An entropy value algorithm-based power battery voltage fault online diagnosis processing method

Abstract

The present application relates to the technical field of battery fault diagnosis, in particular to a power battery voltage fault online diagnosis processing method based on entropy value algorithm, comprising obtaining time sequence values of each single body voltage of the power battery of the electric vehicle to be diagnosed, establishing a voltage data matrix, sliding from top to bottom through the voltage matrix according to a pre-set constant length sliding window, performing kick-out and reconstruction on the data in each sliding window according to the data processing principle, then calculating the entropy value of each single body voltage, performing Z-score conversion on the calculation result to obtain an abnormal evaluation coefficient, identifying abnormal values of the coefficient, if there are abnormal values, positioning the abnormal single body number and time, and evaluating the abnormal degree, at the same time, conveying corresponding prompts to the driver based on the safety management strategy, if there are no abnormal values, the time window continues to slide down, and the above process is repeated for iterative calculation, the present application can accurately locate the single body voltage fault occurrence time and position, and improve the safety of the electric vehicle to a certain extent.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of battery fault diagnosis, and particularly relates to a power battery voltage fault online diagnosis processing method based on an entropy value algorithm. BACKGROUND

[0002] In recent years, the electric vehicle industry has developed rapidly, and vehicle safety accidents have been common. The power battery is considered to be the culprit, so the power battery fault diagnosis of electric vehicles is of great significance. Most of the faults of the power battery system can be traced back to one or more specific single batteries. Single battery faults mainly include overvoltage, undervoltage, short circuit, overtemperature, and poor consistency. These faults can be reflected in the single battery voltage parameter in time, and mainly manifest as inconsistent fluctuations in voltage. Therefore, fast and accurate monitoring and evaluation of single battery voltage abnormal fluctuations can more comprehensively reflect the safety of the power battery system. Various entropy methods can effectively evaluate the disorder of the system and are widely used in the field of battery fault diagnosis, but the precision of the model needs to be sacrificed for calculation efficiency.

[0003] Based on this, the present application provides a power battery voltage fault online diagnosis processing method based on an entropy value algorithm. SUMMARY

[0004] In order to make up for the shortcomings of the prior art and solve the above technical problems, the present application provides a power battery voltage fault online diagnosis processing method based on an entropy value algorithm.

[0005] A power battery voltage fault online diagnosis processing method based on an entropy value algorithm, the method comprising the following steps:

[0006] Step 100: Obtain the time series value of each single battery voltage data in the automobile battery pack to be diagnosed, and construct a voltage data matrix A, wherein:

[0007]

[0008] Wherein, m represents the time series length, n represents the total number of single batteries in the battery pack, a i,j represents the voltage value of the jth single battery at the ith index, i=1,2,…,m;j=1,2,…,n;

[0009] Step 101: Pre-set the length and width of the initial sliding time window, and use the sliding window to cut the data in the voltage data matrix region by region to form a cut voltage matrix B;

[0010] Step 102: According to the data processing principle, the voltage values in the window are kicked out and reconstructed to form a new voltage matrix D;

[0011] Step 103: Calculate the Shannon entropy value of each single battery based on the reconstructed data using the optimized entropy value algorithm;

[0012] Step 104: Convert each single Shannon entropy value based on the optimized Z-score method to obtain the single voltage abnormal fluctuation evaluation coefficient;

[0013] Step 105: Identify abnormal single based on safety evaluation strategy;

[0014] Step 106: Determine whether there is an abnormal single, if so, execute step 107, otherwise execute step 108;

[0015] Step 107: If there is an abnormal value, locate the position and occurrence time of the abnormal single, and further determine the abnormal degree, and send the corresponding prompt to the driver;

[0016] Step 108: If there is no abnormal value, the time window moves down, and steps 101-106 are repeated for the next iteration calculation.

[0017] Preferably, the preset initial sliding time window length in step 101 is k, and the width is the total number of single batteries n, and the voltage data matrix B taken by each sliding window is:

[0018]

[0019] Wherein

[0020] Preferably, the method for processing the voltage value data in the window in step 102 is specifically:

[0021] S200: Real-time monitor the voltage difference of each single battery at continuous time, that is, calculate the voltage difference matrix C of each single battery voltage data at continuous time in the time window matrix B, wherein:

[0022]

[0023] Wherein c i,j represents the difference between the voltage value of the jth single battery at the i+1th index and the voltage value at the ith index, i=1,2,…,k-1; j=1,2,…,n;

[0024] S201: Find all data rows whose single battery voltage difference is less than or equal to 0.001V and continuous more than k / 10 times in matrix C;

[0025] S202: Kick out the data segment corresponding to the row index of matrix B that meets the condition of S201;

[0026] S203: sequentially splice the remaining data in time sequence, and take the average of the two end points as the splicing point to form a new voltage matrix D, wherein:

[0027]

[0028] wherein

[0029] Preferably, the specific steps of calculating the Shannon entropy value of each single battery based on the reconstructed data matrix D in step 103 are as follows:

[0030] S300: find the maximum and minimum values in the time window, i.e. find the minimum value d min and the maximum value d max in the matrix D as follows:

[0031]

[0032] S301: divide the region (d min , d max ) formed by the two maximum and minimum values into l intervals:

[0033]

[0034] wherein L = 1, 2, …, l.

