Battery fault detection method and device, computer equipment, readable storage medium and program product

Abstract

The invention relates to a battery fault detection method and device and a computer readable storage medium. The method comprises the following steps: acquiring related data of a battery module; according to the related data, impedance and an impedance mean value of each single battery in the battery module in a preset frequency band are calculated; calculating correlation coefficients of each single battery under different preset frequency bands; calculating a difference value between a mean value generated by the correlation coefficients of the plurality of single batteries under different preset frequency bands and the correlation coefficient of each single battery, and sorting the difference values to obtain a difference value ranking numerical value of each single battery under different preset frequency bands; calculating a standard deviation and a mean value of the difference ranking numerical values of each single battery under different preset frequency bands, and determining a preset score threshold value of the single battery; scoring the single batteries; and screening out the faulty single battery according to the score of each single battery. By adopting the method, the faulted single battery in the battery module can be quickly and accurately detected.

Description

Technical Field

[0001] The present application relates to the technical field of battery detection, and in particular to a battery fault detection method, apparatus, computer equipment, computer-readable storage medium, and computer program product. Background Art

[0002] The rapid development of battery technology has promoted the continuous progress in the fields of renewable energy, electric vehicles, etc. However, battery abnormality problems still restrict the development of these technologies.

[0003] At present, the existing technologies are divided into battery internal parameter monitoring and data-driven abnormal diagnosis methods. The former monitors and analyzes the internal parameters of lithium-ion batteries. The changes in these parameters can reflect the health status and performance degradation of the battery. The latter uses machine learning and data mining technology to realize automatic identification and prediction of abnormal battery conditions. However, both methods are based on abnormal diagnosis of single cells. When a battery module fails, only the single cells can be tested one by one, and it is impossible to quickly and effectively detect which single cell in the battery module is abnormal, which easily leads to waste of resources during the detection process. Summary of the invention

[0004] Based on this, it is necessary to provide a battery fault detection method, device, computer equipment, computer readable storage medium and computer program product that can quickly and accurately detect faulty single cells in a battery module in response to the above technical problems.

[0005] In a first aspect, the present application provides a battery fault detection method, comprising:

[0006] Acquire relevant data of the battery module; the relevant data includes the voltage and current of each single battery in the battery module;

[0007] Calculating the impedance and the impedance mean of each single battery in the battery module in a preset frequency range according to the relevant data;

[0008] Calculating the correlation coefficient of each single battery in different preset frequency ranges according to the impedance and the impedance average;

[0009] Calculating the difference between the mean value of the correlation coefficients of the plurality of single cells in different preset frequency bands and the correlation coefficient of each single cell, and sorting the difference values ​​to obtain the difference ranking value of each single cell in the different preset frequency bands;

[0010] Calculating the standard deviation and mean of the difference ranking values ​​of each of the single cells in the different preset frequency ranges, and determining the preset threshold of the score of the single cell according to the standard deviation and the mean;

[0011] Scoring the single cell according to the difference ranking value of the single cell in the entire preset frequency range and the preset scoring threshold, to obtain the score of the single cell;

[0012] According to the score of each single battery, the single battery with fault is screened out.

[0013] In one embodiment, calculating the impedance and the impedance mean of each single battery in the battery module in a preset frequency range according to the relevant data includes:

[0014] Based on wavelet transform, the voltage and current of each single battery are represented by waveforms in the time-frequency domain;

[0015] Comparing the time-frequency data of the voltage and current in the waveform state to obtain the impedance of each single battery;

[0016] According to the impedance of each single battery, an impedance average of the single battery is obtained.

[0017] In one embodiment, calculating the correlation coefficient of each single battery in different preset frequency ranges according to the impedance and the impedance mean value includes:

[0018] For each single cell and each preset frequency segment, according to the impedance and the impedance mean of the single cell, calculate the impedance covariance between the single cell and each remaining single cell in the preset frequency segment, and the impedance standard deviation of the single cell;

[0019] The correlation coefficient of the single cell in the preset frequency range is calculated according to the impedance standard deviation of the single cell and the impedance covariance corresponding to the single cell.

[0020] In one embodiment, determining the preset threshold of the score of the single battery according to the standard deviation and the mean value includes:

[0021] The mean value plus the standard deviation of a preset multiple is used as the upper limit of the preset threshold value of the score;

[0022] The standard deviation obtained by subtracting a preset multiple from the mean is used as the lower limit of the preset scoring threshold.

[0023] In one embodiment, scoring the single cell according to the difference ranking value of the single cell in the entire preset frequency range and the preset scoring threshold to obtain the score of the single cell includes:

[0024] If the difference ranking value of the single battery in the entire preset frequency range exceeds the preset scoring threshold, the difference between the difference ranking value and the upper and lower limits of the preset scoring threshold are calculated, and the smallest difference is added by one as the score of the single battery;

[0025] If the difference ranking value of the single cell in the entire preset frequency range does not exceed the preset score threshold, the score of the single cell is set to zero.

[0026] In one embodiment, screening out faulty single cells according to the score of each single cell includes:

[0027] Taking the score of each single battery as a data point, determining the radius and reachable distance of each data point;

[0028] For each data point, the local reachable density of the data point is calculated according to the reachable distance of the remaining data points within the range formed by the radius of the data point;

[0029] Calculate the local outlier factor of each data point according to the local reachable density of the data point and the local reachable density of the remaining data points within the range formed by the radius of the data point;

[0030] If the local outlier factor of the data point is greater than a preset value, the single cell represented by the data point is a faulty single cell.

