Battery pack fault diagnosis method and system based on feature decomposition and data masking

Abstract

The present invention discloses a battery pack fault diagnosis method and system based on eigendecomposition and data masking; the method comprises: obtaining voltage detection quantity by using redundant cross measurement topology; decoupling the voltage detection quantity according to a set time window to obtain a two-dimensional analysis matrix; standardizing the two-dimensional analysis matrix to obtain a standardized matrix; obtaining a covariance matrix according to the standardized matrix; performing eigenvalue decomposition on the covariance matrix to obtain a maximum eigenvalue and an eigenvector corresponding to the maximum eigenvalue; using the maximum eigenvalue as a fault indicator of the battery pack, and using the maximum eigenvalue as a fault indicator of the battery pack to determine whether a fault occurs in the battery pack; and screening elements with a decreasing trend in the eigenvector corresponding to the maximum eigenvalue to determine whether a voltage sensor fault, a short circuit fault, and an electrical connection fault occur in the battery pack.

Description

Technical Field

[0001] The present invention relates to the technical field of battery pack fault diagnosis, and in particular to a battery pack fault diagnosis method and system based on feature decomposition and data masking. Background Art

[0002] The statements in this section merely mention background art related to the present invention and do not necessarily constitute prior art.

[0003] The renewable energy power generation industry has developed rapidly because it can reduce carbon emissions. However, renewable energy power generation methods such as wind power, photovoltaics, and solar thermal power generation have strong randomness, volatility, and intermittency, which leads to insufficient power grid absorption, seriously restricting the development of the renewable energy power generation industry. In order to improve the power grid's absorption capacity for renewable energy power generation, electrochemical energy storage systems have been widely used and have rapidly developed into the core support of the renewable energy power generation industry.

[0004] Lithium-ion batteries have been widely used in electrochemical energy storage systems due to their high energy density, long cycle life, and low self-discharge rate. However, the safety management methods for lithium-ion batteries are far from mature, the battery system fault diagnosis effect is poor, and there is a lack of fast and effective short circuit and connection fault diagnosis methods, which leads to safety accidents such as battery fires and explosions in energy storage systems. Therefore, designing an early fault diagnosis method for battery systems is an urgent problem to be solved.

[0005] Common battery failures include short circuit failures, electrical connection failures, and voltage sensor failures. Voltage sensor failures include "large detection errors" and "unchanged detection values". Early short circuit failures are usually caused by the growth of lithium dendrites inside the battery, which manifests as reduced battery voltage, reduced available capacity, and severe battery heating. Electrical connection failures are usually caused by oxidation of the battery connector surface or loose installation positions, which manifests as small battery voltage measurements and severe heating of the connection parts. Failures with large sensor detection errors are usually caused by harsh temperatures and electromagnetic environments, which manifest as increased voltage detection errors and large abnormal fluctuations. Failures with unchanged sensor detection values ​​are usually caused by chip crashes caused by harsh environments, which manifests as the sensor measurement value remaining the same as the last measured value and no longer changing. Summary of the invention

[0006] In order to solve the deficiencies of the prior art, the present invention provides a battery pack fault diagnosis method and system based on feature decomposition and data masking;

[0007] On the one hand, a battery pack fault diagnosis method based on eigendecomposition and data masking is provided;

[0008] The battery pack fault diagnosis method based on feature decomposition and data masking includes:

[0009] A redundant cross-measurement topology is used to obtain voltage detection quantities, wherein the voltage detection quantities include: the voltages of all battery cells themselves and the voltages of all battery assemblies; the battery assembly refers to a battery cell and an adjacent previous resistor, and a battery cell and an adjacent next resistor;

[0010] According to the set time window, the voltage detection quantity is decoupled to obtain a two-dimensional analysis matrix; the two-dimensional analysis matrix is ​​standardized to obtain a standardized matrix; according to the standardized matrix, a covariance matrix is ​​obtained; the covariance matrix is ​​subjected to eigenvalue decomposition to obtain the maximum eigenvalue and the eigenvector corresponding to the maximum eigenvalue;

[0011] The maximum characteristic value is used as a fault indicator of the battery pack to determine whether the battery pack fails;

[0012] In the eigenvector corresponding to the maximum eigenvalue, elements with a decreasing trend are screened. If only the i-th element passes the screening, it means that the i-th voltage sensor fails; if the 2i-1-th element and the 2i-th element pass the screening, it means that the i-th battery cell has a short circuit fault; if the 2i-th element and the 2i+1-th element pass the screening, it means that an electrical connection fault occurs between the i-th battery cell and the i+1-th battery cell, where i is a positive integer greater than zero and represents the sequence number of the element.