[0035] S302: calculate the frequency matrix F of each single battery voltage data falling into the above l intervals in the matrix D, to obtain the frequency matrix F, wherein:

[0036]

[0037] wherein f i,j represents the number of all data of the single battery j in the matrix D falling into the interval ;

[0038] S303: divide the frequency by the total number to obtain the region probability matrix P of each single data in each region, wherein

[0039]

[0040] wherein represents the probability of the voltage data of the single battery j in the matrix D falling into the i-th interval, and the Shannon entropy of each single battery calculated by the matrix B is:

[0041] H(B) = [H1, H2, L, H j , L, H n ]

[0042] S304: according to Calculate the Shannon entropy value of each monomer, that is, the Shannon entropy value calculated by the time window B of the jth monomer battery.

[0043] Preferably, the monomer voltage abnormal fluctuation evaluation coefficient is obtained by converting the monomer Shannon entropy value based on the optimized Z-score method in step 104. The specific steps are as follows:

[0044] S400: Obtain the Shannon entropy sequence H(B) of each monomer battery,

[0045] S401: Calculate the mean value μ of the Shannon entropy of each monomer battery H , and the calculation formula is:

[0046]

[0047] S402: Remove the data exceeding 2*μ H from the remaining monomer battery Shannon entropy sequence; that is, remove the Shannon entropy value exceeding 2*μ H from the Shannon entropy sequence H(B) to form a new Shannon entropy sequence H(B)' = [H1, H2, L, H g ]; Calculate the mean value and standard deviation of the sequence H(B)', and the calculation formula is:

[0048]

[0049] S403: Calculate the median value H me of the Shannon entropy of each monomer battery; that is, calculate the median value H me of the Shannon entropy sequence H(B) of each monomer battery;

[0050] S404: Standardize the Shannon entropy value of each monomer battery according to ; that is, calculate the abnormal evaluation coefficient of each monomer battery in the current time window according to the Z-score theory.

[0051] Preferably, the abnormal value safety evaluation strategy in step 105 is:

[0052] S500: Identify the monomer battery with |AF| of 3.5 in each time window iteration calculation as a fault monomer with voltage abnormal fluctuation;

[0053] S501: If |AF|≤3.5, execute S502 and S503;

[0054] S502: No voltage fault alarm;

[0055] S503: Perform the next iteration calculation;

[0056] S504: if 3.5<|AF|≤4, then S505 and S506 are executed, or whether S507 is judged to execute S510, or S503;

[0057] S505: voltage secondary fault alarm;

[0058] S506: calculate secondary fault duration;

[0059] S507: whether to exceed threshold;

[0060] S508: if 4<|AF|, then S509 and S510 are executed;

[0061] S509: voltage primary fault alarm;

[0062] S510: the vehicle needs to be checked immediately.

[0063] Preferably, in step 108, if there is no abnormal value, the time window is moved down once, and steps 101 to 106 are repeated for the next iteration calculation.

[0064] A power battery voltage fault online diagnosis processing device based on an entropy value algorithm, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the program:

[0065] Obtain the time sequence value of each single battery voltage data in the automobile battery pack to be diagnosed, and construct a voltage data matrix A;

[0066] Step 101: preset the length and width of the initial sliding time window, and use the sliding window to cut the voltage data matrix in each region to form a cut voltage matrix B;

[0067] Step 102: according to the data processing principle, the voltage values in the window are kicked out and reconstructed to form a new voltage matrix D;

[0068] Step 103: based on the reconstructed data, the Shannon entropy value of each single battery is calculated by using the optimized entropy value algorithm;

[0069] Step 104: based on the optimized Z-score method, the single voltage abnormal fluctuation evaluation coefficient is obtained by transforming the single Shannon entropy value;

[0070] Step 105: based on the safety evaluation strategy, the abnormal single body is identified;

[0071] Step 106: determine whether there is an abnormal single body, if there is, execute step 107, otherwise execute step 108;

[0072] Step 107: If there is an abnormal value, the position and occurrence time of the abnormal monomer are located, and the abnormal degree is further determined, and a corresponding prompt is sent to the driver;

[0073] Step 108: If there is no abnormal value, the time window is moved down, and steps 101-106 are repeated for the next iteration calculation.