[0031] In a second aspect, the present application also provides a battery fault detection device, comprising:

[0032] An acquisition module, used to acquire relevant data of the battery module; the relevant data includes the voltage and current of each single battery in the battery module;

[0033] A calculation module, used for calculating the impedance and the impedance mean of each single battery in the battery module in a preset frequency range according to the relevant data;

[0034] The calculation module is further used to calculate the correlation coefficient of each single battery in different preset frequency ranges according to the impedance and the impedance mean;

[0035] The calculation module is further used to calculate the difference between the mean value generated by the correlation coefficients of multiple single cells in different preset frequency bands and the correlation coefficient of each single cell, and to sort the differences to obtain the difference ranking value of each single cell in the different preset frequency bands;

[0036] The calculation module is further used to calculate the standard deviation and mean of the difference ranking values ​​of each of the single cells in the different preset frequency ranges, and determine the preset threshold of the score of the single cell according to the standard deviation and mean;

[0037] A scoring module, used to score the single cell according to the difference ranking value of the single cell in the entire preset frequency range and the preset scoring threshold, so as to obtain the score of the single cell;

[0038] The screening module is used to screen out faulty single cells according to the score of each single cell.

[0039] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0040] Acquire relevant data of the battery module; the relevant data includes the voltage and current of each single battery in the battery module;

[0041] Calculating the impedance and the impedance mean of each single battery in the battery module in a preset frequency range according to the relevant data;

[0042] Calculating the correlation coefficient of each single battery in different preset frequency ranges according to the impedance and the impedance average;

[0043] Calculating the difference between the mean value of the correlation coefficients of the plurality of single cells in different preset frequency bands and the correlation coefficient of each single cell, and sorting the difference values ​​to obtain the difference ranking value of each single cell in the different preset frequency bands;

[0044] Calculating the standard deviation and mean of the difference ranking values ​​of each of the single cells in the different preset frequency ranges, and determining the preset threshold of the score of the single cell according to the standard deviation and the mean;

[0045] Scoring the single cell according to the difference ranking value of the single cell in the entire preset frequency range and the preset scoring threshold, to obtain the score of the single cell;

[0046] According to the score of each single battery, the single battery with fault is screened out.

[0047] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the following steps are implemented:

[0048] Acquire relevant data of the battery module; the relevant data includes the voltage and current of each single battery in the battery module;

[0049] Calculating the impedance and the impedance mean of each single battery in the battery module in a preset frequency range according to the relevant data;

[0050] Calculating the correlation coefficient of each single battery in different preset frequency ranges according to the impedance and the impedance average;

[0051] Calculating the difference between the mean value of the correlation coefficients of the plurality of single cells in different preset frequency bands and the correlation coefficient of each single cell, and sorting the difference values ​​to obtain the difference ranking value of each single cell in the different preset frequency bands;

[0052] Calculating the standard deviation and mean of the difference ranking values ​​of each of the single cells in the different preset frequency ranges, and determining the preset threshold of the score of the single cell according to the standard deviation and the mean;

[0053] Scoring the single cell according to the difference ranking value of the single cell in the entire preset frequency range and the preset scoring threshold, to obtain the score of the single cell;

[0054] According to the score of each single battery, the single battery with fault is screened out.

[0055] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0056] Acquire relevant data of the battery module; the relevant data includes the voltage and current of each single battery in the battery module;

[0057] Calculating the impedance and the impedance mean of each single battery in the battery module in a preset frequency range according to the relevant data;

[0058] Calculating the correlation coefficient of each single battery in different preset frequency ranges according to the impedance and the impedance average;

[0059] Calculating the difference between the mean value of the correlation coefficients of the plurality of single cells in different preset frequency bands and the correlation coefficient of each single cell, and sorting the difference values ​​to obtain the difference ranking value of each single cell in the different preset frequency bands;

[0060] Calculating the standard deviation and mean of the difference ranking values ​​of each of the single cells in the different preset frequency ranges, and determining the preset threshold of the score of the single cell according to the standard deviation and the mean;

[0061] Scoring the single cell according to the difference ranking value of the single cell in the entire preset frequency range and the preset scoring threshold, to obtain the score of the single cell;

[0062] According to the score of each single battery, the single battery with fault is screened out.

[0063] The battery fault detection method, device, computer equipment, computer-readable storage medium and computer program product described above first obtain the voltage and current of each single cell in the battery module, and calculate the impedance of each single cell according to the voltage and current of each single cell. Calculate the impedance mean according to the impedance of each single cell. Calculate the correlation coefficient of each single cell in different preset frequency bands according to the impedance and the impedance mean of the single cell. Calculate the difference between the correlation coefficient of each single cell in different preset frequency bands and the mean of the correlation coefficients of all single cells in the corresponding preset frequency band according to the correlation coefficient. Sort the difference of the single cells in different preset frequency bands to obtain the difference ranking values ​​of the single cells in different preset frequency bands. Calculate the standard deviation and mean of the ranking values ​​in different preset frequency bands according to the difference ranking values ​​of the single cells, and then obtain the score preset threshold of the single cell according to the standard deviation and mean of the ranking values. Score the single cell according to the difference ranking value of the single cell in the entire preset frequency band and the score preset threshold, and screen out the single cell with faults in the battery module according to the score of each single cell. Therefore, the present application can process different single cells in the battery module to obtain a score for each single cell, and quickly and accurately find the single cell in the battery module that has a different change trend from other single cells, i.e., a faulty single cell, based on the score of each single cell. This can effectively reduce the waste of resources caused by the need to check each single cell one by one during the battery module fault detection process. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0065] FIG1 is a schematic diagram of a flow chart of a battery failure detection method in one embodiment;