[0013] On the other hand, a battery pack fault diagnosis system based on eigendecomposition and data masking is provided;

[0014] The battery pack fault diagnosis system based on feature decomposition and data masking includes:

[0015] An acquisition module is configured to: acquire voltage detection quantities using a redundant cross measurement topology; the voltage detection quantities include: voltages of all battery cells themselves and voltages of all battery assemblies; the battery assembly refers to a battery cell and an adjacent previous resistor, and a battery cell and an adjacent next resistor;

[0016] The processing module is configured to: decouple the voltage detection quantity according to the set time window to obtain a two-dimensional analysis matrix; perform standardization processing on the two-dimensional analysis matrix to obtain a standardized matrix; obtain a covariance matrix according to the standardized matrix; perform eigenvalue decomposition on the covariance matrix to obtain a maximum eigenvalue and an eigenvector corresponding to the maximum eigenvalue;

[0017] A first judgment module is configured to: use the maximum characteristic value as a fault indicator of the battery pack to determine whether the battery pack fails;

[0018] The second judgment module is configured as follows: in the eigenvector corresponding to the maximum eigenvalue, elements with a decreasing trend are screened; if only the i-th element passes the screening, it means that the i-th voltage sensor fails; if the 2i-1-th element and the 2i-th element pass the screening, it means that the i-th battery cell has a short circuit failure; if the 2i-th element and the 2i+1-th element pass the screening, it means that an electrical connection failure occurs between the i-th battery cell and the i+1-th battery cell, wherein i is a positive integer greater than zero, and i represents the sequence number of the element.

[0019] On the other hand, there is also provided an electronic device, comprising:

[0020] a memory for non-transitory storage of computer-readable instructions; and

[0021] a processor for executing the computer readable instructions,

[0022] When the computer-readable instructions are executed by the processor, the method described in the first aspect is executed.

[0023] On the other hand, a storage medium is provided, which non-temporarily stores computer-readable instructions, wherein when the non-temporary computer-readable instructions are executed by a computer, the instructions of the method described in the first aspect are executed.

[0024] On the other hand, a computer program product is provided, comprising a computer program, wherein the computer program is used to implement the method described in the first aspect when running on one or more processors.

[0025] The above technical solution has the following advantages or beneficial effects:

[0026] The present invention uses a redundant interleaved measurement topology to detect the voltage values ​​of the battery and the connection point, and obtains the fault type and fault location by characteristic decomposing the measured values, thereby completing the fault differentiation and diagnosis. The advantages are as follows:

[0027] (1) The present invention can be implemented using common measurement circuits or front-end chips in engineering, has no special requirements on voltage sampling speed, and has low implementation cost;

[0028] (2) The present invention adopts a redundant measurement topology, and the cell voltage of each battery cell is collected by two adjacent voltage sensors, thereby improving the voltage sampling accuracy and robustness of the battery management system;

[0029] (3) The characteristic decomposition method used in the present invention has inherent orthogonality and is insensitive to the battery voltage change caused by the current, but is sensitive to the uncoordinated change components therein, which can accurately detect battery faults while reducing the false alarm rate;

[0030] (4) The present invention establishes a data masking mechanism based on a time window. The fault indicator will respond immediately when a fault occurs, and then recover after a certain period of time. At this time, the method can detect new faults and avoid the fault detection channel being blocked by old faults, affecting the timeliness of fault diagnosis and avoiding missed faults. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0032] Figure 1 The redundant cross-measurement topology of the first embodiment;

[0033] Figure 2 This is a schematic diagram of a voltage sensor circuit according to Embodiment 1;