[0074] A computer readable storage medium, having stored thereon a computer program, the program being executed by a processor to implement the following steps:

[0075] Step 100: Obtain the time series value of each monomer battery voltage data in the automobile battery pack to be diagnosed, and construct a voltage data matrix A;

[0076] Step 101: The length and width of the initial sliding time window are preset, and the voltage data matrix is cut off region by region using a sliding window to form a cut-off voltage matrix B;

[0077] Step 102: According to the data processing principle, the voltage values in the window are kicked out and reconstructed to form a new voltage matrix D;

[0078] Step 103: Based on the reconstructed data, the Shannon entropy value of each monomer battery is calculated using an optimized entropy value algorithm;

[0079] Step 104: Based on the optimized Z-score method, the monomer voltage abnormal fluctuation evaluation coefficient is obtained by transforming the Shannon entropy value of each monomer;

[0080] Step 105: Abnormal monomer identification based on safety evaluation strategy;

[0081] Step 106: Determine whether there is an abnormal monomer, if there is, execute step 107, otherwise execute step 108;

[0082] Step 107: If there is an abnormal value, the position and occurrence time of the abnormal monomer are located, and the abnormal degree is further determined, and a corresponding prompt is sent to the driver;

[0083] Step 108: If there is no abnormal value, the time window is moved down, and steps 101-106 are repeated for the next iteration calculation.

[0084] The beneficial effects of the present application are as follows:

[0085] 1. Compared with the existing battery fault diagnosis method based on Shannon entropy, the beneficial effects of the present application are: since the method adopts an entropy value calculation method based on interval probability, and the interval changes with the change of voltage data, the size of Shannon entropy is extremely susceptible to multiple voltage data segments with long duration and no fluctuation, thereby the misdiagnosis rate is high, based on this, the present application increases the steps of kicking out and reconstructing the original data, and the data processing principle ensures that the fault data with no abnormal fluctuation in the kicked-out data itself, and the data segment that is easy to cause misdiagnosis is kicked out, thereby greatly improving the fault diagnosis accuracy;

[0086] The mean and standard deviation in the original Z-score method are easily affected by the extreme value data during calculation, in order to further improve the accuracy of the method in identifying abnormal monomers, the mean value is replaced by the median value in the standardization process of Shannon entropy, and the abnormal value is removed when calculating the standard deviation, thereby further reducing the misdiagnosis rate;

[0087] The present application uses the principle of sliding time window, and locates the position of abnormal monomers and the time of abnormal occurrence by calculating the voltage data in the sliding window, thereby reducing the calculation amount, and realizing online application, and the method is different from the existing fault diagnosis method of calculating the data of the whole life cycle. BRIEF DESCRIPTION OF DRAWINGS

[0088] The present application will be further described below in combination with the drawings.

[0089] Figure 1 A power battery voltage fault online diagnosis processing method flowchart based on the entropy value algorithm provided by the present application is shown in the figure.

[0090] Figure 2 A data kicking-out and reconstruction principle flowchart provided by the present application is shown in the figure.

[0091] Figure 3 An optimized entropy value algorithm flowchart provided by the present application is shown in the figure.

[0092] Figure 4 An optimized Z-score algorithm flowchart provided by the present application is shown in the figure.

[0093] Figure 5 An abnormal value diagnosis processing flowchart provided by the present application is shown in the figure. DETAILED DESCRIPTION

[0094] The technical scheme of the present application will be described in detail below in combination with the drawings and specific embodiments, and it should be understood that the embodiments and specific features in the embodiments of the present application are detailed descriptions of the technical scheme of the present application, and are not limitations of the technical scheme of the present application, and the technical features in the embodiments and the embodiments of the present application can be combined with each other without conflict.

[0095] An entropy value algorithm-based power battery voltage fault online diagnosis processing method, the method comprises the following steps:

[0096] Step 100: Obtain the time series value of each single battery voltage data in the automobile battery pack to be diagnosed, and construct a voltage data matrix A, wherein:

[0097]

[0098] Wherein, m represents the time series length, n represents the total number of single batteries in the battery pack, a i,j represents the voltage value of the jth single battery at the ith index, i=1,2,…,m; j=1,2,…,n;

[0099] Step 101: The length and width of the initial sliding time window are preset, and the voltage data matrix is cut off in each region by using the sliding window to form the cut-off voltage matrix B;

[0100] Step 102: According to the data processing principle, the voltage value in the window is kicked out and reconstructed to form a new voltage matrix D;

[0101] Step 103: Based on the reconstructed data, the Shannon entropy value of each single battery is calculated by using the optimized entropy value algorithm;

[0102] Step 104: Based on the optimized Z-score method, the single voltage abnormal fluctuation evaluation coefficient is obtained by transforming the single Shannon entropy value;

[0103] Step 105: Based on the safety evaluation strategy, the abnormal single is identified;

[0104] Step 106: Determine whether there is an abnormal single, if there is, execute step 107, otherwise execute step 108;

[0105] Step 107: If there is an abnormal value, the position and occurrence time of the abnormal single are located, and the abnormal degree is further determined, and the corresponding prompt is sent to the driver;

[0106] Step 108: If there is no abnormal value, the time window moves down, and steps 101-106 are repeated for the next iteration calculation.

