A state of charge error online estimation method and system

Abstract

The application relates to the technical field of batteries and discloses a state-of-charge error online estimation method and system, wherein an initial state-of-charge error value corresponding to the power-on moment of a battery management system is obtained through an OCV lookup table algorithm; a real-time state-of-charge error of the battery is calculated through an ampere-hour integral algorithm; the real-time calculation process of the ampere-hour integral algorithm is detected, the real-time state-of-charge error is corrected through a Rint model algorithm when a Rint model correction condition is triggered; the real-time state-of-charge error is calibrated when a calibration algorithm is triggered; until the power-off moment of the battery management system is reached, the real-time state-of-charge error of the battery at this moment is saved and output; a theoretical basis for SOC error estimation is provided, online SOC error estimation is realized, a higher-precision algorithm is selected during estimation, and the SOC error is reduced; the current SOC error calculation can be calibrated, so that different battery packs and different aging conditions can be matched.

Description

Technical Field

[0001] This invention relates to the field of battery technology, and in particular to an online method and system for estimating state of charge error. Background Technology

[0002] Currently, in the development of Battery Management Systems (BMS) for electric vehicles, State of Charge (SOC) represents the remaining capacity, which is the ratio of the remaining capacity after a period of use or long-term disuse to its capacity in a fully charged state, usually expressed as a percentage. Battery SOC cannot be directly measured and can only be estimated using parameters such as battery terminal voltage, charging / discharging current, and internal resistance. These parameters are affected by various uncertainties such as battery aging, environmental temperature changes, and vehicle driving conditions, and there are also various deviations in estimation methods and parameter acquisition. CN109581242 A, "Error Estimation Method and System for State of Charge (SOC)," proposes an error estimation method and system for SOC. This scheme considers various factors affecting SOC estimation error and proposes a calculation model for algorithm error; it imports the errors corresponding to all identified error sources into the algorithm to obtain the SOC error. While this scheme comprehensively considers the sources of error, it fails to analyze them specifically in conjunction with practical engineering applications, resulting in low accuracy in SOC error estimation. Therefore, the existing technology needs improvement. Summary of the Invention

[0003] This invention provides an online method and system for estimating state of charge (SOC) error, which solves the problem of low accuracy in estimating SOC error of batteries in battery management systems in the prior art.

[0004] To address the aforementioned technical problems, the first aspect of this invention provides an online method for estimating the state of charge error, comprising:

[0005] The initial value of the state of charge error corresponding to the power-on time of the battery in the battery management system is obtained, and when it is detected that the power-on meets the OCV correction condition, the initial value of the state of charge error is updated by the OCV lookup table algorithm. The OCV correction condition is that the battery has been resting for a first preset time.

[0006] The state-of-charge error of the battery is updated in real time using the ampere-hour integral algorithm and the initial value of the state-of-charge error to obtain the real-time state-of-charge error.

[0007] When the update process triggers the Rint model correction condition, the real-time state of charge error is corrected using the Rint model algorithm.

[0008] When the battery reaches the preset conditions, the real-time state of charge error is cleared to zero and used as the initial value of the state of charge error. The state of charge error of the battery is then updated in real time using the ampere-hour integration algorithm until the power-off time of the battery management system is reached. The real-time state of charge error corresponding to the power-off time of the battery management system is then saved and output.

[0009] Furthermore, the Rint model correction condition is that when the update process reaches a second preset time, the first state of charge error is less than the second state of charge error; wherein, the first state of charge error is calculated by using the ampere-hour integral algorithm and the initial value of the state of charge error to calculate the state of charge error of the battery; the second state of charge error is calculated by using the Rint model algorithm to calculate the state of charge error of the battery.

[0010] Furthermore, the preset conditions include a first preset condition and a second preset condition; wherein, the first preset condition is that the battery is in a fully charged state; the second preset condition is that the battery is in a discharged state; when the battery reaches the first preset condition, the battery completes one charging process; when the battery reaches the second preset condition, the battery completes one discharging process.

[0011] Furthermore, before saving and outputting the real-time state of charge error corresponding to the power-off time of the battery management system until the power-off time of the battery management system is reached, the method further includes:

[0012] The update process is monitored, and the real-time state of charge error is calibrated when the calibration algorithm is triggered.

[0013] Furthermore, the calibration algorithm includes an OCV lookup table calibration algorithm, an ampere-hour integration calibration algorithm, and a Rint model calibration algorithm; wherein,

[0014] The calibration conditions corresponding to the OCV lookup table calibration algorithm include a first calibration condition, a second calibration condition, and a third calibration condition; the first calibration condition is that the initial value of the state of charge error is calculated by the OCV lookup table algorithm; the second calibration condition is that the battery completes one full charge or discharge process; the third calibration condition is that the battery completes another full charge or discharge process, and the real-time state of charge error of this process is calculated only by the ampere-hour integration algorithm.

[0015] The ampere-hour integration calibration algorithm includes a fourth calibration condition, a fifth calibration condition, and a sixth calibration condition; the fourth calibration condition is that the battery completes one full charge or discharge process; the fifth calibration condition is that the battery completes another full charge or discharge process, and the completed content is the same as that of the fourth calibration condition; the sixth calibration condition is that the battery completes another full charge or discharge process, and the completed content is different from that of the fourth calibration condition.

[0016] The Rint model calibration algorithm includes a seventh calibration condition, an eighth calibration condition, and a ninth calibration condition; the seventh calibration condition is that the battery completes one full charge or discharge process corresponding to the seventh calibration condition; the eighth calibration condition is that during one full charge or discharge process corresponding to the seventh calibration condition, the second state of charge error calculated by the Rint model algorithm is used as the real-time second state of charge error; the ninth calibration condition is that the battery completes another full charge or discharge process.

