Construction method of real-time complete ocv-soc curve of lithium battery based on cloud data

Abstract

The application discloses a kind of based on cloud data's lithium battery real-time complete OCV-SOC curve construction method, comprising the following steps: step one, battery model is constructed;Step two, identification discharge segment OCV, identification discharge segment OCV based on analogy method;Step three, OCV-SOC model is constructed, by analyzing the relationship between the lithium intercalation rate of battery electrode in electrode potential model and SOC, electrode potential expression is improved, obtains OCV-SOC model;Step four, based on the identification complete OCV-SOC model in charging stage, including ohmic resistance change trend analysis, charging stage characteristic analysis, complete OCV-SOC solution, OCV-SOC relationship real-time update.Affirmative effect: the battery OCV of the present application is obtained without disassembling real vehicle battery pack, corrects cloud SOC value, and the method is simple, easy to implement, solves the problem of low precision of cloud data and poor battery state estimation accuracy.

Description

Technical Field

[0001] This invention relates to a method for constructing the OCV-SOC curve of a power battery, and particularly to a method for constructing a real-time complete OCV-SOC curve of a lithium battery based on cloud data, belonging to the field of power battery technology. Background Technology

[0002] Accurate estimation of battery state of charge (SOC) can prevent overcharging and over-discharging, improve battery performance, and extend battery life. Among numerous SOC estimation algorithms, the open-circuit voltage method and the Kalman filter algorithm are the most commonly used. The open-circuit voltage method obtains the open-circuit voltage (OCV) corresponding to different SOCs through long-term static tests, and then establishes the relationship between OCV and SOC to achieve SOC estimation based on open-circuit voltage. The Kalman filter algorithm establishes corresponding state equations and observation equations based on the battery model. Through the observed and measured values ​​of the system output, it performs the optimal estimation of the system state in the sense of minimum variance, thereby achieving SOC estimation. The establishment of the observation equation still uses the OCV-SOC curve as a benchmark and the terminal voltage as a feedback signal to achieve closed-loop correction of the SOC estimate. The accuracy of the OCV-SOC relationship has a direct impact on battery SOC estimation, and the accuracy of battery state of health (SOH) estimation also depends on the OCV-SOC relationship. Therefore, accurate OCV-SOC curves are crucial for improving the accuracy of battery state estimation.

[0003] Since long-term static operation conditions are generally not present in real-world vehicle engineering, many researchers initially studied the relationship between OCV and SOC using laboratory data. However, experimental environments or simulations are often too idealized and difficult to simulate the complex and variable operating conditions of real-world vehicles. In recent years, research on the OCV-SOC relationship of lithium-ion batteries under actual operating conditions has gradually increased. For example, some studies identify OCV based on segmented real-world vehicle data, stitch the identification results into a long OCV-capacity (Ampere-hour, Ah) curve, and then use a laboratory-constructed OCV-Ah database to look up the ends of the identified OCV-Ah curve to obtain a complete OCV-Ah curve. Other studies identify OCV-SOC curves based on segmented real-world vehicle data, construct a real-world vehicle data segment set based on data fragments, establish a reference dataset based on measured SOC and capacity, and stitch the identification results of the real-world vehicle data segment set into a complete OCV-SOC curve according to the reference dataset. However, previous studies have all required the establishment of large reference databases and have calculated OCV using battery capacity as the sole parameter, which significantly increases time costs. Furthermore, the initial SOC value of the data fragments greatly affects the identification results. In reality, data uploaded to big data centers, i.e., the cloud, typically suffers from reduced accuracy, and the aforementioned studies did not consider SOC correction when identifying OCV-SOC. Additionally, these studies failed to account for the real-time changes in the voltage plateau trend in the middle of the OCV-SOC curve caused by different phase transitions during each charge-discharge cycle.

[0004] Therefore, it is impossible to construct a model that can adapt to different temperatures and accurately describe OCV-SOC, and it is impossible to construct a complete OCV-SOC curve based on the charging stage and realize the real-time updating of the OCV-SOC curve. Summary of the Invention

[0005] Objective: To address the problems existing in the prior art, this invention provides a method for constructing a real-time complete OCV-SOC curve for lithium batteries based on cloud data. This invention uses an analogy method to identify the OCV-SOC relationship of discharge segments in cloud data and compares it with OCV-SOC curves measured in the laboratory to analyze the battery characteristics under real-vehicle operating conditions and experimental tests. By analyzing the characteristics of the battery's OCV-SOC curve, a model that can adapt to different temperatures and accurately describe OCV-SOC is constructed. By analyzing the characteristics of the charging stage, a complete OCV-SOC curve for the charging stage is reconstructed, and real-time updates of the OCV-SOC curve are achieved.