[0034] Figure 3 This is a schematic diagram of a data mask in Embodiment 1;

[0035] Figure 4 This is a fault diagnosis flow chart of Embodiment 1;

[0036] Figure 5 C of Example 1 k The maximum eigenvalue (2# battery short circuit);

[0037] Figure 6 The rules of the elements of the corresponding characteristic vectors of Example 1 (2# battery short circuit);

[0038] Figure 7 C of Example 1 k The maximum characteristic value of (electrical connection failure between 2# and 3# batteries);

[0039] Figure 8 The rules of the elements of the corresponding characteristic vectors of Example 1 (electrical connection failure between 2# and 3# batteries);

[0040] Fig. 9 C of Example 1 k The maximum characteristic value of (3# voltage sensor failure);

[0041] Fig.10 Element rules of the corresponding characteristic vectors in Example 1 (3# voltage sensor failure). DETAILED DESCRIPTION

[0042] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0043] The present invention belongs to the field of battery management technology and discloses a battery system early fault diagnosis method, which can detect, distinguish and locate whether a battery system has a battery short circuit fault, an electrical connection fault and a voltage sensor fault. The method mainly comprises the following steps: using a redundant cross measurement topology to detect the battery cell voltage and the voltage of the connection resistor; according to a specified time window, the voltage detection amount is clipped to obtain X k Matrix; for X k Perform feature decomposition to obtain the fault indicator D k If D k If the value is lower than the alarm threshold, it means that the system has the above faults. According to the feature decomposition results, analyze X k The eigenvector corresponding to the maximum eigenvalue of is used to obtain the fault type and fault location according to the element position that meets the fault characteristics.

[0044] Embodiment 1

[0045] This embodiment provides a battery pack fault diagnosis method based on feature decomposition and data masking;

[0046] The battery pack fault diagnosis method based on feature decomposition and data masking includes:

[0047] S101: using a redundant cross measurement topology to obtain voltage detection quantities; the voltage detection quantities include: voltages of all battery cells themselves and voltages of all battery assemblies; the battery assembly refers to a battery cell and an adjacent previous resistor, and a battery cell and an adjacent next resistor;

[0048] S102: Decoupling the voltage detection quantity according to the set time window to obtain a two-dimensional analysis matrix; standardizing the two-dimensional analysis matrix to obtain a standardized matrix; obtaining a covariance matrix according to the standardized matrix; performing eigenvalue decomposition on the covariance matrix to obtain a maximum eigenvalue and an eigenvector corresponding to the maximum eigenvalue;

[0049] S103: using the maximum characteristic value as a fault indicator of the battery pack to determine whether the battery pack fails;

[0050] S104: In the eigenvector corresponding to the maximum eigenvalue, elements with a decreasing trend are screened. If only the i-th element passes the screening, it means that the i-th voltage sensor fails; if the 2i-1-th element and the 2i-th element pass the screening, it means that the i-th battery cell has a short circuit failure; if the 2i-th element and the 2i+1-th element pass the screening, it means that an electrical connection failure occurs between the i-th battery cell and the i+1-th battery cell, where i is a positive integer greater than zero and represents the sequence number of the element.

[0051] Furthermore, the S101: using a redundant cross measurement topology to obtain a voltage detection value specifically includes:

[0052] The battery pack includes: a negative electrode of the battery pack, a first connection point, a resistor R 0,1 , the second connection point, the first battery cell, the third connection point, the resistor R 1,2 , 4th connection point, 2nd battery cell, 5th connection point, resistor R 2,3 , 6th connection point, 3rd battery cell, 7th connection point, resistor R 3,4 , the 8th connection point, the 4th battery cell...the n-1th battery cell, the 2n-1th connection point, the resistor R n-1,n , 2nth connection point, nth battery cell, 2n+1th connection point, resistor R n,n+1 , the 2n+2th connection point, the positive electrode of the battery pack; n is greater than or equal to 2;