[0017] Furthermore, the step of detecting the update process and calibrating the real-time state of charge error when the calibration algorithm is triggered includes:

[0018] When the update process is detected to simultaneously satisfy the first calibration condition, the second calibration condition, and the third calibration condition, the OCV lookup table calibration algorithm is triggered, and the real-time state of charge error is calibrated using the following formula:

[0019] e(OVC0)=e1-Q1 / Q2*e2

[0020] In the formula, e(OVC0) is the real-time state of charge error after calibration by the OCV lookup table calibration algorithm; e1 and Q1 are the real-time state of charge error and discharge capacity of the process corresponding to the second calibration condition, respectively; e2 and Q2 are the real-time state of charge error and discharge capacity of the process corresponding to the second calibration condition, respectively.

[0021] Furthermore, the step of detecting the update process and calibrating the real-time state of charge error when the calibration algorithm is triggered includes:

[0022] When the update process is detected to simultaneously satisfy the fourth and fifth calibration conditions and without triggering other calibration algorithms, the ampere-hour integration calibration algorithm is triggered, and the discharge capacity of the processes corresponding to the fourth and fifth calibration conditions is statistically analyzed to obtain the average error; the average error is the error of an average charge of 10AH.

[0023] The average error is recalibrated, and the calibrated average error is used as the real-time state-of-charge error after calibration by the ampere-hour integral calibration algorithm.

[0024] Furthermore, the step of detecting the update process and calibrating the real-time state of charge error when the calibration algorithm is triggered includes:

[0025] When it is detected that the update process simultaneously satisfies the fourth calibration condition, the sixth calibration condition, and the discharge capacity of the process corresponding to the fourth and sixth calibration conditions is less than the preset capacity, and no other calibration algorithm is triggered, the ampere-hour integral calibration algorithm is triggered, and the calculation error of capacity aging SOH is used as the real-time state of charge error for calibration. The calibration result is the real-time state of charge error calibrated by the ampere-hour integral calibration algorithm.

[0026] Furthermore, the step of detecting the update process and calibrating the real-time state of charge error when the calibration algorithm is triggered includes:

[0027] When it is detected that the update process simultaneously satisfies the seventh, eighth, and ninth calibration conditions, and the discharge capacity corresponding to the seventh, eighth, and ninth calibration conditions is less than the preset capacity, and no other calibration algorithm is triggered, the Rint model calibration algorithm is triggered, and the real-time state-of-charge error is calibrated using the following formula:

[0028] e(Rint) = e1 - Q1 / Q2 * e2

[0029] In the formula, e(Rint) is the real-time state-of-charge error after calibration by the Rint model calibration algorithm.

[0030] A second aspect of the present invention provides an online estimation system for state of charge error, comprising:

[0031] The data acquisition module is used to acquire the initial value of the state of charge error corresponding to the power-on time of the battery management system, and when it is detected that the power-on meets the OCV correction condition, the initial value of the state of charge error is updated by the OCV lookup table algorithm. The OCV correction condition is that the battery has been resting for a first preset time.

[0032] The error update module is used to update the state-of-charge error of the battery in real time using the ampere-hour integration algorithm and the initial value of the state-of-charge error, so as to obtain the real-time state-of-charge error.

[0033] The error correction module is used to correct the real-time state of charge error using the Rint model algorithm when the update process triggers the Rint model correction condition.

[0034] The iterative update module is used to clear the real-time state of charge error to zero and use it as the initial value of the state of charge error when the battery reaches the preset conditions, so as to update the state of charge error of the battery in real time through the ampere-hour integration algorithm until the power-off time of the battery management system is reached, save the real-time state of charge error corresponding to the power-off time of the battery management system and output it.

[0035] Compared with the prior art, the beneficial effects of the embodiments of the present invention are as follows:

[0036] This invention provides an online estimation method and system for state of charge (SCC) error. The method includes: obtaining an initial SCC error value corresponding to the power-on time of the battery management system; updating the initial SCC error value using an OCV lookup table algorithm when the power-on condition is detected to be met, wherein the OCV correction condition is that the battery has been stationary for a first preset time; updating the SCC error of the battery in real time using an ampere-hour integral algorithm and the initial SCC error value to obtain a real-time SCC error; correcting the real-time SCC error using a Rint model algorithm when the update process triggers a Rint model correction condition; clearing the real-time SCC error to zero and using it as the initial SCC error value when the battery reaches the preset condition, so as to update the SCC error of the battery in real time using the ampere-hour integral algorithm until the power-off time of the battery management system is reached; saving and outputting the real-time SCC error corresponding to the power-off time of the battery management system. This invention provides a theoretical basis for SOC error estimation and enables online SOC error estimation; it selects a higher precision algorithm during estimation to reduce SOC error; and it can calibrate the calculation of the current SOC error to match different battery packs and different aging conditions. Attached Figure Description

[0037] To more clearly illustrate the technical solution of the present invention, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0038] Figure 1 This is a flowchart of an online method for estimating state of charge error according to a certain embodiment of the present invention;

[0039] Figure 2 This is an error diagram corresponding to positive and negative COV bias according to a certain embodiment of the present invention;

[0040] Figure 3 The SOH provided in a certain embodiment of the present invention m SOHerror Relationship with SOC error;

[0041] Figure 4 This is a Rint battery model diagram provided in a certain embodiment of the present invention;

[0042] Figure 5 This is a device diagram of an online state-of-charge error estimation system provided in a certain embodiment of the present invention. Detailed Implementation

[0043] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings and examples. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0044] It should be understood that the step numbers used in the text are for ease of description only and are not intended to limit the order in which the steps are performed.