[0006] Technical solution: A method for constructing a real-time complete OCV-SOC curve for a lithium battery based on cloud data, comprising the following steps:

[0007] Step 1: Construct a battery model;

[0008] Step 2: Identify the discharge segment OCV based on the analogy method;

[0009] Step 3: Construct the OCV-SOC model. By analyzing the relationship between the lithium insertion rate of the battery electrode and SOC in the electrode potential model, the electrode potential expression is improved to obtain the OCV-SOC model.

[0010] Step 4: Identify the complete OCV-SOC model based on the charging stage, including analysis of the ohmic internal resistance change trend, analysis of charging stage characteristics, complete OCV-SOC solution, and real-time update of OCV-SOC relationship.

[0011] Furthermore, the battery model constructed in step one is a first-order RC equivalent circuit model, and the equations of this model are as follows:

[0012]

[0013] In the formula, U OCV I is the open-circuit voltage, R0 is the operating current, R1 is the internal resistance in ohms, C1 is the polarization internal resistance, C1 is the polarization capacitor, U1 represents the polarization voltage, i.e., the voltage across R1C1, and U is the terminal voltage.

[0014] Furthermore, in step two, the discharge segment OCV is identified based on the analogy method, using the functional relationship between the terminal voltage U and time t during the zero-input response stage under the battery hybrid power pulse (HPPC) condition. The specific functional relationship is as follows:

[0015]

[0016] In the formula, e is the natural constant, and τ1=R1C1 is the time constant;

[0017] Under HPPC operating conditions, the terminal voltage U during the zero-input response phase follows an exponential function relationship with time t. The expression for the exponential function is:

[0018]

[0019] In the formula, y0 is U OCV The generalized representation of , yes In the generalized representation, A1 and t1 represent the generalized coefficients, and x represents the generalized dependent variable;

[0020] Comparing equations (2) and (3), the battery OCV can be obtained as follows:

[0021] y0=U OCV (4)

[0022] The data segments in the cloud data where the current fluctuates around 0A are considered as the zero-input response stage. The voltage and acquisition time data from at least three consecutive data segments where the current fluctuates around 0A are selected and substituted into equation (3) for fitting, and then the parameter y0 is obtained. The open-circuit voltage U of the battery pack is obtained according to equation (4). OCV ;

[0023] The average open-circuit voltage U of a single cell is calculated based on the battery pack assembly method. OCV,dis The calculation formula is as follows:

[0024]

[0025] In the formula, n is the number of batteries connected in series in the battery pack.

[0026] Furthermore, in step three, the relationship between the lithium insertion rate and SOC of the battery electrode in the electrode potential model is analyzed, and the electrode potential expression is improved to obtain the OCV-SOC model. The specific improvement method is as follows:

[0027] First, the electrode potential model expression consists of three main parts, as follows:

[0028]

[0029] In the formula, U(x) is the electrode potential, x is the lithium insertion rate, and a1, b1, b2, c1, c2, d i e i and f i (i is the number of terms in the hyperbolic tangent (tanh) function) are the relevant parameters of the electrode potential at the corresponding temperature, and all are rational numbers greater than 0;

[0030] Analyzing the relationship between the lithium insertion rate at the battery cathode and the state of charge (SOC), we replace x in equation (6) with 1-s, thereby establishing a model describing the OCV-SOC relationship of the full battery, which is specifically expressed as follows:

[0031]

[0032] In the formula, s is the battery SOC.

[0033] Furthermore, the expression (7) of the full-cell OCV-SOC relationship model is decomposed into a constant term ①, an exponential term ②, and a tangent function term ③; wherein, in the constant term ①, a1 is used to represent the vertical shift of the open-circuit voltage curve of the battery under different temperatures and aging conditions; in the exponential term ②, and The terms describe the changing trends at both ends of the open-circuit voltage curve; the hyperbolic tangent function term ③ is used to describe the voltage plateau in the middle part of the open-circuit voltage curve caused by the phase transition. It is described by one or more hyperbolic tangent functions, and the number of terms increases with the increase of the corresponding plateau.

[0034] Furthermore, in step four, the trend analysis of the ohmic internal resistance variation is based on the relationship curves of the ohmic internal resistance R0 and SOC measured in the laboratory at different temperatures, and the variation law of R0 is analyzed.

[0035] Furthermore, the charging stage characteristic analysis includes the analysis of charging current variation trend, the analysis of the relationship between charging voltage and open circuit voltage under charging conditions, and the analysis of charging voltage curve trend within different temperature ranges.

[0036] Furthermore, the charging current variation trend analysis is based on the collected cloud data to analyze the variation of the average charging current of a single battery under a typical charging condition.

