 to step 211.

[0059] Step 201: Collect N groups of sample data when the battery has an arc strike.

[0060] In this step, an arc generation device can be used to control the battery to have an arc strike anomaly, and N groups of sample data of the battery with an arc strike anomaly are collected at a preset frequency. Among them, each group of sample data includes sampling data of n sampling points, and the sampling data of each sampling point includes voltage data and current data. N is a positive integer, for example, N is 3000, 4000 or 5000, etc.

[0061] Since the sample data collected in this step is used for subsequent determination of the preset matrix and the preset mean matrix , therefore, in order to ensure the determination of the preset matrix and the preset mean matrix It is applicable to determine whether an arc strike anomaly occurs in the battery to be detected. In the embodiments of the present application, preferably, the sampling frequency when sampling the battery to be detected in step 110 is the same as the sampling frequency when sampling the battery with an arc strike anomaly in this step, that is, the sampling frequencies of the sampling data and the sample data are the same.

[0062] Step 202: Convert each group of sample data into second frequency domain data, and determine the second high-frequency component according to the second frequency domain data.

[0063] This step is similar to step 120. Therefore, the principle and implementation method of this step can refer to step 120 and will not be elaborated here. Correspondingly, the second high-frequency component includes a voltage high-frequency component and a current high-frequency component.

[0064] Step 203: Determine the second time domain data according to each group of sample data respectively.

[0065] Among them, the data types of the data included in the second time domain data are the same as those of the data included in the first time domain data, and the second time domain data includes data of M data types. This step is similar to step 130. Therefore, the principle and implementation method of this step can refer to step 130 and will not be elaborated here.

[0066] Step 204: Combine the second high-frequency component and the second time domain data into an N×(M + 2) matrix, and the data types of the data in the same column of the N×(M + 2) matrix are the same.

[0067] Among them, since the second frequency domain data includes a voltage high-frequency component and a current high-frequency component, and the second time domain data includes data of M data types, there are a total of (M + 2) data. Therefore, in this step, the second high-frequency component and the second time domain data determined by each group of sample data are used as a row of data in the N×(M + 2) matrix. Since there are N groups of sample data, N rows of data in the N×(M + 2) matrix are formed.

[0068] Step 205: Determine the average value of each column of data in the N×(M + 2) matrix respectively, and obtain (M + 2) average values corresponding to (M + 2) data types.

[0069] Since the data types of the data in each column of the N×(M + 2) matrix are the same, in this step, by determining the average value of each column of data, (M + 2) average values corresponding to (M + 2) data types can be obtained.

[0070] Step 206: Determine the matrix composed of (M + 2) average values as the preset mean matrix .

[0071] Among them, the matrix It includes the (M + 2) averages determined in step 205.

[0072] Step 207: Determine the difference between each data in the N×(M + 2) matrix and the average value corresponding to the data type of this data, to obtain an N×(M + 2) mean-removed matrix .

[0073] Specifically, determine the difference between each data in the N×(M + 2) matrix composed of the second high-frequency component and the second time-domain data obtained in step 204 and the average value corresponding to the data type of this data determined in step 205. In the embodiments of the present application, through the above operations, data position offset can be eliminated, and the operation of determining the covariance matrix based on the mean-removed matrix can be simplified.

[0074] Step 208: Determine the covariance matrix through the formula . .

[0075] Wherein, the matrix is the transpose matrix of the matrix .

[0076] Step 209: Determine the eigenvalues and eigenvectors corresponding to the eigenvalues of the covariance matrix .

[0077] In this step, determine all the eigenvalues and all the eigenvectors corresponding to the eigenvalues of the covariance matrix .

[0078] Step 210: Determine the eigenvector corresponding to the largest eigenvalue of the covariance matrix as the matrix W .

[0079] Step 211: Determine the transpose matrix of the matrix W as the preset matrix .

[0080] In the embodiments of the present application, by collecting sample data during battery arcing and determining the preset matrix and the preset mean matrix , the numerical value of the arc characteristics can be extracted. Subsequently, based on the preset matrix and the preset mean matrix , when determining whether the battery to be detected has an abnormal arcing, the detection accuracy can be improved.

