CN117741477A

CN117741477A - Method and device for detecting short circuit fault in battery

Info

Publication number: CN117741477A
Application number: CN202410023322.1A
Authority: CN
Inventors: 李斌; 李冠争; 李超; 王帅; 曾锴迪; 徐科; 张智达
Original assignee: Tianjin University; State Grid Tianjin Electric Power Co Ltd
Current assignee: Tianjin University; State Grid Tianjin Electric Power Co Ltd
Priority date: 2024-01-08
Filing date: 2024-01-08
Publication date: 2024-03-22

Abstract

The present disclosure provides a method and apparatus for detecting a short circuit fault in a battery. The method comprises the following steps: determining model attribute information of an equivalent circuit model related to the battery; determining an intermediate battery cell according to the voltage values of all battery cells in the battery; calculating a target state of charge of the target battery cell based on model attribute information of the equivalent circuit model, a voltage value of the target battery cell and a current value of the target battery cell; calculating the intermediate state of charge of the intermediate battery cell based on model attribute information of the equivalent circuit model, a voltage value of the intermediate battery cell and a current value of the intermediate battery cell; calculating the Euclidean distance between the target battery cell and the middle battery cell based on the target state of charge and the voltage value of the target battery cell, and the voltage value of the middle state of charge and the voltage value of the middle battery cell; and determining the internal short circuit fault information of the target battery cell according to the Euclidean distance between the target battery cell and the middle battery cell.

Description

Method and device for detecting short circuit fault in battery

Technical Field

The disclosure relates to the technical field of battery detection, and in particular relates to a method and a device for detecting short circuit faults in a battery.

Background

The short circuit fault in the battery refers to the short circuit fault between the anode and the cathode in the battery or in the electrolyte in the battery, and the short circuit fault can cause the problems of overheating and leakage of the battery, and even cause safety accidents such as explosion or fire disaster. For example, when an internal short circuit fault occurs in the lithium ion battery, a micro channel is formed between the cathode and the anode of the lithium ion battery, so that some electrons in the cathode metal oxide directly flow to the anode, and the electrons directly flowing to the anode may cause confusion of the electron flow in the battery, thereby causing overcharge, overdischarge or overheat of the battery, and further increasing the risk of explosion or fire and other safety accidents of the battery.

However, in the related technology for detecting the internal short-circuit fault of the battery, since the characteristics of the internal short-circuit fault of the battery at the initial stage are often not obvious enough, the situation that the internal short-circuit fault of the battery is difficult to detect in time easily occurs, and thus, corresponding treatment measures are difficult to take in time to eliminate potential safety hazards. This actually increases the risk of causing a series of problems such as degradation of battery performance, shortening of battery life, degradation of battery safety, and the like. In view of the foregoing, there is a need for a solution that can detect short-circuit faults in batteries more sensitively and rapidly.

Disclosure of Invention

To address at least one or more of the technical problems mentioned above, the present disclosure proposes, in one or more aspects, a method and apparatus for detecting a short circuit fault in a battery.

In a first aspect, the present disclosure provides a method of detecting a short circuit fault in a battery. The method comprises the following steps: determining model attribute information of an equivalent circuit model related to the battery; determining an intermediate battery cell according to the voltage values of all battery cells in the battery, wherein the battery comprises a plurality of battery cells, and the voltage value of the intermediate battery cell is the median of the voltage values of all battery cells; calculating a target state of charge of the target battery cell based on model attribute information of the equivalent circuit model, a voltage value of the target battery cell and a current value of the target battery cell, wherein the target battery cell is any battery cell in all battery cells; calculating the intermediate state of charge of the intermediate battery cell based on model attribute information of the equivalent circuit model, a voltage value of the intermediate battery cell and a current value of the intermediate battery cell; calculating the Euclidean distance between the target battery cell and the middle battery cell based on the target state of charge and the voltage value of the target battery cell, and the voltage value of the middle state of charge and the voltage value of the middle battery cell; and determining the internal short circuit fault information of the target battery cell according to the Euclidean distance between the target battery cell and the middle battery cell.

In a second aspect, the present disclosure provides an apparatus for detecting a short circuit fault in a battery, the apparatus comprising: an equivalent circuit model determination module for determining model attribute information of an equivalent circuit model of the battery; the middle battery cell determining module is used for determining the middle battery cell according to the voltage values of all battery cells in the battery, wherein the voltage value of the middle battery cell is the median of the voltage values of all battery cells; a target state of charge calculation module for calculating a target state of charge of the target battery cell based on model attribute information of the equivalent circuit model, a voltage value of the target battery cell, and a current value of the target battery cell; the middle charge state calculation module is used for calculating the middle charge state of the middle battery cell based on the model attribute information of the equivalent circuit model, the voltage value of the middle battery cell and the current value of the middle battery cell; the Euclidean distance calculating module is used for calculating the Euclidean distance between the target battery monomer and the middle battery monomer based on the target state of charge and the voltage value of the target battery monomer, and the middle state of charge and the voltage value of the middle battery monomer, wherein the target battery monomer is any battery monomer in all battery monomers; and the internal short circuit fault information determining module is used for determining the internal short circuit fault information of the target battery cell according to the Euclidean distance between the target battery cell and the middle battery cell.

According to the method and the device for detecting the short circuit fault in the battery, the middle battery cell determined through the voltage value can reflect the situation of the normal battery cell, and the Euclidean distance between the target battery cell and the middle battery cell can be calculated based on the determined model attribute information of the internal short circuit model related to the battery, the voltage value and the current value of the target battery cell and the voltage value and the current value of the middle battery cell. The method and the device not only can integrate the two parameters of the charge state and the voltage, more comprehensively, sensitively and quickly analyze the battery state of the target battery cell and the battery state of the middle battery cell, but also can detect the internal short circuit fault information of the target battery cell more effectively by calculating the Euclidean distance between the target battery cell and the middle battery cell, namely by calculating the Euclidean distance between the target battery cell and the normal battery cell. In addition, the method and the device have small calculated amount of Euclidean distance, and have small calculation resources which are needed to be consumed for diagnosing the internal short circuit faults of all the battery monomers, thereby being beneficial to reducing the burden of hardware.

In addition, the method and the device for detecting the internal short circuit fault of the battery, provided by the embodiment of the invention, comprehensively adopt two parameters of the charge state and the voltage, and compared with the single charge state or the single voltage, the method and the device not only can provide more comprehensive battery state information, can more rapidly identify the abnormal characteristics of the battery, but also have higher stability and environmental adaptability, so that the internal short circuit fault of the battery can be more sensitively and more rapidly detected.

Drawings

The above, as well as additional purposes, features, and advantages of exemplary embodiments of the present disclosure will become readily apparent from the following detailed description when read in conjunction with the accompanying drawings. Several embodiments of the present disclosure are illustrated by way of example, and not by way of limitation, in the figures of the accompanying drawings and in which like reference numerals refer to similar or corresponding parts and in which:

fig. 1 shows a schematic flow chart of a method of detecting an intra-battery short fault according to one embodiment of the present disclosure.

Fig. 2 shows an equivalent circuit model circuit schematic diagram of a battery according to an embodiment of the present disclosure.

Fig. 3 shows a schematic flow chart of a method of detecting an intra-battery short fault according to another embodiment of the present disclosure.

Fig. 4 shows a schematic flow chart of determining model attribute information of an equivalent circuit model related to a battery shown in fig. 3.

Fig. 5A shows the euclidean distance results for an internal short circuit condition where a cell is selected to be connected in parallel with a 100 Ω resistor in one embodiment of the present disclosure.

Fig. 5B shows a euclidean distance result in the case of an internal short circuit with a selected cell parallel 51 Ω resistor in accordance with another embodiment of the present disclosure.

Fig. 5C shows a euclidean distance result in the case of an internal short circuit where a cell is selected to be connected in parallel with a 10Ω resistor in accordance with another embodiment of the present disclosure.

Fig. 6 shows a schematic block diagram of an apparatus for detecting a short-circuit fault in a battery according to an embodiment of the present disclosure.

