WO2019037012A1

WO2019037012A1 - Methods for estimating state of charge of battery and battery pack, and battery management system, battery and electric car using methods for estimating state of charge

Info

Publication number: WO2019037012A1
Application number: PCT/CN2017/098780
Authority: WO
Inventors: 肖鹏飞; 房乔华
Original assignee: 罗伯特·博世有限公司; 肖鹏飞
Priority date: 2017-08-24
Filing date: 2017-08-24
Publication date: 2019-02-28
Also published as: CN111033930A; CN111033930B

Abstract

Provided is a method for estimating a state of charge of a battery. The method comprises: monitoring a current of a battery; when the current is 0, recording a terminal voltage of the battery; building a relationship model between two terminal voltages respectively corresponding to any two consecutive recording times; using recursive least squares algorithm to calculate an open-circuit voltage; and determining a state of charge of the battery according to the open-circuit voltage. Further provided are a method for estimating a state of charge of a battery pack, and a battery management system, a battery and an electric car using the estimation methods. The solution of the present invention has the advantages of short waiting time, high estimation accuracy, low calculation complexity, easy application, and no need for additional hardware, etc.

Description

Method for estimating state of charge of battery and battery pack, battery management system, battery and electric vehicle using the state of charge estimation method

[Technical Field]

The present invention relates to a method for estimating the state of charge of a battery and a battery pack, a battery management system, a battery, and an electric vehicle.

【Background technique】

With the ever-increasing development and demand for mobile devices, electric vehicles, hybrid vehicles, energy storage devices, etc., the demand for secondary batteries as energy resources is rapidly increasing. In general, secondary batteries include nickel-cadmium batteries, nickel-hydrogen batteries, lithium ion batteries, and the like.

Lithium-ion batteries (LiB) have been widely used in consumer electronics and automotive applications. For example, a battery pack for an electric or hybrid electric vehicle typically includes a plurality of lithium ion batteries connected in parallel and/or in series. Due to the complexity of the lithium ion battery reaction process, a sophisticated battery management system is needed, where accurate estimation of the state of charge (ie, SOC) is critical.

An important part of battery management is the estimation of remaining capacity, and the remaining capacity estimation has always been one of the more difficult problems in the industry, and it has become one of the bottlenecks in the promotion and application of new energy. Since the SOC is the internal state parameter of the battery, it cannot be obtained by direct measurement. It can only be indirectly estimated by measuring relevant parameters such as voltage and current, such as the Ampere method and the open circuit voltage method. Due to factors such as charge and discharge current, voltage, temperature, internal resistance, and degradation, there is no uniform method for SOC estimation.

In the patents US 684 1972, CN 102246029 and CN 103344919, a SOC calibration or reset method is described which requires the battery SOC to be charged or discharged to a certain range. The basic theory behind these prior art techniques is to reset the SOC to the lowest or highest boundary when the open circuit voltage reaches a minimum or maximum limit. However, these prior art techniques do not allow real-time estimation of all SOC ranges, and even a substantial change to the SOC is often required when performing estimation or calibration.

Therefore, an improved SOC estimation or calibration method is desired.

[Summary of the Invention]

According to an aspect of the invention, a method of calibrating/estimating the state of charge of a battery in a short time is presented. For all battery modules in the battery pack, the calibration/estimation time of the state of charge can be as short as 2 to 3 minutes, so that the battery management system can significantly increase its calibration update frequency for the state of charge. In addition, since the method of the present invention is proposed to be estimated and calibrated in an idle state (ie, when the output current of the battery is zero), dynamic behavior changes of the battery (eg, due to temperature and aging) do not significantly affect the present invention. The accuracy of the method. In addition, the method of the present invention has the advantages of low computational complexity, easy application, no additional hardware, and the like.

Specifically, a technical solution of the present invention provides a method for estimating a state of charge of a battery, the method comprising: monitoring a current of the battery; and when the current is 0, recording a terminal voltage of the battery, wherein in the first recording time t _k recording a first end of the battery voltage U _k, while in the second recording time t _{k + 1} second record of the battery terminal voltage U _{k + 1,} and wherein said first Recording time and the second recording time are two recording times adjacent to each other; establishing a relationship model between the first terminal voltage and the second terminal voltage, wherein the relationship model is included as a parameter to be solved in the relationship model An open circuit voltage of the battery; calculating the open circuit voltage by a recursive least squares method; and determining a state of charge of the battery according to the open circuit voltage.

