A method and system for detecting short - circuit faults in a lithium battery

Abstract

A lithium battery short-circuit fault detection method and system provided by the present application combines statistical information (cumulative sum) with a model-based observer fault estimation to detect minute changes hidden in environmental noise in fault characteristics, and is used for an early diagnosis method of battery short-circuit. Specifically, an equivalent model of battery short-circuit and a fuzzy observer of the battery are used in combination. According to the residual between the actual value of the state of charge of the battery measured by the equivalent model of battery short-circuit and the estimated value of the state of charge of the battery estimated by the fuzzy observer, the battery short-circuit fault is determined. This fault detection method has the characteristics of model universality, design universality and implementation universality.

Description

Technical Field

[0001] This application relates to the field of intelligent fault diagnosis of battery energy storage systems, and specifically relates to a method and system for detecting short - circuit faults of lithium batteries. Background Art

[0002] Lithium - ion batteries are widely used in the energy storage field. Soft short - circuits in lithium - ion batteries may lead to thermal runaway of the battery, which is a potential safety threat. In order to diagnose and detect soft short - circuits at an early stage, a variety of research methods have been proposed. Diagnostic methods based on abnormal changes in external characteristic parameters of the battery, that is, diagnosing through parameters such as abnormal terminal voltage, temperature, and SOC of the battery; soft short - circuit detection methods based on consistency differences, that is, diagnosing by comparing the differences in external characteristic parameters of the battery; model - based detection methods, which transform the internal short - circuit detection problem into a state estimation or parameter identification problem with the help of the physical model of the battery; internal short - circuit detection methods based on machine learning, which use machine learning algorithms to detect internal short - circuits and further analyze the relationships between internal short - circuits and thermal runaway, and between internal short - circuits and short - circuit triggering conditions.

[0003] Although most of the current research work has recently contributed to the diagnosis and detection of soft short - circuits, these methods have limitations in technology or usage scenarios. General fault detection methods based on analytical models need to have strong universality (model universality, design universality, and implementation universality). Summary of the Invention

[0004] Based on the above problems, in order to timely diagnose the thermal runaway faults of batteries in large - scale energy storage power stations, a battery short - circuit detection method based on a fuzzy observer and a battery short - circuit equivalent model is proposed.

[0005] In the first aspect, this application provides a method for detecting short - circuit faults of lithium batteries, including:

[0006] Input the state of charge of the current battery at all times into a preset battery short - circuit equivalent model, and the battery short - circuit equivalent model outputs the battery state parameters of the next moment of the battery;

[0007] For the current lithium battery, use a fuzzy observer to estimate the battery state estimation parameters of the battery;

[0008] Determine the short - circuit fault of the battery according to the residual between the battery state parameters and the battery state estimation parameters.

[0009] Preferably, the method for detecting short - circuit faults of lithium batteries further includes:

[0010] Model and calculate for all states of charge of a sample battery to obtain the battery short - circuit equivalent model.

[0011] Preferably, the modeling and calculation for all charge states of a sample battery to obtain the battery short-circuit equivalent model includes:

[0012] Obtaining a state of charge of the sample battery at a next moment according to the load current of the sample battery at a current moment and the initial state of charge of the sample battery;

[0013] A battery short-circuit equivalent model is established according to the state of charge of the sample battery at a current moment and the state of charge at a next moment.

[0014] Preferably, obtaining the state of charge of the sample battery at the next moment according to the load current of the sample battery at the current moment and the initial state of charge of the sample battery includes:

[0015] Integrating the load current of the sample battery at the current moment to obtain a current current integral value;

[0016] The state of charge of the sample battery at the next moment is obtained according to the initial state of charge and the current current integral value.

[0017] Preferably, the lithium battery short circuit fault detection method further includes:

[0018] The fuzzy observer is obtained according to the battery voltage state of charge curve and a preset robust observer.

[0019] Preferably, the fuzzy observer is obtained according to the battery voltage state of charge curve and a preset robust observer, including:

[0020] Obtaining an optimal weighting function for the battery state of charge according to the battery voltage state of charge curve and a preset Gaussian function;

[0021] The fuzzy observer is obtained according to the optimal weighting function and the robust observer.

[0022] Preferably, obtaining the optimal weighting function of the battery state of charge according to the battery voltage state of charge curve and a preset Gaussian function includes:

[0023] Obtaining an optimization coefficient of the Gaussian function according to the battery voltage state of charge curve and a preset Gaussian function;

[0024] According to the optimization coefficient, the optimal parameters of the Gaussian function are determined, and then the optimal weighting function of the battery state of charge is determined.

[0025] In a second aspect, the present application provides a lithium battery short circuit fault detection system, comprising:

[0026] Parameter calculation module: Input the state of charge of the current battery at all times into a preset battery short - circuit equivalent model, and the battery short - circuit equivalent model outputs the battery state parameters of the next moment of the battery;

[0027] Parameter estimation module: For the current lithium - ion battery, use a fuzzy observer to estimate the battery state estimation parameters of the battery;

[0028] Fault detection module: Determine the short - circuit fault of the battery according to the residual between the battery state parameters and the battery state estimation parameters.

[0029] Meanwhile, the present invention also provides a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the above - mentioned method is implemented.

[0030] Meanwhile, the present invention also provides a computer - readable storage medium, and the computer - readable storage medium stores a computer program for executing the above - mentioned method.

