Battery SOC estimation algorithm, apparatus and device, and storage medium

Abstract

The invention provides a battery SOC estimation algorithm, device and equipment and a storage medium, and relates to the technical field of lithium battery SOC estimation.The SOC calculated on the basis of a discharge current test and the SOC, obtained on the basis of a Thevenin model, of a lithium battery under the same condition are compared with a variance threshold value and a distance threshold value, so that the corrected state of charge and the fused state of charge are obtained, and the SOC estimation accuracy is improved. The corrected charge state and the fused charge state are combined, output data of the two models are fused, an obtained SOC value is corrected through an error correction step in the fusion process, abnormal numerical values are eliminated, correction is carried out, and the state of the battery is continuously monitored. Therefore, the data is more accurate and closer to the actual SOC of the lithium battery.

Description

Technical Field

[0001] The present invention relates to the technical field of lithium battery SOC estimation, specifically to a SOC estimation algorithm, device, equipment and storage medium for batteries. Background Art

[0002] The state estimation (SOC) of a lithium battery refers to the assessment of the current battery charge level, usually expressed as a percentage. The accurate estimation of SOC is crucial for the performance of the battery management system (BMS) because it directly affects the battery's usage efficiency, safety and lifespan. The main methods for estimating the SOC of lithium batteries include the open circuit voltage method, the coulomb counting method, the voltage-current analysis method, the equivalent circuit model method, etc. Among them, the coulomb counting method calculates the change in electric quantity by continuously tracking the current during the charging and discharging process of the battery, and the equivalent circuit model method is based on the Thevenin model to obtain the SOC of the lithium battery.

[0003] The coulomb counting method relies on the accurate measurement of the charging and discharging current of the battery, and any error in the current sensor will directly affect the calculation result of the SOC. Therefore, if the measurement device is not accurate enough or in the case of high-frequency fluctuations, its cumulative error may lead to inaccurate SOC estimation. The equivalent circuit model method usually requires the establishment of a complex circuit model, which requires in-depth research and modeling of the electrochemical characteristics of the battery. Both methods have their own limitations, and using a single method to evaluate the SOC of a lithium battery will result in a large deviation between the evaluation result and the actual value.

[0004] The above information disclosed in the background art section is only used to enhance the understanding of the background of the present disclosure, so it may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0005] The purpose of the present invention is to provide a SOC estimation algorithm, device, equipment and storage medium for batteries to solve the problems raised in the above background art.

[0006] To achieve the above purpose, the present invention provides the following technical solutions:

[0007] A SOC estimation algorithm for a battery, the specific steps include:

[0008] S1. When the lithium battery is discharging, collect the lithium battery discharge parameters, and the lithium battery discharge parameters include the rated capacity of the lithium battery, the initial state of charge, and the discharge current;

[0009] S2. Conduct a correlation analysis on the discharge parameters of the lithium battery to generate a charge consumption amount, which is used to reflect the percentage of the lithium battery's power consumption calculated based on the accumulated amount of the discharge current over time under the discharge state of the lithium battery. Further, conduct a correlation analysis on the charge consumption amount and the initial state of charge to generate a charge estimation state based on the discharge current, which is used to reflect the percentage of the remaining power of the lithium battery calculated based on the charge consumption amount;

[0010] S3. Conduct a correlation analysis on the charge estimation state to generate a charge change rate, which is used to reflect the amount of charge change of the lithium battery per unit time. Conduct a correlation analysis on the charge change rate with a ten-second window to generate a mean charge change, and conduct a correlation analysis on the mean charge change to generate a charge change variance. The mean charge change is used to reflect the amount of change in the average charge of the lithium battery within a ten-second window, and the charge change variance is used to reflect the stability of the charge change every ten seconds;