[0037] Analysis of the relationship between charging voltage and open-circuit voltage under charging conditions: Based on the expression for open-circuit voltage under low charging rate conditions in equation (8), the trends of the open-circuit voltage curve and the charging voltage curve are analyzed.

[0038] U OCV,C =U C -I C R0 (8)

[0039] In the formula, U OCV,C For OCV under charging conditions, U C I is the charging voltage. C This is the charging current;

[0040] The analysis of charging voltage curve trends in different temperature ranges involves first randomly sampling charging segment voltages from cloud data within different temperature ranges and calculating the average charging voltage of a single cell; then analyzing the changing trends of the average charging voltage curves at different temperatures; and finally comparing and analyzing the average charging voltage curves with the OCV-SOC curves measured in the laboratory at that temperature.

[0041] Furthermore, the complete OCV-SOC solution includes the following steps:

[0042] First, the average charging voltage curve of the charging segment is based on the OCV-SOC curve measured in the laboratory at the corresponding temperature of the segment, and the SOC value is corrected by shifting it laterally by a certain distance k.

[0043] Then, the curve with the corrected SOC value is compared with the OCV-SOC curve measured in the laboratory at the corresponding temperature of that segment. By shifting it longitudinally by a certain distance b, it is made to coincide as much as possible with the laboratory OCV-SOC curve, thereby obtaining the U value at the corresponding SOC. OCV,C ;

[0044] Substituting the OCV-SOC data measured in the laboratory at the corresponding temperature into equation (7), and fitting it using the least squares method, we obtain a specific expression for the OCV-SOC relationship curve based on the improved electrode potential model. Based on the obtained expression, following the principle of "adding on the left and subtracting on the right, adding on the top and subtracting on the bottom," we add a parameter b in the vertical direction and subtract a parameter k in the horizontal direction to obtain the charging voltage-SOC relationship curve model. The specific expression is:

[0045]

[0046] In the formula, b is the difference between the charging voltage and OCV in the charging stage, and k is the SOC correction value for the charging segment.

[0047] By inputting the data from the charging segment, the values ​​of b and k can be obtained;

[0048] Finally, if the shifted curve covers the 0-1SOC range and meets the open-circuit voltage characteristics, then the following steps are unnecessary; otherwise, the following steps are required:

[0049] The OCV-SOC of the shifted curve segment that conforms to the open-circuit voltage characteristic is spliced ​​and fused with the OCV-SOC measured in the laboratory at the corresponding temperature to obtain the complete OCV-SOC curve.

[0050] Furthermore, the OCV-SOC relationship is updated in real time. First, the cloud data is divided into multiple "charge-discharge" units according to the charging stage. Each "charge-discharge" unit consists of a complete charging stage and a discharging stage from the charging stage to the next charging stage. Then, based on the complete OCV-SOC data obtained from Equation (7) and the charging segment, the least squares method is used for fitting, and the OCV-SOC curve relationship parameters of each unit are updated, thereby realizing the real-time update of the OCV-SOC relationship curve.

[0051] Beneficial Effects: This invention obtains the OCV-SOC relationship of cloud data discharge segments through analogy; then, it analyzes the battery characteristics under different operating temperatures and aging conditions, and constructs a model that can accurately describe the OCV-SOC relationship; finally, based on charging stage data and the constructed relationship model, it solves for the complete OCV-SOC curve and realizes real-time updating of the OCV-SOC curve. This invention employs a SOC-OCV model that can adapt to different temperatures, constructing and updating complete OCV-SOC curves at different temperatures. The proposed method can not only obtain the battery OCV and correct the cloud SOC value without disassembling the battery pack, but is also simple and easy to implement, overcoming the problems of low accuracy of cloud data and poor accuracy of battery state estimation caused by directly using measurement information. Attached Figure Description

[0052] Figure 1 This is a trend diagram of the change in current and voltage during the HPPC cyclic process of the present invention;

[0053] Figure 2 This is a comparison chart of the fitting and laboratory test open-circuit voltages under different data point numbers according to the present invention;

[0054] Figure 3 This is a trend diagram of the battery OCV-SOC curve at different temperatures according to the present invention;

[0055] Figure 4 This is a trend chart of the battery OCV-SOC curve under different charge-discharge cycles according to the present invention;

[0056] Figure 5 This is a graph showing the fitting results of the improved electrode potential model of this invention;

[0057] Figure 6 This is a graph showing the variation trend of the battery's internal resistance at different temperatures according to the present invention.

[0058] Figure 7 This is a graph showing the trend of average current variation of a single cell in the charging segment at different temperatures according to the present invention.