Preferably, in the above estimation method, when the current of the battery is not 0 in the recording or calculation process, the estimation of the state of charge is abandoned.

Preferably, in the above estimation method, the terminal voltage of the battery is recorded at fixed time intervals.

Preferably, the above estimating method further comprises: continuously monitoring the terminal voltage of the battery, and recording the terminal voltage of the battery only when the change of the terminal voltage exceeds a preset threshold.

Preferably, in the above estimation method, the preset threshold is 1 mV.

Preferably, in the above estimation method, the relationship model is as follows:

Where [OCV a b] is the parameter matrix to be solved in the relational model, and OCV represents the open circuit voltage of the battery.

Preferably, according to the above relationship, for each cell in the battery pack, a linear relationship is constructed: Y=A·X, and the parameter matrix A=[OCV a b] is obtained by a least squares recursion method:

Where Y=[U _m U _m+1 ... U _n ],

Where m and n are constants and m < n,

Is a recursive coefficient matrix whose convergence value corresponds to the parameter matrix to be solved, K is the gain matrix, P is the covariance matrix, and ε is the error matrix.

Preferably, the linear relationship Y=A·X can also be constructed as follows:

Where Y=[U _m U _m+1 ... U _n ],

Where m and n are constants and m < n,

Preferably, in the above estimation method, if it is within a preset time

If the convergence is not reached, the calculation of the recursive least squares method is stopped.

Preferably, after the estimation method estimates the open circuit voltage OCV, the state of charge of the battery is determined according to the estimated open circuit voltage by looking up the SOC-OCV data table, wherein the SOC-OCV data table and the type and manufacture of the battery cell Process related. Preferably, in the above estimation method, the battery is a lithium ion battery.

Another aspect of the present invention provides a method for estimating a state of charge of a battery pack, the battery pack including one or more battery modules, the method comprising: monitoring a current of the battery pack; 0, for each of the one or more battery modules in the battery pack, the first terminal voltage U _{k,j of the} battery module is recorded at the first recording time t _k , and the second record is recorded Recording, at time t _k+1 , the second terminal voltage U _{k+1,j of} the battery module, wherein the first recording time and the second recording time are two recording times adjacent to each other and wherein j is a positive integer Representing a serial number of the battery module in the battery pack; for the jth battery module, establishing a relationship model between the first terminal voltage and the second terminal voltage, wherein the relationship model is included An open circuit voltage of the battery module of one of the parameters to be solved in the relationship model; the open circuit voltage is calculated by a recursive least squares method; and the state of charge of the battery module is determined according to the open circuit voltage .

Another technical solution of the present invention provides a battery management system, comprising: a monitoring unit for monitoring current of a battery; and a recording unit, configured to record a terminal voltage of the battery when the current is 0, wherein The recording unit is configured to record the first terminal voltage U _{k of} the battery at the first recording time t _{k and to} record the second terminal voltage U _{k+1 of} the battery at the second recording time t _k+1 The first recording time and the second recording time are two recording times adjacent to each other; the model establishing unit is configured to establish a relationship model between the first terminal voltage and the second terminal voltage, wherein the relationship model includes An open circuit voltage of the battery of one of the parameters to be solved in the relationship model; a calculation unit for calculating the open circuit voltage by using a recursive least squares method; and a determining unit for determining, according to the open circuit voltage, The state of charge of the battery is determined.

Yet another aspect of the present invention provides a battery, the battery including a battery management system, wherein the battery management system is configured to perform an estimation method as previously described.

Yet another aspect of the present invention provides an electric vehicle including the aforementioned battery.