[0066] FIG2 is a schematic flow chart of a battery failure detection method in another embodiment;

[0067] FIG3 is an impedance diagram of a single cell in a battery module according to an embodiment;

[0068] FIG4 is a distribution diagram of local outlier factors of single cells in a battery module in one embodiment;

[0069] FIG5 is a block diagram of a battery fault detection device in one embodiment;

[0070] FIG. 6 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0071] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0072] In one embodiment, as shown in FIG1 , a battery fault detection method is provided. This embodiment uses the method applied to a terminal as an example for illustration. It is understandable that the method can also be applied to a server, and can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. In this embodiment, the method includes the following steps 102 to 114:

[0073] Step 102, obtaining relevant data of the battery module; the relevant data includes the voltage and current of each single battery in the battery module.

[0074] Among them, the battery module is a unit composed of multiple single batteries connected in series or in parallel.

[0075] Specifically, the terminal can collect the voltage and current data of each single battery in the battery module through a high-frequency collector, wherein the high-frequency collector is an electronic measuring instrument mainly used for collecting data.

[0076] Step 104, calculating the impedance and the impedance mean of each single battery in the battery module in a preset frequency range according to the relevant data.

[0077] Among them, impedance hinders the current in the circuit and is a complex number consisting of a real part and an imaginary part.

[0078] In some embodiments, the terminal uses wavelet transform to process the collected voltage and current data of the single cell to obtain the waveform of the voltage and current of each single cell in the time-frequency domain. The time-frequency data of the voltage of each single cell is divided by the time-frequency data of the current to obtain the impedance of the single cell at that moment. The impedance collected at each frequency collection point of each single cell in the preset frequency range is divided by the number of collection points to obtain the impedance mean of the single cell in the preset frequency range. A preset number of frequency collection points are set in each frequency range to collect the impedance of the single cell at the frequency point.

[0079] Step 106: Calculate the correlation coefficient of each single battery in different preset frequency ranges according to the impedance and the impedance mean.

[0080] Among them, the correlation coefficient is the Pearson correlation coefficient, which is a statistical indicator for measuring the degree of linear correlation between two variables.

[0081] In some embodiments, for each single cell and each preset frequency segment, the terminal calculates the impedance standard deviation of each single cell in each preset frequency segment; calculates the impedance covariance between each single cell and each remaining single cell in each preset frequency segment. Based on the impedance standard deviation of the single cell and the impedance covariance corresponding to the single cell, the correlation coefficient of each single cell in each preset frequency segment is calculated.

[0082] The correlation coefficient of a single cell in a preset frequency band is the average value of all correlation coefficients calculated between the single cell and each of the remaining single cells in the preset frequency band.

[0083] Step 108 , calculating the difference between the mean value of the correlation coefficients of the plurality of single cells at different preset frequency bands and the correlation coefficient of each single cell, and sorting the differences to obtain a ranking value of the difference of each single cell at different preset frequency bands.

[0084] The difference between the mean value of the correlation coefficients of multiple single cells and the correlation coefficient of each single cell is the DTW (dynamic time warping) distance, which is used to measure the similarity between the two.

[0085] In some embodiments, the terminal calculates the mean value of the correlation coefficient of each single cell in the battery module in each preset frequency band, and the specific calculation formula is shown in (1), and then calculates the difference between the correlation coefficient of each single cell and the mean value of the correlation coefficient, that is, the DTW distance, and the specific calculation formula is shown in (2). The DTW distance of each single cell in each preset frequency band is sorted to obtain the difference ranking value of each single cell in different preset frequency bands.

[0086]

[0087]

[0088] in, is the correlation coefficient of the ith single cell in each preset frequency band, is the number of single cells in the battery module, and t is the frequency acquisition point in the preset frequency band.

[0089] Step 110 , calculating the standard deviation and mean of the difference ranking values ​​of each single battery in different preset frequency ranges, and determining the preset threshold of the score of the single battery according to the standard deviation and the mean.

[0090] In some embodiments, the terminal calculates the standard deviation and mean of the difference ranking values ​​of the single cells according to the difference ranking values ​​of the single cells in different preset frequency bands, and then determines the preset score threshold of each single cell according to the difference ranking value of each single cell.

[0091] Step 112, scoring the single cell according to the difference ranking value of the single cell in the entire preset frequency range and the preset scoring threshold, to obtain the score of the single cell.

[0092] The entire preset frequency band includes all preset frequency bands.

[0093] In some embodiments, the terminal recalculates the difference between the mean value of the correlation coefficients of multiple single cells in the entire preset frequency range and the correlation coefficient of each single cell in the entire preset frequency range, and reorders the differences to obtain a new difference ranking value for each single cell in the entire preset frequency range. Combine the new difference ranking value of each single cell with the score preset threshold corresponding to the single cell, score the single cell, and obtain the score of each single cell.

[0094] Step 114 , screening out faulty single cells according to the score of each single cell.

[0095] In some embodiments, the terminal uses the score of each single battery as a data point to determine the reachable distance of each data point. The local reachable density of each data point is calculated based on the reachable distance of the remaining data points in the range formed by the reachable distance of each data point. The local outlier factor of each data point is calculated based on the local reachable density of each data point and the local reachable density of the remaining data points in the range formed by the reachable distance of the data point.