[0053] A first voltage sensor is connected between the first connection point and the third connection point; a second voltage sensor is connected between the second connection point and the fourth connection point; a third voltage sensor is connected between the third connection point and the fifth connection point; a fourth voltage sensor is connected between the fourth connection point and the sixth connection point; a fifth voltage sensor is connected between the fifth connection point and the seventh connection point; a sixth voltage sensor is connected between the sixth connection point and the eighth connection point; and so on;

[0054] Connecting the 2n-1th voltage sensor between the 2n-1th connection point and the 2n+1th connection point; connecting the 2nth voltage sensor between the 2nth connection point and the 2n+2th connection point;

[0055] The output voltage of the first voltage sensor is The output voltage of the second voltage sensor is The output voltage of the third voltage sensor is The output voltage of the fourth voltage sensor is The output voltage of the fifth voltage sensor is The output voltage of the sixth voltage sensor is And so on; the output voltage of the 2n-1th voltage sensor is The output voltage of the 2nth voltage sensor is

[0056] Furthermore, the S101: using a redundant cross measurement topology to obtain a voltage detection value includes:

[0057] Get the voltage detection data vector Where n is the number of battery cells;

[0058] According to the measurement topology and Kirchhoff's voltage law, the decomposition of the measurement data vector is:

[0059]

[0060] Among them, V c is the battery terminal voltage, R n-+,n is the connection resistance between the n-1th battery cell and the nth battery cell, R 0,1 is the connection resistance of the negative terminal of the battery pack, R n,n++ is the connection resistance of the positive terminal of the battery pack.

[0061] It should be understood that the internal structures of the first battery cell, the second battery cell, the third battery cell ... the n-1th battery cell are the same as the nth battery cell, and the first battery cell includes: a plurality of batteries connected in parallel.

[0062] like Figure 1 As shown, through the redundant cross measurement topology, the voltage of the battery and the connection point is measured to obtain the measurement data vector Where N is the number of batteries connected in series. In this measurement topology, the voltage value measured by each voltage sensor is close to the single cell voltage value of the battery; the voltage sensors are divided into two groups according to the staggered principle, and the voltage detected by each group of voltage sensors meets the characteristics of end-to-end connection. Therefore, this topology can use the single cell voltage measurement circuit or front-end chip commonly used in engineering to measure the battery voltage, such as Figure 2 shown.

[0063] like Figure 2 As shown, the battery pack includes: a negative electrode of the battery pack, a first connection point, a resistor R 0,1 , the second connection point, the first battery cell BAT1, the third connection point, the resistor R 1,2 , the 4th connection point, the second battery cell BAT2, the 5th connection point, the resistor R 2,3 , the 6th connection point, the 3rd battery cell BAT3, the 7th connection point, the resistor R 3,4 , 8th connection point, 4th battery cell BAT4, 9th connection point, resistor R 4,5 , 10th connection point, 4th battery cell BAT5, 11th connection point, resistor R 5,6 , the 12th connection point and the positive terminal of the battery pack;

[0064] The first connection point is connected to the port No. 1 of the CON2 chip through the first connection of the second group of voltage sensors; the second connection point is connected to the port No. 2 of the CON2 chip through the first connection of the first group of voltage sensors; the third connection point is connected to the port No. 3 of the CON2 chip through the second connection of the second group of voltage sensors;

[0065] The fourth connection point is connected to the No. 4 port of the CON2 chip through the second connection of the first group of voltage sensors; the fifth connection point is connected to the No. 5 port of the CON2 chip through the third connection of the second group of voltage sensors; the sixth connection point is connected to the No. 6 port of the CON2 chip through the third connection of the first group of voltage sensors;

[0066] The seventh connection point is connected to the No. 7 port of the CON2 chip through the fourth connection line of the second group of voltage sensors; the eighth connection point is connected to the No. 8 port of the CON2 chip through the fourth connection line of the first group of voltage sensors; the ninth connection point is connected to the No. 9 port of the CON2 chip through the fifth connection line of the second group of voltage sensors;

[0067] The 10th connection point is connected to the 10th port of the CON2 chip through the 5th connection of the 1st group of voltage sensors; the 11th connection point is connected to the 11th port of the CON2 chip through the 6th connection of the 2nd group of voltage sensors; the 12th connection point is connected to the 12th port of the CON2 chip through the 6th connection of the 1st group of voltage sensors;

