A lithium battery fault diagnosis method, device and electronic device based on internal short-circuit current observation

Abstract

The present invention discloses a lithium battery fault diagnosis method, device and electronic device based on internal short-circuit current observation. The method comprises: constructing an equivalent circuit model including a short-circuit resistor, and using the equivalent circuit model including the short-circuit resistor to construct a state space equation including the short-circuit resistor; constructing a proportional-integral observer based on the state space equation including the short-circuit resistor; using the proportional-integral observer to observe the internal short-circuit current of the lithium battery in use; using a cumulative sum method to perform statistical analysis on the internal short-circuit current of the lithium battery to obtain a cumulative sum statistical result, and performing internal short-circuit fault diagnosis on the lithium battery according to the cumulative sum statistical result.

Description

Technical Field

[0001] The present invention relates to the technical field of battery fault diagnosis, and in particular to a lithium battery fault diagnosis method, device and electronic equipment based on internal short-circuit current observation. Background Art

[0002] Power batteries are the main energy storage devices for new energy vehicles. Their performance and cost directly affect the power, economy and safety of electric vehicles. With the rapid development of new energy vehicles, lithium batteries have outstanding advantages in energy density, cycle life, environmental protection, etc., and have become the main choice of power batteries. In order to ensure the safety and reliability of electric vehicles, the safe operation of lithium-ion batteries is crucial.

[0003] Unexpected conductive connection between the positive and negative electrodes inside the battery can easily cause an internal short circuit failure in the lithium battery. This failure may be caused by a variety of factors, including mechanical damage, electrochemical instability, diaphragm damage, etc. Internal short circuit failure may cause the battery to overheat, which in turn causes a thermal runaway reaction. Thermal runaway is a chain reaction in which the temperature inside the battery rises rapidly, which may lead to decomposition of battery materials, fire, or even explosion. Therefore, battery safety management and fault prevention are important considerations in battery design and use. However, the prior art has the following technical disadvantages:

[0004] 1) Existing lithium battery internal short-circuit fault detection methods are difficult to diagnose small internal short-circuit faults.

[0005] Although known observer-based methods are able to detect internal faults by monitoring the operating status of the battery, when the internal short-circuit current is small, it may be difficult for the original methods to sense and accurately diagnose.

[0006] 2) The existing internal short-circuit current measurement method is not suitable for online monitoring of battery status.

[0007] Although the internal short-circuit current can be measured quickly and accurately using a specific test setup, this method is not suitable for online monitoring, that is, it cannot detect the battery's status and potential internal short-circuit problems in real time during normal battery use. Online monitoring is essential to detect faults in a timely manner and take measures to prevent battery damage or dangerous events. Summary of the invention

[0008] The present invention provides a lithium battery fault diagnosis method, device and electronic equipment based on internal short circuit current observation, so as to solve the technical problem in the prior art that the battery status and potential internal short circuit cannot be detected in real time during normal use of the battery.

[0009] An embodiment of the present invention provides a lithium battery fault diagnosis method based on internal short-circuit current observation, comprising: constructing an equivalent circuit model including a short-circuit resistor, and using the equivalent circuit model including the short-circuit resistor to construct a state-space equation including the short-circuit resistor; constructing a proportional-integral observer based on the state-space equation including the short-circuit resistor; using the proportional-integral observer to observe the internal short-circuit current of a lithium battery in use; using a cumulative sum method to perform statistical analysis on the internal short-circuit current of the lithium battery to obtain a cumulative sum statistical result, and performing internal short-circuit fault diagnosis on the lithium battery according to the cumulative sum statistical result.

[0010] Preferably, the constructing of the equivalent circuit model including the short-circuit resistor comprises: obtaining the equivalent circuit model including the short-circuit resistor by connecting in parallel a short-circuit resistor for simulating a short-circuit fault in a lithium battery on the basis of a dual RC battery equivalent circuit model for simulating a transient response of a lithium battery.