[0031] As can be seen from the above technical solutions, a lithium - ion battery short - circuit fault detection method and system provided by the present application combines statistical information (cumulative sum) with a model - based observer fault estimation to detect the tiny changes hidden in environmental noise in the fault characteristics for the early diagnosis method of battery short - circuit. Specifically, a battery short - circuit equivalent model and a battery fuzzy observer are combined. According to the residual between the actual value of the state of charge of the battery measured by the battery short - circuit equivalent model and the estimated value of the state of charge of the battery estimated by the fuzzy observer, the battery short - circuit fault is determined. This fault detection method has the characteristics of model universality, design universality, and implementation universality.

[0032] To make the above - mentioned and other purposes, features, and advantages of the present invention more obvious and understandable, the following specific preferred embodiments are given and described in detail in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0034] Figure 1 It is a schematic flow chart of a lithium - ion battery short - circuit fault detection method in an embodiment of the present application.

[0035] Figure 2 It is a schematic flow chart of the weighted - function self - adjusting state and fault estimator method for battery short - circuit detection in an embodiment of the present application.

[0036] Figure 3 Schematic diagram of the battery short - circuit equivalent circuit model in the embodiment of the present application.

[0037] Figure 4 Schematic diagram of the open - circuit voltage - SOC curve of the lithium - ion battery in the embodiment of the present application.

[0038] Figure 5 Schematic diagram of the design process of the fuzzy observer based on the TS fuzzy system in the embodiment of the present application.

[0039] Figure 6 Schematic diagram of the fuzzification process of the OCV - SOC curve in the embodiment of the present application.

[0040] Figure 7 Schematic diagram of the self - adjusting fault estimation of the weight function in the embodiment of the present application.

[0041] Figure 8 Schematic diagram of the structure of a lithium - battery short - circuit fault detection system in the embodiment of the present application.

[0042] Figure 9 Schematic diagram of the structure of the electronic device in the embodiment of the application. Detailed implementation manners

[0043] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Apparently, the described embodiments are some, but not all, of the embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts shall fall within the protection scope of the present application.

[0044] Since most of the current research work has recently contributed to soft - short - circuit diagnosis and detection, but these methods have limitations in terms of technology or usage scenarios. The general fault - detection method based on the analysis model needs to have strong universality (model universality, design universality, and implementation universality). The present application provides a weighted - function self - adjusting state and fuzzy observer for battery short - circuit detection. Considering the slow - change characteristic of the battery SOC, a systematic method is proposed based on the genetic algorithm to construct a self - adjusting mechanism to cope with the non - linear open - circuit voltage (OCV) - SOC curve. To detect the subtle changes hidden in the environmental noise in the fault characteristics, statistical information (cumulative sum) is combined with the model - based observer fault estimation.

[0045] Based on the above, the present application also provides a lithium battery short-circuit fault detection device for implementing the lithium battery short-circuit fault detection method provided in one or more embodiments of the present application. The lithium battery short-circuit fault detection device can be communicatively connected to user client devices. There can be multiple user client devices, and the lithium battery short-circuit fault detection device can specifically access the client devices through an application server.

[0046] Among them, the lithium battery short-circuit fault detection device can receive a lithium battery short-circuit fault detection instruction from the client device, obtain battery parameters such as the state of charge of the battery from the lithium battery short-circuit fault detection instruction, obtain the actual value of the battery parameters of the battery through a battery short-circuit equivalent model, obtain the estimated value of the battery parameters of the battery through a fuzzy observer, and determine whether the battery is in a fault state based on the residual between the actual value and the estimated value of the battery parameters. If the residual is 0, the battery is operating normally; if the residual is 1, the battery is in a short-circuit fault. Then, the lithium battery short-circuit fault detection device can send the battery fault state to the client device for display.

[0047] It can be understood that the client device can include a smart phone, a tablet electronic device, a portable computer, a desktop computer, and a personal digital assistant (PDA), etc.

[0048] In another actual application scenario, the part for performing the lithium battery short-circuit fault detection can be executed in the classification processing center as described above, or all operations can be completed in the client device. Specifically, it can be selected according to the processing capacity of the client device and the limitations of the user usage scenario, etc. The present application does not limit this. If all operations are completed in the client device, the client device can further include a processor for performing specific processing of the lithium battery short-circuit fault detection.

[0049] The above-mentioned client device can have a communication module (i.e., a communication unit) and can be communicatively connected to a remote server to achieve data transmission with the server. For example, the communication unit can send a lithium battery short-circuit fault detection instruction to the server of the classification processing center so that the server can perform lithium battery short-circuit fault detection processing according to the lithium battery short-circuit fault detection instruction. The communication unit can also receive the unit parameter configuration result returned by the server. The server can include a server on the task scheduling center side, and in other implementation scenarios, it can also include a server of an intermediate platform, such as a server of a third-party server platform communicatively linked to the task scheduling center server. The server can include a single computer device, or a server cluster composed of multiple servers, or a server structure of a distributed device.

[0050] Any suitable network protocol can be used for communication between the above-mentioned server and the client device, including network protocols that have not been developed as of the filing date of this application. The network protocol can, for example, include TCP / IP protocol, UDP / IP protocol, HTTP protocol, HTTPS protocol, etc. Of course, the network protocol can also, for example, include RPC protocol (Remote Procedure Call Protocol) and REST protocol (Representational State Transfer) used on top of the above-mentioned protocols, etc.