[0045] It should be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0046] The terms “comprising” and “including” indicate the presence of the described feature, whole, step, operation, element and / or component, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or collections thereof.

[0047] The term “and / or” refers to any combination of one or more of the associated listed items, as well as all possible combinations, and includes these combinations.

[0048] In one embodiment, such as Figure 1 As shown, the first aspect of the present invention provides an online estimation method for state of charge error, comprising:

[0049] S1. Obtain the initial value of the state of charge error corresponding to the power-on time of the battery management system, and when it is detected that the power-on meets the OCV correction condition, update the initial value of the state of charge error through the OCV lookup table algorithm. The OCV correction condition is that the battery has been resting for a first preset time.

[0050] Specifically, when the battery management system (BMS) is powered on, the battery's state is assessed. If it does not meet the OCV correction condition, the state-of-charge (SOC) error stored in the BMS memory is used as the initial value of the SOC error. If the battery has been in a resting state for a first preset time and meets the OCV correction condition, the SOC of the battery is estimated by looking up a table using the correspondence between OCV and SOC. The first preset time is determined based on actual conditions and is not specifically limited here. It is assumed that there is a good linear relationship between the OCV curves during sampling, and its expression is as follows:

[0051] SOC ref =f(OCV)

[0052] In the formula, f(OCV) is the OCV lookup result calculated using a linear interpolation algorithm; SOC ref The actual value for SOC reference.

[0053] Due to voltage sampling errors, the SOC obtained from the table also contains errors. Assuming the battery is fully charged and has sufficient resting time for OCV correction, the open-circuit voltage sampling value is OCV. m The true open-circuit voltage is OCV0, and the error between the two is Verror.

[0054] OCV0 = OCV m -V error

[0055] At this point, the estimation error of SOC is SOC. error It can be represented as:

[0056] SOC error =f(OCV) m )-f(OCV m -V error )

[0057] The following example uses the sampling accuracy of a certain sampling chip, whose voltage sampling accuracy is shown in Table 1. It is known that the OCV range corresponding to 0-100% SOC of a certain battery cell is 2.966V-4.092V, as shown in Table 2. Within the OCV range, the corresponding voltage sampling accuracy is ±3.9mV, as shown in Table 1 below:

[0058]

[0059]

[0060] The OCV-SOC table for this battery cell is shown below:

[0061] SOC (%) 0 5 10 15 20 25 30 35 40 45 50 OCV(V) 2.966 3.317 3.416 3.447 3.488 3.523 3.553 3.584 3.604 3.619 3.635 SOC (%) 55 60 65 70 75 80 85 90 95 100 OCV(V) 3.655 3.682 3.721 3.778 3.831 3.881 3.93 3.981 4.035 4.092

[0062] Substituting the voltage sampling accuracy into Equation 2 for the SOC estimation error, we can calculate the errors for positive and negative voltage bias as follows: SOC error+ and SOC error- :

[0063] SOC error+ =f(OCV) m )-f(OCV m -3.9mV)

[0064] SOC error- =f(OCV) m )-f(OCV m +3.9mV)

[0065] Assuming the current open-circuit voltage sampling voltage is the OCV voltage corresponding to Table 2, the errors corresponding to positive and negative COV bias are as follows: Figure 2 As shown. To protect the battery, during charging, the potential for a lower State of Charge (SOC) needs to be considered to prevent overcharging; during discharging, the potential for a higher SOC needs to be considered to prevent over-discharging. Therefore, the positive and negative bias errors of SOC need to be considered separately for different usage scenarios in actual use. However, in reality, the OCV-SOC curve is affected by temperature T and battery aging SOH. Therefore, to reduce the estimation error, the OCV curve formula can be optimized:

[0066] SOC ref_T,SOH =f(OCV,T,SOH)

[0067] In the formula, f(OCV,T,SOH) is the OCV calculated from a table, taking into account temperature and aging; SOC... ref_T,SOH The SOC reference value takes into account the effects of temperature and aging.

[0068] At this point, the estimation error of SOC is SOC. error It can be represented as:

[0069] SOC error =f(OCV) m ,T,SOH)-f(OCV m -V error ,T,SOH)

[0070] Since temperature has a relatively small impact on the OCV-SOC curve, to simplify the calculation, we can consider only the effect of SOH error on the calculation results:

[0071] SOC error =f(OCV) m ,T,SOH m )-f(OCV m -V error ,T,SOHm -SOH error )

[0072] In the formula, SOH m This is the calculated value of SOH; SOH error This represents the error corresponding to the SOH calculation value. Since the SOH error can be positive or negative, it needs to be considered in conjunction with the positive or negative deviation of OCV. There are four possible permutations as shown in the table below. The maximum value is taken as the larger error, and the minimum value is taken as the smaller error:

[0073]

[0074] Whether the battery has been left to stand for a first preset time or not, the initial value of the battery's state of charge error is assigned using different methods, ensuring the accuracy of the initial value and facilitating the accurate calculation of the battery's real-time state of charge error in the future.

[0075] S2. The state-of-charge error of the battery is updated in real time using the ampere-hour integral algorithm and the initial value of the state-of-charge error to obtain the real-time state-of-charge error.