[0096] Furthermore, the terminal determines whether the local outlier factor of each data point is much larger than a preset value. If the local outlier factor of the data point is close to the preset value, the single battery represented by the data point is a normal battery; if the local outlier factor of the data point is much larger than the preset value, the single battery represented by the data point is a faulty single battery.

[0097] In the above-mentioned battery fault detection method, the voltage and current of each single cell in the battery module are obtained, and the impedance of each single cell is calculated according to the voltage and current of each single cell. The impedance mean is calculated according to the impedance of each single cell. According to the impedance of the single cell and the impedance mean, the correlation coefficient of each single cell in different preset frequency bands is calculated. According to the correlation coefficient, the difference between the correlation coefficient of each single cell in different preset frequency bands and the mean of the correlation coefficients of all single cells in the corresponding frequency band is calculated. The difference values ​​of the single cells in different preset frequency bands are sorted to obtain the difference ranking values ​​of the single cells in different preset frequency bands. According to the difference ranking values ​​of the single cells, the standard deviation and mean of the ranking values ​​in different preset frequency bands are calculated, and then the score preset threshold of the single cell is obtained according to the standard deviation and mean of the ranking values. According to the difference ranking values ​​of the single cell in the entire preset frequency band and the score preset threshold, the single cell is scored, and according to the score of each single cell, the single cell with faults in the battery module is screened out. Therefore, the present application can process different single cells in the battery module to obtain a score for each single cell, and quickly and accurately find the single cell in the battery module that has a different change trend from other single cells, i.e., a faulty single cell, based on the score of each single cell. This can effectively reduce the waste of resources caused by the need to check each single cell one by one during the battery module fault detection process.

[0098] In some exemplary embodiments, the impedance and impedance mean of each single cell in the battery module in a preset frequency range are calculated based on relevant data, including: based on wavelet transform, the voltage and current of each single cell are represented by a waveform in the time-frequency domain; the time-frequency data of the voltage and current in the waveform state are compared to obtain the impedance of each single cell; based on the impedance of each single cell, the impedance mean of the single cell is obtained.

[0099] In actual implementation, the terminal will provide excitation to the battery module through the Arbin (battery test equipment) device, and use a high-frequency data collector to collect the voltage and current data of each single battery in the battery module. The optimal bandwidth and center frequency of the wavelet function are selected using the optimization algorithm. The specific expression of the wavelet function is shown in formula (3).

[0100]

[0101] in, is the frequency bandwidth, is the center frequency, t is the frequency point, a is the terminal-defined scale factor, and b is the wavelet transform point, i.e., the moment when the excitation signal changes from nothing to something.

[0102] Specifically, the optimization algorithm is used to select the optimal bandwidth and center frequency of the wavelet function. Specifically, the sparsity of the wavelet coefficients can be measured by the Shannon entropy (information entropy) of these wavelet coefficients using sparsity evaluation. The Shannon entropy under different bandwidths is calculated. When the Shannon entropy is the smallest, it means that the current parameters are most suitable. The specific expression of Shannon entropy is shown in formula (4).

[0103]

[0104] in, , is the Shannon entropy, is the probability of the wavelet coefficient; is the wavelet coefficient at the jth scale; M is the total number of wavelet coefficients.

[0105] Furthermore, the terminal passes The preset frequency range to be analyzed, i.e., the preset frequency band, is obtained. The voltage and current data are subjected to wavelet transformation within the preset frequency band to obtain the voltage and current waveforms of the single cell. The voltage time-frequency data of each single cell obtained is compared with the current time-frequency data to obtain the impedance of the single cell. The impedance collected at each frequency collection point of each single cell in the preset frequency band is divided by the number of collection points to obtain the impedance mean of the single cell in the preset frequency band. The specific calculation formula is shown in (5).

[0106]

[0107] in, is the impedance of the ith single cell at the jth frequency collection point, and n is the number of frequency collection points in the preset frequency band.

[0108] In the above embodiment, the voltage and current data of the single battery in the battery module are transformed by wavelet and expressed in the form of waveform, so that the impedance of the single battery is easier to calculate and the change trend is easier to observe.

[0109] In some exemplary embodiments, the correlation coefficient of each single cell in different preset frequency bands is calculated based on the impedance and the impedance mean, including: for each single cell and each preset frequency band, the impedance covariance between the single cell and each remaining single cell and the impedance standard deviation of the single cell in the preset frequency band are calculated based on the impedance and the impedance mean of the single cell; the correlation coefficient of the single cell in the preset frequency band is calculated based on the impedance standard deviation of the single cell and the impedance covariance corresponding to the single cell.

[0110] In some embodiments, the terminal calculates the impedance covariance between each single cell and each remaining single cell in each preset frequency band based on the impedance of each single cell at different frequency collection points in each preset frequency band and the impedance average of each single cell in the preset frequency band. The specific calculation formula is shown in (6).

[0111]

[0112] in, is the impedance of the ith single cell at the jth frequency acquisition point, is the impedance of the kth single cell at the jth frequency acquisition point, is the average impedance of the ith single cell in the preset frequency range, is the impedance mean value of the kth single cell in the preset frequency range, and n is the number of frequency collection points in the preset frequency range.

[0113] The terminal calculates the impedance standard deviation of each single cell in the preset frequency band based on the impedance of each single cell at different frequency collection points in each preset frequency band and the impedance mean of each single cell in the preset frequency band. The specific calculation formula is shown in (7).

[0114]

[0115] in, is the impedance of the ith single cell at the jth frequency acquisition point, is the impedance mean value of the ith single cell in the preset frequency range, and n is the number of frequency acquisition points in the preset frequency range.