[0051] A lithium battery short-circuit fault detection method, system, electronic device, and computer-readable storage medium provided by this application combines statistical information (cumulative sum) with model-based observer fault estimation to detect subtle changes hidden in environmental noise in fault characteristics for an early battery short-circuit diagnosis method. Specifically, it combines a battery short-circuit equivalent model and a battery fuzzy observer, and determines a battery short-circuit fault based on the residual between the actual value of the state of charge of the battery measured by the battery short-circuit equivalent model and the estimated value of the state of charge of the battery estimated by the fuzzy observer. This fault detection method has the characteristics of model universality, design universality, and implementation universality.

[0052] Specifically, it will be described separately through the following multiple embodiments and application examples.

[0053] To diagnose the battery thermal runaway fault of a large-scale energy storage power station in a timely manner, an embodiment of a lithium battery short-circuit fault detection method provided by this application is referred to Figure 1 The lithium battery short-circuit fault detection method specifically includes the following content:

[0054] Step 100: Input the state of charge of the current battery at all times into a preset battery short-circuit equivalent model, and the battery short-circuit equivalent model outputs the battery state parameters of the next moment of the battery;

[0055] Step 200: For the current lithium battery, use a fuzzy observer to estimate the battery state estimation parameters of the battery;

[0056] Step 300: Determine the short-circuit fault of the battery according to the residual between the battery state parameters and the battery state estimation parameters.

[0057] In this embodiment, considering the slow change characteristic of the battery SOC, a systematic method is proposed based on the genetic algorithm to construct a self-regulating mechanism to cope with the non-linear open-circuit voltage (OCV)-SOC curve. To detect subtle changes hidden in environmental noise in fault characteristics, statistical information (cumulative sum) is combined with model-based observer fault estimation, and the overall strategy architecture is asFigure 2 As shown, first establish a battery short - circuit equivalent model, and then fuzzify the OCV - SOC curve and the robust observer to obtain a fuzzy observer.

[0058] The fuzzy observer is essentially a weighted - function self - adjusting robust observer. The estimated values of the battery parameters are obtained through the fuzzy observer. The battery short - circuit equivalent model is as follows Figure 3 . The resistance R0 represents the ohmic resistance, which includes the resistances of the contacts, electrodes, and electrolyte. The double - RC loop characterizes the charge transfer effect, diffusion effect, and double - layer behavior inside the lithium - ion battery and can simulate the transient response of the battery. In addition, compared with the single - RC and triple - RC structures, the double - RC network is a good trade - off between model error and model complexity. The actual values of the battery parameters are obtained through the battery short - circuit equivalent model. Whether the battery is in a fault state is judged by the residual between the actual values and the estimated values of the battery parameters. If the residual is 0, the battery operates normally; if the residual is 1, the battery is in a short - circuit fault.

[0059] Since I in = I batt + I sc , the battery equivalent circuit model with a short - circuit resistance in the i - th SOC interval is expressed as: x(k + 1) = Ax(k)+B f f(k)+B d d(k)

[0060] y(k) = Cx(k)+Du(k)+D f f(k)+D d d(k) (7)

[0061] where x(k) ∈ R n is the state vector; y(k) ∈ R p is the output; u(k) ∈ R m is the known input, corresponding to I batt ; is the battery fault, corresponding to I sc ; is the disturbance belonging to l2[0, ∞]; B d and D d are constant real matrices of appropriate dimensions.

[0062] Design the fault estimator as a proportional - integral observer. The integral term can not only ensure the robust state estimation when a fault occurs but also provide fault estimation simultaneously. The fuzzy observer can be expressed as:

[0063]

[0064]

[0065]

[0066] Among them, is the estimated state vector; is the observer output; is the estimated f(k); L ∈ R n×p and; is the observer gain.

[0067] The error dynamics between the model expression (7) and the observer expression (8) can be described by Equation (9):

[0068]

[0069] Where:

[0070]

[0071]

[0072]

[0073] And △f(k) = f(k + 1) - f(k) belongs to l2[0, ∞]. Ii is the symbol of the identity matrix with i×i dimensions. 0 is the zero matrix with the corresponding dimensions.

[0074] The design principle of the observer is to determine so that the error dynamic model Equation (9) satisfies the following two objectives:

[0075] (1) is Hurwitz stable. The eigenvalues of its discrete-time system are inside the unit circle;

[0076] (2) The fault estimation error e f (k) is insensitive to That is, the smaller e f (k) is, the better.

[0077] As can be seen from the above description, a lithium battery short-circuit fault detection method provided by an embodiment of the present application combines statistical information (cumulative sum) with model-based observer fault estimation in order to detect minute changes hidden in environmental noise in fault characteristics, and is used for the early diagnosis method of battery short circuits. Specifically, the battery short-circuit equivalent model and the battery fuzzy observer are combined and used. According to the residual between the actual value of the state of charge of the battery measured by the battery short-circuit equivalent model and the estimated value of the state of charge of the battery estimated by the fuzzy observer, the battery short-circuit fault is determined. This fault detection method has the characteristics of model universality, design universality, and implementation universality.

[0078] In an embodiment of a lithium battery short - circuit fault detection method provided in the present application, the lithium battery short - circuit fault detection method further includes:

[0079] Modeling and calculating for all states of charge of a sample battery to obtain the battery short - circuit equivalent model.