[0076] Specifically, the ampere-hour integral algorithm is mainly suitable for operating conditions where the initial SOC value is known. When the current sensor accuracy is high, the SOC estimation has good accuracy. Its calculation expression is:

[0077]

[0078] In the formula, SOC0 is the initial SOC value; η is the charge / discharge efficiency; I is the charge / discharge current; t is the integration period; N is the number of integration periods starting from SOC0; SOH is the capacity aging state; Q rate This refers to the battery's rated capacity.

[0079] The battery's rated capacity is readily available at the factory and serves as a standard reference value; its error is not considered here. The charge / discharge efficiency error is generally above 99%, but for simplified calculations, its error is temporarily disregarded, and its default value is set to 100%. The integration period is determined by the MCU's Timer function, and its error is relatively small; therefore, its error is also temporarily disregarded here. Thus, when the errors of Qrate, η, and t are not considered, the SOC error of the ampere-hour integration algorithm is related to the errors of its initial value, SOH, and current:

[0080] SOC0 = SOC 0m -SOC 0error

[0081] SOH=SOH m -SOH error

[0082] I(k)=I(k) m -I(k) error

[0083] In the formula, SOC0, SOH, and I(k) are the initial value of SOC, the actual value corresponding to the capacity aging, and the sampling current, respectively; SOC 0m SOH m I(k) m These are the corresponding values ​​used in the calculation, SOC. 0error SOH error I(k) error These are the corresponding errors.

[0084] Ignoring the errors of Qrate, η, and t, the calculated value of the current SOC is SOC. cal for:

[0085]

[0086] Subtracting the error, we can obtain the true value:

[0087]

[0088] Simplifying the above two formulas, we get:

[0089]

[0090] Therefore, the SOC error calculated by the ampere-hour integration method can be mainly divided into three parts. For ease of description later, we will use A-B+C to represent this; A is the initial SOC error, when SOH... errror and I(k) error When SOC is 1, according to the above formula, the SOC error is equal to the initial SOC error; B is the middle part of the above formula, when SOC 0errror and I(k) error When it is 0, the SOC error is related to the current ampere-hour integral and the calculated SOH value. m and SOH error SOH error Related; C is the third part of the above formula. When the initial error of SOC and the estimation error of SOH are 0, the estimation error of SOC is mainly related to the error of the current.

[0091] In B, the SOC error is proportional to the currently calculated cumulative integral; when the battery goes from fully charged to fully discharged, the maximum SOC error can be calculated, at which point the following condition is met:

[0092]

[0093] At this point, the error of SOC can be expressed as:

[0094]

[0095] SOH in the above formula m SOH error The relationship between the error and SOC is shown in the graph. Figure 3 As shown, with the SOH estimation error remaining constant, the SOC error gradually increases with the aging of the SOH. For the ampere-hour integration algorithm, to control the SOC error within 3%, the SOH estimation accuracy needs to be <±3%.

[0096] In C, to evaluate the impact of current integration on SOC error, a standard ammeter is needed to calibrate the current sampling chip: The current error is obtained by looking up the current detected by the Hall sensor in a table, and then integrated into C to calculate the SOC error caused by the current sampling error. A simpler approach is to import a specific operating condition, performing 50 charge-discharge cycles on the battery within a certain range, then discharging the battery completely, and calculating the SOC error at the end of the operating condition. Since the charging capacity ≈ discharging capacity during overcharge cycles, the charging capacity and SOC error corresponding to 50 cycles can be statistically analyzed to obtain the error e_chg10Ah corresponding to 10AH (calibrable). In this case, the SOC error caused by the current error can be expressed as:

[0097] SOC error =SOC error (A)-SOC error (B)+SOC error (C)

[0098] The initial value of SOC, SOH, and the magnitude of the current have errors in two directions. When the error directions of parts A, B, and C are uncertain, the SOC error can be written as the absolute value of the three, that is:

[0099] SOC error =±(|SOC) error (A)|+|SOC error (B)|+|SOC error (C)|).

[0100] S3. When the update process triggers the Rint model correction condition, the real-time state of charge error is corrected using the Rint model algorithm.

[0101] The Rint model correction condition is that when the update process reaches a second preset time, the first state-of-charge error is less than the second state-of-charge error; wherein the first state-of-charge error is calculated by using the ampere-hour integral algorithm and the initial value of the state-of-charge error to calculate the state-of-charge error of the battery; and the second state-of-charge error is calculated by using the Rint model algorithm to calculate the state-of-charge error of the battery.

[0102] Specifically, the Rint model algorithm is a simplified battery model, such as... Figure 4 As shown, the SOC is estimated by estimating the OCV under dynamic operating conditions using the relationship between internal resistance and voltage. Because the Rint model does not consider the effect of voltage polarization, it is generally applicable to operating conditions above 10℃ and where a small current (typically <0.1C) is applied for a period of time (refer to the resting time of OCV). Figure 4 As shown, taking the discharge direction as an example:

[0103] OCV m =V m +I m ×R 0m

[0104] In the formula, V m I m R 0m These represent the sampling voltage, sampling current, and equivalent ohmic internal resistance, respectively; OCV m Let R be the calculated open-circuit voltage. This takes into account the aging of the internal resistance. 0m It can be written as the following expression:

[0105] R 0m =R 0_BOL ×SOHR m

[0106] In the formula, R 0_BOL and SOHR m These represent the internal resistance during the BOL stage and its corresponding internal resistance aging state, respectively, and the estimated SOC value is calculated accordingly. cal :

[0107] SOC cal =f(OCV) m )=f(V m +I m ×R 0_BOL ×SOHR m )

[0108] In the formula, R 0_BOL This is the initial value of the internal resistance, which is related to temperature and SOC.