[0116] The terminal will calculate the Pearson correlation coefficient between the single cell and each of the remaining single cells based on the impedance standard deviation of each single cell in the preset frequency band and the impedance covariance between the single cell and each of the remaining single cells. The specific calculation formula is shown in formula (8). The average of all Pearson correlation coefficients of the single cell in the preset frequency band is the Pearson correlation coefficient of the single cell.

[0117]

[0118] In the above embodiment, the terminal intuitively displays the correlation of each single battery in the battery module through the Pearson correlation coefficient, making subsequent analysis easier.

[0119] In some exemplary embodiments, the preset score threshold of the single cell is determined based on the standard deviation and the mean, including: taking the mean plus the preset multiple of the standard deviation as the upper limit of the preset score threshold; taking the mean minus the preset multiple of the standard deviation as the lower limit of the preset score threshold.

[0120] In actual implementation, the terminal will add the mean of the difference ranking values ​​of each single cell plus the standard deviation of the single cell at a preset multiple as the upper limit of the preset threshold value of the score of the single cell; and subtract the standard deviation of the single cell at a preset multiple from the mean of the difference ranking values ​​of each single cell as the lower limit of the preset threshold value of the score of the single cell.

[0121] The calculation formula for the mean is shown in (9), and the calculation formula for the standard deviation is shown in (10).

[0122]

[0123]

[0124] in, is the difference ranking value of the ith single cell in the preset frequency range, and n is the number of single cells in the battery module.

[0125] In the above embodiment, by setting the preset scoring threshold, it is possible to quickly identify which single cells deviate from the normal level.

[0126] In some exemplary embodiments, the single cell battery is scored according to the difference ranking value of the single cell battery in the entire preset frequency band and the preset score threshold to obtain the score of the single cell battery, including: if the difference ranking value of the single cell battery in the entire preset frequency band exceeds the preset score threshold, then the difference between the difference ranking value and the upper and lower limits of the preset score threshold are calculated, and the smallest difference is added by one as the score of the single cell battery; if the difference ranking value of the single cell battery in the entire preset frequency band does not exceed the preset score threshold, then the score of the single cell battery is set to zero.

[0127] In actual implementation, the terminal will determine the relationship between the difference ranking value of each single battery in the entire preset frequency range and the preset score threshold of the battery cell, and score each single battery.

[0128] Specifically, if the difference ranking value of a single cell in the entire preset frequency range exceeds the preset score threshold, the difference between the difference ranking value and the upper and lower limits of the preset score threshold are calculated, and the smallest difference is added by one as the score of the single cell. The specific score calculation formula is shown in (11).

[0129]

[0130] in, Set the upper limit of the scoring threshold. is the lower limit of the preset scoring threshold, Corresponding to the average value of single battery, is the standard deviation of the corresponding single cell.

[0131] If the difference ranking value of the single battery in the entire preset frequency range does not exceed the preset scoring threshold, the score of the single battery is set to zero.

[0132] In the above embodiment, by scoring the deviation degree of each single cell, the specific situation of each single cell is quantified, so that the subsequent data processing is simpler.

[0133] In some exemplary embodiments, faulty single cells are screened out based on the score of each single cell, including: taking the score of each single cell as a data point, determining the radius and reachable distance of each data point; for each data point, calculating the local reachable density of the data point based on the reachable distance of the remaining data points within the range formed by the radius of the data point; calculating the local outlier factor of each data point based on the local reachable density of the data point and the local reachable density of the remaining data points within the range formed by the radius of the data point; if the local outlier factor of the data point is greater than a preset value, the single cell represented by the data point is a faulty single cell.

[0134] In actual implementation, the terminal will take the score of each single battery as a data point. Given a data point k, calculate the distance from the data point to all other data points, and find the P data points closest to the data point. Let P be a parameter, which represents the number of nearest neighboring data points selected. For each data point k, the radius of the data point is the distance to the Pth nearest neighboring data point.

[0135] For each data point q within the circular range of radius R belonging to data point k, the reachable distance of data point k to data point q is defined as shown in (12).

[0136]

[0137] in, express The Pth nearest neighboring data point. is the distance from data point k to data point q, is the distance from data point q to the Pth nearest neighboring data point of k.

[0138] For each data point k, the calculation formula of the local reachable density is shown in (13).

[0139]

[0140] in, is the set of all adjacent data points within the radius of data point k.

[0141] Furthermore, the local outlier factor of data point k is calculated through the local reachable density of data point k and the local reachable density of the data points adjacent to data point k. The specific calculation formula is shown in (14).

[0142]

[0143] in, is the local reachability density of the neighboring data points q of data point k, is the local reachable density of data point k, is the set of all adjacent data points within the radius of data point k.

[0144] Furthermore, if the local outlier factor of data point k is much larger than the preset value, indicating that its density is much lower than the density of its adjacent data points, the single cell represented by the data point k is a faulty single cell; if the local outlier factor of data point k is close to the preset value, it means that the single cell represented by the data point k is a normal single cell.

[0145] In the above embodiment, by calculating the local outlier factor of each single cell, the single cell with a fault can be directly determined, making the detection process more simple and accurate.

[0146] To explain the battery fault detection method in the present application in detail, an example is given below. The specific flow chart is shown in Figure 2. By way of example, the present application is described with a specific lithium battery fault detection method.