[0080] In this embodiment, according to Kirchhoff's circuit law, I in = I batt + I sc is the actual input current of the battery, whether there is a short - circuit fault or not. If R sc approaches infinity, then I sc ≈ 0, I in ≈ I batt , indicating that the battery is in good operating condition. Otherwise, an internal short - circuit or external short - circuit fault occurs. The state of charge SOC ∈ [0%, 100%] of the battery can be obtained by modeling and calculating through the classical Coulomb counting method to obtain the battery short - circuit equivalent model.

[0081] In an embodiment of a lithium battery short - circuit fault detection method provided in the present application, the modeling and calculating for all states of charge of a sample battery to obtain the battery short - circuit equivalent model includes:

[0082] Obtaining the state of charge of the sample battery at the next moment according to the load current of the sample battery at the current moment and the initial state of charge of the sample battery;

[0083] Establishing a battery short - circuit equivalent model according to the state of charge of the sample battery at the current moment and the state of charge at the next moment.

[0084] In this embodiment, as Figure 4 shown, its average value V OC (SOC) is usually a non - linear monotonically increasing function of SOC. In each SOC interval, it can be approximated as (a i and b i are constant within the i - th SOC interval). V1 and V2 are the voltages across capacitors C1 and C2 respectively. V batt is the terminal voltage of the lithium - ion battery. I batt is the battery load current. According to the reference direction in Figure 2 , “+” represents the discharge process, while “ - ” represents the charging process. I sc is the SC current flowing into the equivalent SC resistor Rsc, and its value is Integrate the load current of the sample battery at the current moment to obtain the current current integral value; according to the initial state of charge and the current current integral value, obtain the state of charge of the sample battery at the next moment, that is: the battery SOC ∈ [0%, 100%] can be modeled and calculated by the classical Coulomb counting method:

[0085]

[0086] where η is the charge-discharge efficiency, usually approximated as 1; C n is the nominal capacity of the battery, in Ah; soc(t) is the soc at time point t based on its initial value Soc(t0).

[0087] Using Kirchhoff's law, the battery dynamic model can be described by the following discrete state space representation:

[0088]

[0089] where:

[0090]

[0091] Since I in = I batt + I sc , the equivalent circuit model of the battery with a short-circuit resistance in the i-th SOC interval is expressed as:

[0092] x(k + 1) = Ax(k) + BI batt (k) + B f I sc (k)

[0093] y(k) = Cx(k) + DI batt (k) + D f I sc (k) (6)

[0094] where, B f = B; C = [-1 -1 a i ; D = D f = -R0. The state vector is x(k) = [V1(k), V1(k), soc(k)]'. The model output is y(k) = V batt (k) - b i .

[0095] Considering the modeling error and measurement noise, the battery short-circuit model equation (6) can be further extended to equation (7), and its state equation and output equation both consider bounded disturbances:

[0096] x(k + 1) = Ax(k) + B f f(k) + Bd d(k)

[0097] y(k) = Cx(k) + Du(k) + D f f(k) + D d d(k) (7)

[0098] where x(k) ∈ R n is the state vector; y(k) ∈ R p is the output; u(k) ∈ R m is the known input, corresponding to I batt ; is the battery fault, corresponding to I sc ; is the disturbance belonging to l2[0, ∞]; B d and D d are constant real matrices of appropriate dimensions.

[0099] In an embodiment of a lithium battery short - circuit fault detection method provided in this application, the lithium battery short - circuit fault detection method further includes:

[0100] Obtaining the fuzzy observer according to the battery voltage state - of - charge curve and a preset robust observer.

[0101] In this embodiment, a method of hybridizing the linear intervals of the OCV - SOC curve and constructing a weighted function self - adjusting fault estimator is proposed. The establishment of the TS fuzzy system includes the modeling and observer / controller design processes, and uses a set of local LTI models to simulate the nonlinear system. These models are interpolated using nonlinear weighted functions or membership functions and can fuse all linear subsystems.

[0102] In an embodiment of a lithium battery short - circuit fault detection method provided in this application, obtaining the fuzzy observer according to the battery voltage state - of - charge curve and a preset robust observer includes:

[0103] Obtaining the optimal weighted function of the state - of - charge of the battery according to the battery voltage state - of - charge curve and a preset Gaussian function;

[0104] Obtaining the fuzzy observer according to the optimal weighted function and the robust observer.

[0105] In this embodiment, the overall design process of this fuzzy observer is as Figure 5 shown. First, the OCV - SOC curve is fuzzified. Figure 6 The fuzzification process of the battery OCV - SOC curve is shown.

[0106] Figure 6(b): Consider \(g\) (\(g\in N\) and \(g\geq2\)) linear segments in the OCV - SOC curve, and fit the curve of each linear segment into the form of Equation (3). The determination of \(g\) depends on the non - linear level of the OCV - SOC curve.

[0107]

[0108] When using this method, the linear segments do not need to be continuous. This method has two main advantages: compared with the classical model using small discrete steps, this model is simpler; this model has better robustness because this model does not require a strong approximation of the linearity of the whole curve.