[0109] Considering V m I m and SOHR m Error, true SOC value real The expression is

[0110] SOC real =f(V m -V error +R 0_BOL ×(Im -I error )×(SOHR m -SOHR error ))

[0111] SOC estimate cal Compared with the true value of SOC real By subtracting the difference, the SOC error can be obtained.

[0112] During the battery's charging and discharging process, the ampere-hour integral algorithm essentially runs continuously. Only after each charging / discharging cycle reaches a second preset time is the state-of-charge (SOC) error calculated by the Rint model algorithm compared with the SOC error calculated by the ampere-hour integral algorithm at that moment. The smaller one is taken as the real-time SOC error for that current moment, and the algorithm corresponding to this comparison result is used to continue calculation until the charging or discharging process of the battery ends. This application selects a higher-precision algorithm during estimation to reduce SOC error.

[0113] S4. When the battery reaches the preset conditions, the real-time state of charge error is cleared to zero and used as the initial value of the state of charge error. The state of charge error of the battery is updated in real time through the ampere-hour integration algorithm until the power-off time of the battery management system is reached. The real-time state of charge error corresponding to the power-off time of the battery management system is saved and output.

[0114] In one embodiment, the preset conditions include a first preset condition and a second preset condition; wherein, the first preset condition is that the battery is in a fully charged state; the second preset condition is that the battery is in a discharged state; when the battery reaches the first preset condition, the battery completes one charging process; when the battery reaches the second preset condition, the battery completes one discharging process.

[0115] Specifically, when the battery is fully charged, the highest single-cell voltage is greater than or equal to the full charge voltage, the charging current is less than or equal to the first preset current, and this lasts for a third preset time. When the battery is discharged, the lowest single-cell voltage is less than or equal to the discharge cutoff voltage, the discharge current is less than or equal to the second preset current, and this lasts for a fourth preset time. Among these, the full charge voltage, the first preset current, and the third preset time during the charging cutoff phase are calibration values, as are the discharge cutoff voltage, the second preset current, and the fourth preset time during the discharge cutoff phase. These calibration values ​​are determined based on actual conditions and are not specifically limited here.

[0116] This application divides the power-on to power-off process of the battery management system into a cyclic charging and discharging process. During each charging and discharging process, the state-of-charge (SOC) error is updated in real time using an ampere-hour integral algorithm, and a higher-precision algorithm is selected during estimation to reduce SOC error. The charging and discharging process of the battery is monitored, and when the Rint model correction conditions are met, the Rint model algorithm is used to correct the real-time SOC error of the battery. When the calibration algorithm is triggered, an appropriate calibration algorithm is used to calibrate the real-time SOC error of the battery to improve the accuracy of online SOC error estimation and achieve matching for different battery packs and different aging conditions. When the battery is fully charged or discharged, the real-time SOC error is cleared to zero and the next discharge or charging process of the battery is initiated, realizing accurate estimation of SOC error throughout the entire battery life cycle.

[0117] In one embodiment, before saving and outputting the real-time state of charge error corresponding to the power-off time of the battery management system, the method further includes: detecting the update process and calibrating the real-time state of charge error when a calibration algorithm is triggered. The calibration algorithm includes an OCV lookup table calibration algorithm, an ampere-hour integral calibration algorithm, and a Rint model calibration algorithm.

[0118] The calibration conditions corresponding to the OCV lookup table calibration algorithm include a first calibration condition, a second calibration condition, and a third calibration condition; the first calibration condition is that the initial value of the state of charge error is calculated by the OCV lookup table algorithm; the second calibration condition is that the battery completes one full charge or discharge process; the third calibration condition is that the battery completes another full charge or discharge process, and the real-time state of charge error of this process is calculated only by the ampere-hour integration algorithm.

[0119] In one embodiment, the update process is monitored, and the real-time state of charge error is calibrated when the calibration algorithm is triggered, including:

[0120] When the update process is detected to simultaneously satisfy the first calibration condition, the second calibration condition, and the third calibration condition, the OCV lookup table calibration algorithm is triggered, and the real-time state of charge error is calibrated using the following formula:

[0121] e(OVC0)=e1-Q1 / Q2*e2

[0122] In the formula, e(OVC0) is the real-time state of charge error after calibration by the OCV lookup table calibration algorithm; e1 and Q1 are the real-time state of charge error and discharge capacity of the process corresponding to the second calibration condition, respectively; e2 and Q2 are the real-time state of charge error and discharge capacity of the process corresponding to the second calibration condition, respectively.

[0123] Specifically, when initial SOC error and current sampling error are not considered, the SOC estimation error is proportional to the ampere-hour integral capacity. To simplify the calculation, the current sampling error can be ignored when the throughput (capacity integral in one current direction) between two full charge or discharge cycles is less than twice the rated capacity. This application uses e(OVC0) to replace the theoretically calculated value (lookup table value) at OCV0 under temperature and SOH conditions to calibrate the SOC error calculation. Because OCV0 may not be completely consistent with the OCV value in the lookup table, this application takes the point in the lookup table OCV that is closest to OCV0 to improve the accuracy of SOC estimation.

[0124] The ampere-hour integration calibration algorithm includes a fourth calibration condition, a fifth calibration condition, and a sixth calibration condition; the fourth calibration condition is that the battery completes one full charge or discharge process; the fifth calibration condition is that the battery completes another full charge or discharge process, and the completed content is the same as that of the fourth calibration condition; the sixth calibration condition is that the battery completes another full charge or discharge process, and the completed content is different from that of the fourth calibration condition.