[0068] Port 13 of the CON2 chip is connected to port C0 of the front-end chip No. 2; port 14 of the CON2 chip is connected to port C0 of the front-end chip No. 1; port 15 of the CON2 chip is connected to port C1 of the front-end chip No. 2; port 16 of the CON2 chip is connected to port C1 of the front-end chip No. 1;

[0069] Port 17 of the CON2 chip is connected to port C2 of the front-end chip No. 2; port 18 of the CON2 chip is connected to port C2 of the front-end chip No. 1; port 19 of the CON2 chip is connected to port C3 of the front-end chip No. 2; port 20 of the CON2 chip is connected to port C3 of the front-end chip No. 1;

[0070] Port 21 of the CON2 chip is connected to port C4 of the front-end chip No. 2; port 22 of the CON2 chip is connected to port C4 of the front-end chip No. 1; port 23 of the CON2 chip is connected to port C5 of the front-end chip No. 2; port 24 of the CON2 chip is connected to port C5 of the front-end chip No. 1.

[0071] The cell voltage of each battery cell and the connection resistance at each location are included in the measurement values ​​of the two voltage sensors. Although the changes in battery voltage, current and connection point voltage are random, their change rules are constant, which can be used to complete fault detection and differentiation.

[0072] Exemplarily, the front-end chip No. 1 and the front-end chip No. 2 may be the LT68xx series of ADI and the MC33771 series of NXP.

[0073] Furthermore, if Figure 3 As shown, S102: decoupling the voltage detection quantity according to the set time window to obtain a two-dimensional analysis matrix, which specifically includes:

[0074] Create a time window with a width of w and widen V in the time dimension s ;

[0075] Set the moving step of the time window; the time window is the data mask;

[0076] The time window is moved from left to right on the voltage detection data vector according to the set moving step size. Each time it slides, the data in the time window is organized into a two-dimensional analysis matrix X k ;

[0077]

[0078] Among them, x k,i Represents the measurement value of the i-th sensor at the k-th moment in the time window.

[0079] Incorporating masking mechanisms into data analysis (e.g. Figure 3 ), the analysis process is separated from the V outside the time window s Data decoupling, finally abstracting the two-dimensional analysis matrix X k .

[0080] Further, S102: performing standardization processing on the two-dimensional analysis matrix to obtain a standardized matrix, specifically including:

[0081] Simplify formula (1) to:

[0082] x=As1+Bs2;

[0083] in:

[0084]

[0085] Combined with formula (2), the vector x is expanded in the time dimension to obtain the two-dimensional analysis matrix X k :

[0086] X k =S 1,k A T +S 2,k B T ;

[0087] in:

[0088]

[0089] Represents the voltage of the i-th battery at the k-th moment in the time window.

[0090]

[0091] R i,i++,k Represents the equivalent connection resistance between the i-th battery and the i+1-th battery at the k-th moment in the time window. R 0,1 Represents the total negative equivalent connection resistance, R n,n+1 Represents the total positive equivalent connection resistance.

[0092] X k Standardize and get the standardized matrix

[0093]

[0094]

[0095]

[0096] Where μ0 is X k The mean vector of and are the mean vectors of s1 and s2 respectively, Yes X k The standard deviation diagonal matrix of

[0097] Further, S102: according to the standardized matrix, a covariance matrix is ​​obtained, which specifically includes:

[0098]

[0099] Among them, C k represents the covariance matrix.

[0100] Further, S102: performing eigenvalue decomposition on the covariance matrix to obtain the maximum eigenvalue and the eigenvector corresponding to the maximum eigenvalue, specifically including:

[0101] And the covariance matrix C k Perform eigenvalue decomposition and get:

[0102]

[0103] Among them, Λ k =diag{λ 1,k ,λ 2,k ,…,λ n-1,k} is Ck Eigenvalues ​​in decreasing order, V k is the eigenvector in the corresponding order.

[0104] Further, S103: using the maximum characteristic value as a fault indicator of the battery pack to determine whether the battery pack fails, specifically including: if the maximum characteristic value is less than a set threshold, it indicates that the battery pack has not failed, otherwise it indicates that the battery pack has failed.