[0059] Figure 8 This is a comparison chart of the cloud-based charging voltage curve and the OCV-SOC curve at different laboratory temperatures.

[0060] Figure 9 This is a schematic diagram of the principle of solving the OCV-SOC curve based on the charging segment of the present invention;

[0061] Figure 10 This is a diagram showing the splicing and fusion result of the present invention;

[0062] Figure 11 These are OCV-SOC curves after correction for different charging segments according to the present invention;

[0063] Figure 12 This is a diagram showing the updated results of the complete OCV-SOC curve during the charging phase of this invention. Detailed Implementation

[0064] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, but the scope of protection of the present invention is not limited thereto.

[0065] A method for constructing a real-time complete OCV-SOC curve for a lithium battery based on cloud data includes the following steps:

[0066] Step 1: Construct a battery model;

[0067] The battery model is constructed using a first-order RC equivalent circuit model to describe its characteristics. However, the model is not limited to this; it is used here merely for ease of explanation. The mathematical equation describing this model can be expressed as:

[0068]

[0069] In the formula, U OCV I is the open-circuit voltage, R0 is the operating current, R1 is the internal resistance in ohms, C1 is the polarization internal resistance, C1 is the polarization capacitor, U1 represents the polarization voltage, i.e., the voltage across R1C1, and U is the terminal voltage.

[0070] Step 2: Identify the discharge segment OCV based on the analogy method;

[0071] like Figure 1 As shown, the DE stage is the resting stage after pulse discharge – the “rebound rise” stage, where the terminal voltage rises rapidly and then gradually stabilizes, which is the “rebound characteristic” of the battery. The process of polarized capacitor discharge is the zero-input response stage of the RC circuit.

[0072] The functional relationship between the terminal voltage U and time t during the zero-input response stage under battery hybrid power pulse operation is as follows:

[0073]

[0074] In the formula, e is the natural constant, and τ1=R1C1 is the time constant.

[0075] Under HPPC operating conditions, the terminal voltage U during the zero-input response phase follows an exponential function relationship with time t. The expression for the exponential function is:

[0076]

[0077] In the formula, y0 is U OCV The generalized representation of , where A1 and t1 represent generalized coefficients, and x represents the generalized dependent variable.

[0078] Comparing equations (2) and (3), the battery OCV can be obtained as follows:

[0079] y0=U OCV (4)

[0080] The continuous data segments with current in the range of 0A to 0.9A in the cloud data of lithium iron phosphate batteries are regarded as the zero-input response stage. The voltage and acquisition time data of 5, 10, 15, 20 and 25 consecutive data segments with current in the range of 0A to 0.9A are selected and substituted into equation (3) for fitting, and then the parameter y0 is obtained. The open circuit voltage U of the battery pack is obtained according to equation (4). OCV .

[0081] Considering that the lithium iron phosphate battery pack is assembled in a 14-parallel, 180-series configuration, the average open-circuit voltage of a single cell can be expressed as:

[0082]

[0083] The above analysis of battery characteristics under actual vehicle operating conditions, such as Figure 2 The figure shows a comparison between the average open-circuit voltage of a single cell obtained by fitting at different numbers of data points and the OCV-SOC curve characteristics tested in the laboratory. Figure 2 As can be seen from the figure, during actual vehicle operation, most points of the average open-circuit voltage of the individual cells obtained by fitting fall near the OCV-SOC curve measured in the laboratory, and the trend of the fitted scatter points is almost consistent with the trend of the OCV-SOC curve measured in the laboratory. The number of fitted points has little impact on the identification of open-circuit voltage, indicating that the method of using cloud data to identify parameters by analogy with the zero-input response stage of HPPC is feasible. The figure also shows that there are basically no open-circuit voltage scatter points in the SOC range of 0 to 0.2. This is because in actual vehicle operation, the battery capacity is basically maintained above SOC 0.2, and it is difficult to completely discharge the battery.

[0084] Step 3: Construct the OCV-SOC model, including OCV-SOC curve characteristic analysis. By analyzing the relationship between the lithium insertion rate of the battery electrode and SOC in the electrode potential model, the electrode potential expression is improved to obtain the OCV-SOC model.

[0085] The OCV-SOC curve characteristic analysis is based on the OCV-SOC relationship curves at different temperatures and under different charge-discharge cycles, to analyze the OCV-SOC curve characteristics of lithium-ion batteries under different temperatures and aging conditions.

[0086] like Figure 3 The figure shows the OCV-SOC relationship curves of a single cell in a lithium iron phosphate battery pack at three temperatures: 5℃, 25℃, and 55℃. As can be seen from the figure, the OCV-SOC curves are almost identical in shape at different temperatures within the range of 0.1 to 0.95 SOC, especially the trends of curve characteristics are basically consistent. The vertical shifts of the curves at different temperatures are also small, indicating that the OCV-SOC curve characteristics of the battery remain almost unchanged.