[0125] In one embodiment, the update process is monitored, and the real-time state of charge error is calibrated when the calibration algorithm is triggered, including:

[0126] When the update process is detected to simultaneously satisfy the fourth and fifth calibration conditions and without triggering other calibration algorithms, the ampere-hour integration calibration algorithm is triggered, and the discharge capacity of the processes corresponding to the fourth and fifth calibration conditions is statistically analyzed to obtain the average error; the average error is the error of an average charge of 10AH.

[0127] The average error is recalibrated, and the calibrated average error is used as the real-time state-of-charge error after calibration by the ampere-hour integral calibration algorithm.

[0128] Specifically, if the current operating condition meets the fourth and fifth calibration conditions and no other SOC calibration methods are triggered during this period, the SOC error is approximately equal to the integral error of the current. By statistically analyzing the integral amount during this period, the average charging error of 10Ah can be obtained, thus allowing for recalibration of e_chg10Ah. To reduce the impact of random errors, the throughput requirement can be increased, for example, by more than 10 times the rated capacity.

[0129] The throughput here refers to the battery's charge and discharge capacity, generally used to count the number of battery cycles. Since the battery's charging and discharging capacities are roughly equal, in practical use, capacity in one direction can be used for statistics. For example, capacity in the discharge direction can be used. When the discharge capacity reaches 1 times the rated capacity, the battery can be considered to have completed an equivalent full charge-slow discharge cycle.

[0130] In one embodiment, detecting the update process and calibrating the real-time state of charge error when the calibration algorithm is triggered further includes:

[0131] When it is detected that the update process simultaneously satisfies the fourth calibration condition, the sixth calibration condition, and the discharge capacity of the process corresponding to the fourth and sixth calibration conditions is less than the preset capacity, and no other calibration algorithm is triggered, the ampere-hour integral calibration algorithm is triggered, and the calculation error of capacity aging SOH is used as the real-time state of charge error for calibration. The calibration result is the real-time state of charge error calibrated by the ampere-hour integral calibration algorithm.

[0132] Specifically, if the current operating condition meets the fourth and sixth calibration conditions, and the total capacity throughput is less than twice the rated capacity during this period, and no other SOC calibration methods are triggered, then the SOC error at this time is approximately the calculation error of the capacity aging SOH. The error caused by the SOH error in the ampere-hour integration algorithm can be calibrated by calibrating the estimated error of the SOH at this time.

[0133] The Rint model calibration algorithm includes a seventh calibration condition, an eighth calibration condition, and a ninth calibration condition; the seventh calibration condition is that the battery completes one full charge or discharge process corresponding to the seventh calibration condition; the eighth calibration condition is that during one full charge or discharge process corresponding to the seventh calibration condition, the second state of charge error calculated by the Rint model algorithm is used as the real-time second state of charge error; the ninth calibration condition is that the battery completes another full charge or discharge process.

[0134] In one embodiment, the update process is monitored, and the real-time state of charge error is calibrated when the calibration algorithm is triggered, including:

[0135] When it is detected that the update process simultaneously satisfies the seventh, eighth, and ninth calibration conditions, and the discharge capacity corresponding to the seventh, eighth, and ninth calibration conditions is less than the preset capacity, and no other calibration algorithm is triggered, the Rint model calibration algorithm is triggered, and the real-time state-of-charge error is calibrated using the following formula:

[0136] e(Rint) = e1 - Q1 / Q2 * e2

[0137] In the formula, e(Rint) is the real-time state-of-charge error after calibration by the Rint model calibration algorithm.

[0138] Specifically, when the seventh, eighth, and ninth calibration conditions are met simultaneously, and no other SOC calibration methods have been triggered, the total capacity throughput is less than twice the rated capacity during this period. The SOC error at this time is approximately the SOC error estimated by the Rint model. By comprehensively considering the voltage sampling error and the current sampling error, the SOHR error can be calibrated, thereby more accurately evaluating the SOC error calculated by the Rint model.

[0139] This application embodiment addresses the issue of low accuracy in estimating the State of Charge (SOC) error of batteries in a battery management system by designing an online estimation method for SOC error. This implements the online SOC error estimation method and system provided by this invention. The method includes: obtaining an initial value of the SOC error corresponding to the battery's power-on moment in the battery management system; and updating the initial value of the SOC error using an OCV lookup table algorithm when the current power-on meets the OCV correction condition, wherein the OCV correction condition is that the battery has been stationary for a first preset time; updating the battery's SOC error in real time using an ampere-hour integral algorithm and the initial value of the SOC error to obtain the real-time SOC error; and triggering R during the update process. When the Rint model is used to correct the conditions, the real-time state of charge (SOC) error is corrected using the Rint model algorithm. When the battery reaches the preset conditions, the real-time SOC error is cleared to zero and used as the initial value of the SOC error. The battery's SOC error is then updated in real time using the ampere-hour integration algorithm until the power-off time of the battery management system is reached. The real-time SOC error corresponding to the power-off time of the battery management system is saved and output. This technical solution provides a theoretical basis for SOC error estimation and enables online SOC error estimation. A higher precision algorithm is selected during estimation to reduce SOC error. The calculation of the current SOC error can be calibrated to match different battery packs and different aging conditions.

[0140] It should be noted that although the steps in the flowchart above are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order requirement for the execution of these steps, and they can be executed in other orders.

[0141] In another embodiment, such as Figure 5 As shown, a second aspect of the present invention provides an online estimation system for state of charge error, comprising:

[0142] The data acquisition module 10 is used to acquire the initial value of the state of charge error corresponding to the power-on time of the battery management system, and when it is detected that the power-on meets the OCV correction condition, the initial value of the state of charge error is updated by the OCV lookup table algorithm. The OCV correction condition is that the battery has been resting for a first preset time.