[0011] S4. Set a variance threshold, and determine whether the charge change variance exceeds the variance threshold. If it exceeds the variance threshold, proceed to step S5. If it does not exceed the variance threshold, conduct a correlation analysis on the charge estimation state and the model estimation state, and output a fused charge state, which is used to reflect the charge state generated through a comprehensive analysis of the percentage of the remaining power of the lithium battery calculated based on the discharge current and the percentage of the remaining power of the lithium battery measured through a model experiment on the lithium battery;

[0012] S5. Error correction. Conduct a correlation analysis on the charge estimation state and the model estimation state, and output a distance parameter between the charge estimation state and the model estimation state, which is used to reflect the Euclidean distance between the charge estimation state and the model estimation state;

[0013] S6. Set a distance threshold. When the distance parameter is greater than the distance threshold, output a corrected charge state, which is used to reflect the remaining power of the lithium battery obtained from the corrected model estimation state. Feed the corrected charge state back to the battery management system and continuously monitor the battery state.

[0014] Furthermore, collect the discharge parameters of the lithium battery. The rated capacity C rated of the lithium battery is the battery capacity of the lithium battery, and the initial state of charge SOC0 is the percentage of the initial remaining power of the lithium battery. The discharge current I T is the current magnitude at the T-th second during the discharge process of the lithium battery. The subscript T is used to index the time, with the unit of seconds and taking integer values.

[0015] Furthermore, conduct a correlation analysis on the discharge parameters of the lithium battery to generate a charge consumption amount CX T The charge consumption amount CXT Used to reflect the percentage of power consumption of a lithium battery after T seconds calculated based on the accumulated amount of discharge current over time. The formula is as follows:

[0016]

[0017] Perform a correlation analysis on the charge consumption CX T and the initial state of charge SOC0 to generate the estimated state of charge SOC based on the discharge current T , and the formula is as follows:

[0018] SOC T = SOC0 - CX T

[0019] The estimated state of charge SOC T is used to reflect the percentage of the remaining power of the lithium battery at the T-th second calculated based on the charge consumption. The subscript T is used for indexing time.

[0020] Furthermore, perform a correlation analysis on the estimated state of charge to generate the rate of change of charge ΔSOC T , and the formula is as follows:

[0021] ΔSOC T = SOC T - SOC T-1

[0022] The rate of change of charge ΔSOC T is used to reflect the change amount of the percentage of charge of the lithium battery per unit time at the T-th second;

[0023] Perform a correlation analysis on the rate of change of charge with a ten-second window to generate the average rate of change of charge Perform a correlation analysis on the average rate of change of charge to generate the variance of the rate of change of charge σ i , and the formula is as follows:

[0024]

[0025] The average rate of change of charge is the average change amount of charge at the i-th second of collecting the discharge parameters of the lithium battery. The variance of the rate of change of charge σ i is used to reflect the smoothness of the change of the lithium battery's power, where i is used for indexing time.

[0026] Furthermore, set the variance threshold σ th to 0.01. When σ i > σ th , execute S5; when σ i ≤ σ th, perform a correlation analysis on the state of charge estimation SOC T and the model estimation state MSOC T to output the fused state of charge RSOC T , and the formula is:

[0027]

[0028] Among them, the model estimation state MSOC T is the percentage of the remaining power of the lithium battery at the Tth second measured by conducting a model experiment on the lithium battery. The fused state of charge RSOC T is used to reflect the percentage state of the remaining charge of the lithium battery generated by the comprehensive analysis of the percentage of the remaining power of the lithium battery calculated based on the discharge current and the percentage of the remaining power of the lithium battery measured by conducting a model experiment on the lithium battery.

[0029] Furthermore, for error correction, perform a correlation analysis on the state of charge estimation state and the model estimation state, and output the distance parameter d between the state of charge estimation SOC T and the model estimation state MSOC T , and the formula is:

[0030] d = |SOC T - MSOC T |

[0031] The distance parameter d is used to reflect the Euclidean distance between the state of charge estimation state and the model estimation state.