[0147] First, the terminal will provide excitation to the battery module through the Arbin (battery test equipment) device. The excitation is a square wave with an amplitude of 0.1C, a period of 15s, and a duty cycle of 1 / 3. A high-frequency data collector is used to collect the voltage and current data of each single cell in the battery module. The optimization algorithm is used to select the optimal bandwidth and center frequency of the wavelet function. The specific optimization algorithm is: The optimization algorithm is used to select the optimal bandwidth and center frequency of the wavelet function. The specific optimization algorithm is: Using sparsity evaluation, the sparsity of the wavelet coefficients can be measured by the Shannon entropy (information entropy) of these wavelet coefficients, and the Shannon entropy under different bandwidths is calculated. When the Shannon entropy is the smallest, it means that the current parameters are most suitable. The specific expression of Shannon entropy is shown in formula (4). The specific expression of the wavelet function is shown in formula (3).

[0148] Furthermore, the terminal passes The preset frequency range to be analyzed, i.e., the preset frequency band, is obtained. The voltage and current data are subjected to wavelet transformation within the preset frequency band to obtain the voltage and current waveforms of the single cell. The voltage time-frequency data of each single cell is compared with the current time-frequency data to obtain the impedance of the single cell. The specific impedance is shown in FIG3 . The impedance collected at each frequency collection point of each single cell in the preset frequency band is divided by the number of collection points to obtain the impedance mean of the single cell in the preset frequency band. The specific calculation formula is shown in (5).

[0149] The terminal will calculate the impedance covariance between each single cell and each remaining single cell in each preset frequency band according to the impedance of each single cell at different frequency collection points in each preset frequency band and the average impedance of each single cell in the preset frequency band. The specific calculation formula is shown in (6). The impedance standard deviation of each single cell in the preset frequency band is calculated according to the impedance of each single cell at different frequency collection points in each preset frequency band and the average impedance of each single cell in the preset frequency band. The specific calculation formula is shown in (7).

[0150] Then, according to the impedance standard deviation of each single cell and the impedance covariance between the single cell and each remaining single cell in the preset frequency range, the Pearson correlation coefficient between the single cell and each remaining single cell is calculated. The specific calculation formula is shown in formula (8). The average of all Pearson correlation coefficients of the single cell in the preset frequency range is the Pearson correlation coefficient of the single cell.

[0151] The terminal will calculate the mean value of the correlation coefficient of each single cell in the battery module in each preset frequency band. The specific calculation formula is shown in (1). Then, the difference between the correlation coefficient of each single cell and the mean value of the correlation coefficient is calculated, i.e., the DTW distance. The specific calculation formula is shown in (2). The DTW distance of each single cell in each preset frequency band is sorted to obtain the difference ranking value of each single cell in different preset frequency bands.

[0152] The terminal will add the mean of the difference ranking values ​​of each single cell plus the standard deviation of the single cell at a preset multiple as the upper limit of the preset threshold of the score of the single cell; and subtract the standard deviation of the single cell at a preset multiple from the mean of the difference ranking values ​​of each single cell as the lower limit of the preset threshold of the score of the single cell. The calculation formula of the mean is shown in (9), and the calculation formula of the standard deviation is shown in (10).

[0153] The terminal will determine the relationship between the difference ranking value of each single battery in the entire preset frequency range and the preset score threshold of the battery cell, and score each single battery.

[0154] Specifically, if the difference ranking value of the single cell in the entire preset frequency range exceeds the preset scoring threshold, the difference between the difference ranking value and the upper and lower limits of the preset scoring threshold are calculated, and the smallest difference is added by one as the score of the single cell. The specific scoring calculation formula is shown in (11). If the difference ranking value of the single cell in the entire preset frequency range does not exceed the preset scoring threshold, the score of the single cell is set to zero.

[0155] The terminal will take the score of each single battery as a data point. Given a data point k, calculate the distance from the data point to all other data points, and find the P data points closest to the data point. Let P be a parameter, which represents the number of nearest neighboring data points selected. For each data point k, the radius of the data point is the distance to the Pth nearest neighboring data point.

[0156] For each data point q within the circular range of radius R belonging to data point k, the reachable distance of data point k to data point q is defined as shown in (12). For each data point k, the calculation formula of the local reachable density is shown in (13). Furthermore, the local outlier factor of data point k is calculated by the local reachable density of data point k and the local reachable density of the data points adjacent to data point k. The specific calculation formula is shown in (14).

[0157] Furthermore, if the local outlier factor of data point k is much larger than 1, indicating that its density is much lower than the density of its adjacent data points, the single cell represented by data point k is a faulty single cell; if the local outlier factor of data point k is close to 1, it means that the single cell represented by data point k is a normal single cell. The distribution diagram of the local outlier factor of the specific single cell is shown in FIG4 .

[0158] The present application proposes a battery fault detection method, which compares the impedance changes of each single cell in the battery module during the charging process. It can diagnose whether the single cell is abnormal by monitoring the differences in the local outlier factors of each single cell. Compared with the existing technology, the data required for the fault detection technology of the present application is relatively simple and the detection is highly efficient, reducing the consumption of resources during the detection process.

[0159] The present invention provides a battery fault detection method. The present invention is applicable to various battery types and different working conditions, and these advantages are particularly useful for electric vehicles, so that the faulty single battery in the battery module can be detected quickly and accurately.

[0160] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0161] Based on the same inventive concept, the embodiment of the present application also provides a battery fault detection device for implementing the battery fault detection method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more battery fault detection device embodiments provided below can refer to the limitations of the battery fault detection method above, and will not be repeated here.

[0162] In an exemplary embodiment, as shown in FIG5 , a battery fault detection device is provided, including: an acquisition module 501, a calculation module 502, a scoring module 503 and a screening module 504, wherein:

[0163] The acquisition module is used to acquire relevant data of the battery module; the relevant data includes the voltage and current of each single battery in the battery module.