[0143] Error update module 20 is used to update the state of charge error of the battery in real time using the ampere-hour integration algorithm and the initial value of the state of charge error, so as to obtain the real-time state of charge error;

[0144] Error correction module 30 is used to correct the real-time state of charge error by means of Rint model algorithm when the update process triggers the Rint model correction condition.

[0145] The iterative update module 40 is used to clear the real-time state of charge error to zero and use it as the initial value of the state of charge error when the battery reaches the preset conditions, so as to update the state of charge error of the battery in real time through the ampere-hour integration algorithm until the power-off time of the battery management system is reached, save the real-time state of charge error corresponding to the power-off time of the battery management system and output it.

[0146] It should be noted that the modules in the aforementioned online state-of-charge error estimation system can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the computer device's memory as software, so that the processor can call and execute the corresponding operations of each module. For specific limitations regarding the online state-of-charge error estimation system, please refer to the limitations of the online state-of-charge error estimation method described above; both have the same function and role, and will not be repeated here.

[0147] In summary, this invention provides an online estimation method and system for state of charge (SCC) error. The method includes: obtaining an initial SCC error value corresponding to the power-on time of the battery management system; updating the initial SCC error value using an OCV lookup table algorithm when the power-on condition is detected to be met, wherein the OCV correction condition is that the battery has been stationary for a first preset time; updating the SCC error of the battery in real time using an ampere-hour integral algorithm and the initial SCC error value to obtain a real-time SCC error; correcting the real-time SCC error using a Rint model algorithm when the update process triggers a Rint model correction condition; clearing the real-time SCC error to zero and using it as the initial SCC error value when the battery reaches the preset condition, so as to update the SCC error of the battery in real time using the ampere-hour integral algorithm until the power-off time of the battery management system is reached; saving and outputting the real-time SCC error corresponding to the power-off time of the battery management system. This invention provides a theoretical basis for SOC error estimation and enables online SOC error estimation; it selects a higher precision algorithm during estimation to reduce SOC error; and it can calibrate the calculation of the current SOC error to match different battery packs and different aging conditions.

[0148] The various embodiments in this specification are described in a progressive manner. For directly identical or similar parts of the embodiments, refer to each other. Each embodiment focuses on its differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments. It should be noted that the technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as the combination of these technical features does not contradict each other, it should be considered within the scope of this specification.

[0149] The embodiments described above are merely preferred embodiments of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various improvements and substitutions without departing from the technical principles of this invention, and these improvements and substitutions should also be considered within the scope of protection of this application. Therefore, the scope of protection of this patent application should be determined by the scope of the claims.

Claims

1. A method of estimating state-of-charge error on-line, characterized by, The method comprises the following steps: acquiring an initial state-of-charge error corresponding to the time when the battery is powered on by a battery management system, and updating the initial state-of-charge error by an OCV lookup table algorithm when it is detected that the present power-on meets an OCV correction condition, wherein the OCV correction condition is that the battery is stationary for a first preset time; updating the state-of-charge error of the battery in real time by an ampere-hour integral algorithm and the initial state-of-charge error to obtain a real-time state-of-charge error; correcting the real-time state-of-charge error by a Rint model algorithm when it is detected that the updating process triggers a Rint model correction condition; when the battery meets a preset condition, clearing the real-time state-of-charge error and taking it as the initial state-of-charge error, updating the state-of-charge error of the battery in real time by the ampere-hour integral algorithm, detecting the updating process, and calibrating the real-time state-of-charge error when the calibration algorithm is triggered until the power-off time of the battery management system is reached, saving the real-time state-of-charge error corresponding to the power-off time of the battery management system and outputting it; the preset condition comprises a first preset condition and a second preset condition; wherein the first preset condition is that the battery is in a full-charge state; the second preset condition is that the battery is in an empty state; when the battery meets the first preset condition, the battery completes a charging process; when the battery meets the second preset condition, the battery completes a discharging process; the calibration algorithm comprises an OCV lookup table calibration algorithm, an ampere-hour integral calibration algorithm, and a Rint model calibration algorithm; wherein, the calibration condition corresponding to the OCV lookup table calibration algorithm comprises a first calibration condition, a second calibration condition, and a third calibration condition; the first calibration condition is that the initial state-of-charge error is calculated by the OCV lookup table algorithm; the second calibration condition is that the battery completes a full-charge or empty process; the third calibration condition is that the battery again completes a full-charge or empty process, and the real-time state-of-charge error of the process is calculated only by the ampere-hour integral algorithm; the ampere-hour integral calibration algorithm comprises a fourth calibration condition, a fifth calibration condition, and a sixth calibration condition; the fourth calibration condition is that the battery completes a full-charge or empty process; the fifth calibration condition is that the battery again completes a full-charge or empty process, and the completion content is the same as that of the fourth calibration condition; the sixth calibration condition is that the battery again completes a full-charge or empty process, and the completion content is different from that of the fourth calibration condition; the Rint model calibration algorithm comprises a seventh calibration condition, an eighth calibration condition, and a ninth calibration condition; the seventh calibration condition is that the battery completes a full-charge or empty process; the eighth calibration condition is that the Rint model correction condition is triggered in the full-charge or empty process corresponding to the seventh calibration condition; the ninth calibration condition is that the battery again completes a full-charge or empty process.