[0109] Figure 6 (c): Assign membership functions to each linear segment, that is, each linear segment is weighted by the function \(\pi(SOC)\) to reflect the degree of influence on the global non - linear behavior of the OCV - SOC curve. In addition, in order to achieve a smooth transition between different linear segments, \(\pi\) is selected as a Gaussian - type function. Therefore, \(\pi(Soc)\) is determined by its mean value \(\mu\) and variance \(\sigma\). 2 For any value of \(soc\), assume:

[0110] \(\pi\) i (\(soc\)) \(\geq0\) and

[0111] Figure 6 (d): After the above two steps, the OCV - SOC curve can be expressed as:

[0112]

[0113] Where: Therefore, for any SOC value, \(h\) i (\(Soc\)) satisfies:

[0114] \(h\) i (\(soc\)) \(\geq0\) and

[0115] Figure 6 (e): Select the optimal \((\mu,\sigma\) i \()\) for \(\pi\) of each membership function (\(i = 1,2,\cdots,g\)). 2 )

[0116] Design a fault estimator using the following theorem:

[0117] Theorem: Let a specified performance level of \(H\) be \(\gamma\), and a circular domain be given. ∞ If there exist two symmetric positive - definite matrices and two matrices Satisfy Equation (10) and Equation (11):​

[0118]

[0119]

[0120] When

[0121] represents the symmetric terms of the matrix, followed by the error dynamics (9)

[0122] where represents the symmetric terms of the matrix. The error dynamic model equation (9) satisfies H ∞ performance index The eigenvalues belong to gain matrix is determined by In short, the design of the linear fault estimator is subject to two linear matrix inequalities. By adjusting the circular domain the observer performance can be adjusted.

[0123] Therefore, according to the previously proposed method, a robust observer is first established for each linear segment. That is, the gain vector of each sub-observer is determined Then, based on equation (8), g linear robust observers can be directly mixed into equation (12), where the subscript (·)fuzzy represents the elements of the TS fuzzy observer.

[0124]

[0125]

[0126]

[0127] Two important points should be noted. First, according to the design method of the TS fuzzy system, the optimal MF of different linear observers is exactly the same as that in the fuzzy process of the battery OCV-SOC curve

[31] . Second, that is, the slow change characteristic of SOC makes where is included in the estimated state vector Therefore, is the so-called fuzzy variable in the TS fuzzy observer of the battery equivalent circuit model, and the obtained TS fuzzy observer is essentially a weighted function self-regulating robust observer. The schematic diagram of the obtained TS fuzzy observer is as Figure 7 shown.

[0128] In an embodiment of a lithium battery short circuit fault detection method provided in this application, the obtaining of the optimal weighted function of the battery state of charge according to the battery voltage state of charge curve and a preset Gaussian function includes:

[0129] Based on the battery voltage state of charge curve and a preset Gaussian function, obtain the optimization coefficient of the Gaussian function;

[0130] Based on the optimization coefficient, determine the optimal parameters of the Gaussian function, and further determine the optimal weighted function of the battery state of charge.

[0131] In this embodiment, Figure 6 (e): For π of each membership function i (i = 1, 2,..., g), select the optimal (μ, σ 2 ). The determination coefficient R used as the optimization criterion 2 is calculated as follows:

[0132]

[0133] where n ocv is the number of data points used in the optimization process; V OC (soc) is the original average OCV - SOC data, as Figure 6 (a) shows. Finally, as Figure 6 (f) shows, when R 2 is infinitely close to 1, obtain the optimal (μ, σ 2 ) of the membership function, and the optimal membership function obtained after the optimization process will be used for subsequent observer fusion.

[0134] Second, in order to timely diagnose the battery thermal runaway fault of a large - scale energy storage power station, this application provides an embodiment of a lithium - battery short - circuit fault detection system. Refer to Figure 8 , the lithium - battery short - circuit fault detection system specifically includes the following:

[0135] Parameter calculation module 01: Input the state of charge of the current battery at all times into a preset battery short - circuit equivalent model, and the battery short - circuit equivalent model outputs the battery state parameters of the next moment of the battery;

[0136] Parameter estimation module 02: For the current lithium - battery, use a fuzzy observer to estimate the battery state estimation parameters of the battery;

[0137] Fault detection module 03: Determine the short - circuit fault of the battery according to the residual between the battery state parameters and the battery state estimation parameters.

[0138] In this embodiment, the parameter calculation module 01 is used to establish an equivalent battery short - circuit model and obtain the actual values of battery parameters through the equivalent battery short - circuit model. The parameter estimation module 02 is used to design a fuzzy observer and obtain the estimated values of battery parameters through the fuzzy observer. The fault detection module 03 calculates the residual between the actual values of battery parameters of the parameter calculation module 01 and the estimated values of battery parameters of the parameter estimation module 02, and determines the battery fault status through the residual. When the residual is 1, the battery has a fault; when the residual is 0, the battery is operating normally.

[0139] As can be seen from the above description, for a lithium - battery short - circuit fault detection system provided by an embodiment of the present application, in order to detect the tiny changes hidden in environmental noise in the fault characteristics, statistical information (cumulative sum) is combined with a model - based observer fault estimation for the early diagnosis method of battery short - circuit. Specifically, an equivalent battery short - circuit model and a battery fuzzy observer are combined and used. According to the residual between the actual value of the state of charge of the battery measured by the equivalent battery short - circuit model and the estimated value of the state of charge of the battery estimated by the fuzzy observer, the battery short - circuit fault is determined. This fault detection method has the characteristics of model universality, design universality, and implementation universality.

[0140] At the hardware level, in order to diagnose the battery thermal runaway fault of a large - scale energy storage power station in a timely manner, an embodiment of an electronic device for all or part of the content of a lithium - battery short - circuit fault detection method provided by the present application is as follows:

[0141] Figure 9 This is a schematic block diagram of the system composition of the electronic device 9600 according to an embodiment of the present application. As Figure 9 shown, the electronic device 9600 may include a central processing unit 9100 and a memory 9140; the memory 9140 is coupled to the central processing unit 9100. It should be noted that this Figure 9 is exemplary; other types of structures can also be used to supplement or replace this structure to implement telecommunication functions or other functions.