[0032] Furthermore, set the distance threshold d th to 0.05. When d > d th , output the corrected state of charge XSOC T , and the formula is:

[0033] XSOC T = MSOC T + k * (MSOC T - SOC T )

[0034] Among them, k is the correction factor, and its value is 0.5;

[0035] When d ≤ d th , let

[0036] Merge the corrected state of charge XSOC T and the fused state of charge RSOC T and feedback the data to the battery management system and continuously monitor the battery state.

[0037] The present invention also provides a device for estimating the state of charge (SOC) of a battery, which is used to execute the SOC estimation algorithm of the battery, including:

[0038] A discharge parameter acquisition module, which is used to acquire the discharge parameters of the lithium battery when the lithium battery is discharging, and the discharge parameters of the lithium battery include the rated capacity of the lithium battery, the initial state of charge, and the discharge current;

[0039] A discharge parameter analysis module, which is used to perform a correlation analysis on the discharge parameters of the lithium battery to generate a charge consumption amount, and the charge consumption amount is used to reflect the percentage of the lithium battery power consumption calculated by the cumulative amount of the discharge current over time under the discharge state of the lithium battery. Further, a correlation analysis is performed on the charge consumption amount and the initial state of charge to generate a charge estimation state based on the discharge current;

[0040] A charge estimation state analysis module, which is used to perform a correlation analysis on the charge estimation state to generate a charge change rate, and the charge change rate is used to reflect the charge change amount of the lithium battery per unit time. A correlation analysis is performed on the charge change rate with a ten-second window to generate a charge change average value, and a correlation analysis is performed on the charge change average value to generate a charge change variance;

[0041] A variance judgment module, which is used to judge whether the charge change variance exceeds a variance threshold. If it exceeds the variance threshold, the error correction module is executed. If it does not exceed the variance threshold, a correlation analysis is performed on the charge estimation state and the model estimation state, and a fused charge state is output;

[0042] An error correction module, which is used to perform a correlation analysis on the charge estimation state and the model estimation state, and output a distance parameter between the charge estimation state and the model estimation state;

[0043] A distance threshold judgment module, which is used to set a distance threshold. When the distance parameter is greater than the distance threshold, a corrected charge state is output, and the charge correction state is used to reflect the remaining power of the lithium battery obtained from the corrected model estimation state. The charge correction state is fed back to the battery management system and the battery state is continuously monitored.

[0044] The present invention also provides a device and a storage medium, on which computer program instructions are stored, and the computer program instructions are executed by a processor to perform the SOC estimation algorithm of the battery.

[0045] Compared with the prior art, the beneficial effects of the present invention are:

[0046] In the present invention, the SOC calculated based on the discharge current test and the SOC of the lithium battery under the same conditions obtained based on the Thevenin model are compared with the variance threshold and the distance threshold, so as to obtain the corrected state of charge and the fused state of charge, and the data is merged and fed back to the battery management system to continuously monitor the battery state. The merged corrected state of charge and fused state of charge not only fuse the output data of the two models, but also correct the obtained SOC value through an error correction step during the fusion process, eliminating abnormal values and making corrections, making the data more accurate and closer to the actual SOC of the lithium battery. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 It is a schematic diagram of the overall algorithm flow of the present invention;

[0048] Figure 2 It is a schematic diagram of the overall device system of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0049] To make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the following further elaborates on the present invention in detail with reference to specific embodiments.

[0050] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should have the ordinary meaning understood by those of ordinary skill in the field to which the present invention belongs. The "first", "second" and similar terms used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. The terms such as "including" or "comprising" mean that the elements or objects appearing before this word cover the elements or objects listed after this word and their equivalents, without excluding other elements or objects. The terms such as "connected" or "coupled" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The terms such as "upper", "lower", "left" and "right" are only used to represent relative positional relationships, and when the absolute position of the object being described changes, the relative positional relationship may also change accordingly.