[0107] In step 101, the length of the initial sliding time window is k, and the width is the total number of single batteries n. The voltage data matrix cut off by the sliding window each time is B, wherein:

[0108]

[0109] Wherein

[0110] The method for processing the voltage value data in the window in step 102 is specifically:

[0111] S200: Real-time monitoring of voltage differences of each single battery at continuous time points, that is, calculating a voltage difference matrix C of each single battery voltage data at continuous time points in the time window matrix B, wherein:

[0112]

[0113] Wherein c i,j represents the difference between the voltage value of the jth single battery at the i+1th index and the voltage value at the ith index, i = 1, 2, …, k-1; j = 1, 2, …, n;

[0114] S201: Finding all data rows in matrix C whose single battery voltage difference is less than or equal to 0.001V for more than k / 10 times in succession;

[0115] S202: Kicking out the data segment corresponding to the row index of matrix B that meets the condition of S201;

[0116] S203: Splicing the remaining data in sequence, taking the average of the two end points as the splicing point, to form a new voltage matrix D, wherein:

[0117]

[0118] Wherein

[0119] The specific steps of calculating the Shannon entropy value of each single battery based on the reconstructed data matrix D in step 103 are:

[0120] S300: Finding the maximum and minimum values in the time window, that is, finding the minimum value d min and the maximum value d max in matrix D as:

[0121]

[0122] S301: Dividing the region (d min , d max ) formed by the two extreme values into l intervals:

[0123]

[0124] Wherein L = 1, 2, …, l.

[0125] S302: Calculating the frequency matrix F of each single battery voltage data falling into the above l intervals in matrix D, to obtain the frequency matrix F, wherein:

[0126]

[0127] Where f i,j This indicates that all data of individual cell j in matrix D fall within the interval The number of;

[0128] S303: Dividing the frequency by the total number yields the regional probability matrix P for each individual data point in each region, where...

[0129]

[0130] in Let represent the probability that the voltage data of cell j in matrix D falls into the i-th interval, and let be the Shannon entropy of each cell calculated from matrix B:

[0131] H(B) = [H1,H2,L,H] j ,L,H n ]

[0132] S304: According to Calculate the Shannon entropy value for each cell, which represents the Shannon entropy value of the j-th cell calculated by time window B.

[0133] In step 104, the Shannon entropy values ​​of each individual cell are transformed using the optimized Z-score method to obtain the individual cell voltage anomaly fluctuation assessment coefficients. The specific steps are as follows:

[0134] S400: Obtain the Shannon entropy sequence H(B) for each individual battery cell.

[0135] S401: Calculate the mean Shannon entropy μ of each individual cell. H The calculation formula is:

[0136]

[0137] S402: Kick out H(B) containing more than 2*μ H After analyzing the data, the standard deviation of the remaining single-cell Shannon entropy sequence is calculated; that is, the Shannon entropy sequence H(B) exceeding 2*μ is removed. H The Shannon entropy values ​​are used to form a new Shannon entropy sequence H(B)'=[H1,H2,L,H g ], The mean and standard deviation of sequence H(B)' are calculated using the following formulas:

[0138]

[0139] S403: Calculate the median H of the Shannon entropy for each individual cell. me That is, calculate the median H of the Shannon entropy sequence H(B) for each individual cell. me ;

[0140] S404: According to Standardize the Shannon entropy value of each single battery; that is, calculate the abnormal evaluation coefficient of each single battery in the current time window according to the Z-score theory.

[0141] The safety evaluation strategy of the abnormal value in step 105 is:

[0142] S500: Identify the single battery with 3.5<|AF| in the iterative calculation of each time window as a fault single battery with voltage abnormal fluctuation;

[0143] S501: If |AF|≤3.5, execute S502 and S503;

[0144] S502: No voltage fault alarm;

[0145] S503: Perform the next iteration calculation;

[0146] S504: If 3.5<|AF|≤4, execute S505 and S506, or judge whether to execute S510 according to S507, or S503;

[0147] S505: Voltage secondary fault alarm;

[0148] S506: Calculate the secondary fault duration;

[0149] S507: Whether to exceed the threshold;

[0150] S508: If 4<|AF|, execute S509 and S510;

[0151] S509: Voltage primary fault alarm;

[0152] S510: The vehicle needs to be checked immediately;

[0153] In step 108, if there is no abnormal value, the time window is moved down once, and steps 101 to 106 are repeated to perform the next iteration calculation.