[0142] In one embodiment, the lithium - battery short - circuit fault detection function can be integrated into the central processing unit. Among them, the central processing unit can be configured to perform the following controls:

[0143] Step 100: Input the state of charge of the current battery at all times into a preset equivalent battery short - circuit model, and the equivalent battery short - circuit model outputs the battery state parameters of the next moment of the battery;

[0144] Step 200: For the current lithium - battery, use a fuzzy observer to estimate the battery state estimation parameters of the battery;

[0145] Step 300: Determine the short - circuit fault of the battery according to the battery state parameters and the residual of the battery state estimation parameters.

[0146] In this embodiment, considering the slow - change characteristic of the battery SOC, a systematic method is proposed based on the genetic algorithm to construct a self - regulating mechanism to cope with the non - linear open - circuit voltage (OCV) - SOC curve. To detect the tiny changes hidden in the environmental noise in the fault characteristics, the statistical information (cumulative sum) is combined with the model - based observer fault estimation. The overall strategy architecture is as Figure 2 shown. First, establish a battery short - circuit equivalent model, and then fuzzify the OCV - SOC curve and the robust observer to obtain a fuzzy observer.

[0147] The fuzzy observer is essentially a weighted - function self - regulating robust observer. The estimated values of the battery parameters are obtained through the fuzzy observer. The battery short - circuit equivalent model is as Figure 3 . The resistance R0 represents the ohmic resistance, which includes the resistance of the contacts, electrodes, and electrolyte. The double - RC circuit characterizes the charge - transfer effect, diffusion effect, and double - layer behavior inside the lithium - ion battery and can simulate the transient response of the battery. In addition, compared with the single - RC and triple - RC structures, the double - RC network is a good trade - off between model error and model complexity. The actual values of the battery parameters are obtained through the battery short - circuit equivalent model. Whether the battery is in a fault state is judged by the residual between the actual values and the estimated values of the battery parameters. If the residual is 0, the battery operates normally; if the residual is 1, the battery is in a short - circuit fault.

[0148] Since i in = batt + sc , the battery equivalent circuit model with a short - circuit resistance in the i - th SOC interval is expressed as: x(k + 1)=Ax(k)+B f f(k)+B d d(k)

[0149] y(k)=Cx(k)+Du(k)+D f f(k)+D d d(k) (7)

[0150] where, x(k)∈R n is the state vector; y(k)∈R p is the output; u(k)∈R m is the known input, corresponding to I batt ; is the battery fault, corresponding to I sc ; is the disturbance belonging to l2[0, ∞]; B d and D dis a constant real matrix of appropriate dimension.

[0151] The fault estimator is designed as a proportional-integral observer. The integral term can not only ensure robust state estimation when a fault occurs, but also provide fault estimation simultaneously. The fuzzy observer can be expressed as:

[0152]

[0153]

[0154]

[0155] where is the estimated state vector; is the observer output; is the estimated f(k); L ∈ R n×p and; is the observer gain.

[0156] The error dynamics between the model expression (7) and the observer expression (8) can be described by Equation (9):

[0157]

[0158] where:

[0159]

[0160]

[0161]

[0162] and △f(k) = f(k + 1) - f(k) is the symbol belonging to l2[0, ∞] Ii is the identity matrix with i × i dimension. 0 is the zero matrix with the corresponding dimension.

[0163] The design principle of the observer is to determine so that the error dynamics model Equation (9) satisfies the following two objectives:

[0164] (1) is Hurwitz stable. The eigenvalues of its discrete-time system are inside the unit circle;

[0165] (2) The fault estimation error e f (k) is insensitive to i.e., the smaller e f (k) is, the better.

[0166] As can be seen from the above description, an electronic device provided by an embodiment of the present application combines statistical information (cumulative sum) with a model-based observer fault estimation to detect minute changes hidden in environmental noise in fault features, for an early battery short-circuit diagnosis method. Specifically, it combines a battery short-circuit equivalent model and a battery fuzzy observer, and determines a battery short-circuit fault based on the residual between the actual value of the state of charge of the battery measured by the battery short-circuit equivalent model and the estimated value of the state of charge of the battery estimated by the fuzzy observer. This fault detection method has the characteristics of model universality, design universality, and implementation universality.

[0167] In another embodiment, the lithium battery short-circuit fault detection device can be separately configured from the central processing unit 9100. For example, the lithium battery short-circuit fault detection device can be configured as a chip connected to the central processing unit 9100, and the lithium battery short-circuit fault detection function is realized through the control of the central processing unit.

[0168] As Figure 9 shown, the electronic device 9600 may further include: a communication module 9110, an input unit 9120, an audio processor 9130, a display 9160, and a power supply 9170. It should be noted that the electronic device 9600 does not necessarily have to include Figure 9 all the components shown in Figure 9 ; in addition, the electronic device 9600 may further include Figure 9 components not shown in

[0169] As Figure 9 shown, the central processing unit 9100 is sometimes also referred to as a controller or an operation control, and may include a microprocessor or other processor devices and / or logic devices. The central processing unit 9100 receives inputs and controls the operations of the various components of the electronic device 9600.