Detailed Description

The following description of the technical solutions in the embodiments of the present disclosure will be made clearly and completely with reference to the accompanying drawings in the embodiments of the present disclosure, and it is apparent that the described embodiments are some embodiments of the present disclosure, but not all embodiments. Based on the embodiments in this disclosure, all other embodiments that a person skilled in the art would obtain without making any inventive effort are within the scope of protection of this disclosure.

For the purposes of promoting an understanding of the principles and advantages of the disclosure, reference will now be made to the embodiments illustrated in the drawings and specific language will be used to describe the same.

According to a first aspect of the present disclosure, a method of detecting a short circuit fault in a battery is provided. It will be appreciated that internal short circuit faults may occur in different types of batteries. As just one example, the foregoing battery and the battery referred to hereinafter may be a lithium ion battery or may be a sodium ion battery, and may not be limited thereto.

Fig. 1 shows a schematic flow chart of a method of detecting an intra-battery short fault according to one embodiment of the present disclosure. As shown in FIG. 1, the method 100 includes operations S110-S160.

Operation S110: model attribute information of an equivalent circuit model associated with the battery is determined.

The idea behind building an equivalent circuit model here is to simplify a complex circuit into a simple circuit while maintaining the equivalent properties of the circuit for ease of analysis or design of the circuit. For a battery, the equivalent circuit model is a circuit consisting of electrical components (e.g., resistance, capacitance, inductance, etc.) and a voltage source (or current source) that is capable of approximating the dynamic behavior of the battery. Establishing an equivalent circuit model of the battery is helpful for better understanding of the dynamic behavior of the battery, analyzing the performance of the battery, and the like.

It can be understood that the batteries of different types have different structures, working principles and characteristics, and those skilled in the art can reasonably select the equivalent circuit model of the battery according to actual needs and specific use scenarios. Taking a lithium ion battery as an example, in one embodiment, the equivalent circuit model may be selected as a second-order RC equivalent circuit model. The voltage changes slowly with time when the battery is discharged or charged due to the polarization characteristic of the lithium battery, and the voltage needs to be recovered to a stable value for a relatively long time when the charging and discharging are stopped. This is similar to the dynamic behavior exhibited by a capacitor cell and a resistor of one of the energy storage cells in the circuit after forming an RC loop. Therefore, when an equivalent circuit model of the lithium ion battery is established, the dynamic characteristics of the lithium ion battery can be simulated by adopting the series connection of a direct current resistor and an RC loop. The second-order RC equivalent circuit (namely two RC loops) adopted by the embodiment can meet the precision requirement and can not bring too much calculation burden.

According to an embodiment of the present disclosure, the equivalent circuit model of the battery is a second-order RC equivalent circuit model, which is composed of one voltage source, two RC loops, and one resistor connected in series. As shown in fig. 2, U _OC Represents the open circuit voltage of the battery, R ₀ Representing the ohmic resistance of the cell; r is R ₁ And R is ₂ Represents the polarization internal resistance of the battery, C ₁ And C ₂ Representing the polarization capacitance of the battery; i represents charge and discharge current when the battery works; u (U) _t Representing the terminal voltage of the battery that can be measured directly.

The expression of the equivalent circuit model is as follows:

in formula (0), U ₁ And U ₂ Voltages of two RC loops, U _OC Can be determined from the State of Charge (SOC) of the battery, I and U _t The current and voltage of the battery during operation can be measured by the sensor. In the equivalent circuit model above, I and U _t Can be measured by a sensor, U _oc Is a function of SOC, which can be written as U _oc,t ＝f(SOC _t ) Wherein SOCt is the SOC value of the battery at time t. Thus, if U is determined _oc Functional relation with state of charge SOC and R for the rest ₀ 、R ₁ 、R ₂ 、C ₁ 、C ₂ And carrying out parameter identification to determine the model attribute information of the equivalent circuit model.

Further, as shown in fig. 1, the method 100 further includes:

operation S120: and determining the middle battery cell according to the voltage values of all the battery cells in the battery. The battery comprises a plurality of battery cells, and the middle battery cell is used for representing the condition of a normal battery cell so as to detect the battery cell with internal short circuit fault later.

According to one embodiment of the present disclosure, the voltage value of the middle battery cell may be the median of the voltage values of all battery cells. For example, there are 5 battery cells in one series battery module, and the battery cells 1, 2, 3, 4, and 5 are arranged in order of voltage values from low to high, respectively. The middle cell is the middle cell 3 having a voltage value equal to the voltage value of all the cells. According to another embodiment of the present disclosure, the middle battery cell may also be determined by other ways to be used for characterizing the situation of the normal battery cell, for example, it may be selected as a certain battery cell corresponding to the mode of the voltage values of all battery cells, and may not be limited thereto.

In an application scenario, there are usually a plurality of battery cells in the series battery module, and the middle battery cell determined based on the above manner can not be affected by the highest voltage value and the lowest voltage value, and compared with other battery cells, the situation of the normal battery cells in the series battery module can be better reflected.

Operation S130: and calculating the target state of charge of the target battery cell based on the model attribute information of the equivalent circuit model, the voltage value of the target battery cell and the current value of the target battery cell. The target battery monomer is any battery monomer in all battery monomers. Preferably, the voltage value of the target battery cell may be the lowest value among the voltage values of all battery cells.

The State of Charge (SOC), which as the name implies refers to the available State of Charge remaining in the battery, may represent the ratio of the remaining Charge of the battery to the fully charged Charge, which is typically a percentage. A state of charge SOC of 0% may be considered as a fully discharged battery and a state of charge SOC of 100% may be considered as a fully charged battery.

Based on the equivalent circuit model mentioned above, in this model the values of capacitance and resistance are typically related to the state of charge SOC of the battery. For example only, as the battery discharges (SOC decreases), the electrolyte concentration within the battery may decrease, the active material on the plates becomes depleted, and these changes may cause an increase in the resistance inside the battery. In addition, the movement and distribution of charge can also affect the capacitance of the battery. Therefore, under different state of charge SOC conditions, the internal characteristics of the battery may change, thereby affecting the parameters of the equivalent circuit model.

And (S140) calculating the intermediate charge state of the intermediate battery cell based on the model attribute information of the equivalent circuit model, the voltage value of the intermediate battery cell and the current value of the intermediate battery cell.

In the embodiment of the disclosure, the intermediate state of charge of the intermediate battery cell may be used to reflect the state of charge of the normal battery cell, so as to detect the internal short circuit fault of the target battery cell later.

According to an alternative embodiment of the present disclosure, the target state of charge of the target cell and the intermediate state of charge of the intermediate cell may be calculated based on electrochemical impedance spectra. It can be appreciated that the calculation of the state of charge may be selected by other methods, and those skilled in the art may reasonably select according to actual needs and application scenarios, and the like, which is not limited herein.

Impedance is an important parameter of lithium ion batteries, which varies with battery temperature, state of charge, SOC, aging, etc. Electrochemical impedance spectroscopy (electrochemical impedance spectroscopy, EIS) is a technique for performing impedance measurement by applying small amplitude sinusoidal perturbations of different frequencies to a system, contains impedance information over a very wide frequency range, can reflect information of various complex internal electrochemical reactions and electrode interface structures, has the advantages of non-destructive and in-situ measurement, and is therefore more suitable for calculating intermediate states of charge and intermediate states of charge.

Operation S150 calculates a euclidean distance between the target battery cell and the intermediate battery cell based on the target state of charge and the voltage value of the target battery cell, and the intermediate state of charge and the voltage value of the intermediate battery cell.

State of charge, SOC, and voltage are relatively real-time battery parameters, the changes of which can generally be reacted in a short time. In the initial stage of internal short circuit of the battery, both the voltage change caused by the internal short circuit fault and the state of charge (SOC) change can be detected rapidly, and by detecting the two parameters simultaneously, the abnormal state of the battery can be identified more sensitively and rapidly, so that the internal short circuit fault of the battery can be detected more sensitively and accurately. In addition, when the battery is internally short-circuited, the change of the state of charge (SOC) and the change of the voltage are generally complementary, and the comprehensive analysis of the state of charge (SOC) and the voltage of the battery is facilitated, so that the state of the battery can be more comprehensively, stably and accurately analyzed, and the internal short-circuit fault of the battery can be more accurately detected.