[0081] Figure 3 Fig. shows the structural schematic diagram of the battery arcing abnormal detection device provided by the embodiments of the present application. The specific implementation of the battery arcing abnormal detection device is not limited in the specific embodiments of the present application.

[0082] As shown Figure 3 in FIG. 300, the battery arc-drawing anomaly detection device 300 may include: a processor 302 and a memory 304.

[0083] Among them, the memory 304 is used to store a computer program 306. The memory 304 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk memory. The computer program 306 may include computer-executable instructions.

[0084] The processor 302 is used to execute the computer program 306 to implement the battery arc-drawing anomaly detection method embodiment described above.

[0085] The processor 302 may be a central processing unit CPU, or a specific integrated circuit ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of the present application. One or more processors included in the battery arc-drawing anomaly detection device 300 may be of the same type of processor, such as one or more CPUs; or may be of different types of processors, such as one or more CPUs and one or more ASICs.

[0086] An embodiment of the present application provides a computer-readable storage medium, and the storage medium stores a computer program, and when the computer program is executed by a processor, the battery arc-drawing anomaly detection method embodiment described above is implemented.

[0087] An embodiment of the present application provides a computer program, and the computer program can be executed by a processor to implement the battery arc-drawing anomaly detection method embodiment described above.

[0088] An embodiment of the present application provides a computer program product, and the computer program product includes a computer program, and when the computer program is executed by a processor, the battery arc-drawing anomaly detection method embodiment described above is implemented.

[0089] In several embodiments provided by the present application, if any function is implemented in the form of a software functional module / unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, all or part of the technical solution of the present application can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be an electronic device such as a personal computer, a server, etc.) to execute all or part of the steps of the methods described in various embodiments of the present application. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks, or optical discs that can store computer program codes.

[0090] The algorithms or displays provided herein are not inherently related to any particular computer, virtual system, or other device. Various general-purpose systems can also be used in conjunction with the teachings provided herein. Based on the above description, the structure required to construct such systems is obvious. In addition, the embodiments of the present application are not directed to any particular programming language. It should be understood that the content of the present application described herein can be implemented using various programming languages, and the descriptions made above regarding specific languages are for the purpose of disclosing the best mode of the present application.

[0091] It should be noted that the above embodiments illustrate the present application rather than limit the present application, and those skilled in the art can design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in the claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The present application can be implemented by means of hardware including several different elements and by means of a properly programmed computer. In a claim listing several devices, several of the units or modules among these devices can be embodied by the same hardware item. The use of the words first, second, and third, etc. does not denote any order. These words can be interpreted as names. The steps in the above embodiments, unless otherwise specified, should not be construed as limiting the order of execution.

[0092] The above-described embodiments merely represent several implementation manners of the present application, and their descriptions are relatively specific and detailed, but should not be construed as limiting the patent scope of the present application. It should be pointed out that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.

Claims

1. A method for detecting abnormal arcing of a battery, characterized in that, The method includes: Sampling the battery to be detected to obtain a set of sampling data, where the sampling data includes sampling data of n sampling points, and the sampling data of each sampling point includes voltage and current. n is a positive integer and n > 1; Converting the sampling data into first frequency-domain data and determining a first high-frequency component according to the first frequency-domain data, where the first high-frequency component includes a voltage high-frequency component and a current high-frequency component; Determining first time-domain data according to the sampling data, where the first time-domain data includes at least one of the instantaneous maximum voltage change amplitude, the instantaneous maximum current change amplitude, the maximum instantaneous voltage change rate, the maximum instantaneous current change rate, the voltage standard deviation, and the current standard deviation; Performing dimensionality reduction fusion on the multi-dimensional matrix composed of the first high-frequency component and the first time-domain data to obtain a single fusion index value; Determining whether the battery to be detected has an arcing anomaly according to the fusion index value.

2. The method according to claim 1, characterized in that, The performing dimensionality reduction fusion on the multi-dimensional matrix composed of the first high-frequency component and the first time-domain data to obtain a single fusion index value includes: By the formula dimensionality reduction and fusion are performed on the multi-dimensional matrix, where is the fusion index value, is a preset matrix, is the multi-dimensional matrix, is a preset mean matrix, and and are determined according to N groups of sample data collected during battery arcing. Each group of the sample data includes sampling data of n sampling points, and N is a positive integer.