[Description of the Drawings]

The disclosure of the present invention will become more apparent from the drawings. Those skilled in the art will readily appreciate that the drawings are for illustrative purposes only and are not intended to limit the scope of the invention. In the picture:

1 shows a method of estimating a state of charge of a battery according to an embodiment of the present invention;

2 illustrates a method of estimating a state of charge of a battery pack in accordance with an embodiment of the present invention;

FIG. 3 illustrates a method of estimating a state of charge of a battery pack according to an embodiment of the present invention; FIG.

Figure 4 illustrates a battery management system in accordance with one embodiment of the present invention;

Figure 5 illustrates a comparison of estimated open circuit voltages to actual open circuit voltages in accordance with a method in accordance with one embodiment of the present invention;

Figure 6 illustrates the relationship between the state of charge and the open circuit voltage in accordance with one embodiment of the present invention;

Figure 7 illustrates estimated errors for different preloads prior to SOC estimation calibration, in accordance with one embodiment of the present invention;

Figure 8 illustrates estimation errors at different SOC points, in accordance with one embodiment of the present invention.

DETAILED DESCRIPTION

The following description describes specific embodiments of the invention in order to illustrate the invention Some conventional aspects have been simplified or omitted to teach the principles of the invention. Those skilled in the art will appreciate that variations from these embodiments are intended to fall within the scope of the present invention. Those skilled in the art will appreciate that the features described below can be joined in various ways to form multiple variations of the invention. Therefore, the invention is not limited to the specific embodiments described below, but only by the claims and their equivalents.

In the description of the following description, for the convenience of description, it will mainly be developed around a lithium ion battery. However, those skilled in the art will readily appreciate that the methods and technical solutions described herein are equally applicable to other secondary batteries including, but not limited to, nickel cadmium batteries, nickel hydride batteries, and the like.

FIG. 1 illustrates a method 1000 of estimating a state of charge of a battery in accordance with an embodiment of the present invention.

In step 110, the current of the battery is monitored.

In step 120, when the current is zero, the recording the battery voltage of the battery (i.e., voltage), wherein the first recording a first recording time t _k U _k terminal voltage of the battery, whereas in the second recording time t _{k +1} records the second terminal voltage U _{k+1 of} the battery, wherein the first recording time and the second recording time are two recording times adjacent to each other.

In step 130, a relationship model between the first terminal voltage and the second terminal voltage is established, wherein the relationship model includes an open circuit voltage (OCV) of the battery as one of the parameters to be solved in the relationship model.

In step 140, the open circuit voltage is calculated using a recursive least squares method.

In step 150, the state of charge of the battery is determined based on the open circuit voltage.

In the method 1000 described above, the various method steps 110 to 150 are shown in a sequential manner. It should be noted that those skilled in the art will appreciate that the above methods may not be performed in other sequences as shown.

For example, in one implementation, step 130 can be performed prior to

steps

110 and 120. In such an implementation, the relationship model can be predefined. That is to say, the execution of step 130 only needs to select an appropriate relational model according to the type of battery from a plurality of predefined relational models. This operation may occur before the current of the monitoring battery described in step 110 and the recording terminal voltage described in step 120.

In a preferred embodiment, the current of the battery is continuously monitored in step 110, and once the current of the battery is found to be not zero during recording or calculation, the estimate of the state of charge is discarded. In this way, it can be ensured that the method of the invention is always performed in an idle state (ie, when the output current of the battery is zero), so that the dynamic behavior of the battery (eg, due to temperature and aging) does not significantly affect the accuracy of the method. .

In one embodiment, the terminal voltage of the battery is recorded at a fixed time interval at step 120, which is preferably less than or equal to 1 s. In another embodiment, the estimating method 1000 further includes continuously monitoring the terminal voltage of the battery. And in step 120, the terminal voltage of the battery is recorded only when the change in the terminal voltage exceeds a predetermined threshold (preferably, more than 1 mV).

In a preferred embodiment, the relationship model in step 130 can be as follows:

Where [OCV a b] is the parameter matrix to be solved in the relational model, OCV represents the open circuit voltage of the battery, and a and b are coefficients of the least squares method. t _k represents the relaxation time of the data sample, where k is a positive integer representing the sequence number of the sample point.