Euclidean distance (Euclidean Distance) is typically used to measure the true distance between two points in a multidimensional space. In the embodiment of the disclosure, the euclidean distance between the target battery cell and the middle battery cell is used for measuring the real distance between the target battery cell and the middle battery cell, namely the real distance between the target battery cell and the normal battery cell.

Operation S160: and determining the internal short circuit fault information of the target battery cell according to the Euclidean distance between the target battery cell and the middle battery cell.

In some application scenarios, euclidean distance may be used to reflect the similarity or difference between features of two samples. A smaller euclidean distance generally indicates greater similarity between the two samples, and a larger euclidean distance generally indicates greater variability between the two samples.

In the embodiment of the present disclosure, the larger the euclidean distance between the target battery cell and the intermediate battery cell, the larger the similarity between the target battery cell and the intermediate battery cell, that is, the closer the target battery cell is to the normal battery cell. The smaller the euclidean distance between the target cell and the intermediate cell, the greater the difference between the target cell and the intermediate cell, i.e., the closer the target cell is to the internal short-circuit cell.

According to the method for detecting the short circuit fault in the battery, the Euclidean distance between the target battery monomer and the middle battery monomer is calculated based on the target state of charge and the voltage value of the target battery monomer, and the middle state of charge and the voltage value of the middle battery monomer. The method can not only integrate the two parameters of the charge state and the voltage, more comprehensively, sensitively and rapidly analyze the battery state of the target battery cell and the battery state of the middle battery cell, but also can more effectively detect the internal short circuit fault information of the target battery cell by calculating the Euclidean distance between the target battery cell and the middle battery cell, namely by calculating the Euclidean distance between the target battery cell and the normal battery cell. In addition, the method has small calculated amount of Euclidean distance, and less calculation resource is needed to be consumed for diagnosing the internal short circuit faults of all the battery monomers, thereby being beneficial to reducing the burden of hardware.

In addition, the method for detecting the internal short circuit fault of the battery provided by the embodiment of the disclosure comprehensively adopts the two parameters of the state of charge and the voltage, so that compared with the state of charge or the voltage adopted independently, the method can provide more comprehensive state information of the battery, can more rapidly identify the abnormal characteristics of the battery, has higher stability and environmental adaptability, and can more sensitively and rapidly detect the internal short circuit fault of the battery.

Fig. 3 shows a schematic flow chart of a method of detecting an intra-battery short fault according to another embodiment of the present disclosure. It is to be understood and appreciated that the method 300 shows more details of the method 100 and can be considered a viable or preferred embodiment of the method 100, and thus the description above with respect to the method 100 applies equally to the method 300 shown in fig. 3 below. As further shown in FIG. 3, the method 300 includes operations S310-S380.

Operation S310: model attribute information of an equivalent circuit model associated with the battery is determined. The model attribute information comprises fitting relation attributes and parameter value information of an equivalent circuit model.

Fig. 4 shows a schematic flow chart of determining model attribute information of an equivalent circuit model related to a battery shown in fig. 3. According to an embodiment of the present disclosure, as shown in fig. 4, the above operation S310 includes operations S311 to S313.

Operation S311: voltage response information corresponding to a state of charge of the battery is acquired.

According to embodiments of the present disclosure, electrochemical impedance spectroscopy tests may be performed on the battery using an electrochemical workstation to obtain voltage response information related to state of charge, SOC. When the electrochemical impedance spectrum test is carried out, a small-amplitude alternating current excitation signal (such as a sine wave current signal) can be applied according to a sine rule when the battery is in an open circuit state, and the voltage response of the battery at different time scales can be studied by changing the frequency of the sine wave current, so that information about the internal dynamic characteristics of the battery and the like can be obtained by observing the relation between the input current and the voltage response.

Specifically, a small amplitude sine wave current signal i can be injected into the battery under the selected state of charge SOC condition to obtain the voltage response U of the battery _t Wherein the kth value of the current signal i is denoted as i _k Voltage response U _t Is denoted as U _t,k . For example, the selected state of charge SOC may include 0%,5%,10%,15%, …,100%. More preferably, the frequency of the sine wave current signal i is 25mHz-1kHz.

It should be noted that the state of charge SOC is an indicator for measuring the amount of charge stored in the battery, and thus the internal electrochemical state of the battery may change with the change of the state of charge SOC, thereby affecting the voltage response of the battery. According to embodiments of the present disclosure, the voltage response U of the battery under selected SOC conditions can be obtained in the above manner _t 。

Operation S312: and processing the voltage response information according to a recursive least square algorithm to obtain the parameter value information of the equivalent circuit model.

It will be appreciated that the impedance spectrum is the current signal i and the voltage response U _t By spectral analysis of the phase differences and amplitudes of the current and voltage, information related to the parameter value information of the equivalent circuit model of the battery can be deduced, while by electrochemical impedance spectroscopy testing of the battery under selected SOC conditions, it can be determined that the battery is under selected SOC conditionsInformation related to parameter value information of the equivalent circuit model.

According to an embodiment of the present disclosure, the method for performing parameter identification on the equivalent circuit model may be selected as a recursive least square method, and of course, those skilled in the art may reasonably select other parameter identification methods according to actual needs, application scenarios, and the like, which is not limited herein. In the following, only a recursive least square method is taken as an example, and the equivalent circuit model is subjected to parameter identification.

Based on the equivalent circuit model formula in the formula (0), the electrical quantity characteristics can be expressed as:

SOC based on ampere-hour integration is defined as:

(1), Is the derivative of the first RC loop voltage, < >>As a derivative of the second RC loop voltage, SOC _k And SOC (System on chip) _k-1 SOC values of the battery at k time and k-1 time respectively, I _k For the current of the battery at time k, C _n For the rated capacity of the battery, Δt is the sampling interval, ζ is the coulomb coefficient of the battery, C _n Δt, and ζ are constants.

The electrical quantity characteristic in the equation (1) can be expressed as:

in the formula (2), τ ₁ ＝R ₁ C ₁ ,τ ₂ ＝R ₂ C ₂ . Wherein U is _t (s) is the battery terminal voltage in the Laplace domain, U _oc (s) is the open-cell voltage in the Laplace domain, I(s) is the cell current in the Laplace domain, τ ₁ Indicating the time constant, τ, of the first RC loop ₂ Indicating the time constant of the second RC loop.

Further, the equation (2) may be converted into a discrete time domain by a relationship of the laplace domain s and the discrete time domain z, and the relationship conversion equation of the laplace domain s and the discrete time domain z is:

in equation (3), s is a laplace operator, z is a discrete time-domain operator, and Δt is a sampling interval.

Thus, the discrete time domain expression of the electrical quantity characteristic in equation (2) may be:

formula (4), I _k For the value of the current signal i at time k, U _t,k For the voltage response U at time k _t Value of U _oc,k Open circuit voltage U at time k _oc Wherein 5 parameter values are defined, respectively θ _1,k 、θ _2,k 、θ _3,k 、θ _4,k And theta _5,k They are equivalent circuit model parameters R with the battery ₀ 、R ₁ 、R ₂ 、C ₁ 、C ₂ The physical quantity of interest can be obtained by substituting equation (3) into equation (2) _1,k 、θ _2,k 、θ _3,k 、θ _4,k And theta _5,k And R is R ₀ 、R ₁ 、R ₂ 、C ₁ 、C ₂ Is:

from the z discrete time domain operator characteristic, the representation mode of the electric quantity characteristic in the discrete time domain in the formula (4) is changed into:

U _t,k ＝(1-θ _1,k -θ _2,k )U _oc,k +θ _1,k U _t,k-1 +θ _2,k U _t,k-2 +θ _3,k I _k +θ _4,k I _k-1 +θ _5,k I _k-2 (6)

in formula (6), U _t,k-1 Voltage response U for time k-1 _t Value of U _t,k-2 Voltage response U for time k-2 _t Value of I _k-1 Is the value of the current signal I at time k-1, I _k-2 The value of the current signal i at time k-2.