3. The method according to claim 2, wherein The sampling frequency of the sampling data and the sample data is the same.

4. The method according to claim 2, wherein The preset matrix and the preset mean matrix are determined through the following steps: Collecting the N groups of sample data when the battery arcs; Converting each group of the sample data into second frequency-domain data and determining a second high-frequency component according to the second frequency-domain data, where the second high-frequency component includes a voltage high-frequency component and a current high-frequency component; Respectively determining second time-domain data according to each group of the sample data, where the data types of the second time-domain data are the same as those of the data included in the first time-domain data, and the second time-domain data includes data of M data types; Forming an N×(M + 2) matrix with the second high-frequency component and the second time-domain data, and the data types of the data in the same column of the N×(M + 2) matrix are the same; Respectively determining the average value of the data in each column of the N×(M + 2) matrix to obtain (M + 2) average values corresponding to (M + 2) data types; Determine the matrix composed of the (M + 2) averages as the preset mean matrix ; Determine the difference between each data in the N×(M + 2) matrix and the average value corresponding to the data type of the data, to obtain an N×(M + 2) mean-removed matrix ; Determine the covariance matrix through the formula ;​ Determine the covariance matrix and the eigenvectors corresponding to the eigenvalues Determine the eigenvector corresponding to the maximum eigenvalue of the covariance matrix as matrix W ; Determine the transpose matrix of the said matrix W as the said preset matrix .

5. The method according to claim 1, characterized in that The determining whether the battery to be detected has an arcing anomaly according to the fusion index value includes: If the fusion index value is greater than a first preset threshold, determining that the battery to be detected has an arcing anomaly.

6. The method according to claim 1, wherein Before the determining whether the battery to be detected has an arcing anomaly according to the fusion index value, the method further includes: Repeatedly executing the step of sampling the battery to be detected to obtain a set of sampling data until the step of performing dimensionality reduction fusion on the multi-dimensional matrix composed of the first high-frequency component and the first time-domain data to obtain a single fusion index value for multiple times to obtain multiple fusion index values; The determining whether the battery to be detected has an arcing anomaly according to the fusion index value includes: If there are k consecutive fusion index values that are all greater than the first preset threshold, determining that the battery to be detected has an arcing anomaly, where k is a positive integer and k ≥ 2.

7. The method according to claim 6, wherein After if there are k consecutive fusion index values that are all greater than the first preset threshold, determining that the battery to be detected has an arcing anomaly, the method further includes: Repeat the step of sampling the battery to be detected to obtain a set of sampling data until the step of performing dimensionality reduction and fusion on the multi-dimensional matrix composed of the first high-frequency component and the first time-domain data to obtain a single fusion index value for multiple times, and obtain multiple fusion index values; If there are s consecutive fusion index values that are all less than the second preset threshold, it is determined that the arcing anomaly of the battery to be detected is lifted, where the second preset threshold is less than or equal to the first preset threshold, s is a positive integer, and s≥2.

8. The method according to claim 1, wherein The determining the first time-domain data according to the sampling data includes: Determine the instantaneous maximum voltage change amplitude through the formula wherein is the voltage of two adjacent sampling points in the sampling data; and are the voltages of two adjacent sampling points in the sampling data Determine the instantaneous maximum change amplitude of the current through the formula wherein , and are the currents of two adjacent sampling points in the sampling data; Determine the instantaneous maximum rate of change of voltage through the formula where is the sampling time interval between two adjacent sampling points in the sampling data; ​ Determine the instantaneous maximum change rate of current through the formula ;​ Determine the voltage standard deviation through the formula where , is the voltage at the i-th sampling point in the sampling data, is the average value of all voltages in the sampling data; Determine the current standard deviation through the formula where is the current at the i-th sampling point in the sampling data, and is the average value of all currents in the sampling data.

9. A battery arc abnormal detection device, comprising a memory, a processor, and a computer program stored on the memory, characterized in that, The processor executes the computer program to implement the battery arcing anomaly detection method according to any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the battery arcing anomaly detection method according to any one of claims 1 to 8.

Images