[0105] It should be understood that C k The maximum eigenvalue of reflects the coordination of each measured voltage change. When this value decreases, it means that there are uncoordinated components in the voltage change law, which are usually caused by battery failure. Therefore, use C k The maximum eigenvalue of is taken as the fault indicator of the battery system and is denoted as D k =max 1≤j≤n-1 |λ j,k |=λ 1,k .

[0106] It should be understood that the step S104: selecting elements with a decreasing trend in the eigenvector corresponding to the maximum eigenvalue, k In the eigenvector corresponding to the maximum eigenvalue of X, each element represents k The synergy between each voltage and other voltages in the filter is analyzed, and the elements with decreasing trends are screened and their serial numbers are checked.

[0107] Through the above steps, the detection, differentiation and location of battery system battery short circuit fault, electrical connection fault and voltage sensor fault can be completed. The flow chart is as follows: Figure 4 shown.

[0108] According to the experimental results, when the second battery has a short circuit fault, C k The maximum eigenvalue of Figure 5 As shown, the corresponding elements of the eigenvector are Figure 6 As shown in the figure; when there is an electrical connection failure between the second and third batteries, C k The maximum eigenvalue variation law of Figure 7 As shown, the corresponding elements of the eigenvector are Figure 8 As shown; when the third voltage sensor fails, C k The maximum eigenvalue variation law of Fig. 9 As shown, the corresponding elements of the eigenvector are Fig.10 shown.

[0109] Embodiment 2

[0110] This embodiment provides a battery pack fault diagnosis system based on feature decomposition and data masking;

[0111] The battery pack fault diagnosis system based on feature decomposition and data masking includes:

[0112] An acquisition module is configured to: acquire voltage detection quantities using a redundant cross measurement topology; the voltage detection quantities include: voltages of all battery cells themselves and voltages of all battery assemblies; the battery assembly refers to a battery cell and an adjacent previous resistor, and a battery cell and an adjacent next resistor;

[0113] The processing module is configured to: decouple the voltage detection quantity according to the set time window to obtain a two-dimensional analysis matrix; perform standardization processing on the two-dimensional analysis matrix to obtain a standardized matrix; obtain a covariance matrix according to the standardized matrix; perform eigenvalue decomposition on the covariance matrix to obtain a maximum eigenvalue and an eigenvector corresponding to the maximum eigenvalue;

[0114] A first judgment module is configured to: use the maximum characteristic value as a fault indicator of the battery pack to determine whether the battery pack fails;

[0115] The second judgment module is configured as follows: in the eigenvector corresponding to the maximum eigenvalue, elements with a decreasing trend are screened; if only the i-th element passes the screening, it means that the i-th voltage sensor fails; if the 2i-1-th element and the 2i-th element pass the screening, it means that the i-th battery cell has a short circuit failure; if the 2i-th element and the 2i+1-th element pass the screening, it means that an electrical connection failure occurs between the i-th battery cell and the i+1-th battery cell, wherein i is a positive integer greater than zero, and i represents the sequence number of the element.

[0116] It should be noted that the acquisition module, processing module, first judgment module and second judgment module correspond to steps S101 to S104 in Embodiment 1, and the examples and application scenarios implemented by the modules and the corresponding steps are the same, but are not limited to the contents disclosed in Embodiment 1. It should be noted that the modules as part of the system can be executed in a computer system such as a set of computer executable instructions.

[0117] The description of each embodiment in the above embodiments has different emphases. For parts not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0118] The proposed system can be implemented in other ways. For example, the system embodiment described above is only illustrative, and the division of the modules is only a logical function division. In actual implementation, there may be other division methods, such as multiple modules can be combined or integrated into another system, or some features can be ignored or not executed.

[0119] Embodiment 3

[0120] This embodiment also provides an electronic device, including: one or more processors, one or more memories, and one or more computer programs; wherein the processor is connected to the memory, and the one or more computer programs are stored in the memory. When the electronic device is running, the processor executes the one or more computer programs stored in the memory so that the electronic device executes the method described in the above embodiment one.