[0011] Preferably, the severity of the short-circuit fault in the lithium battery is simulated by adjusting the resistance value of the short-circuit resistor; accordingly, the resistance value of the short-circuit resistor is inversely proportional to the severity of the short-circuit fault in the lithium battery.

[0012] Preferably, the state space equation comprises:

[0013] x(k+1)=Ax(k)+BI batt (k)+B f I sc (k)

[0014] y(k)=Cx(k)+DI batt (k)+D f I sc (k)

[0015] Among them, B f =B, C = [-1, -1, a i ], D=D f = -R0;

[0016] Where k represents the discrete time step, x(k) represents the state vector of the current time step, x(k+1) represents the state vector of the next time step, y(k) represents the output, R0 represents the series resistance, I batt (k) represents the battery current vector; I sc (k) represents the short-circuit current vector; A represents the state transfer matrix, B represents the input matrix of the battery current, and B f represents the input matrix of short-circuit current, C represents the output matrix, and D represents the direct transfer matrix, which describes the effect of input battery current on output. Similarly, D f Indicates the effect of short-circuit current on output, a iis the output matrix parameter.

[0017] Preferably, the proportional-integral observer comprises:

[0018]

[0019] In the formula, Represents the estimated value of the state vector at the current time step; represents the output of the observer; represents the estimated value of the battery fault f(k); L represents the proportional gain matrix and F represents the integral gain matrix; represents the estimated value of the state vector for the next time step, u(k) represents the input vector, represents the cumulative sum of the output errors computed from the initial time step.

[0020] Preferably, the internal short circuit fault diagnosis of the lithium battery based on the cumulative sum statistical results includes: obtaining the maximum value of the cumulative sum of observation results of the lithium battery in a healthy state, and using the maximum value of the cumulative sum as a fault diagnosis threshold; by comparing the cumulative sum statistical results with the fault diagnosis threshold, and when the cumulative sum statistical results are higher than the fault diagnosis threshold, determining that the lithium battery has an internal short circuit fault; when determining that the lithium battery has an internal short circuit fault, the internal short circuit current observed each time by the proportional-integral observer is integrated by using the cumulative sum to obtain the severity of the internal short circuit fault of the lithium battery.

[0021] Preferably, the method of using the cumulative sum to perform statistical analysis on the internal short-circuit current of the lithium battery to obtain the cumulative sum statistical result includes: calculating the sufficient statistics of the internal short-circuit current of the lithium battery based on a Gaussian distribution process; and obtaining the cumulative sum statistical result using the sufficient statistics of the internal short-circuit current of the lithium battery.

[0022] Preferably, the cumulative and statistical results include:

[0023]

[0024] in, represents the cumulative sum statistic at time step k The difference between the minimum value of the cumulative sum statistic of all previous time steps; N represents the total sample; represents the cumulative sum statistic at time step k, Represents the minimum value of the cumulative sum statistic from time step 1 to time step k: j represents a time step from time step 1 to time step k.

[0025] An embodiment of the present invention also provides a lithium battery fault diagnosis device based on internal short-circuit current observation, including: a construction module, used to construct an equivalent circuit model including a short-circuit resistor, and use the equivalent circuit model including the short-circuit resistor to construct a state-space equation including the short-circuit resistor; and based on the state-space equation including the short-circuit resistor, construct a proportional-integral observer; a current observation module, used to use the proportional-integral observer to observe the internal short-circuit current of the lithium battery in use; a fault diagnosis module, used to perform statistical analysis on the internal short-circuit current of the lithium battery using a cumulative sum method to obtain a cumulative sum statistical result, and perform internal short-circuit fault diagnosis on the lithium battery based on the cumulative sum statistical result.

[0026] An embodiment of the present invention also provides an electronic device, characterized in that it includes a memory, a processor, and a computer program stored in the memory and executable on the processor; when the computer program is executed by the processor, the steps of the lithium battery fault diagnosis method based on internal short-circuit current observation as described above are implemented.