[0087] like Figure 4 The figure shows the OCV-SOC relationship curves of a single cell in a lithium iron phosphate battery pack after 100, 300, and 500 charge-discharge cycles at 25°C. As can be seen from the figure, the OCV-SOC relationship curves under different aging conditions show almost no change, and the trends of the curve characteristics are basically consistent.

[0088] By analyzing the relationship between the lithium insertion rate at the battery cathode and the state of charge (SOC) in the electrode potential model, the electrode potential expression is improved, thus obtaining the OCV-SOC model. The specific improvement method is as follows:

[0089] First, the electrode potential model expression consists of three main parts, and the general expression is as follows:

[0090]

[0091] In the formula, U(x) is the electrode potential, x is the lithium insertion rate, and a1, b1, b2, c1, c2, d i e i and f i (i is the number of terms in the hyperbolic tangent (tanh) function) are the relevant parameters of the electrode potential at the corresponding temperature, and all are rational numbers greater than 0;

[0092] Analyzing the relationship between the lithium insertion rate at the battery cathode and the state of charge (SOC), we replace x in equation (6) with 1-s, thereby establishing a model describing the OCV-SOC relationship of the full battery, which is specifically expressed as follows:

[0093]

[0094] In the formula, s is the battery SOC;

[0095] Furthermore, equation (7) can be decomposed into a constant term ①, an exponential term ②, and a tangent function term ③. In the constant term ①, a1 represents the vertical shift of the open-circuit voltage curve under different temperatures and aging conditions; in the exponential term ②... and The terms describe the changing trends at both ends of the open-circuit voltage curve; the hyperbolic tangent function term ③ is used to describe the voltage plateau in the middle part of the open-circuit voltage curve due to the phase transition. It can be described by one or more hyperbolic tangent functions. As the number of corresponding plateaus increases, the number of terms also increases.

[0096] from Figure 3 and Figure 4 It can be seen that the OCV-SOC curve of lithium iron phosphate battery has two obvious voltage plateaus, so the model describing the curve should contain two hyperbolic tangent function terms.

[0097] like Figure 5 The figure shows the fitting results of the improved electrode potential model. As can be seen from the figure, the improved electrode potential model can well describe the voltage plateau and trend of the battery open-circuit voltage curve, and the fitted curve is very smooth.

[0098] Step 4: Identify the complete OCV-SOC model based on the charging stage, including analysis of the ohmic internal resistance change trend, analysis of charging stage characteristics, complete OCV-SOC solution, and real-time update of OCV-SOC relationship.

[0099] The analysis of the variation trend of ohmic internal resistance is based on the relationship curves of ohmic internal resistance R0 and SOC measured in the laboratory at different temperatures, which analyzes the variation law of R0.

[0100] like Figure 6 The figure shows the ohmic resistance variation curves of a single cell in a lithium iron phosphate battery pack at three temperatures: 5℃, 25℃, and 55℃. As can be seen from the figure, the ohmic resistance varies very little within the range of 0.2 to 1 SOC. Under actual vehicle operating conditions, the SOC is generally above 0.2 at the start of charging, meaning the ohmic resistance of the battery can be considered a fixed value during the charging phase.

[0101] The charging stage characteristic analysis includes the analysis of charging current variation trend, the analysis of the relationship between charging voltage and open circuit voltage under charging conditions, and the analysis of charging voltage curve trend within different temperature ranges.

[0102] Charging current variation trend analysis: Based on the collected cloud data, the average current of individual cells in the battery pack under a typical charging condition is analyzed.

[0103] like Figure 7 The figure shows the temperature variations of individual cells in a lithium iron phosphate battery pack within the ranges of 34–42℃, 31–40℃, 31–38℃, 34–40℃, 21–28℃, 23–31℃, 18–25℃, 12–21℃, and 11–20℃. The figure shows that the charging current remains almost constant across different temperature ranges, and the current range is 5–6A. Since the capacity of a single lithium iron phosphate cell is 29Ah, it can be determined that the battery pack is being charged using a constant current, low-rate charging method.

[0104] Analysis of the relationship between charging voltage and open-circuit voltage under charging conditions: Based on the expression for open-circuit voltage under low-rate conditions in equation (8), the relationship between the open-circuit voltage and charging voltage during the charging segment is analyzed.

[0105] U OCV,C =U C -I C R0 (8)

[0106] In the formula, U OCV,C For OCV under charging conditions, U C I is the charging voltage. C This is the charging current.