[0121] It should be understood that in this embodiment, the processor may be a central processing unit CPU, and the processor may also be other general-purpose processors, digital signal processors DSP, application-specific integrated circuits ASIC, off-the-shelf programmable gate arrays FPGA or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0122] The memory may include a read-only memory and a random access memory, and provide instructions and data to the processor. A portion of the memory may also include a non-volatile random access memory. For example, the memory may also store information about the device type.

[0123] In the implementation process, each step of the above method can be completed by an integrated logic circuit of hardware in a processor or an instruction in the form of software.

[0124] The method in the first embodiment can be directly embodied as a hardware processor, or a combination of hardware and software modules in the processor. The software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware. To avoid repetition, it will not be described in detail here.

[0125] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with this embodiment can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0126] Embodiment 4

[0127] This embodiment further provides a computer-readable storage medium for storing computer instructions. When the computer instructions are executed by a processor, the method described in the first embodiment is completed.

[0128] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A battery pack fault diagnosis method based on feature decomposition and data masking, characterized in that: include: Use redundant cross-measurement topology to obtain voltage detection value; The voltage detection quantity includes: the voltage of all battery cells themselves and the voltage of all battery assemblies; the battery assembly refers to the battery cell and the adjacent previous resistor, and the battery cell and the adjacent next resistor; According to the set time window, the voltage detection quantity is decoupled to obtain a two-dimensional analysis matrix; the two-dimensional analysis matrix is ​​standardized to obtain a standardized matrix; according to the standardized matrix, a covariance matrix is ​​obtained; the covariance matrix is ​​subjected to eigenvalue decomposition to obtain the maximum eigenvalue and the eigenvector corresponding to the maximum eigenvalue; The maximum characteristic value is used as a fault indicator of the battery pack to determine whether the battery pack fails; In the eigenvector corresponding to the maximum eigenvalue, select the elements with decreasing trend. If only elements pass the screening, then the A voltage sensor fails; if the elements and elements pass the screening, then the A short circuit occurs in a battery cell; if the elements and elements pass the screening, then the The battery and There is an electrical connection failure between the battery cells, among which, is a positive integer greater than zero, Indicates the ordinal number of the element.

2. The battery pack fault diagnosis method based on feature decomposition and data masking as claimed in claim 1, characterized in that: Use redundant cross-measurement topology to obtain voltage detection values, including: A battery pack, comprising: a negative electrode of the battery pack, a first connection point, a resistor connected in sequence , 2nd connection point, 1st battery cell, 3rd connection point, resistor , 4th connection point, 2nd battery cell, 5th connection point, resistor , 6th connection point, 3rd battery cell, 7th connection point, resistor , 8th connection point, 4th battery cell...n-1th battery cell, 2n-1th connection point, resistor , 2nth connection point, nth battery cell, 2n+1th connection point, resistor , the 2n+2th connection point, the positive electrode of the battery pack; n is greater than or equal to 2; Connecting the 2n-1th voltage sensor between the 2n-1th connection point and the 2n+1th connection point; connecting the 2nth voltage sensor between the 2nth connection point and the 2n+2th connection point; The output voltage of the first voltage sensor is ; The output voltage of the second voltage sensor is ; The output voltage of the third voltage sensor is ; The output voltage of the fourth voltage sensor is ; The output voltage of the fifth voltage sensor is ; The output voltage of the sixth voltage sensor is ; And so on; the output voltage of the 2n-1th voltage sensor is ; The output voltage of the 2nth voltage sensor is .

3. The battery pack fault diagnosis method based on feature decomposition and data masking as claimed in claim 2, characterized in that: Use a redundant cross-measurement topology to obtain voltage detection quantities, including: Get the voltage detection data vector ,in, is the number of battery cells; According to the measurement topology and Kirchhoff's voltage law, the decomposition of the measurement data vector is: （1） in, is the battery terminal voltage, is the connection resistance between the n-1th battery cell and the nth battery cell, is the connection resistance of the negative terminal of the battery pack, is the connection resistance of the positive terminal of the battery pack.