[0027] The beneficial effect of the present invention is that during the use of the lithium-ion battery, the internal short-circuit current is estimated by an online observation method, thereby detecting the internal short-circuit fault, formulating corresponding control and maintenance strategies, and improving the safety and reliability of the power battery system. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 It is a flow chart of a lithium battery minor fault diagnosis method based on internal short-circuit current observation provided by the present invention;

[0029] Figure 2 It is a schematic diagram of an internal short-circuit model provided by the present invention;

[0030] Figure 3 It is a schematic diagram of using a proportional-integral observer to observe an internal short-circuit current provided by the present invention;

[0031] Figure 4 It is a schematic diagram of the structure of lithium battery minor fault diagnosis based on internal short-circuit current observation provided by the present invention;

[0032] Figure 5 It is a schematic diagram of the accumulation and result of 1000 internal short-circuit faults provided by the present invention;

[0033] Figure 6 It is a flow chart of a lithium battery fault diagnosis method based on internal short-circuit current observation provided by the present invention;

[0034] Figure 7 It is a schematic diagram of a lithium battery fault diagnosis device based on internal short-circuit current observation provided by the present invention. DETAILED DESCRIPTION

[0035] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0036] The present invention proposes a lithium battery minor fault diagnosis method based on internal short-circuit current observation. The method adopts a lithium battery short-circuit equivalent circuit model and proposes a short-circuit current observation method based on a proportional integral observer. The method can accurately estimate the short-circuit current value online and evaluate the severity of the internal short circuit of the lithium battery. It can also accurately identify the occurrence of minor internal short-circuit current through the cumulative sum method, which has important application value for the identification and elimination of early faults.

[0037] Figure 6 is a flow chart of a lithium battery fault diagnosis method based on internal short-circuit current observation provided by the present invention, such as Figure 6 As shown, it includes: step S1: constructing an equivalent circuit model including a short-circuit resistor, and using the equivalent circuit model including the short-circuit resistor to construct a state-space equation including the short-circuit resistor; step S2: constructing a proportional-integral observer based on the state-space equation including the short-circuit resistor; step S3: using the proportional-integral observer to observe the internal short-circuit current of the lithium battery in use; step S4: using the cumulative sum method to perform statistical analysis on the internal short-circuit current of the lithium battery to obtain the cumulative sum statistical result, and diagnosing the internal short-circuit fault of the lithium battery according to the cumulative sum statistical result.

[0038] Furthermore, the construction of the equivalent circuit model including the short-circuit resistor includes: obtaining the equivalent circuit model including the short-circuit resistor by connecting a short-circuit resistor for simulating a short-circuit fault in a lithium battery in parallel on the basis of a dual RC battery equivalent circuit model for simulating a transient response of a lithium battery.

[0039] Specifically, the severity of the short-circuit fault in the lithium battery is simulated by adjusting the resistance value of the short-circuit resistor; accordingly, the resistance value of the short-circuit resistor is inversely proportional to the severity of the short-circuit fault in the lithium battery.

[0040] Furthermore, the state space equation includes:

[0041] x(k+1)=Ax(k)+BI batt (k)+B f I sc (k)

[0042] y(k)=Cx(k)+DI batt (k)+D f I sc (k)

[0043] Among them, B f =B, C = [-1, -1, ai ],D=D f = -R0;

[0044] Where k represents the discrete time step, x(k) represents the state vector of the current time step, x(k+1) represents the state vector of the next time step, y(k) represents the output, R0 represents the series resistance, I batt (k) represents the battery current vector; I sc (k) represents the short-circuit current vector; A represents the state transfer matrix, B represents the input matrix of the battery current, and B f represents the input matrix of short-circuit current, C represents the output matrix, and D represents the direct transfer matrix, which describes the effect of input battery current on output. Similarly, D f Indicates the effect of short-circuit current on output, a i is the output matrix parameter.