[0154] Specifically, assuming that there are 95 single batteries in a battery box of an electric vehicle to be detected, 200 sampling data are selected, and a time series matrix A of all single batteries is established as follows:

[0155]

[0156] Wherein a i,j represents the voltage value of the jth single battery at the ith index, i=1, 2,…, 200; j=1, 2,…, 95;

[0157] In step S101, the length of the preset initial sliding window is 100 and the width is 95, the voltage time sequence matrix is intercepted by the sliding window in each region, and the voltage data matrix B intercepted for the first time is:

[0158]

[0159] Based on the above voltage data matrix B, step S102 is performed according to the flowchart shown in FIG. 2, that is, the data processing principle is specifically as follows: Figure 2

[0160] In step S200, the voltage difference of each single battery at continuous time is monitored in real time, that is, the voltage difference matrix C of the voltage data of each single battery at continuous time in the time sliding window matrix B is calculated:

[0161]

[0162] In step S201, the row index of all data in which the voltage difference of each single battery in the matrix C is less than or equal to 0.001 V and is continuous for more than 10 times is found out;

[0163] In step S202, the data segment corresponding to the row index meeting the condition of step S201 in the matrix B is kicked out;

[0164] In step S203, the remaining data is sequentially spliced in time sequence, the average value of the two end point data is taken as the splicing point, and a new voltage matrix D is formed:

[0165]

[0166] Wherein

[0167] Step S103 is performed according to the flowchart shown in FIG. 3, that is, the optimized entropy value algorithm is specifically as follows: Figure 3

[0168] In step S300, the minimum value d min and the maximum value d max in the matrix D are calculated as follows:

[0169]

[0170] In step S301, the region (d min , d max ) formed by the two extreme values is divided into 10 intervals:

[0171]

[0172] Wherein L = 1, 2, …, 10.

[0173] ​​In step S302, the frequency of each single battery voltage data falling into the above 10 intervals is calculated to obtain matrix F:

[0174]

[0175] wherein f i,j represents the number of data of single battery j in matrix D falling into interval .

[0176] In step S303, the probability matrix P is calculated according to matrix F:

[0177]

[0178] wherein represents the probability of voltage data of single battery j in matrix D falling into the i-th interval.

[0179] In step S304, the Shannon entropy of each single battery calculated from matrix B is:

[0180] H(B)=[H1,H2,L,H j ,L,H 95 ]

[0181] wherein represents the Shannon entropy value of the j-th single battery in the time window B.

[0182] Step S104 is performed according to the flowchart shown in Figure 4 , that is, the optimized Z-score algorithm is:

[0183] In step S401, the mean value of the Shannon entropy sequence H(B) obtained in step S400 is calculated, and the calculation formula is:

[0184]

[0185] In step S402, the Shannon entropy values exceeding 2*μ H in the Shannon entropy sequence H(B) are kicked out to form a new Shannon entropy sequence H(B)'=[H1,H2,L,H g ], The mean value and the standard deviation of the sequence H(B)' are calculated, and the calculation formulas are respectively:

[0186]

[0187] In step S403, the median value H me of the Shannon entropy sequence is calculated.

[0188] In step S404, the abnormal evaluation coefficient of each single battery in the current time window is calculated according to the Z-score theory, and the calculation formula is:

[0189]

[0190] Finally, steps S105, S106, S107 and S108 are performed according to the flowchart shown in the figure, that is, when the power battery voltage fails, that is, an abnormal value occurs, the safety evaluation strategy is as follows: Figure 5

[0191] In step S501, based on a large amount of real failure and normal vehicle data analysis and verification, the single battery of 3.5<|AF| is identified as a fault single battery with abnormal voltage fluctuation.

[0192] In step S502, if |AF|≤3.5, there is no voltage abnormal fluctuation fault in the single battery, and the system does not alarm voltage failure, and step S503 will be performed, that is, the next iteration calculation is performed, and the second intercepted voltage data matrix B is:

[0193]

[0194] Further, steps S102-S106 are repeated.

[0195] In step S504, if 3.5<|AF|≤4, the system issues a voltage secondary failure alarm, and in step S506, the secondary failure alarm duration is calculated, and if it exceeds the threshold, step S510 is executed, otherwise step S503 is executed.

[0196] In step S508, if 4<|AF|, the system issues a voltage primary failure alarm, prompting the driver to stop the vehicle immediately and check.

[0197] The above process exemplifies the complete calculation process when the sliding window intercepts the data once, that is, the power battery voltage fault online diagnosis processing method based on the entropy value algorithm provided by the embodiment of the application, which can realize real-time and accurate evaluation of voltage abnormal fluctuation fault by repeatedly performing the above steps for iteration calculation when implemented online.