4. The battery pack fault diagnosis method based on feature decomposition and data masking as claimed in claim 3 is characterized in that: According to the set time window, the voltage detection quantity is decoupled to obtain a two-dimensional analysis matrix, which specifically includes: Create a width of The time window is expanded by the time dimension ; Set the moving step of the time window; the time window is the data mask; The time window is slid from left to right on the voltage detection data vector according to the set moving step size. With each sliding step, the voltage detection data in the time window is decoupled from the voltage detection data outside the time window to obtain a two-dimensional analysis matrix ; （2） in, Indicates the first At this moment, The measured value of a sensor.

5. The battery pack fault diagnosis method based on feature decomposition and data masking as claimed in claim 4, characterized in that: The two-dimensional analysis matrix is ​​standardized to obtain a standardized matrix, which specifically includes: Simplify formula (1) to: ； in: ； ； ； ； is the battery terminal voltage, is the connection resistance between the n-1th battery cell and the nth battery cell, is the connection resistance of the negative terminal of the battery pack, is the connection resistance of the positive terminal of the battery pack; Combined with formula (2), we get: ； in: Indicates the first The moment The voltage of the battery; Indicates the first At this moment, The battery and The equivalent connection resistance between the batteries is Represents the total negative equivalent connection resistance, Represents the total positive equivalent connection resistance; right Standardize and get the standardized matrix : in, for The mean vector of , , and They are and The mean vector of ; yes The standard deviation diagonal matrix of .

6. The battery pack fault diagnosis method based on feature decomposition and data masking as claimed in claim 5, characterized in that: According to the standardized matrix, the covariance matrix is ​​obtained, which includes: in, represents the covariance matrix; represents a two-dimensional analysis normalization matrix; represents a time window; Perform eigenvalue decomposition on the covariance matrix to obtain the maximum eigenvalue and the eigenvector corresponding to the maximum eigenvalue, including: And the covariance matrix Perform eigenvalue decomposition and get: ； in, yes The eigenvalues ​​in decreasing order, is the eigenvector in the corresponding order.

7. The battery pack fault diagnosis method based on feature decomposition and data masking as claimed in claim 1 or 6, characterized in that: The maximum characteristic value is used as a fault indicator of the battery pack to determine whether the battery pack fails. Specifically, if the maximum characteristic value is less than a set threshold, it indicates that the battery pack has not failed; otherwise, it indicates that the battery pack has failed.

8. A battery pack fault diagnosis system based on feature decomposition and data masking, characterized in that: include: An acquisition module is configured to: acquire a voltage detection value by using a redundant cross measurement topology; The voltage detection quantity includes: the voltage of all battery cells themselves and the voltage of all battery assemblies; the battery assembly refers to the battery cell and the adjacent previous resistor, and the battery cell and the adjacent next resistor; The processing module is configured to: decouple the voltage detection quantity according to the set time window to obtain a two-dimensional analysis matrix; perform standardization processing on the two-dimensional analysis matrix to obtain a standardized matrix; obtain a covariance matrix according to the standardized matrix; perform eigenvalue decomposition on the covariance matrix to obtain a maximum eigenvalue and an eigenvector corresponding to the maximum eigenvalue; A first judgment module is configured to: use the maximum characteristic value as a fault indicator of the battery pack to determine whether the battery pack fails; The second judgment module is configured to: select elements with a decreasing trend in the eigenvector corresponding to the maximum eigenvalue, if only the first elements pass the screening, then the A voltage sensor fails; if the elements and elements pass the screening, then the A short circuit occurs in a battery cell; if the elements and elements pass the screening, then the The battery and There is an electrical connection failure between the battery cells, among which, is a positive integer greater than zero, Indicates the ordinal number of the element.

9. An electronic device, comprising: a memory for non-transitory storage of computer readable instructions; as well as a processor for executing the computer readable instructions, When the computer-readable instructions are executed by the processor, the method described in any one of claims 1 to 7 is executed.

10. A storage medium, characterized in that: The computer-readable instructions are non-transitory stored, wherein when the non-transitory computer-readable instructions are executed by a computer, the instructions of the method according to any one of claims 1 to 7 are executed.

Images