[0164] The calculation module is used to calculate the impedance and the impedance mean value of each single battery in the battery module in a preset frequency range according to the relevant data.

[0165] The calculation module is further used to calculate the correlation coefficient of each single battery in different preset frequency ranges according to the impedance and the impedance average.

[0166] The calculation module is further used to calculate the difference between the mean value generated by the correlation coefficients of multiple single cells in different preset frequency bands and the correlation coefficient of each single cell, and sort the differences to obtain the difference ranking value of each single cell in the different preset frequency bands.

[0167] The calculation module is further used to calculate the standard deviation and mean of the difference ranking values ​​of each of the single cells in the different preset frequency bands, and determine the preset score threshold of the single cell according to the standard deviation and mean.

[0168] A scoring module, used to score the single cell according to the difference ranking value of the single cell in the entire preset frequency range and the preset scoring threshold, so as to obtain the score of the single cell;

[0169] The screening module is used to screen out faulty single cells according to the score of each single cell.

[0170] In some embodiments, the above-mentioned calculation module is also used to represent the voltage and current of each single cell using a waveform in the time-frequency domain based on wavelet transform; compare the time-frequency data of the voltage and current in the waveform state to obtain the impedance of each single cell; and obtain the impedance mean of the single cell according to the impedance of each single cell.

[0171] In some embodiments, the above-mentioned calculation module is also used to calculate, for each single cell and each preset frequency band, the impedance covariance between the single cell and each remaining single cell in the preset frequency band, and the impedance standard deviation of the single cell according to the impedance and the impedance mean of the single cell; and calculate the correlation coefficient of the single cell in the preset frequency band according to the impedance standard deviation of the single cell and the impedance covariance corresponding to the single cell.

[0172] In some embodiments, the device also includes a determination module for using the mean plus a preset multiple of the standard deviation as the upper limit of the preset score threshold; and using the mean minus the preset multiple of the standard deviation as the lower limit of the preset score threshold.

[0173] In some embodiments, the above-mentioned scoring module is also used to calculate the difference between the difference ranking value and the upper and lower limits of the preset scoring threshold respectively if the difference ranking value of the single cell in the entire preset frequency band exceeds the preset scoring threshold, and add one to the smallest difference as the score of the single cell; if the difference ranking value of the single cell in the entire preset frequency band does not exceed the preset scoring threshold, the score of the single cell is set to zero.

[0174] In some embodiments, the above-mentioned screening module is also used to take the score of each single cell as a data point to determine the radius and reachable distance of each data point; for each data point, the local reachable density of the data point is calculated according to the reachable distance of the remaining data points within the range formed by the radius of the data point; the local outlier factor of each data point is calculated according to the local reachable density of the data point and the local reachable density of the remaining data points within the range formed by the radius of the data point; if the local outlier factor of the data point is greater than a preset value, the single cell represented by the data point is a faulty single cell.

[0175] Each module in the above battery fault detection device can be implemented in whole or in part by software, hardware or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each module.

[0176] In an exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be shown in FIG6. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit, and an input device. The processor, the memory, and the input / output interface are connected via a system bus, and the communication interface, the display unit, and the input device are connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner may be implemented through WIFI, a mobile cellular network, near field communication (NFC) or other technologies. When the computer program is executed by the processor, a battery fault detection method is implemented. The display unit of the computer device is used to form a visually visible picture, which may be a display screen, a projection device, or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.

[0177] Those skilled in the art will understand that the structure shown in FIG. 6 is merely a block diagram of a partial structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0178] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0179] Acquire relevant data of the battery module; the relevant data includes the voltage and current of each single battery in the battery module;

[0180] Calculating the impedance and the impedance mean of each single battery in the battery module in a preset frequency range according to the relevant data;

[0181] Calculating the correlation coefficient of each single battery in different preset frequency ranges according to the impedance and the impedance average;

[0182] Calculating the difference between the mean value of the correlation coefficients of the plurality of single cells in different preset frequency bands and the correlation coefficient of each single cell, and sorting the difference values ​​to obtain the difference ranking value of each single cell in the different preset frequency bands;

[0183] Calculating the standard deviation and mean of the difference ranking values ​​of each of the single cells in the different preset frequency ranges, and determining the preset threshold of the score of the single cell according to the standard deviation and the mean;

[0184] Scoring the single cell according to the difference ranking value of the single cell in the entire preset frequency range and the preset scoring threshold, to obtain the score of the single cell;

[0185] According to the score of each single battery, the single battery with fault is screened out.

[0186] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0187] Acquire relevant data of the battery module; the relevant data includes the voltage and current of each single battery in the battery module;

[0188] Calculating the impedance and the impedance mean of each single battery in the battery module in a preset frequency range according to the relevant data;

[0189] Calculating the correlation coefficient of each single battery in different preset frequency ranges according to the impedance and the impedance average;

[0190] Calculating the difference between the mean value of the correlation coefficients of the plurality of single cells in different preset frequency bands and the correlation coefficient of each single cell, and sorting the difference values ​​to obtain the difference ranking value of each single cell in the different preset frequency bands;

[0191] Calculating the standard deviation and mean of the difference ranking values ​​of each of the single cells in the different preset frequency ranges, and determining the preset threshold of the score of the single cell according to the standard deviation and the mean;

[0192] Scoring the single cell according to the difference ranking value of the single cell in the entire preset frequency range and the preset scoring threshold, to obtain the score of the single cell;

[0193] According to the score of each single battery, the single battery with fault is screened out.