[0170] Among them, the memory 9140, for example, can be one or more of a buffer, a flash memory, a hard drive, a removable medium, a volatile memory, a non-volatile memory, or other suitable devices. It can store the above information related to failures, and can also store programs for executing relevant information. And the central processing unit 9100 can execute the program stored in the memory 9140 to implement information storage or processing, etc.

[0171] The input unit 9120 provides inputs to the central processing unit 9100. The input unit 9120 is, for example, a key or a touch input device. The power supply 9170 is used to provide power to the electronic device 9600. The display 9160 is used for displaying display objects such as images and texts. The display can be, for example, an LCD display, but is not limited thereto.

[0172] The memory 9140 may be a solid-state memory, for example, a read-only memory (ROM), a random access memory (RAM), a SIM card, etc. It may also be a memory that stores information even when power is off, can be selectively erased and has more data. An example of such a memory is sometimes referred to as an EPROM, etc. The memory 9140 may also be some other type of device. The memory 9140 includes a buffer memory 9141 (sometimes referred to as a buffer). The memory 9140 may include an application / function storage unit 9142 for storing application programs and function programs or the processes for operating the electronic device 9600 by the central processing unit 9100.

[0173] The memory 9140 may also include a data storage unit 9143 for storing data, such as contacts, digital data, pictures, sounds, and / or any other data used by the electronic device. The driver storage unit 9144 of the memory 9140 may include various drivers of the electronic device for communication functions and / or for performing other functions of the electronic device (such as a messaging application, an address book application, etc.).

[0174] The communication module 9110 is a transmitter / receiver 9110 that transmits and receives signals via the antenna 9111. The communication module (transmitter / receiver) 9110 is coupled to the central processing unit 9100 to provide input signals and receive output signals, which may be the same as in the case of a conventional mobile communication terminal.

[0175] Based on different communication technologies, multiple communication modules 9110 may be provided in the same electronic device, such as a cellular network module, a Bluetooth module, and / or a wireless local area network module, etc. The communication module (transmitter / receiver) 9110 is also coupled to the speaker 9131 and the microphone 9132 via the audio processor 9130 to provide an audio output via the speaker 9131 and receive an audio input from the microphone 9132, thereby implementing normal telecommunication functions. The audio processor 9130 may include any suitable buffer, decoder, amplifier, etc. In addition, the audio processor 9130 is also coupled to the central processing unit 9100, so that recording can be performed on the local machine through the microphone 9132, and the sound stored on the local machine can be played through the speaker 9131.

[0176] Embodiments of the present application also provide a computer-readable storage medium capable of implementing all steps in the lithium battery short-circuit fault detection method in the above embodiments. A computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, it implements all steps of the lithium battery short-circuit fault detection method with the execution subject being a server or a client in the above embodiments. For example, when the processor executes the computer program, the following steps are implemented:

[0177] Step 100: Input the state of charge (SOC) of the current battery at all times into a preset battery short - circuit equivalent model, and the battery short - circuit equivalent model outputs the battery state parameters of the next moment of the battery.

[0178] Step 200: For the current lithium - ion battery, use a fuzzy observer to estimate the battery state estimation parameters of the battery.

[0179] Step 300: Determine the short - circuit fault of the battery according to the battery state parameters and the residual of the battery state estimation parameters.

[0180] In this embodiment, considering the slow - change characteristic of the battery SOC, a systematic method is proposed based on the genetic algorithm to construct a self - regulating mechanism to cope with the non - linear open - circuit voltage (OCV) - SOC curve. In order to detect the tiny changes hidden in the environmental noise in the fault characteristics, statistical information (cumulative sum) is combined with the model - based observer fault estimation. The overall strategy architecture is as Figure 2 shown. First, establish a battery short - circuit equivalent model, and then fuzzify the OCV - SOC curve and the robust observer to obtain a fuzzy observer.

[0181] The fuzzy observer is essentially a weighted - function self - regulating robust observer. The battery parameter estimation value of the battery is obtained through the fuzzy observer. The battery short - circuit equivalent model is as Figure 3 . The resistance R0 represents the ohmic resistance, which includes the resistance of the contacts, electrodes, and electrolyte. The double - RC circuit characterizes the charge transfer effect, diffusion effect, and double - layer behavior inside the lithium - ion battery and can simulate the transient response of the battery. In addition, compared with the single - RC and triple - RC structures, the double - RC network is a good trade - off between model error and model complexity. The actual value of the battery parameters of the battery is obtained through the battery short - circuit equivalent model. Whether the battery is in a fault state is judged by the residual between the actual value and the estimated value of the battery parameters. If the residual is 0, the battery operates normally; if the residual is 1, the battery is in a short - circuit fault.

[0182] Since I in =I batt +I sc , the equivalent circuit model of the battery with a short - circuit resistance in the i - th SOC interval is expressed as: x(k + 1) = Ax(k)+B f f(k)+B d d(k)

[0183] y(k) = Cx(k)+Du(k)+D f f(k)+D d d(k) (7)

[0184] where, x(k)∈Rn is the state vector; y(k) ∈ R p is the output; u(k) ∈ R m is the known input, corresponding to I batt ; is the battery fault, corresponding to I sc ; is the disturbance belonging to l2[0, ∞]; B d and D d are constant real matrices of appropriate dimensions.