[0198] Based on the same inventive concept as the power battery voltage fault online diagnosis processing method based on the entropy value algorithm in the foregoing embodiment, the present application also provides a power battery voltage fault online diagnosis processing device based on the entropy value algorithm, which stores a computer program. The program is executed by the processor to realize the steps of any method of the foregoing power battery voltage fault online diagnosis processing method based on the entropy value algorithm, specifically as follows:

[0199] ​It includes a bus architecture, the bus can include any number of interconnected buses and bridges, the bus links various circuits including one or more processors represented by a processor and a memory represented by a memory together. The bus can also connect various other circuits such as peripheral devices, voltage stabilizers, sensors, cameras and power management circuits together, which are well known in the art, therefore, the present application will not be further described.

[0200] Embodiment three

[0201] Based on the same inventive concept as the aforementioned embodiment, the present application also provides a computer readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the following steps:

[0202] Obtain the time series value of each single battery voltage data in the automobile battery pack to be diagnosed, and construct a voltage data matrix A;

[0203] Step 101: preset the length and width of the initial sliding time window, and use the sliding window to cut the data of the voltage data matrix in each region to form a cut voltage matrix B;

[0204] Step 102: according to the data processing principle, the voltage value in the window is kicked out and reconstructed to form a new voltage matrix D;

[0205] Step 103: based on the reconstructed data, the Shannon entropy value of each single battery is calculated by using the optimized entropy value algorithm;

[0206] Step 104: based on the optimized Z-score method, the single voltage abnormal fluctuation evaluation coefficient is obtained by transforming the single Shannon entropy value;

[0207] Step 105: based on the safety evaluation strategy, the abnormal single is identified;

[0208] Step 106: determine whether there is an abnormal single, if there is, execute step 107, otherwise execute step 108;

[0209] Step 107: if there is an abnormal value, the position and occurrence time of the abnormal single are located, and the abnormal degree is further determined, and the corresponding prompt is sent to the driver;

[0210] Step 108: if there is no abnormal value, the time window moves down, and steps 101-106 are repeated for the next iteration calculation.

[0211] A computer readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the following steps:

[0212] Step 100: Obtain the time series value of each single battery voltage data in the automobile battery pack to be diagnosed, and construct a voltage data matrix A;

[0213] Step 101: Pre-set the length and width of the initial sliding time window, and use the sliding window to cut the voltage data matrix in each region to form a cut voltage matrix B;

[0214] Step 102: According to the data processing principle, the voltage values in the window are kicked out and reconstructed to form a new voltage matrix D;

[0215] Step 103: Based on the reconstructed data, the Shannon entropy value of each single battery is calculated by using the optimized entropy value algorithm;

[0216] Step 104: Based on the optimized Z-score method, the single voltage abnormal fluctuation evaluation coefficient is obtained by transforming the single Shannon entropy value;

[0217] Step 105: Based on the safety evaluation strategy, the abnormal single is identified;

[0218] Step 106: Determine whether there is an abnormal single, if there is, execute step 107, otherwise execute step 108;

[0219] Step 107: If there is an abnormal value, the position and occurrence time of the abnormal single are located, and the abnormal degree is further determined, and the corresponding prompt is sent to the driver;

[0220] Step 108: If there is no abnormal value, the time window is moved down, and steps 101-106 are repeated for the next iteration calculation.

[0221] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer usable program code.

[0222] The present application is described with reference to flowcharts and / or block diagrams according to the method, device (system) and computer program product of the embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of the flows and / or blocks in the flowchart and / or block diagram can be realized by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device realize the functions described in the flowchart and / or block diagram.Figure 1 apparatuses that implement the functions specified in the flowchart Figure 1 block or blocks.

[0223] These computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture including instructions which implement the flowchart Figure 1 apparatuses that implement the functions specified in the flowchart Figure 1 block or blocks.

[0224] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions that execute on the computer or other programmable apparatus provide steps for implementing the flowchart Figure 1 apparatuses that implement the functions specified in the flowchart Figure 1 block or blocks.

[0225] Obviously, numerous modifications and variations of the present application are possible in light of the above teachings. It is therefore to be understood that within the scope of the apparatuses that implement the functions specified in the flowchart