[0107] Depend on Figure 6 and Figure 7 It can be seen that I of each charging segment CR0 can be considered as a fixed value, and based on equation (8), U OCV,C with U C The difference between them is a fixed value I C If R0, then the OCV-SOC curve of the charging segment can be obtained by shifting the charging voltage curve downward by a certain distance in the vertical direction.

[0108] The analysis of charging voltage curve trends in different temperature ranges involves first randomly sampling charging segment voltages from cloud data within different temperature ranges and calculating the average charging voltage of a single cell; then analyzing the changing trends of the average charging voltage curves at different temperatures; and finally comparing and analyzing the average charging voltage curves with the OCV-SOC curves measured in the laboratory at that temperature.

[0109] like Figure 8 The figure shows a comparison of the average charging voltage curves of individual cells in lithium iron phosphate batteries within the temperature ranges of 34–42℃, 31–40℃, 31–38℃, 34–40℃, 21–28℃, 23–31℃, 18–25℃, 12–21℃, and 11–20℃ with the laboratory OCV-SOC curves. The figure shows that during the charging phase of lithium iron phosphate batteries, the individual cell charging voltage curves initially show an increase due to varying initial SOCs, resulting in poor curve consistency. However, once the battery charging stabilizes, the average charging voltage curve and the open-circuit voltage curve measured by HPPC show a generally consistent trend, differing only in numerical value. Similarly, the figure also shows a horizontal difference between the SOC of the charging segment and the laboratory-measured SOC. This is due to acquisition errors in the cloud data acquisition platform, leading to errors in the calculated SOC for the charging segment.

[0110] The complete OCV-SOC solution mainly includes the following steps:

[0111] First, the average charging voltage curve of the charging segment is used as a reference, based on the OCV-SOC curve measured in the laboratory at the corresponding temperature of that segment. The SOC value is then corrected by horizontally shifting it by a certain distance k. For example... Figure 9 (b) shows the result of the charging voltage curve after being horizontally shifted by k.

[0112] Then, the curve with the corrected SOC value is compared with the OCV-SOC curve measured in the laboratory at the corresponding temperature of that segment. By shifting it longitudinally by a certain distance b, it is made to coincide as much as possible with the laboratory OCV-SOC curve, thereby obtaining the U value at the corresponding SOC. OCV,C ;like Figure 9 As shown in (c), the charging voltage curve after horizontal translation is the result of vertical translation. The charging voltage curve after horizontal and vertical translation basically coincides with the OCV-SOC measured in the laboratory during the plateau period.

[0113] Substitute the OCV-SOC data measured in the laboratory at the corresponding temperature into equation (7) and fit it using the least squares method to obtain the OCV-SOC relationship curve expression based on the improved electrode potential model. Based on the obtained expression, following the principle of "adding on the left and subtracting on the right, adding on the top and subtracting on the bottom", add a parameter b in the vertical direction and subtract a parameter k in the horizontal direction to obtain the charging voltage and SOC relationship curve model. Substitute the data of the charging segment into the equation to obtain the values ​​of b and k.

[0114] Substituting the OCV-SOC data of the lithium iron phosphate battery at 25℃ into equation (7), we obtain the expression for the OCV-SOC relationship curve based on the improved electrode potential model:

[0115]

[0116] Therefore, the relationship curve model between charging voltage and SOC can be specifically expressed as follows:

[0117]

[0118] In the formula, b is the difference between the charging voltage and OCV in the charging stage, and k is the SOC correction value for the charging segment.

[0119] use Figure 8 Data from nine charging segments at different temperatures were used for verification. Considering the differences between the initial and later charging curves, charging voltage and SOC data within the range of 0.4 to 0.8 were selected and substituted into equation (10). Of course, a function similar to equation (10) needs to be established for each temperature. The fitting parameter values ​​for each charging stage are shown in Table 1.

[0120] Table 1 Fitting results of partial charging segments of lithium iron phosphate batteries

[0121]

[0122] like Figure 11 The figure shows the OCV-SOC curves after correction for different charging segments. The corrected SOC is obtained by subtracting the corresponding lateral shift k from the original SOC for each charging stage, and the open-circuit voltage value corresponding to the SOC for that stage is obtained by subtracting the corresponding longitudinal shift b from the corresponding charging voltage.

[0123] Finally, if the shifted curve covers the 0-1SOC range and meets the open-circuit voltage characteristics, then the following steps are unnecessary; otherwise, the following steps are required:

[0124] The OCV-SOC of the shifted curve segment that conforms to the open-circuit voltage characteristics is spliced ​​and fused with the OCV-SOC measured in the laboratory at the corresponding temperature to obtain the complete OCV-SOC.