[0051] Embodiment:

[0052] Please refer to Figure 1 , the present invention provides a technical solution:

[0053] The SOC estimation algorithm of the battery, the specific steps include:

[0054] S1. When the lithium battery is discharging, collect the lithium battery discharge parameters, and the lithium battery discharge parameters include the rated capacity of the lithium battery, the initial state of charge, and the discharge current;

[0055] Collect the lithium battery discharge parameters, the rated capacity of the lithium battery C ratedLet \(C\) be the battery capacity of the lithium battery, and \(SOC_0\) be the initial state of charge, which represents the percentage of the initial remaining power of the lithium battery. In the present invention, the initial state of charge \(SOC_0\) is 100%, and the discharge current is \(I\). T The current magnitude during the discharge process of the lithium battery at the \(T\)-th second, which is measured by an ammeter. The subscript \(T\) is used to index time, with the unit of seconds and taking integer values.

[0056] S2. Conduct a correlation analysis on the discharge parameters of the lithium battery to generate a charge consumption amount. The charge consumption amount is used to reflect the percentage of the lithium battery's power consumption calculated based on the accumulated amount of the discharge current over time under the discharge state of the lithium battery. That is, based on the data analysis and processing of the discharge current, a charge consumption amount is generated, and further a correlation analysis is conducted on the charge consumption amount and the initial state of charge to generate a charge estimation state based on the discharge current. The charge estimation state is used to reflect the percentage of the remaining power of the lithium battery calculated based on the charge consumption amount, that is, the current percentage of the remaining power of the lithium battery generated by the data analysis and processing of the discharge current.

[0057] Conduct a correlation analysis on the discharge parameters of the lithium battery to generate a charge consumption amount \(CX\). T The charge consumption amount \(CX\). T is used to reflect the percentage of the power consumption of the lithium battery after \(T\) seconds calculated based on the accumulated amount of the discharge current over time. The formula is as follows:[[]]END]]

[0058]

[0059] Conduct a correlation analysis on the charge consumption amount \(CX\). T and the initial state of charge \(SOC_0\) to generate a charge estimation state \(SOC\) based on the discharge current. T The formula is as follows:[[]]END]]

[0060] \(SOC\). T \(= SOC_0 - CX\). T

[0061] The charge estimation state \(SOC\). T is used to reflect the percentage of the remaining power of the lithium battery at the \(T\)-th second calculated based on the charge consumption amount. The subscript \(T\) is used to index time.

[0062] S3. Conduct a correlation analysis on the state of charge estimation to generate a state of charge change rate. The state of charge change rate is used to reflect the amount of change in the state of charge of the lithium battery per unit time and is an important indicator for evaluating the remaining power of the lithium battery. Conduct a correlation analysis on the state of charge change rate with a ten-second window to generate a mean state of charge change. For example, the window from 0 - 10 seconds is the first window, 1 - 11 seconds is the second window, 2 - 12 seconds is the third window, and so on. This windowed analysis improves the breadth of the analyzed data and the analysis accuracy, and then conduct a correlation analysis on the mean state of charge change to generate a state of charge change variance. The mean state of charge change is used to reflect the amount of change in the average state of charge of the lithium battery within a ten-second window, and the state of charge change variance is used to reflect the stability of the state of charge change every ten seconds;

[0063] The analysis of the state of charge change rate, the mean state of charge change, and the state of charge change variance helps to deeply understand the discharge characteristics and performance stability of the battery.

[0064] Conduct a correlation analysis on the state of charge estimation state to generate a state of charge change rate ΔSOC T , and the formula is:

[0065] ΔSOC T = SOC T - SOC T-1

[0066] The state of charge change rate ΔSOC T is used to reflect the percentage change in the state of charge of the lithium battery per unit time at the T-th second. Among them, SOC T-1 is used to reflect the percentage of the remaining power of the lithium battery at the (T - 1)-th second calculated based on the consumed state of charge, and the subscript T - 1 is used for indexing time.