Claims

1. A power battery voltage fault online diagnosis processing method based on an entropy value algorithm, characterized in that: The method comprises the following steps: Step 100: Obtain the time sequence value of the voltage data of each single battery in the automobile battery pack to be diagnosed, and construct a voltage data matrix A, wherein: ， wherein, denotes the length of the time sequence, denotes the total number of single batteries in the battery pack, denotes the voltage value of the th single battery at the th index, =1,2,…, ; =1,2,…, ; Step 101: The length and width of the initial sliding time window are preset, and the voltage data matrix is cut off region by region by using the sliding window to form a cut-off voltage matrix B; Step 102: data kick-out and reconstruction of the voltage values in the window according to data processing principles to form a new voltage matrix D ; wherein the method for processing the voltage value data in the window in step 102 is: S200: Real-time monitoring of the voltage difference of each single battery at continuous time, calculating the voltage difference matrix C of each single battery voltage data at continuous time in the time window matrix B; S201: Find all data rows whose voltage difference is less than or equal to 0.001V for more than k / 10 times continuously in matrix C; S202: Kick out the data segment corresponding to the row index in matrix B that meets the condition of S201; S203: sequentially splice the remaining data in time sequence, take the mean of the two end point data as the splicing point, and form a new voltage matrix D ; Step 103: Calculate the Shannon entropy value of each single battery based on the reconstructed data using the optimized entropy value algorithm; In step 103, the data matrix is reconstructed based on the data matrix D The step of calculating the Shannon entropy value of each single battery by using the optimized entropy value algorithm is as follows: S300: Find the maximum and minimum values in the time window, find the minimum value d D in the matrix min and the maximum value d max ; S301: divide the region (d min , d max ) composed of the two extreme values into intervals S302: Calculate the matrix D The frequency matrix F is obtained by counting the frequency of each monomer cell voltage data falling into the above interval. S303: The frequency is divided by the total number to obtain the regional probability matrix P of each single data in each region; S304: According to Calculate the Shannon entropy value for each monomer, where , representing a matrix D Medium single cell battery j The voltage data falls into the first i The probability of each interval; Step 104: The single battery voltage abnormal fluctuation evaluation coefficient is obtained by transforming the single battery Shannon entropy value based on the optimized Z-score method; wherein the specific steps of transforming the single battery Shannon entropy value based on the optimized Z-score method in step 104 to obtain the single battery voltage abnormal fluctuation evaluation coefficient are as follows: S400: Obtain the shannon entropy sequence of each single battery , S401: Calculate the mean value of the Shannon entropy of each single battery ; S402: kick out more than calculating the standard deviation of the sequence of the remaining monomer battery shannon entropy after the data; S403: Calculate the median of the Shannon entropy of each single battery ; S404: According to standardizing the Shannon entropy value of each monomer battery; Step 105: Abnormal single battery identification based on safety evaluation strategy; Step 106: Determine whether there is an abnormal single battery, if yes, execute step 107, otherwise execute step 108; Step 107: If there is an abnormal value, locate the position and occurrence time of the abnormal single battery, and further determine the abnormal degree, and send the corresponding prompt to the driver; Step 108: If there is no abnormal value, move the time window down, and repeat steps 101-106 for the next iteration calculation.

2. The processing method for online diagnosis of voltage fault of power battery based on entropy value algorithm according to claim 1, characterized in that: The preset initial sliding time window length in step 101 is , the width is the same as the total number of single batteries , and the voltage data matrix taken by each sliding window is B, wherein: ， wherein wherein, denotes the matrix the length of the time series, denotes the length of the time series of the matrix A, denotes the width of the total number of monoblocs, .

3. The method according to claim 2, wherein the method is characterized in that: The method for processing the voltage value data in the window in step 102 is as follows: S200: Real-time monitoring of the voltage difference of each single battery at continuous time, calculating the voltage difference matrix C of each single battery voltage data at continuous time in the time window matrix B, wherein: ， wherein represents the voltage value of the th single cell at the th index and the voltage value of the th index, = 1, 2, …, ; = 1, 2, …, ; S201: Find out the single battery voltage difference in matrix C is less than or equal to 0.001V continuously more than Row index of all data more than 10 S202: Kick out the data segment corresponding to the row index in matrix B that meets the condition of S201; S203: sequentially splice the remaining data in time sequence, take the mean of the two end point data as the splicing point, and form a new voltage matrix D wherein: ， wherein wherein, denotes the matrix D the timing length of the matrix denotes the matrix the timing length of the matrix is the total number of monoblocs, .

4. The method according to claim 3, characterized in that: Step 103 of reconstructing the data matrix D The specific steps of calculating the Shannon entropy value of each single battery by using the optimized entropy value algorithm are as follows: S300: Find the maximum and minimum values in the time window, i.e. find the minimum and maximum values in the matrix D and d max are:​ ， S301: divide the region (d min , d max ) composed of the two extreme values into intervals ， wherein = 1, 2,..., ; S302: Calculate the matrix D The frequency matrix F is obtained, wherein each monomer cell voltage data falls into the above interval, and the frequency matrix F is obtained, wherein: ， wherein denotes a matrix D monocell number of all data of the monocell falling into the interval of the interval S303: The frequency is divided by the total number to obtain the regional probability matrix P of each single data in each region, wherein ， wherein represents a matrix D The probability that the voltage data of the single battery in the first interval, and the Shannon entropy of each single battery calculated by the matrix B is: ， S304: According to Calculate the monomer Shannon entropy value, that is, the Shannon entropy value calculated by the time window B for the jth monomer battery.