[0045] Furthermore, the proportional-integral observer comprises:

[0046]

[0047] in, Represents the estimated value of the state vector at the current time step; represents the output of the observer; represents the estimated value of the battery fault f(k); L represents the proportional gain matrix and F represents the integral gain matrix; represents the estimated value of the state vector for the next time step, u(k) represents the input vector, represents the cumulative sum of the output errors computed from the initial time step.

[0048] Specifically, the internal short circuit fault diagnosis of the lithium battery based on the cumulative and statistical results includes: obtaining the maximum value of the cumulative sum of the observation results of the lithium battery in a healthy state, and using the maximum value of the cumulative sum as a fault diagnosis threshold; by comparing the cumulative and statistical results with the fault diagnosis threshold, and when the cumulative and statistical results are higher than the fault diagnosis threshold, determining that the lithium battery has an internal short circuit fault; when determining that the lithium battery has an internal short circuit fault, the internal short circuit current observed by the proportional-integral observer each time is integrated by using the cumulative sum to obtain the severity of the internal short circuit fault of the lithium battery.

[0049] Furthermore, the method of using the cumulative sum to perform statistical analysis on the internal short-circuit current of the lithium battery to obtain the cumulative sum statistical result includes: calculating the sufficient statistics of the internal short-circuit current of the lithium battery based on a Gaussian distribution process; and obtaining the cumulative sum statistical result using the sufficient statistics of the internal short-circuit current of the lithium battery.

[0050] Specifically, the cumulative and statistical results include:

[0051]

[0052] in, represents the cumulative sum statistic at time step k The difference between the minimum value of the cumulative sum statistic of all previous time steps; N represents the total sample; represents the cumulative sum statistic at time step k, Represents the minimum value of the cumulative sum statistic from time step 1 to time step k: j represents a time step from time step 1 to time step k.

[0053] Figure 7 is a schematic diagram of a lithium battery fault diagnosis device based on internal short-circuit current observation provided by the present invention, such as Figure 7 As shown, it includes: a construction module, which is used to construct an equivalent circuit model including a short-circuit resistor, and use the equivalent circuit model including the short-circuit resistor to construct a state-space equation including the short-circuit resistor; and based on the state-space equation including the short-circuit resistor, a proportional-integral observer is constructed; a current observation module, which is used to use the proportional-integral observer to observe the internal short-circuit current of the lithium battery in use; a fault diagnosis module, which is used to perform statistical analysis on the internal short-circuit current of the lithium battery using a cumulative sum method to obtain a cumulative sum statistical result, and perform internal short-circuit fault diagnosis on the lithium battery according to the cumulative sum statistical result.

[0054] An embodiment of the present invention also provides an electronic device, characterized in that it includes a memory, a processor, and a computer program stored in the memory and executable on the processor; when the computer program is executed by the processor, the steps of the lithium battery fault diagnosis method based on internal short-circuit current observation as described above are implemented.

[0055] In view of the existing lithium battery micro-internal short circuit fault diagnosis problem, the present invention proposes a fault diagnosis method based on internal short circuit current observation. The specific implementation steps of the method are as follows: Figure 1 shown.

[0056] Step 1) Construction of internal short circuit model:

[0057] The equivalent circuit model is adopted, in which the double RC circuit characterizes the charge transfer effect, diffusion effect and double electric layer behavior inside the lithium battery to simulate the transient response of the battery. On the basis of the double RC equivalent circuit model, by introducing the short circuit resistor R_sc to simulate the occurrence of short circuit, it can be described as an equivalent circuit model of a healthy battery in parallel with an equivalent short circuit resistor R_sc, such as Figure 2 shown.