[0067] Conduct a correlation analysis on the state of charge change rate with a ten-second window to generate a mean state of charge change Conduct a correlation analysis on the mean state of charge change to generate a state of charge change variance σ i , and the formula is:

[0068]

[0069] The mean state of charge change is the amount of change in the average state of charge at the i-th second when collecting the discharge parameters of the lithium battery. The state of charge change variance σ i is used to reflect the smoothness of the change in the state of charge of the lithium battery, where i is used for indexing time.

[0070] S4. Set the variance threshold, and determine whether the variance of the charge change exceeds the variance threshold. If it exceeds the variance threshold, proceed to step S5. If it does not exceed the variance threshold, perform a correlation analysis on the charge estimation state and the model estimation state, and output the fused charge state. The fused charge state is used to reflect the charge state generated by a comprehensive analysis of the percentage of the remaining battery charge of the lithium battery calculated based on the discharge current and the percentage of the remaining battery charge of the lithium battery measured through model experiments on the lithium battery.

[0071] When setting the variance threshold, first, collect the charge state change rate data of the lithium battery under different temperature working conditions (-10 to 80 degrees Celsius). These data should cover multiple discharge tests to obtain comprehensive performance characteristics. Process the charge state change rate data, calculate the mean and variance of the charge state change rate, and use the empirical rule to set the variance threshold. Set the threshold as the mean plus or minus a certain multiple of the standard deviation, and select it within the interval. In the present invention, the set variance threshold σ th is 0.01. When σ i >σ th , execute S5; when σ i ≤σ th , perform a correlation analysis on the charge estimation state SOC T and the model estimation state MSOC T , and output the fused charge state RSOC T . The formula is as follows:

[0072]

[0073] where the model estimation state MSOC T is the percentage of the remaining battery charge of the lithium battery at the T-th second measured through model experiments on the lithium battery. The fused charge state RSOC T is used to reflect the percentage state of the remaining battery charge of the lithium battery generated by a comprehensive analysis of the percentage of the remaining battery charge of the lithium battery calculated based on the discharge current and the percentage of the remaining battery charge of the lithium battery measured through model experiments on the lithium battery. When conducting model experiments, the equivalent circuit model method is used to measure the percentage state of the remaining battery charge of the lithium battery. Specifically, the Thevenin model is used to obtain the percentage state of the remaining battery charge of the lithium battery.

[0074] When σ i ≤σ th , through data processing of the charge estimation state SOC T and the model estimation state MSOC T , obtain the fused charge state RSOC T . Perform a mean process on the values of the two evaluation models to obtain a more accurate SOC of the lithium battery.

[0075] When σ i > σ th , execute S5 for error correction. When σ i > σ th , it indicates that the data obtained for the state of charge estimation exceeds the normal variance threshold range obtained and set based on experiments, that is, the probability of abnormality of this data is relatively high. Therefore, S5 is to perform a correlation analysis on the state of charge estimation state and the model estimation state, and output the distance parameter between the state of charge estimation state and the model estimation state. The distance parameter is used to reflect the Euclidean distance between the state of charge estimation state and the model estimation state, and is used to determine the gap between the two estimation models;

[0076] Perform a correlation analysis on the state of charge estimation state and the model estimation state, and output the state of charge estimation state SOC T and the model estimation state MSOC T The distance parameter d between them is based on the formula:

[0077] d = |SOC T - MSOC T |

[0078] The distance parameter d is used to reflect the Euclidean distance between the state of charge estimation state and the model estimation state. The distance parameter d is used to reflect the gap between the SOC values obtained by the two models under the same conditions.

[0079] S6. Set a distance threshold. When the distance parameter is greater than the distance threshold, output the corrected state of charge. The corrected state of charge is used to reflect the remaining battery power of the lithium battery obtained from the corrected model estimation state, and feedback the corrected state of charge to the battery management system and continuously monitor the battery state.