5. The entropy value algorithm-based online diagnosis processing method for voltage fault of power battery according to claim 4, characterized in that: The specific steps of transforming the single battery Shannon entropy value based on the optimized Z-score method in step 104 to obtain the single battery voltage abnormal fluctuation evaluation coefficient are as follows: S400: Obtain the shannon entropy sequence of each single battery , S401: Calculate the mean value of the Shannon entropy of each single battery The calculation formula is: ， S402: kick out the monomer battery Shannon entropy sequence remaining after calculating the standard deviation of the data; that is, kick out the Shannon entropy value exceeding in the Shannon entropy sequence H(B) to form a new Shannon entropy sequence , , Calculate the mean and standard deviation of the sequence​ ， ， S403: Calculate the median value H of the Shannon entropy of each single battery me ; that is, calculate the median value of the sequence H(B) of the Shannon entropy of each single battery ; S404: According to Standardize the Shannon entropy value of each single battery; that is, calculate the abnormal evaluation coefficient of each single battery in the current time window according to the Z-score theory.

6. The processing method of claim 5, wherein the processing method is characterized in that: The safety evaluation strategy of the abnormal value in step 105 is as follows: S500: Identify the single battery with 3.5<|AF| in each time window iteration calculation as a fault single battery with voltage abnormal fluctuation; S501: If |AF|≤3.5, execute S502 and S503; S502: No voltage fault alarm; S503: Perform the next iteration calculation; S504: If 3.5<|AF|≤4, execute S505 and S506, or judge whether to execute S510 according to S507, or S503; S505: Voltage secondary fault alarm; S506: Calculate the duration of the secondary fault; S507: Whether to exceed the threshold; S508: If 4<|AF|, S509 and S510 are executed; S509: Voltage primary fault alarm; S510: The vehicle needs to be checked immediately.

7. The processing method of claim 6, wherein the processing method further comprises: In step 108, if there is no abnormal value, the time window is moved down once, and steps 101-106 are repeated for the next iteration calculation.

8. A power battery voltage fault online diagnosis processing device based on an entropy value algorithm, applied to the power battery voltage fault online diagnosis processing method based on the entropy value algorithm in claim 1, characterized in that: The computer program product comprises a memory, a processor and a computer program stored in the memory and executable on the processor, and the processor implements the following steps when executing the program: Obtain the time sequence values of the voltage data of each single battery in the automobile battery pack to be diagnosed, and construct a voltage data matrix A; Step 101: preset the length and width of the initial sliding time window, and use the sliding window to cut the voltage data matrix region by region to form a cut voltage matrix B; Step 102: data kick-out and reconstruction of the voltage values in the window according to the data processing principle, forming a new voltage matrix D ; Step 103: calculate the Shannon entropy value of each single battery based on the reconstructed data using the optimized entropy value algorithm; Step 104: transform the single battery Shannon entropy value based on the optimized Z-score method to obtain the single battery voltage abnormal fluctuation evaluation coefficient; Step 105: identify the abnormal single battery based on the safety evaluation strategy; Step 106: determine whether there is an abnormal single battery, if there is, execute step 107, otherwise execute step 108; Step 107: if there is an abnormal value, locate the position and occurrence time of the abnormal single battery, further determine the abnormal degree, and send a corresponding prompt to the driver; Step 108: if there is no abnormal value, move the time window down, repeat steps 101-106 for the next iteration calculation.

9. A computer-readable storage medium, applied to the online diagnosis processing method of power battery voltage fault based on entropy value algorithm in claim 1, characterized in that: The computer program product comprises a memory, a processor and a computer program stored in the memory and executable on the processor, and the processor implements the following steps when executing the program: Step 100: obtain the time sequence values of the voltage data of each single battery in the automobile battery pack to be diagnosed, and construct a voltage data matrix A; Step 101: preset the length and width of the initial sliding time window, and use the sliding window to cut the voltage data matrix region by region to form a cut voltage matrix B; Step 102: data kick-out and reconstruction of the voltage values in the window according to the data processing principle, forming a new voltage matrix D ; Step 103: calculate the Shannon entropy value of each single battery based on the reconstructed data using the optimized entropy value algorithm; Step 104: transform the single battery Shannon entropy value based on the optimized Z-score method to obtain the single battery voltage abnormal fluctuation evaluation coefficient; Step 105: identify the abnormal single battery based on the safety evaluation strategy; Step 106: determine whether there is an abnormal single battery, if there is, execute step 107, otherwise execute step 108; Step 107: if there is an abnormal value, locate the position and occurrence time of the abnormal single battery, further determine the abnormal degree, and send a corresponding prompt to the driver; Step 108: if there is no abnormal value, move the time window down, repeat steps 101-106 for the next iteration calculation.

Images