[0185] The fault estimator is designed as a proportional-integral observer. The integral term can not only ensure robust state estimation when a fault occurs, but also provide fault estimation simultaneously. The fuzzy observer can be expressed as:

[0186]

[0187]

[0188]

[0189] where, is the estimated state vector; is the observer output; is the estimated f(k); L ∈ R n×p and; is the observer gain.

[0190] The error dynamics between the model expression (7) and the observer expression (8) can be described by Equation (9):

[0191]

[0192] where:

[0193]

[0194]

[0195]

[0196] and △f(k) = f(k + 1) - f(k) is the symbol belonging to l2[0, ∞] Ii is the identity matrix with i×i dimensions. 0 is the zero matrix with the corresponding dimensions.

[0197] The design principle of the observer is to determine so that the error dynamics model Equation (9) satisfies the following two objectives:

[0198] (1) is Hurwitz stable. The eigenvalues of its discrete-time system are inside the unit circle;

[0199] (2) The fault estimation error e f (k) is insensitive to , that is, the smaller e f (k) is, the better.

[0200] As can be seen from the above description, a computer-readable storage medium provided by an embodiment of the present application combines statistical information (cumulative sum) with model-based observer fault estimation to detect minute changes hidden in environmental noise in fault features, and is used for an early diagnosis method of battery short circuit. Specifically, it combines a battery short-circuit equivalent model and a battery fuzzy observer, and determines a battery short-circuit fault according to the residual between the actual value of the state of charge of the battery measured by the battery short-circuit equivalent model and the estimated value of the state of charge of the battery estimated by the fuzzy observer. This fault detection method has the characteristics of model universality, design universality, and implementation universality.

[0201] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a device, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing computer-usable program codes.

[0202] The present invention is described with reference to the flowcharts and / or block diagrams of methods, devices (apparatus), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, and the combination of processes and / or blocks in the flowcharts and / or block diagrams 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 devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate means for realizing the specified functions in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0203] These computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable data processing devices to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured product including an instruction device, and the instruction device realizes the specified functions in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0204] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus, so that a series of operation steps are executed on the computer or other programmable apparatus to produce a computer-implemented process, thereby the instructions executed on the computer or other programmable apparatus provide steps for implementing the functions specified in one process or multiple processes and / or blocks Figure 1 one process or multiple processes and / or blocks Figure 1 steps for implementing the functions specified in one block or multiple blocks.

[0205] In the present invention, specific embodiments are used to illustrate the principles and implementation manners of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; at the same time, for those of ordinary skill in the art, according to the idea of the present invention, there will be changes in the specific implementation manners and application scope. In summary, the content of this specification should not be construed as a limitation to the present invention.

Claims

1. A lithium battery short circuit fault detection method, characterized in that: include: Inputting the state of charge of the current battery at all times into a preset battery short-circuit equivalent model, the battery short-circuit equivalent model outputting the battery state parameter of the battery at the next moment; For the current lithium battery, a fuzzy observer is used to estimate the battery state estimation parameters of the battery; determining a short circuit fault of the battery according to the battery state parameter and a residual of the battery state estimation parameter; The lithium battery short circuit fault detection method further includes: Modeling and calculating all charge states of a sample battery to obtain the battery short-circuit equivalent model; The lithium battery short circuit fault detection method further includes: Obtaining the fuzzy observer according to the battery voltage state of charge curve and a preset robust observer; The fuzzy observer is obtained according to the battery voltage state of charge curve and a preset robust observer, including: Obtaining an optimal weighting function for the battery state of charge according to the battery voltage state of charge curve and a preset Gaussian function; Obtaining the fuzzy observer according to the optimal weighting function and the robust observer; Obtaining an optimal weighting function of the battery state of charge according to the battery voltage state of charge curve and a preset Gaussian function includes: Obtaining an optimization coefficient of the Gaussian function according to the battery voltage state of charge curve and a preset Gaussian function; According to the optimization coefficient, the optimal parameters of the Gaussian function are determined, and then the optimal weighting function of the battery state of charge is determined.

2. The lithium battery short circuit fault detection method according to claim 1, characterized in that: The modeling and calculation for all charge states of a sample battery to obtain the battery short-circuit equivalent model includes: Obtaining a state of charge of the sample battery at a next moment according to the load current of the sample battery at a current moment and the initial state of charge of the sample battery; A battery short-circuit equivalent model is established according to the state of charge of the sample battery at a current moment and the state of charge at a next moment.

3. The lithium battery short circuit fault detection method according to claim 1, characterized in that: The obtaining, based on the load current of the sample battery at the current moment and the initial state of charge of the sample battery, the state of charge of the sample battery at the next moment includes: Integrating the load current of the sample battery at the current moment to obtain a current current integral value; The state of charge of the sample battery at the next moment is obtained according to the initial state of charge and the current current integral value.

4. A lithium battery short circuit fault detection system, based on the lithium battery short circuit fault detection method according to any one of claims 1 to 3, characterized in that: include: Parameter calculation module: inputs the state of charge of the current battery at all times into a preset battery short-circuit equivalent model, and the battery short-circuit equivalent model outputs the battery state parameters of the battery at the next moment; Parameter estimation module: for the current lithium battery, using fuzzy observer to estimate the battery state estimation parameters of the battery; A fault detection module is configured to determine a short circuit fault of the battery based on the battery state parameter and a residual of the battery state estimation parameter.

5. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the lithium battery short circuit fault detection method according to any one of claims 1 to 3 is implemented.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the lithium battery short circuit fault detection method according to any one of claims 1 to 3 is implemented.

Images