θ in formula (6) _1,k 、θ _2,k 、θ _3,k 、θ _4,k And theta _5,k After the 5 parameters are identified, the parameter value R in the equivalent circuit model can be obtained by the formula (5) ₀ 、R ₁ 、R ₂ 、C ₁ 、C ₂ And (5) identifying.

It can be understood that the battery is a highly nonlinear time-varying system in actual operation, and each parameter value in the equivalent circuit model changes along with the time change of the battery operation, so that in order to improve the accuracy of the equivalent circuit model, a parameter identification method can be adopted to solve the problem that the parameter value of the equivalent circuit model changes at any time. The parameter identification is to acquire various data of the battery, such as current and voltage, and the like in the actual work of the battery, then analyze mathematical relation formulas of the various parameters according to an equivalent circuit model, and finally solve specific numerical values of the various parameters of the battery at the current moment according to an algorithm. The parameter identification can truly reflect the parameter values of the equivalent circuit model of the battery at different moments, so that the accuracy of the equivalent circuit model can be improved to a certain extent.

According to one embodiment of the present disclosure, a recursive least squares method may be employed to perform parameter identification on the parameter values of the equivalent circuit model. The basic principle of the recursive least square method is to correct according to the input value of the current moment and the estimated value of the last moment so as to obtain the estimated value of the current moment, so that the recursive least square method is more suitable for carrying out parameter identification on the parameter value of the equivalent circuit model in the embodiment of the disclosure.

The general formula of the recursive least square method is divided into two steps, wherein the first step is to define a recursive least square formula:

in formula (7), y _k Is the observed value of k time phi _k For a vector of observations at times k-1, k-2 and inputs of k, k-1, k-2, θ _k Is the parameter value at time k, ζ _k Representing the model error, ψ is its error covariance matrix.

And the second step is to perform iterative solution calculation, wherein the formula is as follows:

in the formula (8), e _k Is the observed value y _k Prediction error, P _k Is the covariance matrix at time k, L _k To recursively least squares gain, P _k-1 For the covariance matrix at time k-1, θ _k For the estimation result of k time to the parameter value, θ _k-1 Estimation of parameter values for time k-1

According to an embodiment of the present disclosure, to satisfy the calculation of the adaptive recursive least square method, the observed value above is the terminal voltage U _t ，φ _k ＝[1 U _t,k-1 U _t,k-2 I _k I _k-1 I _k-2 ]，θ _k ＝[θ _1,k θ _2,k θ _3,k θ _4,k θ _5,k ]。

After passing through equation (8), an estimate of the parameter value, θ, can be determined _k Which represents θ _1,k 、θ _2,k 、θ _3,k 、θ _4,k And theta _5,k These 5 parameter values. Thus, θ can be determined by the current and voltage _k Further, the parameter value information R of the equivalent circuit model is obtained by the formula (5) ₀ 、R ₁ 、R ₂ 、C ₁ 、C ₂ 。

Operation S313: and determining fitting relation attributes between the charge state of the battery and parameter value information of the equivalent circuit model.

Through the above operation, equivalent circuit model parameter value information under the selected state of charge SOC condition can be obtained. Specifically, in the embodiment of the present disclosure, the parameter value information of the equivalent circuit model related to the corresponding state of charge SOC may be obtained under different conditions that the state of charge of the battery is 0%,5%,10%,15%, …,100%, respectively. Further, since the parameter value information obtained by the parameter identification in the above manner is limited, in order to obtain the parameter value information of the equivalent circuit model more comprehensively and accurately, the obtained parameter value information of the equivalent circuit model under different SOC conditions can be respectively subjected to polynomial fitting with the SOC, so that the fitting relation attribute between the SOC of the battery and the parameter value information of the equivalent circuit model is determined, and further, the mutual updating of the SOC and the parameter value information time can be realized according to the fitting relation attribute, thereby effectively improving the model precision of the equivalent circuit model and being beneficial to more accurately calculating the SOC of the battery.

Specifically, all the parameter value information related to the state of charge SOC can be combined with the state of charge SOC to determine the attribute of the fitting relationship between the state of charge of the battery and the parameter value information of the equivalent circuit model. The polynomial highest term of each parameter value of the equivalent circuit model in the fitting relation attribute can be determined according to a method such as Lagrangian interpolation method. Illustratively, U in embodiments of the present disclosure _oc 、R ₀ 、R ₁ 、C ₁ 、R ₂ 、C ₂ The polynomial highest terms of (c) may be set to 5, 7, 6 th order, respectively.

Based on this, according to the embodiment of the present disclosure, the attribute of the fitting relationship between the state of charge of the battery and the parameter value information of the equivalent circuit model is:

in the formula (9), U _OC R is the open circuit voltage of the battery ₀ Ohmic resistance of the battery; r is R ₁ And R is ₂ For the polarization internal resistance of the battery, C ₁ And C ₂ Is the polarization capacitance of the cell. SOC is the charge state of the battery, and a0 to a5 are U _OC Fitting parameter values with a polynomial of the SOC, wherein b 0-b 7 are R ₀ Fitting parameter values with polynomials of SOC, c 0-c 6 are R ₁ Fitting parameter values with a polynomial of the SOC, wherein d 0-d 6 are C ₁ Fitting parameter values with polynomials of SOC, wherein e 0-e 6 are R ₂ Fitting parameter values with a polynomial of the SOC, wherein f0 to f6 are C ₂ The parameter values are fitted to the polynomials of the SOC.

Through the operations of S311 to S313, model attribute information of the equivalent circuit model related to the battery can be determined. The model attribute information comprises fitting relation attributes and parameter value information of an equivalent circuit model.

Further, as shown in fig. 3, the method 300 further includes:

operation 320: and determining the middle battery cell according to the voltage values of all the battery cells in the battery. The battery comprises a plurality of battery cells, and the middle battery cell is used for representing the condition of a normal battery cell so as to detect the battery cell with internal short circuit fault later.

Operation 330: and calculating the target state of charge of the target battery cell based on the model attribute information of the equivalent circuit model, the voltage value of the target battery cell and the current value of the target battery cell. The target battery monomer is any battery monomer in all battery monomers. Preferably, the voltage value of the target battery cell may be the lowest value among the voltage values of all battery cells.

Operation 340: and calculating the intermediate state of charge of the intermediate battery cell based on the model attribute information of the equivalent circuit model, the voltage value of the intermediate battery cell and the current value of the intermediate battery cell. In the embodiment of the disclosure, the intermediate state of charge of the intermediate battery cell may be used to reflect the state of charge of the normal battery cell, so as to detect the internal short circuit fault of the target battery cell later.

In operations 330 and 340, according to one embodiment of the present disclosure, the method for calculating the target state of charge and/or the intermediate state of charge may use an unscented kalman filtering method, and of course, those skilled in the art may reasonably select other state of charge calculation methods according to actual needs and application scenarios, etc., which is not limited herein. In the following, the calculation of the target state of charge and/or the intermediate state of charge is performed using the unscented kalman filter method as an example only.

Unscented Kalman filtering (Unscented Kalman Filter, UKF) is a filtering algorithm for non-linear system state estimation, and compared with traditional Kalman filtering, unscented Kalman filtering approximates the probability distribution of a non-linear system by introducing unscented transformation (Unscented Transformation), thereby improving the accuracy of the estimation. In general, the iterative process of unscented kalman filtering (Unscented Kalman Filter, UKF) includes:

setting an initial state estimation valueAnd an initial error covariance matrix->E[·]To find the mathematical expectation:

estimating state quantity using posterior at last momentPosterior estimation error covariance matrix +.>The sampling generates 2n+1 sample points, and the formula is:

in the formula (12), x _0,k-1 Is the sampling point at the mean value of k-1 time, x _i,k-1 For sampling points elsewhere at time k-1,state quantity is estimated for a posterior at time k-1,/->Estimating an error covariance matrix for the posterior at the moment k-1, wherein n is a state quantity +.>λ is the scaling factor. More preferably, λ may be set to 1.5.