[0194] In one embodiment, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the following steps:

[0195] Acquire relevant data of the battery module; the relevant data includes the voltage and current of each single battery in the battery module;

[0196] Calculating the impedance and the impedance mean of each single battery in the battery module in a preset frequency range according to the relevant data;

[0197] Calculating the correlation coefficient of each single battery in different preset frequency ranges according to the impedance and the impedance average;

[0198] Calculating the difference between the mean value of the correlation coefficients of the plurality of single cells in different preset frequency bands and the correlation coefficient of each single cell, and sorting the difference values ​​to obtain the difference ranking value of each single cell in the different preset frequency bands;

[0199] Calculating the standard deviation and mean of the difference ranking values ​​of each of the single cells in the different preset frequency ranges, and determining the preset threshold of the score of the single cell according to the standard deviation and the mean;

[0200] Scoring the single cell according to the difference ranking value of the single cell in the entire preset frequency range and the preset scoring threshold, to obtain the score of the single cell;

[0201] According to the score of each single battery, the single battery with fault is screened out.

[0202] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0203] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0204] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0205] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A battery failure detection method, characterized in that: The method comprises: Acquire relevant data of the battery module; the relevant data includes the voltage and current of each single battery in the battery module; Calculating the impedance and the impedance mean of each single battery in the battery module in a preset frequency range according to the relevant data; Calculating the correlation coefficient of each single battery in different preset frequency ranges according to the impedance and the impedance average; Calculating the difference between the mean value of the correlation coefficients of the plurality of single cells in different preset frequency bands and the correlation coefficient of each single cell, and sorting the difference values ​​to obtain the difference ranking value of each single cell in the different preset frequency bands; Calculating the standard deviation and mean of the difference ranking values ​​of each of the single cells in the different preset frequency ranges, and determining the preset threshold of the score of the single cell according to the standard deviation and the mean; Scoring the single cell according to the difference ranking value of the single cell in the entire preset frequency range and the preset scoring threshold, to obtain the score of the single cell; According to the score of each single battery, the single battery with fault is screened out.

2. The method according to claim 1, characterized in that The step of calculating the impedance and the impedance mean of each single battery in the battery module in a preset frequency range according to the relevant data includes: Based on wavelet transform, the voltage and current of each single battery are represented by waveforms in the time-frequency domain; Comparing the time-frequency data of the voltage and current in the waveform state to obtain the impedance of each single battery; According to the impedance of each single battery, an impedance average of the single battery is obtained.

3. The method according to claim 1, characterized in that The step of calculating the correlation coefficient of each single battery in different preset frequency ranges according to the impedance and the impedance mean value includes: For each single cell and each preset frequency segment, according to the impedance and the impedance mean of the single cell, calculate the impedance covariance between the single cell and each remaining single cell in the preset frequency segment, and the impedance standard deviation of the single cell; The correlation coefficient of the single cell in the preset frequency range is calculated according to the impedance standard deviation of the single cell and the impedance covariance corresponding to the single cell.

4. The method according to claim 1, characterized in that The step of determining the preset threshold value of the score of the single battery according to the standard deviation and the mean value includes: The mean value plus the standard deviation of a preset multiple is used as the upper limit of the preset threshold value of the score; The standard deviation obtained by subtracting a preset multiple from the mean is used as the lower limit of the preset scoring threshold.

5. The method according to claim 1, characterized in that Scoring the single cell according to the difference ranking value of the single cell in the entire preset frequency range and the preset scoring threshold to obtain the score of the single cell includes: If the difference ranking value of the single battery in the entire preset frequency range exceeds the preset scoring threshold, the difference between the difference ranking value and the upper and lower limits of the preset scoring threshold are calculated, and the smallest difference is added by one as the score of the single battery; If the difference ranking value of the single cell in the entire preset frequency range does not exceed the preset score threshold, the score of the single cell is set to zero.

6. The method according to claim 1, characterized in that The step of screening out faulty single cells according to the score of each single cell comprises: Taking the score of each single battery as a data point, determining the radius and reachable distance of each data point; For each data point, the local reachable density of the data point is calculated according to the reachable distance of the remaining data points within the range formed by the radius of the data point; Calculate the local outlier factor of each data point according to the local reachable density of the data point and the local reachable density of the remaining data points within the range formed by the radius of the data point; If the local outlier factor of the data point is greater than a preset value, the single cell represented by the data point is a faulty single cell.

7. A battery fault detection device, characterized in that: The device comprises: An acquisition module, used to acquire relevant data of the battery module; the relevant data includes the voltage and current of each single battery in the battery module; A calculation module, used for calculating the impedance and the impedance mean of each single battery in the battery module in a preset frequency range according to the relevant data; The calculation module is further used to calculate the correlation coefficient of each single battery in different preset frequency ranges according to the impedance and the impedance mean; The calculation module is further used to calculate the difference between the mean value generated by the correlation coefficients of multiple single cells in different preset frequency bands and the correlation coefficient of each single cell, and to sort the differences to obtain the difference ranking value of each single cell in the different preset frequency bands; The calculation module is further used to calculate the standard deviation and mean of the difference ranking values ​​of each of the single cells in the different preset frequency ranges, and determine the preset threshold of the score of the single cell according to the standard deviation and mean; A scoring module, used to score the single cell according to the difference ranking value of the single cell in the entire preset frequency range and the preset scoring threshold, so as to obtain the score of the single cell; The screening module is used to screen out faulty single cells according to the score of each single cell.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images