2. The method of claim 1, wherein, The Rint model correction condition is that when the updating process reaches a second preset time, a first state of charge error is less than a second state of charge error; wherein the first state of charge error is obtained by calculating the state of charge error of the battery through the ampere-hour integral algorithm and the state of charge error initial value; and the second state of charge error is obtained by calculating the state of charge error of the battery through the Rint model algorithm.

3. The method of claim 1, wherein, The detection of the updating process and the calibration of the real-time state of charge error when the calibration algorithm is triggered, comprising: When the updating process meets the first calibration condition, the second calibration condition and the third calibration condition, the OCV lookup table calibration algorithm is triggered, and the real-time state of charge error is calibrated by the following formula: e(OVC0) = e1 - Q1 / Q2 * e2 In the formula, e(OVC0) is the real-time state of charge error calibrated by the OCV lookup table calibration algorithm; e1 and Q1 are the real-time state of charge error and the discharge capacity corresponding to the process of the second calibration condition respectively; e2 and Q2 are the real-time state of charge error and the discharge capacity corresponding to the process of the second calibration condition respectively.

4. The method of claim 1, wherein, The detection of the updating process and the calibration of the real-time state of charge error when the calibration algorithm is triggered, comprising: When the updating process meets the fourth calibration condition, the fifth calibration condition, and no other calibration algorithm is triggered, the ampere-hour integral calibration algorithm is triggered, and the average error is obtained by counting the discharge capacity corresponding to the process of the fourth calibration condition and the fifth calibration condition; the average error is the error of average charging 10AH; The average error is recalibrated, and the calibrated average error is taken as the real-time state of charge error calibrated by the ampere-hour integral calibration algorithm.

5. The method of claim 1, wherein, The detection of the updating process and the calibration of the real-time state of charge error when the calibration algorithm is triggered, comprising: When the updating process meets the fourth calibration condition, the sixth calibration condition, and the discharge capacity corresponding to the process of the fourth calibration condition and the sixth calibration condition is less than a preset capacity, and no other calibration algorithm is triggered, the ampere-hour integral calibration algorithm is triggered, and the calculation error of the capacity aging SOH is taken as the real-time state of charge error for calibration, and the calibration result is the real-time state of charge error calibrated by the ampere-hour integral calibration algorithm.

6. The method of claim 3, wherein, The detection of the updating process and the calibration of the real-time state of charge error when the calibration algorithm is triggered, comprising: When the updating process meets the seventh calibration condition, the eighth calibration condition, the ninth calibration condition, and the discharge capacity corresponding to the process of the seventh calibration condition, the eighth calibration condition and the ninth calibration condition is less than the preset capacity, and no other calibration algorithm is triggered, the Rint model calibration algorithm is triggered, and the real-time state of charge error is calibrated by the following formula: e(Rint) = e1 - Q1 / Q2 * e2 In the formula, e(Rint) is the real-time state of charge error calibrated by the Rint model calibration algorithm.

7. A state of charge error online estimation system, characterized by, The method comprises the following steps: a data acquisition module is configured to acquire an initial state-of-charge error corresponding to a power-on time of a battery management system, and update the initial state-of-charge error by an OCV lookup table algorithm when it is detected that the present power-on meets an OCV correction condition, wherein the OCV correction condition is that the battery is stationary for a first preset time; an error updating module is configured to update a state-of-charge error of the battery in real time by an ampere-hour integral algorithm and the initial state-of-charge error to obtain a real-time state-of-charge error; an error correction module is configured to correct the real-time state-of-charge error by a Rint model algorithm when it is detected that an updating process triggers a Rint model correction condition; an iterative updating module is configured to clear the real-time state-of-charge error and take it as the initial state-of-charge error when the battery meets a preset condition, update the state-of-charge error of the battery in real time by the ampere-hour integral algorithm, detect the updating process, calibrate the real-time state-of-charge error when a calibration algorithm is triggered, and save a real-time state-of-charge error corresponding to a power-off time of the battery management system and output the real-time state-of-charge error until the power-off time of the battery management system is reached; the preset condition comprises a first preset condition and a second preset condition; wherein the first preset condition is that the battery is in a full-charge state, and the second preset condition is that the battery is in an empty state; when the battery meets the first preset condition, the battery completes a charging process; when the battery meets the second preset condition, the battery completes a discharging process; the calibration algorithm comprises an OCV lookup table calibration algorithm, an ampere-hour integral calibration algorithm, and a Rint model calibration algorithm; wherein a calibration condition corresponding to the OCV lookup table calibration algorithm comprises a first calibration condition, a second calibration condition, and a third calibration condition; the first calibration condition is that the initial state-of-charge error is calculated by the OCV lookup table algorithm; the second calibration condition is that the battery completes a full-charge or empty process; and the third calibration condition is that the battery again completes a full-charge or empty process, and the real-time state-of-charge error of the process is calculated only by the ampere-hour integral algorithm; the ampere-hour integral calibration algorithm comprises a fourth calibration condition, a fifth calibration condition, and a sixth calibration condition; the fourth calibration condition is that the battery completes a full-charge or empty process; the fifth calibration condition is that the battery again completes a full-charge or empty process, and the completion content is the same as that of the fourth calibration condition; and the sixth calibration condition is that the battery again completes a full-charge or empty process, and the completion content is different from that of the fourth calibration condition; the Rint model calibration algorithm comprises a seventh calibration condition, an eighth calibration condition, and a ninth calibration condition; the seventh calibration condition is that the battery completes a full-charge or empty process; the eighth calibration condition is that the Rint model correction condition is triggered in the full-charge or empty process corresponding to the seventh calibration condition; and the ninth calibration condition is that the battery again completes a full-charge or empty process.

Images