In another preferred embodiment, the relationship model in step 130 can be as follows:

In order to solve the parameter matrix in the relational model, the present invention proposes to use the recursive least squares method for estimation. For example, in step 140, for each cell in the battery pack, a linear relationship is constructed: Y=A·X, and the parameter matrix A=[OCV a b] is obtained by a least squares recursion method:

In one embodiment, Y = [U _m U _m+1 ... U _n ],

Where m and n are constants and m < n,

Is a recursive coefficient matrix, K is the gain matrix, P is the covariance matrix, and ε is the error matrix.

In another embodiment, Y = [U _m U _m+1 ... U _n ],

Where m and n are constants and m < n,

Is a recursive coefficient matrix, K is a gain matrix, P is a covariance matrix, and ε is an error matrix.

Repeat the recursive calculation in the process of solving the parameter matrix until

The calculated value reaches convergence. Converging at this time

That is, it corresponds to the parameter matrix to be solved. If the convergence or verification fails within the preset time limit t _max (preferably ≤ 10 minutes), the calculation is stopped.

In one embodiment, if |OCV-U ₁ |≤|U _m -U ₁ |, the terminal voltage U _{m is} used as an open circuit voltage in step 150 to determine the state of charge of the battery. Otherwise, if |OCV-U ₁ |>|U _m -U ₁ |, then this OCV (ie, open circuit voltage) is used as the basis for determining the state of charge of the battery in step 150.

In step 150, the corresponding SOC (ie, state of charge) may be derived based on the previously determined open circuit voltage by interpolation calculation of the SOC-OCV equation SOC=f (OCV) or SOC-OCV data table. The SOC-OCV data sheet is provided by the manufacturer of the cell and is related to the type and manufacturing process of the cell.

In a specific implementation, the above method 1000 is applied to lithium of different electrode material combinations such as lithium nickel cobalt manganese oxide (NCM), lithium iron phosphate (LFP), lithium manganese oxide (LMO) and other chemicals. Ion battery.

2 illustrates a method 2000 of estimating a state of charge of a battery pack in accordance with an embodiment of the present invention. Here, the battery pack may include one or more battery modules connected in series and/or in parallel.

In step 210, the current of the battery pack is monitored.

In step 220, when the current is 0, for each of the one or more battery modules in the battery pack, the first terminal voltage U _{k,j of the} battery module is recorded at the first recording time t _k , And recording the second terminal voltage U _{k+1,j of the} battery module at the second recording time t _k+1 , wherein the first recording time and the second recording time are two recording times adjacent to each other, and wherein j is a positive integer Represents the serial number of the battery module in the battery pack.

In step 230, a relationship model between the first terminal voltage and the second terminal voltage is established for the jth battery module, wherein the relationship model includes the one of the parameters to be solved in the relationship model. The open circuit voltage of the battery module.

In step 240, the open circuit voltage is calculated using a recursive least squares method.

In step 250, the charge state of the battery module is determined according to the obtained open circuit voltage. state.

In the method 2000 described above, the various method steps 210 to 250 are shown in a sequential manner. It should be noted that those skilled in the art will appreciate that the above methods may not be performed in other sequences as shown.

For example, in one implementation, step 230 can be performed prior to

steps

210 and 220. For example, the relationship model can be pre-defined. That is to say, the execution of step 250 only needs to select an appropriate relational model according to the type of battery from a plurality of predefined relational models. This operation may occur prior to monitoring the current of the battery pack described in step 210 and the recording terminal voltage described in step 220.

In a preferred embodiment, in step 210, the current of each battery module in the battery pack is continuously monitored, and once the current of the battery module is found to be not 0 during recording or calculation, the battery is discarded. Estimation of the state of charge of the battery module. In this way, it can be ensured that the method of the invention is always performed in an idle state (ie, when the output current of the battery module is zero), so that the dynamic behavior change of the battery pack (for example, due to temperature and aging) does not significantly affect the method. The accuracy.

In one embodiment, the terminal voltage of the battery module is recorded at a fixed time interval at step 220, which is preferably less than or equal to 1 s. In another embodiment, the estimation method 2000 further includes continuously monitoring the terminal voltage of the battery module. And in step 220, the terminal voltage of the battery module is recorded only when the change of the terminal voltage exceeds a preset threshold (preferably, more than 1 mV).