[0058] Step 2), Model state space representation

[0059] The battery state of charge SOC∈[0%, 100%] can be calculated by the classical coulomb counting method:

[0060]

[0061] Where η is the coulombic efficiency, which is usually close to 1 for LIBs; Ibatt is the measured battery current (A); soc(t0) is the initial value of the state of charge; C n represents the rated capacity of the battery (Ah); soc(t) represents the required SOC value at time point t based on its initial value soc(t0); I batt (τ) represents the current value at time τ.

[0062] If the short-circuit current is not considered, using Kirchhoff's law, the battery dynamic model can be described by the following state space equation:

[0063]

[0064] V batt (k) = V oc (soc)-V1(k)-V2(k)-R0I in (k)

[0065]

[0066] Among them, V1, V2 are the voltages of the double RC branches, T is the sampling interval, R0 is the series resistance, Voc is the open circuit voltage, V oc =a i ·SOC+b i ; Voc is a function of SOC, a i and b i is the fitting constant term of Voc.

[0067] Considering the short-circuit current, according to Kirchhoff's law, I in =I batt +I sc The mathematical expression (state space expression) of the equivalent circuit model is:

[0068] x(k+1)=Ax(k)+BI batt (k)+B f I sc (k)

[0069] y(k)=Cx(k)+DI batt (k)+D f I sc (k)

[0070] in

[0071] B f =B, C = [-1, -1, a i ], D=D f =-R0

[0072] Where x(k) = [V1(k), V2(k), soc(k)]′ – the state vector of the model;

[0073] y(k)=V batt (k)-b i ——Output of the model.

[0074] Step 3) Design of internal short-circuit current observer

[0075] To construct the observer, first introduce noise into the state-space model:

[0076] x(k+1)=Ax(k)+Bu(k)+B f f(k)+B d d(k)

[0077] y(k)=Cx(k)+Du(k)+D f f(k)+D d d(k)

[0078] Where x(k)∈R n ——the state vector of the model;

[0079] y(k)∈R p ——output value of the model;

[0080] u(k)∈R m ——known input, which corresponds to I batt ;

[0081] - So-called battery failure, which corresponds to I sc ;

[0082] ——Introduced interference and noise, whose value belongs to l2[0,∞];

[0083] B d and D d ——Constant real matrix.

[0084] Then, the proportional-integral observer is designed, which is expressed as:

[0085]

[0086] In the formula, ——the estimated state vector;

[0087] ——output of the observer;

[0088] ——is the estimated value of the battery failure f(k);

[0089] L∈R n×p and ——Observer gain.

[0090] The observer error is recorded as:

[0091]

[0092] in

[0093]

[0094] Δf(k)=f(k+1)-f(k)

[0095] in, represents the error vector at time step k, representing the difference between the actual state and the estimated state; is the state transition matrix, is the observer gain matrix, which is used to adjust the estimation error; is the output matrix; and is the input matrix; is the external disturbance vector.

[0096] The observer matrix is ​​formulated using the pole placement method The value of, specifically, The value of The eigenvalue position of is within the unit circle, so that the error converges and the observation result is obtained.

[0097] The observer error directly affects the accuracy of the state estimation. The dynamic characteristics of the error determine how quickly and how well the observer can converge to the true system state. Therefore, by adjusting the observer gain The error can be minimized and the accuracy of state estimation can be improved.

[0098] The above proportional-integral observer is used to observe the internal short-circuit current I sc , the results are as follows Figure 3 shown.

[0099] It should be pointed out that the proportional-integral observer observing the internal short-circuit current cannot be used directly for fault diagnosis because the noise is too large, so a subsequent accumulation and method is required.

[0100] Step 4) Identification of small internal short circuit faults

[0101] For small internal short-circuit faults, it is difficult to use step changes as the basis for internal short-circuit judgment. Therefore, the cumulative sum method is introduced to analyze the observed internal short-circuit current, such as Figure 4 shown.