[0080] The threshold is usually set to judge stability and trigger the error correction mechanism when a large error occurs. Set the distance threshold d th to 0.05, that is, 5%. When the deviation exceeds 5%, that is, when d > d th , output the corrected state of charge XSOC T , and the formula is:

[0081] XSOC T = MSOC T + k * (MSOC T - SOC T )

[0082] The distance threshold can be set to other values. The smaller the set value, the more accurate the finally output SOC value. k is the correction factor, set to 0.5. The corrected state of charge XSOC T is based on the model estimation state MSOC T and the state of charge estimation state SOCT Perform numerical correction, eliminate abnormal values and make corrections to make the estimated SOC value more accurate;

[0083] When d ≤ d th Let Corrected state of charge XSOC T Take the model estimated state MSOC T And the state of charge estimated state SOC T Of the intermediate value.

[0084] The corrected state of charge XSOC T And the fused state of charge RSOC T Perform data merging and feedback to the battery management system and continuously monitor the battery status. The merged corrected state of charge XSOC T And the fused state of charge RSOC T Not only fuses the output data of the two models, but also corrects the obtained SOC value through an error correction step during the fusion process, eliminates abnormal values and makes corrections, making the data more accurate and closer to the actual SOC of the lithium battery.

[0085] Refer to Figure 2 , The present invention also provides a device for estimating the SOC of a battery, which is used to execute the SOC estimation algorithm of the battery, including:

[0086] Discharge parameter acquisition module, the discharge parameter acquisition module is used to collect the discharge parameters of the lithium battery when the lithium battery is discharging, and the discharge parameters of the lithium battery include the rated capacity of the lithium battery, the initial state of charge, and the discharge current;

[0087] Discharge parameter analysis module, the discharge parameter analysis module is used to perform a correlation analysis on the discharge parameters of the lithium battery to generate a charge consumption amount, and the charge consumption amount is used to reflect the percentage of the battery power consumption calculated by the accumulated amount of the discharge current over time under the discharge state of the lithium battery. Further, a correlation analysis is performed on the charge consumption amount and the initial state of charge to generate a state of charge estimation state based on the discharge current;

[0088] State of charge estimation state analysis module, the state of charge estimation state analysis module is used to perform a correlation analysis on the state of charge estimation state to generate a state of charge change rate, and the state of charge change rate is used to reflect the amount of charge change of the lithium battery per unit time. A correlation analysis is performed on the state of charge change rate with a ten-second window to generate an average state of charge change, and a correlation analysis is performed on the average state of charge change to generate a state of charge change variance;

[0089] Variance judgment module, the variance judgment module is used to judge whether the state of charge change variance exceeds the variance threshold. If it exceeds the variance threshold, the error correction module is executed. If it does not exceed the variance threshold, a correlation analysis is performed on the state of charge estimation state and the model estimation state, and the fused state of charge is output;

[0090] An error correction module is used to perform a correlation analysis on the charge estimation state and the model estimation state, and output a distance parameter between the charge estimation state and the model estimation state.

[0091] A distance threshold judgment module is used to set a distance threshold. When the distance parameter is greater than the distance threshold, it outputs a corrected charge state. The charge correction state is used to reflect the remaining battery power of the lithium battery obtained from the corrected model estimation state, and feedbacks the charge correction state to the battery management system and continuously monitors the battery state.

[0092] The present invention also provides a device and a storage medium, on which computer program instructions are stored, and the computer program instructions are executed by a processor to perform the SOC estimation algorithm of the battery.

[0093] The above formulas are all dimensionless and take their numerical calculations. The formula is a formula obtained by software simulation of a large amount of collected data to be closest to the actual situation. The preset parameters in the formula are set by those skilled in the art according to the actual situation.

[0094] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any arbitrary combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art can realize that the units and algorithm steps of the examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed by hardware or software methods depends on the specific application and design constraints of the technical solution.

[0095] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units. They can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0096] The above is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in this application, and all should be covered by the protection scope of this application.