In a preferred embodiment, the relationship model in step 230 can be as follows:

Where [OCV _j a b] is the parameter matrix to be solved in the relational model, OCV _j represents the open circuit voltage of the jth battery module, and a and b are coefficients of the least squares method. t _k represents the relaxation time of the data sample, where k is a positive integer representing the sequence number of the sample point.

In another preferred embodiment, the relationship model in step 230 can be as follows:

In order to solve the parameter matrix in the relational model, the present invention proposes to use the recursive least squares method for estimation. For example, in step 240, for each cell in the battery pack, a linear relationship is constructed: Y=A·X, and the parameter matrix A=[OCV a b] is obtained by a least squares recursion method:

In one embodiment, Y _j =[U _m,j U _m+1,j ... U _n,j ],

Where m and n are constants and m < n,

In another embodiment, Y _j =[U _m,j U _m+1,j ... U _n,j ],

Where m and n are constants and m < n,

The calculated value reaches convergence. Converging at this time

In one embodiment, if |OCV _j -U _1,j |≤|U _m,j -U _1,j |, then the terminal voltage U _{m,j is} used as an open circuit voltage to determine the jth battery in step 250. The state of charge of the module. Otherwise, if |OCV _j -U _1,j |>|U _m,j -U _1,j |, then in step 250, this OCV _j (ie, open circuit voltage) is used as the charging of the jth battery module. The basis of the state.

In step 250, the interpolation f (OCV) or SOC _{_j} -OCV _j data table is calculated by a formula of SOC-OCV SOC =, to derive the corresponding SOC (i.e., State of Charge) based on the previously determined open circuit voltage. The SOC _j -OCV _j data sheet is provided by the manufacturer of the cell and is related to the type and manufacturing process of the cell.

In a specific implementation, the above method 2000 is applied to lithium of different electrode material combinations such as lithium nickel cobalt manganese oxide (NCM), lithium iron phosphate (LFP), lithium manganese oxide (LMO) and other chemicals. Ion battery.

FIG. 3 shows a detailed flow chart of a method 3000 for estimating the state of charge of a battery pack in an electric vehicle application scenario.

In step 310, the state of the vehicle is determined and current is detected.

In step 320, it is determined whether the car is in an idle state and the detected current is zero. If yes, proceed to step 340. Otherwise, the calibration or estimation of the state of charge is abandoned.

In step 340, the terminal voltage U _k,j and the time t _{k are} recorded.

In step 350, it is determined whether k is greater than m, or _k is greater than the time t is t _m, m may be set to the minimum sampling frequency, and t _m represents the shortest relaxation time. If yes, proceed to step 370. Otherwise, return to step 340.

In step 360, a suitable relationship model is selected from a plurality of predefined relationship models and the relationship model is established in step 370, wherein the relationship model includes the jth battery module as one of the parameters to be solved. Open circuit voltage

After the relationship model is established, in step 380, the parameter matrix to be solved is calculated via a recursive least squares method.

Subsequently, in step 390, it is determined whether the parameter matrix converges. If yes, proceed to step 410. Otherwise, in step 400, further determines whether t _k greater than a preset longest relaxation time t _max. If it is greater than the maximum relaxation time, then return to step 310. Otherwise, proceed to step 410.

In step 410, the parameter matrix obtained by the recursive least squares method is verified and the estimated open circuit voltage values are stored in an array of OCVs.

In step 420, the state of charge of the jth battery module is estimated by the relationship between the SOC and the OCV.

Although not shown, it should be understood that method 3000 may further include the step of comparing and updating the estimated state of charge value to the SOC value in the battery management system after performing step 420.

FIG. 4 illustrates a battery management system 4000 in accordance with one embodiment of the present invention.