[0102] The Monte Carlo method is used to test the model 1000 times, including 500 times in healthy conditions and 500 times in fault conditions. The cumulative sum function is denoted as S. For the short-circuit current observation value I sc , based on the Gaussian distribution process, the cumulative sum statistic s k Denoted as:

[0103]

[0104] Among them, μ f and μ h They are the estimated short-circuit current The average value under both battery failure and healthy conditions. σ h is the estimated short-circuit current The standard deviation in a healthy state. According to literature data, μ f =0.03, μ h =0,σ h =0.006. represents the estimated short-circuit current at time step k.

[0105] Compute the cumulative sum of all individual statistics for the previous N time steps:

[0106]

[0107] Among them, N is the total number of samples.

[0108] To identify significant changes in the cumulative sum statistic, define the statistic

[0109]

[0110] in, is the cumulative sum statistic at time step k, is the minimum value of the cumulative sum statistic from time step 1 to time step k, then, represents the cumulative sum statistic at time step k The difference between the minimum value of the cumulative sum statistics of all previous time steps.

[0111] The cumulative sum of short-circuit failures within 1000 times is as follows: Figure 5As shown. The maximum value of the cumulative sum of the observation results in the healthy state is taken as the threshold T. If the cumulative sum is higher than the threshold T after a fault occurs, it is judged that an internal short circuit fault has occurred and measures should be taken in time. The cumulative sum method can effectively identify the time of fault occurrence, which is more obvious than the step effect when observing the internal short circuit current. Even when a small internal short circuit current occurs, it can be effectively identified.

[0112] In summary, the present invention aims at the problem of online internal short-circuit fault diagnosis of lithium batteries, and proposes an internal short-circuit current estimation method based on an observer, which can effectively identify the internal short-circuit current and diagnose the occurrence of internal short circuit. For the situation where the internal short-circuit current is small and difficult to observe in the early stage of internal short-circuit fault, a statistical method based on cumulative sum is proposed, which uses the cumulative sum to process and integrate each data, and has the ability to identify the severity of the internal short circuit, so that the internal short-circuit fault diagnosis has visualization ability, which helps to solve the problem of early internal short-circuit fault diagnosis, and is of great significance to improving the reliability of the fault diagnosis function of the battery management system.

[0113] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but the scope of the present invention is not limited thereby. Any modification, equivalent substitution and improvement made by those skilled in the art without departing from the scope and essence of the present invention shall be within the scope of the present invention.

Claims

1. A lithium battery fault diagnosis method based on internal short-circuit current observation, characterized in that: include: By connecting a short-circuit resistor for simulating a short-circuit fault in a lithium battery in parallel on the basis of a dual RC battery equivalent circuit model for simulating a transient response of a lithium battery, an equivalent circuit model including the short-circuit resistor for simulating a short-circuit fault in a lithium battery is constructed, and a state space equation including the short-circuit resistor is constructed using the equivalent circuit model including the short-circuit resistor; Based on the state-space equation including the short-circuit resistor, construct a proportional-integral observer; Utilizing the proportional-integral observer, observing the internal short-circuit current of the lithium battery in use; The internal short-circuit current of the lithium battery is statistically analyzed by using a cumulative sum method to obtain a cumulative sum statistical result, and the internal short-circuit fault diagnosis of the lithium battery is performed according to the cumulative sum statistical result, which includes: Calculating sufficient statistics of the internal short-circuit current of the lithium battery based on a Gaussian distribution process; Using sufficient statistics of the internal short-circuit current of the lithium battery, a cumulative and statistical result is obtained; Obtaining the maximum value of the cumulative sum of observation results of the lithium battery in a healthy state, and using the maximum value of the cumulative sum as a fault diagnosis threshold; By comparing the cumulative and statistical results with the fault diagnosis threshold, and when the cumulative and statistical results are higher than the fault diagnosis threshold, it is determined that an internal short circuit fault occurs in the lithium battery; When it is determined that the lithium battery has an internal short circuit fault, the internal short circuit current observed by the proportional-integral observer each time is integrated by using a cumulative sum to obtain the severity of the internal short circuit fault of the lithium battery; The cumulative and statistical results include: in, represents the cumulative sum statistic at time step k The difference between the minimum value of the cumulative sum statistic of all previous time steps; N represents the total sample; represents the cumulative sum statistic at time step k, Represents the minimum value of the cumulative sum statistic from time step 1 to time step k: j represents a time step from time step 1 to time step k.