And calculating a mean weight and a covariance weight, wherein the formula is as follows:

in the formula (13), the amino acid sequence of the compound,for the weight at the mean +.>Is covariance weight at mean, +.>And->For weights at other points and covariance weights, β is the scaling factor and ρ is the scaling factor. Preferably, beta may be setSet to 2, ρ may be set to 0.04.

And (3) carrying out state prediction:

calculating state quantity:

in the formula (14), x _i,k|k-1 To predict the sample value at k-time based on the optimal estimate at k-1 time, A _k-1 For the state transition matrix at time k-1, B _k-1 Matrix for converting input into state at time k-1, x _i,k-1 2n+1 sample points at time k-1 calculated for equation (12), u _k-1 The input quantity at the time of k-1,is the predicted state quantity at time k.

Calculating a covariance matrix:

in the formula (15), P _k ^- Covariance matrix predicted for k time, Q _k Is a noise matrix of the state transition matrix.

And (3) calculating deterministic sampling points:

updating the observed value, wherein the observed value is y _i,k|k-1 Which represents the observation of predicting the k-time from the optimal estimate of the k-1 time,g (x _i,k|k-1 ) The observation equation in the state space equation is:

calculating a Kalman gain K:

K＝P _xy,k P _y,k (19)

in the formula (18), P _y,k Covariance matrix calculated according to observation value at time k, P _xy,k Covariance matrix calculated for k moment according to state value and observation value, R _k Is an observation noise matrix.

In the formula (19), K is a kalman gain.

Status measurement update:

in the formula (20), the amino acid sequence of the compound,for the state quantity estimate at time k +.>Is the predicted state quantity at the moment k, y _k For the observation sample value at time k, +.>The predicted observed quantity at time k.

Covariance matrix measurement update:

P _k ⁺ ＝P _k ^- -KP _y,k K ^T (21)

in the formula (21), P _k ⁺ An updated value for the covariance matrix measurement at time k.

Then, the process returns to equation (12).

For the embodiments of the present disclosure are:

in order to realize estimation of the SOC by the unscented kalman filter, it is necessary to find the corresponding observation equation, state equation, and output equation of the battery system according to the model attribute information of the equivalent circuit model mentioned above.

Namely:

g(·)＝U _oc,k (SOC _k )-U _1,k -U _2,k -I _k R ₀ 。

state quantity x _k-1 ＝[SOC _k-1 U _1,k-1 U _2,k-1 ]Input u _k ＝I _k Observed quantity y _k ＝U _t,k 。

Since the state quantity includes the SOC, the SOC value is estimated as long as the state quantity is estimated.

After the parameter value information and the state space equations of the equivalent circuit model of the battery are obtained, the target state of charge of the target battery monomer and/or the intermediate state of charge of the intermediate battery monomer can be calculated according to the algorithm step of unscented Kalman filtering.

According to an embodiment of the present disclosure, operation 330 may specifically include: when k is>The state quantity at the time k is obtained based on the formula (12-21) at 0Due to->The first value of the vector is SOC, thus yielding the target state of charge of the target cell at time k.

According to an embodiment of the present disclosure, operation 340 may specifically include: when k is>The state quantity at the time k is obtained based on the formula (12-21) at 0Due to->The first value of the vector is SOC, thus yielding the intermediate state of charge of the intermediate cell at time k.

In the above manner, the target state of charge of the target battery cell and/or the intermediate state of charge of the intermediate battery cell may be calculated based on unscented Kalman filtering.

Further, as shown in fig. 3, the method 300 further includes:

operation S350: based on the target state of charge and the voltage value of the target battery cell, and the intermediate state of charge and the voltage value of the intermediate battery cell, the Euclidean distance between the target battery cell and the intermediate battery cell is calculated.

According to an embodiment of the present disclosure, operation S350 includes:

normalizing the voltage value of the target battery cell according to a normalization formula to obtain a target normalized voltage value;

normalizing the voltage value of the middle battery cell according to a normalization formula to obtain a middle normalized voltage value;

based on the target state of charge and the target normalized voltage value, and the intermediate state of charge and the intermediate normalized voltage value, a euclidean distance between the target cell and the intermediate cell is calculated.

Wherein, the normalization formula comprises:

in the formula (22), U _i,k To normalize the pre-processed cell voltage values,the table is the voltage value of the battery cell after normalization processing, U _lower U is the upper cut-off voltage of the battery _upper The lower cutoff voltage of the battery, wherein the cell voltage value includes a voltage value of the target cell or a voltage value of the intermediate cell.

The Euclidean distance between the target battery cell and the middle battery cell is determined according to the following relation:

in the formula (23), D _i,k For Euclidean distance between target cell and intermediate cell, SOC _i,k For target state of charge, SOC _median,k Is of intermediate state of charge, U _i,k For target normalized voltage value, U _median,k And i is the number of battery cells in the battery, and is the middle normalized voltage value.

Further, as shown in fig. 3, the method 300 further includes:

operation 360: and determining the internal short circuit fault information of the target battery cell according to the Euclidean distance between the target battery cell and the middle battery cell.

Wherein, the internal short circuit fault information of the target battery cell is determined according to the following relation:

in the formula (24), τ is a preset threshold value. More preferably, τ may be set to 10%.

Operation 370: and judging whether the internal short circuit fault information of the target battery cell is characterized as a fault or not.

If yes, executing operation S380; if not, operations S313 to S370 are iteratively performed until the internal short circuit information of the target battery cell characterizes a failure.

Operation S380: under the condition that the internal short circuit fault information of the target battery monomer represents faults, alarm information can be generated to alarm.

Corresponding to the above-described processes of iteratively performing operations S313 to S380, in the case where the internal short circuit failure information of the target battery cell is characterized as normal, a detection process including an i-th detection process including:

for each battery cell, updating the i-1 th parameter value information of the equivalent circuit model according to the fitting relation attribute of the state of charge of the battery and the parameter value of the equivalent circuit model and the i-1 th state of charge result of the battery cell to obtain the i-1 th parameter value information of the equivalent circuit model, wherein i is more than 1, and i is an integer, and the battery cell comprises a target battery cell or an intermediate battery cell;

Calculating an ith state of charge of the battery cell based on the ith parameter value and an ith voltage value of the battery cell, wherein the battery cell comprises a target battery cell or an intermediate battery cell;

calculating an ith Euclidean distance between the target battery cell and the middle battery cell based on the ith target state of charge of the target battery cell and the ith voltage value of the target battery cell, and the ith middle state of charge of the middle battery cell and the ith voltage value of the middle battery cell;

according to the ith Euclidean distance, the ith internal short circuit fault information of the target battery cell is determined;

and judging whether the ith internal short circuit fault information of the target battery cell represents a fault or not.

When the above-mentioned detection process is performed for the 1 st time, the offline parameter value information of the equivalent circuit model of the battery may be determined in an offline state, and used as the 1 st parameter value information in the detection process.

According to one embodiment of the present disclosure, the offline parameter value information (i.e., the 1 st parameter value information) of the equivalent model circuit of the lithium ion battery and the fitting relationship attribute between the state of charge of the battery and the parameter value information of the equivalent circuit model may be determined in an offline state by selecting the lithium ion battery sample of the same type, brand, model, etc. as the lithium ion battery in the internal short circuit fault experiment at the beginning of the internal short circuit fault experiment in operations S311 to S313. Then, for each cell, the 1 st state of charge of the cell is calculated based on the 1 st parameter value and the 1 st voltage value of the cell.