[0125] from Figure 9 (c) It can be seen that the instability of the curve change in the early stage of charging and the change in the curve amplitude caused by the increase of ohmic impedance in the end stage are only used to calculate OCV-SOC.

[0126] like Figure 10 The figure shows the result of splicing and fusing intermediate data from the charging segment with laboratory OCV-SOC at the corresponding temperature. As can be seen from the figure, the spliced ​​data can completely cover the 0–1 SOC range.

[0127] The above-mentioned real-time update of the OCV-SOC relationship first divides the cloud data into multiple "charge-discharge" units according to the charging stage. Each "charge-discharge" unit consists of a complete charging stage and a discharge segment from the charging stage to the next charging stage. Then, based on the complete OCV-SOC data obtained from Equation (7) and the charging segment, the least squares method is used for fitting, and the OCV-SOC curve relationship parameters of each unit are updated, thereby realizing the real-time update of the OCV-SOC relationship curve.

[0128] like Figure 12 As shown, Figure 8 The complete OCV-SOC curve update results for the nine charging stages are shown in the figure. As can be seen from the figure, the real-time updated fused OCV-SOC fitting curve can accurately describe the OCV-SOC relationship of each "charge-discharge" unit.

[0129] The embodiments described above are preferred embodiments of the present invention, but the present invention is not limited to the above embodiments. Any obvious improvements, substitutions or modifications that can be made by those skilled in the art without departing from the essence of the present invention shall fall within the protection scope of the present invention.

Claims

1. A method for constructing a real-time complete OCV-SOC curve of a lithium battery based on cloud data, characterized in that, Includes the following steps: Step 1: Construct a battery model; Step 2: Identify the discharge segment OCV based on the analogy method; In step two, the discharge segment OCV is identified based on the analogy method, using the functional relationship between the terminal voltage U and time t during the zero-input response stage under the battery hybrid power pulse condition. The specific functional relationship is as follows: （2） In the formula, e is the natural constant. =R1C1 is the time constant; Under HPPC operating conditions, the terminal voltage U during the zero-input response phase follows an exponential function relationship with time t. The expression for the exponential function is: （3） In the formula, y0 is U OCV The generalized representation of , yes The generalized representation of , and Represents the generalization coefficient. Represents the generalized dependent variable; Comparing equations (2) and (3), the battery OCV is obtained as follows: （4） The data segments in the cloud data where the current fluctuates around 0A are regarded as the zero-input response stage. The voltage and acquisition time data of at least three consecutive data segments where the current fluctuates around 0A are selected and substituted into equation (3) for fitting, and then the parameter y0 is obtained. The open-circuit voltage U of the battery pack is obtained according to equation (4). OCV ; The average open-circuit voltage U of a single cell is calculated based on the battery pack assembly method. OCV,dis The calculation formula is as follows: （5） In the formula, n is the number of batteries connected in series in the battery pack; Step 3: Construct the OCV-SOC model. By analyzing the relationship between the lithium insertion rate of the battery electrode and SOC in the electrode potential model, the electrode potential expression is improved to obtain the OCV-SOC model. In step three, the relationship between the lithium insertion rate and SOC of the battery electrode is analyzed in the electrode potential model. The electrode potential expression is then improved to obtain the OCV-SOC model. The specific improvement method is as follows: First, the electrode potential model expression consists of three parts, as follows: （6） In the formula, U(x) is the electrode potential, x is the lithium intercalation ratio, and a1, b1, b2, c1, c2, d i e i and f i These are the relevant parameters of the electrode potential at the corresponding temperature, and all are rational numbers greater than 0, where i is the number of terms in the hyperbolic tangent (tanh) function; Analyzing the relationship between the lithium insertion rate of the battery cathode and the state of charge (SOC), we change x in equation (6) to 1 - s, thereby establishing a model describing the OCV-SOC relationship of the full battery, which is specifically expressed as follows: （7） In the formula, s is the battery SOC; Step 4: Identify the complete OCV-SOC model based on the charging stage, including analysis of the ohmic internal resistance change trend, analysis of charging stage characteristics, complete OCV-SOC solution, and real-time update of OCV-SOC relationship.

2. The method for constructing a real-time complete OCV-SOC curve of a lithium battery based on cloud data according to claim 1, characterized in that, The battery model constructed in step one is a first-order RC equivalent circuit model, and the equations of this model are as follows: （1） In the formula, U OCV I is the open-circuit voltage, R0 is the operating current, R1 is the internal resistance in ohms, C1 is the polarization internal resistance, C1 is the polarization capacitor, U1 represents the polarization voltage, i.e., the voltage across R1C1, and U is the terminal voltage.