As shown in FIG. 4, the battery management system 4000 includes a monitoring unit 510, a recording unit 520, a model establishing unit 530, a computing unit 540, and a determining unit 550. In the battery management system 4000, the monitoring unit 510 is used to monitor the current of the battery. The recording unit 520 is configured to record the terminal voltage of the battery when the current is 0, wherein the recording unit 520 is configured to record the first terminal voltage U _{k of the} battery at the first recording time t _k and at the second recording time t _{k +1} records the second terminal voltage U _{k+1 of the} battery, wherein the first recording time and the second recording time are two recording times adjacent to each other. The model establishing unit 530 is configured to establish a relationship model between the first terminal voltage and the second terminal voltage, wherein the relationship model includes an open circuit voltage of the battery as one of the parameters to be solved. The calculation unit 540 is configured to calculate the open circuit voltage by using a recursive least squares method. The determining unit 550 is configured to determine the state of charge of the battery based on the calculated open circuit voltage.

Figure 5 shows a comparison of estimated open circuit voltages to actual open circuit voltages in accordance with a method in accordance with one embodiment of the present invention. In order to verify the accuracy of the method of the present invention, experiments were conducted on a 2.8Ah18650 type NCM battery.

An example of a recursive estimate of the open circuit voltage as described in accordance with the method of the present invention is given in FIG. In FIG. 5, the horizontal axis represents time (t), and the vertical axis represents voltage (V). It can be seen that the recursively calculated open circuit voltage (represented by 610) achieves good convergence within 3 minutes of relaxation time. The difference between the estimated open circuit voltage and the OCV value (shown by dashed line 620) with a relaxation time greater than 2 hours is small. In Figure 5, curve 630 represents the terminal voltage of the battery/battery module.

Figure 6 illustrates the relationship of the state of charge to the open circuit voltage in accordance with one embodiment of the present invention. In FIG. 6, the horizontal axis represents SOC (%), and the vertical axis represents OCV, that is, open circuit voltage (V). This relationship can be stored, for example, in advance in the battery management system of the electric vehicle. As mentioned earlier, the SOC value can be easily interpolated using the estimated OCV as an input.

Figure 7 shows an SOC estimation error analysis for different preloads prior to SOC estimation calibration for 2.8 Ah, 18650 type batteries. The various preloading scenarios are as follows:

Preload 1:0.5C discharge, SOC changes from 100% to 55%;

Preload 2: 0.5C discharge, SOC changes from 60% to 55%;

Preloaded 3:1C discharge, SOC changed from 60% to 55%;

Preload 4: 0.125C discharge, SOC changes from 60% to 55%;

Preload 5: 0.5C charging, SOC changes from 55% to 60%;

Preloaded 6:1C charging, SOC changed from 55% to 60%;

Preload 7: 0.125C charging, SOC changed from 55% to 60%.

It can be seen that the estimation method described using the present invention can be applied to various preloading situations with high precision.

Referring to Figure 8, Figure 8 analyzes the SOC estimation error (%) at different SOC points. All calculations for different SOC points converge for a duration of 90 seconds to 200 seconds. In general, In addition to 65%-75%, good accuracy can be achieved over the entire SOC range. In the SOC range of 0% to 60%, high precision can be achieved, ie the maximum error is <1%.

In summary, various embodiments of the present invention propose a solution that can calibrate/estimate the state of charge of a battery in a short time. For all battery modules in the battery pack, the calibration/estimation time of the state of charge can be as short as 2 to 3 minutes, so that the battery management system can significantly increase its calibration update frequency for the state of charge. In addition, since the solution of the present invention is proposed to be estimated and calibrated in an idle state (ie, when the output current of the battery is zero), dynamic behavior changes of the battery (eg, due to temperature and aging) do not significantly affect the present invention. The accuracy of the method. In addition, the solution of the present invention has the advantages of low computational complexity, easy application, no additional hardware, and the like.