When the detection process is executed for the 2 nd time, for each battery cell, the 1 st time parameter value information of the equivalent circuit model is updated according to the fitting relation attribute of the state of charge of the battery and the parameter value of the equivalent circuit model and the 1 st time state of charge result of the battery cell, so as to obtain the 2 nd time parameter value information of the equivalent circuit model. Then, for each cell, the 2 nd state of charge of the cell is calculated based on the 2 nd parameter value and the 2 nd voltage value of the cell.

It can be appreciated that due to U in the equivalent circuit model _oc Is a function of the state of charge SOC, the accuracy of the state of charge SOC affects the accuracy of the open circuit voltage Uoc, and thus the result of parameter identification in the equivalent circuit model and the terminal voltage U _t Is used for the prediction accuracy of (a). Therefore, when the model attribute information of the equivalent circuit model is determined in the ith detection process, the i-1 th parameter value information of the equivalent circuit model can be updated based on the fitting relation attribute of the state of charge of the battery and the parameter value information of the equivalent circuit model and the i-1 st state of charge result of the battery cell, so that the accuracy of the equivalent circuit model is improved. And then calculating the ith charge state result of the battery cell according to the ith parameter value information of the obtained equivalent circuit model.

Because the parameter value information of the equivalent circuit model and the state of charge SOC of each battery cell are dynamically changed at all times, by the setting of operation S380, the method 300 provided by the embodiment of the disclosure can dynamically update the parameter value information of the equivalent circuit model and the state of charge SOC of each battery cell, so as to better reflect the actual condition of the state of charge SOC of each battery cell, thereby further improving the accuracy of detecting the internal short circuit fault of the target battery cell. In addition, operation S380 can also enable the method 300 to perform the next detection process when the internal short-circuit fault of the target battery is not detected, thereby playing a role in monitoring whether the internal short-circuit fault condition exists in the battery in real time.

In order to facilitate a better understanding of the technical solutions of the present disclosure, a detailed description will be given below with reference to specific embodiments and accompanying drawings.

It should be noted that, in some embodiments, the present disclosure performs an internal short circuit fault experiment by adopting a parallel connection manner of resistors with different resistance values and selected battery cells, so as to simulate the internal short circuit fault battery cells with different degrees in a practical application scenario. In addition, the person skilled in the art may select other ways to simulate the internal short circuit fault cell in the practical application scenario to perform the internal short circuit fault experiment, which is not described herein.

According to one embodiment of the present disclosure, the battery is selected as a lithium ion battery, the positive active material of which is nickel cobalt manganese, and the negative active material of which is graphite. 4 battery monomers are selected to form a serial module of the lithium ion battery, wherein the 4 battery monomers are respectively numbered as a monomer 1, a monomer 2, a monomer 3 and a monomer 4, and the serial module is charged with constant current and constant voltage and discharged with constant current so as to simulate normal operation conditions.

First, model attribute information about the lithium ion battery is obtained based on operation S110 or operation S310 under the condition that the state of charge of the battery is 0%,5%,10%,15%, …,100%, respectively. The model attribute information comprises parameter value information of an equivalent circuit model and fitting relation attributes between the charge state of the battery and the parameter value information of the equivalent circuit model.

Further, one battery cell is selected from the series module, and one or more resistors are connected in parallel to the selected battery cell, so that the total resistance of the combination of the selected battery cell and the parallel resistor is smaller than the resistance of other remaining battery cells, and the selected battery cell is set as an internal short-circuit battery cell. And then, under the condition of selecting the parallel resistors of the battery cells, carrying out constant current and constant voltage charging and constant current discharging experiments on the serial modules.

Under the experimental conditions, the current values of the battery cells in the series module are equal, and the total resistance of the combination of the selected battery cells and the parallel resistor is smaller than the resistance of other residual battery cells, so that the partial pressure of the selected battery cells in the series module is smaller than that of other battery cells.

The selected battery cell may be the battery cell 2 in the series module, and the resistance of the parallel resistor may be 100 Ω, 51 Ω, or 10 Ω, for example.

The voltage values of the battery cells 1-4 are then measured, respectively, and the middle battery cell is determined. Specifically, the voltage values of the 4 battery cells can be sequentially arranged from low to high according to the definition of the median, and the average value of the two voltage values in the middle is the voltage value of the middle battery cell. Based on the experimental conditions and principles mentioned above, the voltage division of the battery cells 2 in the series module is smaller than that of the other battery cells due to the parallel resistance of the battery cells 2. Thus, the target cell may be selected as the cell 2 having the lowest voltage value among the respective cells.

Then, based on operation S130 or operation S330, a target state of charge of the target battery cell is calculated; based on operation S140 or operation S340, an intermediate state of charge of the intermediate battery cell is calculated.

Finally, based on operation S150 or operation S350, the euclidean distance between the target battery cell and the intermediate battery cell is determined, and based on operation S160 or operation S360, the internal short circuit fault information of the target battery cell is determined according to the euclidean distance between the target battery cell and the intermediate battery cell.

Fig. 5A shows the euclidean distance results for an internal short circuit condition where a cell is selected to be connected in parallel with a 100 Ω resistor in one embodiment of the present disclosure. Fig. 5B shows a euclidean distance result in the case of an internal short circuit with a selected cell parallel 51 Ω resistor in accordance with another embodiment of the present disclosure. Fig. 5C shows a euclidean distance result in the case of an internal short circuit where a cell is selected to be connected in parallel with a 10Ω resistor in accordance with another embodiment of the present disclosure.

As shown in fig. 5A to 5C, based on the above experimental conditions and principles, in each internal short circuit condition, the euclidean distance value of the battery cell 2 exceeds the threshold τ, and an alarm is given, so that no missed judgment occurs. The Euclidean distance values of other battery monomers do not exceed the threshold value tau, no alarm measures are made, and erroneous judgment does not occur.

In one exemplary application scenario, the battery mentioned above may be used in an electric vehicle. In an electric vehicle, a battery module of the battery may include a plurality of battery cells, and a battery management system (Battery Management System, BMS) of the electric vehicle generally monitors the states of the battery cells and performs control according to a set threshold. Specifically, according to embodiments of the present disclosure, in the event that an internal short circuit fault information characterizing fault of a battery cell is detected, the battery management system may issue an alarm signal and control the current transformer to stop outputting to prevent further damage to the battery. Among these, current transformers are typically electrical devices that can vary the voltage, frequency, phase number, and other power or characteristics of a power supply system. In the electric automobile, under the condition that the battery monomer has an internal short circuit fault, the converter is controlled to stop outputting, so that overdischarge or overcharge of the battery monomer can be prevented, and the integral safety and stability of the battery can be maintained.

According to a second aspect of the present disclosure, there is provided an apparatus for detecting a short circuit fault in a battery. Fig. 6 shows a schematic block diagram of an apparatus for detecting a short-circuit fault in a battery according to an embodiment of the present disclosure.

As shown in fig. 6, the apparatus 600 includes an equivalent circuit model determination module 610, an intermediate battery cell determination module 620, a target state of charge calculation module 630, an intermediate state of charge calculation module 640, a euclidean distance calculation module 650, and an internal short fault information determination module 660.

An equivalent circuit model determination module 610 for determining model attribute information of an equivalent circuit model of the battery. The model attribute information comprises parameter value information of an equivalent circuit model and fitting relation attributes between the charge state of the battery and the parameter value information of the equivalent circuit model.

The middle battery cell determining module 620 is configured to determine a middle battery cell according to the voltage values of all battery cells in the battery, where the voltage value of the middle battery cell is the median of the voltage values of all battery cells.

The target state of charge calculation module 630 is configured to calculate a target state of charge of the target battery cell based on model attribute information of the equivalent circuit model, a voltage value of the target battery cell, and a current value of the target battery cell.

And an intermediate state of charge calculation module 640 for calculating an intermediate state of charge of the intermediate battery cell based on the model attribute information of the equivalent circuit model, the voltage value of the intermediate battery cell, and the current value of the intermediate battery cell.