2. The method according to claim 1, characterized in that The severity of the short circuit fault in the lithium battery is simulated by adjusting the resistance value of the short circuit resistor; Correspondingly, the resistance value of the short-circuit resistor is inversely proportional to the severity of the short-circuit fault in the lithium battery.

3. The method according to claim 1, characterized in that The state space equations include: in, ; in, represents a discrete time step, Represents the state vector of the model at the current time step; represents the state vector for the next time step, Indicates output; represents the series resistance; represents the battery current vector; represents the short-circuit current vector; A represents the state transfer matrix, B represents the input matrix of the battery current, represents the input matrix of short-circuit current, C represents the output matrix, and D represents the direct transfer matrix, which describes the effect of input battery current on output. Similarly, Indicates the effect of short-circuit current on output, is the output matrix parameter.

4. The method according to claim 3, characterized in that The proportional-integral observer comprises: In the formula, Represents the estimated value of the state vector at the current time step; represents the output of the observer; Indicates battery failure An estimated value of; represents the proportional gain matrix and represents the integral gain matrix; represents the estimated value of the state vector for the next time step, represents the input vector, represents the cumulative sum of the output errors computed from the initial time step.

5. A lithium battery fault diagnosis device based on internal short-circuit current observation, characterized in that: include: A construction module is used to construct an equivalent circuit model including a short-circuit resistor for simulating short-circuit faults in lithium batteries by connecting a short-circuit resistor in parallel to a dual RC battery equivalent circuit model for simulating transient response of lithium batteries, and to construct a state space equation including the short-circuit resistor using the equivalent circuit model including the short-circuit resistor; and to construct a proportional-integral observer based on the state space equation including the short-circuit resistor; A current observation module, used to observe the internal short-circuit current of the lithium battery in use by using the proportional-integral observer; A fault diagnosis module, used for performing statistical analysis on the internal short-circuit current of the lithium battery by using a cumulative sum method to obtain a cumulative sum statistical result, and performing internal short-circuit fault diagnosis on the lithium battery according to the cumulative sum statistical result; The fault diagnosis module is specifically used to calculate the sufficient statistics of the internal short-circuit current of the lithium battery based on the Gaussian distribution process; and obtain the cumulative and statistical results by using the sufficient statistics of the internal short-circuit current of the lithium battery; Obtaining the maximum value of the cumulative sum of observation results of the lithium battery in a healthy state, and using the maximum value of the cumulative sum as a fault diagnosis threshold; By comparing the cumulative sum statistical result with the fault diagnosis threshold, and when the cumulative sum statistical result is higher than the fault diagnosis threshold, it is judged that the lithium battery has an internal short circuit fault; when it is judged that the lithium battery has an internal short circuit fault, the internal short circuit current observed by the proportional-integral observer each time is integrated by using the cumulative sum to obtain the severity of the internal short circuit fault of the lithium battery; The cumulative and statistical results include: in, represents the cumulative sum statistic at time step k The difference between the minimum value of the cumulative sum statistic of all previous time steps; N represents the total sample; represents the cumulative sum statistic at time step k, Represents the minimum value of the cumulative sum statistic from time step 1 to time step k: j represents a time step from time step 1 to time step k.

6. An electronic device, characterized in that: It comprises a memory, a processor and a computer program stored in the memory and executable on the processor; when the computer program is executed by the processor, the steps of the lithium battery fault diagnosis method based on internal short-circuit current observation described in any one of claims 1 to 4 are implemented.

Images