Claims

A method for estimating a state of charge of a battery, the method comprising:

Monitoring the current of the battery;

When the current is 0, the terminal voltage of the battery is recorded, wherein the first terminal voltage U k of the battery is recorded at the first recording time t k , and the battery is recorded at the second recording time t k+1 a second terminal voltage U k+1 , wherein the first recording time and the second recording time are two recording times adjacent to each other;

Establishing a relationship model between the first terminal voltage and the second terminal voltage, wherein the relationship model includes an open circuit voltage of the battery as one of parameters to be solved in the relationship model;

Calculating the open circuit voltage by using a recursive least squares method;

A state of charge of the battery is determined based on the open circuit voltage.
The estimation method according to claim 1, wherein the estimation of the state of charge is discarded when the current of the battery is not 0 during recording or calculation.
The estimation method according to claim 1 or 2, wherein the terminal voltage of the battery is recorded at fixed time intervals.
The estimating method according to claim 1 or 2, further comprising: continuously monitoring the terminal voltage of the battery, and recording the terminal voltage of the battery only when the change in the terminal voltage exceeds a predetermined threshold.
The estimation method according to claim 4, wherein said predetermined threshold is 1 mV.
The estimation method according to claim 1, wherein the relationship model is as follows:

Where [OCV a b] is the parameter matrix to be solved in the relational model, OCV table The open circuit voltage of the battery is shown.
The estimation method according to claim 1, wherein the relationship model is as follows:

Where [OCV a b] is the parameter matrix to be solved in the relational model, and OCV represents the open circuit voltage of the battery.
The estimation method according to claim 6, wherein the calculation of the recursive least squares method is performed by a recursive equation as shown below:

Where Y=[U m U m+1 ... U n ],

m and n are constants, and m<n,
Is a recursive coefficient matrix whose convergence value corresponds to the parameter matrix to be solved, K is the gain matrix, P is the covariance matrix, and ε is the error matrix.
The estimation method according to claim 7, wherein for each of the battery cells, a linear relationship is constructed: Y = A · X, and the parameter matrix A = [OCV is obtained by a least squares recursion method. a b]:

Where Y=[U m U m+1 ... U n ],

m and n are constants, and m<n,
Is a recursive coefficient matrix whose convergence value corresponds to the parameter matrix to be solved, K is the gain matrix, P is the covariance matrix, and ε is the error matrix.
The estimation method according to claim 8 or 9, wherein if it is within a predetermined time
If the convergence is not reached, the calculation of the recursive least squares method is stopped.
The estimation method according to claim 1, wherein the state of charge of said battery is determined based on said open circuit voltage by looking up a SOC-OCV data table, and wherein the type and manufacturing process of the SOC-OCV data table and the battery cell Related.
The estimation method of claim 1, wherein the battery is a lithium ion battery.
A method for estimating a state of charge of a battery pack, the battery pack comprising one or more battery modules, wherein the method comprises:

Monitoring the current of the battery pack;

When the current is 0, for each of the one or more battery modules in the battery pack, the first terminal voltage U k,j of the battery module is recorded at the first recording time t k , And recording the second terminal voltage U k+1,j of the battery module at the second recording time t k+1 , wherein the first recording time and the second recording time are two recording times adjacent to each other And wherein j is a positive integer representing a serial number of the battery module in the battery pack;

Establishing a relationship model between the first terminal voltage and the second terminal voltage for the jth battery module, wherein the relationship model is included as a to-be-solved in the relationship model One of the parameters of the open circuit voltage of the battery module;

Calculating the open circuit voltage by using a recursive least squares method;

And determining a state of charge of the battery module according to the open circuit voltage.
A battery management system, comprising:

a monitoring unit for monitoring the current of the battery;

A recording unit for the current is zero, the terminal voltage of the battery is recorded, wherein the recording unit configured to record the first recording time t k of the first battery terminal voltage U k, whereas in the Recording a second terminal voltage U k+1 of the battery at a recording time t k+1 , wherein the first recording time and the second recording time are two recording times adjacent to each other;

a model establishing unit, configured to establish a relationship model between the first terminal voltage and the second terminal voltage, wherein the relationship model includes an open circuit of the battery as one of parameters to be solved in the relationship model Voltage;

a calculating unit, configured to calculate the open circuit voltage by using a recursive least squares method;

And a determining unit configured to determine a state of charge of the battery according to the open circuit voltage.
A battery, characterized in that the battery comprises a battery management system, wherein the battery management system is configured to perform the estimation method according to any one of claims 1 to 13.
An electric vehicle characterized in that the electric vehicle includes the battery of claim 15.