The euclidean distance calculating module 650 is configured to calculate the euclidean distance between the target battery cell and the intermediate battery cell based on the target state of charge and the voltage value of the target battery cell, and the intermediate state of charge and the voltage value of the intermediate battery cell, where the target battery cell is any battery cell of all battery cells.

And the internal short circuit fault information determining module 660 is configured to determine internal short circuit fault information of the target battery cell according to the euclidean distance between the target battery cell and the middle battery cell.

According to an embodiment of the present disclosure, the apparatus further includes a voltage acquisition sub-module and a current acquisition sub-module.

And the voltage acquisition sub-module is used for acquiring the voltage values of all the battery monomers in the battery.

And the current acquisition sub-module is used for acquiring the current values of all the battery monomers in the battery.

According to an embodiment of the present disclosure, any multiple of the above-mentioned voltage acquisition sub-module and the current acquisition sub-module, or at least part of the functions of any multiple of the above-mentioned voltage acquisition sub-module and the current acquisition sub-module may be implemented by a data acquisition module, where the data acquisition module may be selected to include key functions such as signal amplification, filtering, sampling, digital-to-analog conversion, communication isolation, and the like, include components for collecting voltages such as a resistor voltage-dividing circuit, and include components for collecting currents such as a hall current sensor or a current divider.

According to an embodiment of the present disclosure, any of the above-mentioned equivalent circuit model determining module 610, intermediate battery cell determining module, target state of charge calculating module 630, intermediate state of charge calculating module 640, euclidean distance calculating module 650, and internal short circuit fault information determining module 660, or at least part of the functions of any of them, may be implemented by a processing module, which may be selected as a component having functions of storage, communication, control, analog-to-digital conversion, and the like. For example, the foregoing processing module may be implemented by a single-chip microcomputer, a field programmable gate array, a digital signal processor, or the like.

According to an embodiment of the present disclosure, the apparatus further includes a voltage response acquisition sub-module, a voltage response processing sub-module, and a fitting relationship attribute determination sub-module.

And the voltage response acquisition sub-module is used for acquiring voltage response information corresponding to the charge state of the battery.

And the voltage response processing sub-module is used for processing the voltage response information according to a recursive least square algorithm to obtain the parameter value information of the equivalent circuit model.

And the fitting relation attribute determining submodule is used for determining fitting relation attributes between the charge state of the battery and parameter value information of the equivalent circuit model.

According to an embodiment of the present disclosure, the above apparatus further includes a first calculation sub-module and a second calculation sub-module.

The first computing sub-module is used for processing the parameter value information of the equivalent circuit model, the voltage value of the target battery cell, the current value of the target battery cell and the fitting relation attribute of the state of charge of the battery and the parameter value information of the equivalent circuit model according to the unscented Kalman filtering algorithm to obtain the target state of charge of the target battery cell.

The second calculation sub-module is used for processing the parameter value information of the equivalent circuit model, the voltage value of the middle battery cell, the current value of the middle battery cell and the fitting relation attribute of the state of charge of the battery and the parameter value information of the equivalent circuit model according to the unscented Kalman filtering algorithm to obtain the middle state of charge of the middle battery cell.

According to an embodiment of the present disclosure, the above apparatus further includes a third calculation sub-module, a fourth calculation sub-module, and a fifth calculation sub-module.

And the third calculation sub-module is used for carrying out normalization processing on the voltage value of the target battery cell according to a normalization formula to obtain a target normalized voltage value.

And the fourth calculation sub-module is used for carrying out normalization processing on the voltage value of the middle battery cell according to a normalization formula to obtain a middle normalized voltage value.

And a fifth calculation sub-module for calculating the euclidean distance between the target battery cell and the intermediate battery cell based on the target state of charge and the target normalized voltage value, and the intermediate state of charge and the intermediate normalized voltage value.

According to an embodiment of the present disclosure, the above-mentioned apparatus further includes a comparing sub-module and a determining sub-module.

And the comparison sub-module is used for comparing the Euclidean distance between the target battery monomer and the middle battery monomer with a preset threshold value to obtain a comparison result.

And the determining submodule is used for determining whether the internal short circuit fault information represents a fault or not according to the comparison result.

According to an embodiment of the present disclosure, the above apparatus further comprises an alarm sub-module and an iteration sub-module.

And the alarm sub-module is used for alarming under the condition that the internal short circuit fault information of the target battery monomer represents a fault.

And the iteration sub-module is used for enabling the i to be increased by 1 automatically under the condition that the internal short circuit fault information of the target battery monomer represents normal conditions, and iteratively executing the i detection process until the internal short circuit fault information of the target battery monomer represents faults.

According to an embodiment of the present disclosure, the apparatus further includes a first iteration unit.

The first iteration unit is used for updating the ith-1 th parameter value of the equivalent circuit model according to the fitting relation attribute of the charge state of the battery and the parameter value information of the equivalent circuit model and the ith-1 th charge state result of the battery for each battery cell in the ith execution detection process to obtain the ith parameter value information of the equivalent circuit model, wherein i is more than 1, and i is an integer. The battery cell comprises a target battery cell or an intermediate battery cell.

Any one or more of the above-described modules, sub-modules, units, sub-units may be at least partially enabled to communicate via IIC, RS485, RS232, or CAN, etc. communication protocols, in accordance with embodiments of the present disclosure.

Any number of modules, sub-modules, units, sub-units, or at least some of the functionality of any number of the sub-units according to embodiments of the present disclosure may be implemented in one module. Any one or more of the modules, sub-modules, units, sub-units according to embodiments of the present disclosure may be implemented as split into multiple modules. Any one or more of the modules, sub-modules, units, sub-units according to embodiments of the present disclosure may be implemented at least in part as a hardware circuit, such as a Field Programmable Gate Array (FPGA), a Programmable Logic Array (PLA), a system-on-chip, a system-on-substrate, a system-on-package, an Application Specific Integrated Circuit (ASIC), or in any other reasonable manner of hardware or firmware that integrates or encapsulates the circuit, or in any one of or a suitable combination of three of software, hardware, and firmware. Alternatively, one or more of the modules, sub-modules, units, sub-units according to embodiments of the present disclosure may be at least partially implemented as computer program modules, which when executed, may perform the corresponding functions.

For example, any of the equivalent circuit model determination module 610, the intermediate battery cell determination module 620, the target state of charge calculation module 630, the intermediate state of charge calculation module 640, the euclidean distance calculation module 650, and the internal short fault information determination module 660 may be incorporated in one module/unit/sub-unit or any of the modules/units/sub-units may be split into a plurality of modules/units/sub-units. Alternatively, at least some of the functionality of one or more of these modules/units/sub-units may be combined with at least some of the functionality of other modules/units/sub-units and implemented in one module/unit/sub-unit. According to embodiments of the present disclosure, at least one of the equivalent circuit model determination module 610, the intermediate battery cell determination module 620, the target state of charge calculation module 630, the intermediate state of charge calculation module 640, the euclidean distance calculation module 650, and the internal short fault information determination module 660 may be implemented at least in part as hardware circuitry, such as a Field Programmable Gate Array (FPGA), a Programmable Logic Array (PLA), a system-on-chip, a system-on-substrate, a system-on-package, an Application Specific Integrated Circuit (ASIC), or may be implemented in hardware or firmware in any other reasonable manner of integrating or packaging the circuitry, or in any one of, or in any suitable combination of, software, hardware, and firmware. Alternatively, at least one of the equivalent circuit model determination module 610, the intermediate battery cell determination module 620, the target state of charge calculation module 630, the intermediate state of charge calculation module 640, the euclidean distance calculation module 650, and the internal short fault information determination module 660 may be at least partially implemented as a computer program module that, when executed, may perform the corresponding functions.

It should be noted that, in the embodiment of the present disclosure, the device portion for detecting the short-circuit fault in the battery corresponds to the method portion for detecting the short-circuit fault in the battery in the embodiment of the present disclosure, and the description of the device portion for detecting the short-circuit fault in the battery specifically refers to the method portion for detecting the short-circuit fault in the battery, which is not described herein again.