3. The method for constructing a real-time complete OCV-SOC curve of a lithium battery based on cloud data according to claim 1, characterized in that: The expression (7) of the full-cell OCV-SOC relationship model is broken down into constant terms. Index Item and tangent function term Among them, the constant term In this context, a1 represents the vertical shift of the open-circuit voltage curve under different temperatures and aging conditions; the exponent term... middle and Describe the changing trends at both ends of the open-circuit voltage curve; hyperbolic tangent function term It is used to describe the voltage plateau in the middle part of the open-circuit voltage curve caused by the phase transition. It is described by one or more hyperbolic tangent functions. As the number of corresponding plateaus increases, the number of terms also increases.

4. The method for constructing a real-time complete OCV-SOC curve of a lithium battery based on cloud data according to claim 1, characterized in that: The analysis of the ohmic resistance variation trend in step four is based on the relationship curves of ohmic resistance R0 and SOC measured in the laboratory at different temperatures, and the variation law of R0 is analyzed.

5. The method for constructing a real-time complete OCV-SOC curve of a lithium battery based on cloud data according to claim 1 or 4, characterized in that: The charging stage characteristic analysis in step four includes the analysis of charging current variation trend, the analysis of the relationship between charging voltage and open circuit voltage under charging conditions, and the analysis of charging voltage curve trend within different temperature ranges.

6. The method for constructing a real-time complete OCV-SOC curve of a lithium battery based on cloud data according to claim 5, characterized in that: The charging current variation trend analysis is based on the collected cloud data to analyze the variation of the average current of a single battery cell in a typical charging condition. Analysis of the relationship between charging voltage and open-circuit voltage under charging conditions: Based on the expression for open-circuit voltage under low charging rate conditions in equation (8), the trends of the open-circuit voltage curve and the charging voltage curve are analyzed. （8） In the formula, U OCV,C For OCV under charging conditions, U C I is the charging voltage. C This is the charging current; The analysis of charging voltage curve trends in different temperature ranges involves first randomly sampling charging segment voltages from cloud data within different temperature ranges and calculating the average charging voltage of a single cell; then analyzing the changing trends of the average charging voltage curves at different temperatures; and finally comparing and analyzing the average charging voltage curves with the OCV-SOC curves measured in the laboratory at that temperature.

7. The method for constructing a real-time complete OCV-SOC curve of a lithium battery based on cloud data according to claim 1, characterized in that, The complete OCV-SOC solution in step four includes the following steps: First, the average charging voltage curve of the charging segment is based on the OCV-SOC curve measured in the laboratory at the corresponding temperature of the segment, and the SOC value is corrected by shifting it laterally by a certain distance k. Then, the curve with the corrected SOC value is compared with the OCV-SOC curve measured in the laboratory at the corresponding temperature of that segment. By shifting it longitudinally by a certain distance b, it is made to coincide as much as possible with the laboratory OCV-SOC curve, thereby obtaining the U value at the corresponding SOC. OCV,C ; Substituting the OCV-SOC data measured in the laboratory at the corresponding temperature into equation (7), and fitting it using the least squares method, we obtain a specific expression for the OCV-SOC relationship curve based on the improved electrode potential model. Based on the obtained expression, following the principle of "adding on the left and subtracting on the right, adding on the top and subtracting on the bottom," we add a parameter b in the vertical direction and subtract a parameter k in the horizontal direction to obtain the charging voltage-SOC relationship curve model; the expression is: （9） In the formula, b is the difference between the charging voltage and OCV in the charging stage, and k is the SOC correction value for the charging segment. By inputting the data from the charging segment, the values ​​of b and k can be obtained; Finally, if the shifted curve covers the 0~1 SOC and meets the open-circuit voltage characteristics, then the following steps are unnecessary; otherwise, the following steps are required: The OCV-SOC of the shifted curve segment that conforms to the open-circuit voltage characteristic is spliced ​​and fused with the OCV-SOC measured in the laboratory at the corresponding temperature to obtain the complete OCV-SOC curve.

8. The method for constructing a real-time complete OCV-SOC curve of a lithium battery based on cloud data according to claim 1, characterized in that: In step four, the OCV-SOC relationship is updated in real time. First, the cloud data is divided into multiple "charge-discharge" units according to the charging stage. Each "charge-discharge" unit consists of a complete charging stage and a discharging stage from the charging stage to the next charging stage. Then, based on the complete OCV-SOC data obtained from equation (7) and the charging segment, the least squares method is used for fitting, and the OCV-SOC curve relationship parameters of each unit are updated, thereby realizing the real-time update of the OCV-SOC relationship curve.

Images