Claims

1. SOC estimation algorithm for a battery, characterized in that, The specific steps include: S1. When the lithium battery is discharging, collect the lithium battery discharge parameters, where the lithium battery discharge parameters include the rated capacity of the lithium battery, the initial state of charge, and the discharge current; S2. Conduct a correlation analysis on the lithium battery discharge parameters to generate a charge consumption amount, where the charge consumption amount is used to reflect the percentage of the lithium battery's power consumption calculated by the accumulated amount of the discharge current over time under the lithium battery's discharge state. Further, conduct a correlation analysis on the charge consumption amount and the initial state of charge to generate a charge estimation state based on the discharge current, where the charge estimation state is used to reflect the percentage of the remaining power of the lithium battery calculated based on the charge consumption amount; S3. Conduct a correlation analysis on the charge estimation state to generate a charge change rate, where the charge change rate is used to reflect the amount of charge change of the lithium battery per unit time. Conduct a correlation analysis on the charge change rate with a ten-second window to generate a charge change average value, and conduct a correlation analysis on the charge change average value to generate a charge change variance. The charge change average value is used to reflect the average amount of charge change of the lithium battery within the ten-second window, and the charge change variance is used to reflect the stability of the charge change every ten seconds; S4. Set a variance threshold, and determine whether the charge change variance exceeds the variance threshold. If it exceeds the variance threshold, proceed to step S5. If it does not exceed the variance threshold, conduct a correlation analysis on the charge estimation state and the model estimation state, and output a fused charge state. The fused charge state is input into the battery management system, where the fused charge state is used to reflect the charge state generated by the comprehensive analysis of the percentage of the remaining power of the lithium battery calculated based on the discharge current and the percentage of the remaining power of the lithium battery measured through model experiments on the lithium battery; S5. Conduct a correlation analysis on the charge estimation state and the model estimation state, and output a distance parameter between the charge estimation state and the model estimation state, where the distance parameter is used to reflect the Euclidean distance between the charge estimation state and the model estimation state; S6. Set a distance threshold. When the distance parameter is greater than the distance threshold, output a corrected charge state, where the charge correction state is used to reflect the remaining power of the lithium battery obtained from the corrected model estimation state. Feed the charge correction state back to the battery management system and continuously monitor the battery state.

2. The SOC estimation algorithm of the battery according to claim 1, characterized in that: Collect the discharge parameters of the lithium battery, where the rated capacity C of the lithium battery rated is the battery capacity of the lithium battery, the initial state of charge SOC0 is the percentage of the initial remaining power of the lithium battery, and the discharge current I T is the current magnitude at the T-th second during the discharge process of the lithium battery. The subscript T is used to index time, with the unit of seconds and taking integer values.

3. The SOC estimation algorithm for the battery according to claim 2, characterized in that: Perform a correlation analysis on the discharge parameters of the lithium battery to generate the charge consumption CX T , the charge consumption CX T is used to reflect the percentage of the battery power consumption of the lithium battery after T seconds calculated based on the accumulated amount of the discharge current over time. The formula is as follows: Regarding the charge consumption CX T and the initial state of charge SOC0, a correlation analysis is performed to generate the estimated state of charge SOC based on the discharge current T , and the formula used is: SOC T = SOC0 - CX T State of Charge (SOC) for charge estimation T It is used to reflect the percentage of the remaining battery power at the T-th second calculated based on the charge consumption, and the subscript T is used to index time.

4. The SOC estimation algorithm of the battery according to claim 3, characterized in that: Perform a correlation analysis on the state of charge estimation to generate the rate of change of charge ΔSOC T , and the formula used is as follows: ΔSOC T = SOC T - SOC T-1 Rate of change of charge ΔSOC T It is used to reflect the change amount of the charge percentage of the lithium battery per unit time at the T-th second; Perform a correlation analysis on the charging change rate with a ten-second window to generate the average charging change For the average charging change Perform a correlation analysis to generate the charging change variance σ i , and the formula used is: Average charge change It is the change amount of the average charge at the i-th second of the discharge parameters of the lithium battery being collected. The charge change variance is σ i It is used to reflect the smoothness of the change in the battery power of the lithium battery, where i is the index for time.