It should be noted that, in some application scenarios, the method and apparatus for detecting a short circuit fault in a battery provided by the present disclosure may be applied to other batteries suitable for the foregoing method and apparatus, such as a lithium ion battery or a sodium ion battery, and the disclosure is not limited thereto.

Thus, embodiments of the present disclosure have been described in detail with reference to the accompanying drawings. It should be noted that, in the drawings or the text of the specification, implementations not shown or described are all forms known to those of ordinary skill in the art, and not described in detail. Furthermore, the above definitions of the elements and methods are not limited to the specific structures, shapes or modes mentioned in the embodiments, and may be simply modified or replaced by those of ordinary skill in the art.

It should also be noted that the foregoing describes various embodiments of the present disclosure. These examples are provided to illustrate the technical content of the present disclosure, and are not intended to limit the scope of the claims of the present disclosure. A feature of one embodiment may be applied to other embodiments by suitable modifications, substitutions, combinations, and separations.

Furthermore, unless specifically described or steps must occur in sequence, the order of the above steps is not limited to the list above and may be changed or rearranged according to the desired design. In addition, the above embodiments may be mixed with each other or other embodiments based on design and reliability, i.e. the technical features of the different embodiments may be freely combined to form more embodiments.

While the foregoing embodiments have been described in some detail for purposes of clarity of understanding, it will be understood that the foregoing embodiments are merely illustrative of the invention and are not intended to limit the invention, and that any modifications, equivalents, improvements, etc. that fall within the spirit and principles of the present disclosure are intended to be included within the scope of the present disclosure.

Claims

1. A method of detecting a short circuit fault in a battery, comprising:

determining model attribute information of an equivalent circuit model related to the battery;

determining a middle battery cell according to the voltage values of all battery cells in a battery, wherein the battery comprises a plurality of battery cells, and the voltage value of the middle battery cell is the median of the voltage values of all battery cells;

Calculating a target state of charge of a target battery cell based on model attribute information of the equivalent circuit model, a voltage value of the target battery cell and a current value of the target battery cell, wherein the target battery cell is any battery cell in all battery cells;

calculating the intermediate state of charge of the intermediate battery cell based on model attribute information of the equivalent circuit model, the voltage value of the intermediate battery cell and the current value of the intermediate battery cell;

calculating a Euclidean distance between the target battery cell and the intermediate battery cell based on the target state of charge and the voltage value of the target battery cell, and the intermediate state of charge and the voltage value of the intermediate battery cell;

and determining the internal short circuit fault information of the target battery cell according to the Euclidean distance between the target battery cell and the middle battery cell.

2. The method of claim 1, wherein the voltage value of the target cell comprises a lowest of the voltage values of all of the cells.

3. The method of claim 2, wherein the determining model attribute information of an equivalent circuit model associated with the battery comprises:

Acquiring voltage response information corresponding to the state of charge of the battery;

processing the voltage response information according to a recursive least square algorithm to obtain parameter value information of the equivalent circuit model;

and determining fitting relation attributes between the charge state of the battery and the parameter value information of the equivalent circuit model, wherein the model attribute information comprises the parameter value information of the equivalent circuit model and fitting relation attributes between the charge state of the battery and the parameter value information of the equivalent circuit model.

4. The method of claim 3, wherein the calculating the target state of charge of the target battery cell based on model attribute information of the equivalent circuit model, a voltage value of the target battery cell, and a current value of the target battery cell comprises:

processing parameter value information of the equivalent circuit model, a voltage value of the target battery cell, a current value of the target battery cell and the fitting relation attribute between the state of charge of the battery and the parameter value information of the equivalent circuit model according to an unscented Coleman filtering algorithm to obtain the target state of charge of the target battery cell;

The calculating the intermediate state of charge of the intermediate battery cell based on the model attribute information of the equivalent circuit model, the voltage value of the intermediate battery cell, and the current value of the intermediate battery cell includes:

and processing the parameter value information of the equivalent circuit model, the voltage value of the middle battery cell and the current value of the middle battery cell and the fitting relation attribute between the state of charge of the battery and the parameter value information of the equivalent circuit model according to an unscented Kalman filtering algorithm to obtain the middle state of charge of the middle battery cell.

5. The method of claim 2, wherein the calculating the euclidean distance between the target cell and the intermediate cell based on the target state of charge and the voltage value of the target cell, and the intermediate state of charge and the voltage value of the intermediate cell comprises:

And calculating the Euclidean distance between the target battery cell and the middle battery cell based on the target state of charge and the target normalized voltage value, and the middle state of charge and the middle normalized voltage value.

6. The method of claim 2, wherein determining the internal short circuit fault information of the target cell according to the euclidean distance comprises:

comparing the Euclidean distance between the target battery monomer and the middle battery monomer with a preset threshold value to obtain a comparison result; and

and determining whether the internal short circuit fault information represents a fault or not according to the comparison result.

7. The method of claim 5, wherein the euclidean distance is obtained according to the following relationship:

wherein D is _i,k For the Euclidean distance between the target cell and the intermediate cell, SOC _i,k For the target state of charge, SOC _median,k For the intermediate state of charge, U _i,k For the target normalized voltage value, U _median,k And i is the number of the battery cells for the intermediate normalized voltage value.

8. The method of claim 5, the normalization formula comprising:

in U _i,k For normalizing the voltage value of the battery cell before processing, the ug _i,k The table is the voltage value of the battery cell after normalization processing, U _lower U is the upper cut-off voltage of the battery _upper And the lower cut-off voltage of the battery, and the battery cell voltage value comprises the voltage value of the target battery cell or the voltage value of the middle battery cell.

9. The method of claim 6, further comprising:

alarming under the condition that the internal short circuit fault information of the target battery monomer represents faults;

under the condition that the internal short circuit fault information of the target battery cell represents normal, iteratively executing the following detection process until the internal short circuit fault information of the target battery cell represents fault, wherein the detection process comprises an ith detection process, and the ith detection process comprises:

for each battery cell, updating the ith-1 th parameter value of the equivalent circuit model according to the fitting relation attribute of the charge state of the battery and the parameter value information of the equivalent circuit model and the ith-1 th charge state result of the battery cell to obtain the ith parameter value of the equivalent circuit model, i is more than 1, and i is an integer,

the battery cell comprises a target battery cell or an intermediate battery cell;

Calculating an ith state of charge of the battery cell based on the ith parameter value and the ith voltage value of the battery cell, wherein the battery cell comprises a target battery cell or an intermediate battery cell;

calculating an ith Euclidean distance between a target battery cell and an intermediate battery cell based on an ith target state of charge of the target battery cell and an ith voltage value of the target battery cell, and an ith intermediate state of charge of the intermediate battery cell and an ith voltage value of the intermediate battery cell;

comparing the ith Euclidean distance with a preset threshold value to obtain an ith comparison result; and

and determining whether the ith internal short circuit fault information represents a fault or not according to the ith comparison result.

10. An apparatus for detecting a short circuit fault in a battery, comprising:

an equivalent circuit model determination module for determining model attribute information of an equivalent circuit model of the battery;

the middle battery monomer determining module is used for determining middle battery monomers according to the voltage values of all battery monomers in the battery, wherein the voltage value of the middle battery monomer is the median of the voltage values of all battery monomers;

A target state of charge calculation module for calculating a target state of charge of the target battery cell based on model attribute information of the equivalent circuit model, a voltage value of the target battery cell, and a current value of the target battery cell;

a middle state of charge calculation module for calculating a middle state of charge of the middle battery cell based on model attribute information of the equivalent circuit model, a voltage value of the middle battery cell, and a current value of the middle battery cell;

the Euclidean distance calculating module is used for calculating the Euclidean distance between the target battery cell and the middle battery cell based on the target state of charge and the voltage value of the target battery cell, and the middle state of charge and the voltage value of the middle battery cell, wherein the target battery cell is any battery cell in all battery cells;

and the internal short circuit fault information determining module is used for determining the internal short circuit fault information of the target battery cell according to the Euclidean distance between the target battery cell and the middle battery cell.