5. The SOC estimation algorithm of the battery according to claim 4, characterized in that: Set the variance threshold σ th to 0.

01. When σ i > σ th , execute S5; when σ i ≤ σ th , perform a correlation analysis on the charged state estimate SOC T and the model state estimate MSOC T to output the fused state of charge RSOC T . The formula used is: Among them, the model estimation state MSOC T is the percentage of the remaining power of the lithium battery at the T-th second measured by conducting a model experiment on the lithium battery. The integrated state of charge RSOC T is used to reflect the percentage state of the remaining charge of the lithium battery generated by comprehensively analyzing the percentage of the remaining power of the lithium battery calculated based on the discharge current and the percentage of the remaining power of the lithium battery measured by conducting a model experiment on the lithium battery.

6. The SOC estimation algorithm for the battery according to claim 5, characterized in that: Error correction, perform correlation analysis on the state of charge estimation and the state of model estimation, and output the state of charge estimation SOC T and the state of model estimation MSOC T The distance parameter d between them is based on the formula: d = |SOC T - MSOC T | The distance parameter d is used to reflect the Euclidean distance between the charge estimation state and the model estimation state.

7. The SOC estimation algorithm of the battery according to claim 6, wherein: Set the distance threshold d th to 0.

05. When d > d th , output the corrected state of charge XSOC T , and the formula is as follows: XSOC T = MSOC T + k * (MSOC T - SOC T ) Where k is a correction factor with a value of 0.5; When d ≤ d th , let The corrected state of charge XSOC T and the integrated state of charge RSOC T are used for data merging and feedback to the battery management system, and the battery state is continuously monitored.

8. An SOC estimation device for a battery, which is used to execute the SOC estimation algorithm for the battery described in claim 1, characterized in that, It includes: A discharge parameter collection module, which is used to collect the lithium battery discharge parameters when the lithium battery is discharging, where the lithium battery discharge parameters include the rated capacity of the lithium battery, the initial state of charge, and the discharge current; A discharge parameter analysis module, which is used to conduct a correlation analysis on the lithium battery discharge parameters to generate a charge consumption amount, where the charge consumption amount is used to reflect the percentage of the lithium battery's power consumption calculated by the accumulated amount of the discharge current over time under the lithium battery's discharge state. Further, conduct a correlation analysis on the charge consumption amount and the initial state of charge to generate a charge estimation state based on the discharge current; A charged state estimation status analysis module, which is used to perform a correlation analysis on the charged state estimation status to generate a charge change rate. The charge change rate is used to reflect the amount of charge change of the lithium battery per unit time. A correlation analysis is performed on the charge change rate with a ten-second window to generate a mean charge change, and a correlation analysis is performed on the mean charge change to generate a charge change variance; A variance judgment module, which is used to judge whether the charge change variance exceeds a variance threshold. If it exceeds the variance threshold, the error correction module is executed. If it does not exceed the variance threshold, a correlation analysis is performed on the charged state estimation status and the model estimation status, and a fused charged state is output; An error correction module, which is used to perform a correlation analysis on the charged state estimation status and the model estimation status, and output a distance parameter between the charged state estimation status and the model estimation status; A distance threshold judgment module, which is used to set a distance threshold. When the distance parameter is greater than the distance threshold, a corrected charge state is output. The charge correction state is used to reflect the remaining power of the lithium battery obtained from the corrected model estimation state, and the charge correction state is fed back to the battery management system and the battery state is continuously monitored.

9. A device and a storage medium, characterized in that: Computer program instructions are stored on a device and a storage medium, and when the computer program instructions are executed by a processor, the algorithm described in claim 1 is implemented.

Images