A battery state of health calculation method

Abstract

This invention relates to the field of battery management technology and discloses a method for calculating the state of health (SOH) of a battery. The method includes: S1, acquiring the original SOC-OCV curve and determining the SOC decay boundary point; S2, keeping the OCV value constant, establishing the relationship between OCV and SOH, and forming a SOC-OCV scaling and splicing formula; S3, monitoring the cumulative discharge capacity in real time; when 70% of the current usable capacity is reached, calculating the current SOC value; substituting the estimated SOH value into the SOC-OCV scaling and splicing formula to generate a SOC-OCV comparison curve; S4, obtaining the theoretical open-circuit voltage value corresponding to the current SOC value; S5, when the stability criterion is met, acquiring the cell voltage, comparing it with the theoretical open-circuit voltage, and adaptively adjusting the SOH value; S6, using the adjusted SOH value as a new input, repeating S3 to S5 for continuous iterative calculation to gradually reduce the SOH error. This application overcomes the shortcomings of fixed parameters in adapting to dynamic operating conditions, eliminating long-term accumulated errors while ensuring that SOH can reflect the true health state of the battery in real time, ultimately achieving a balance between accuracy and timeliness in SOH calculation throughout the entire battery lifecycle.

Description

Technical Field

[0001] This invention relates to the field of battery management technology, and more specifically to a method for calculating battery health status. Background Technology

[0002] In the field of electric vehicle technology, State of Health (SOH) is a core indicator for measuring the degree of performance degradation of power batteries, directly determining battery lifespan, safety, and the vehicle's range. Accurate SOH assessment not only helps improve the user's driving experience but also provides crucial information for optimizing battery management strategies and extending battery life.

[0003] Currently, the commonly used SOH calculation methods in the industry mainly fall into two categories. The first category is the ampere-hour integration method (also known as the coulomb counting method). This method calculates SOH by recording the cumulative charge or discharge of the battery and combining it with the rated capacity. However, the ampere-hour integration method is highly dependent on the accuracy of the current sensor and does not consider the nonlinear effects of factors such as charge / discharge rate, ambient temperature, and depth of discharge on the battery aging process. Under long-term operation, it is prone to cumulative errors, resulting in a large deviation between the calculated results and the actual battery health status.

[0004] Another type is the two-point capacity method based on OCV calibration. This method utilizes two opportunities to meet the open-circuit voltage (OCV) calibration conditions, obtaining the corresponding state of charge (SOC) and cumulative capacity difference, and then applying the formula... The calculation is performed. Although the two-point capacity method based on OCV calibration can eliminate some initial errors in principle, the SOC-OCV curve will change significantly as the battery ages. In practical applications, it is difficult to obtain the SOC-OCV mapping relationship after aging in real time, so the accuracy of SOH calculation cannot be guaranteed.

[0005] The limitations of the above methods not only reduce the reliability of battery status monitoring, but may also lead to misjudgments of the remaining battery life, thereby affecting the effective implementation of the vehicle's energy management strategy and even bringing potential safety risks.

[0006] Therefore, the main technical contradiction in the existing technology lies in the need to accurately reflect the dynamic changes of the SOC-OCV curve after battery aging, while avoiding the high cost and difficulty of measuring the complete aging curve in real time in engineering practice. Summary of the Invention

[0007] The present invention aims to provide a method for calculating battery health status, which solves the problems of low reliability of existing technologies for battery status monitoring and difficulty in obtaining SOC-OCV curves after aging in real time, resulting in inaccurate SOH calculation.

[0008] To achieve the above objectives, the present invention adopts the following technical solution: a method for calculating battery health status, comprising the following steps: S1. Obtain the original SOC-OCV curves for the BOL and EOL states of the battery cell, respectively, and determine the SOC attenuation boundary point. ; S2, keeping the OCV value unchanged, establish the relationship between the OCV curve and SOH, and form the SOC-OCV scaling and splicing formula; S3. After the battery system is fully charged and begins to discharge, monitor the cumulative discharge capacity in real time. When the cumulative discharge capacity reaches 70% of the current available capacity, record the discharge capacity and calculate the current SOC value. Substitute the estimated SOH value into the SOC-OCV scaling and splicing formula to generate an SOC-OCV comparison curve that adapts to the current aging level. S4. Based on the current SOC value of the battery, obtain the theoretical open-circuit voltage value corresponding to that SOC through the SOC-OCV comparison curve. ; S5: After the vehicle is parked, the battery management system automatically monitors the battery's resting conditions and collects the individual cell voltage when the stability criteria are met. and the theoretical open-circuit voltage obtained from S4 In comparison, the SOH value is adaptively adjusted based on the direction and magnitude of the voltage deviation; S6. Using the adjusted SOH value as the new input, repeat S3 to S5 to continuously iterate and calculate, gradually reducing the SOH error.

[0009] The principles and advantages of this scheme are: To address the aforementioned technical contradictions, this invention proposes a battery health state calculation method based on segmented scaling and stitching of SOC-OCV curves. Its principle and beneficial effects are as follows: (1) Dynamically adapts to aging curve, eliminating the need for real-time full-curve measurement. By analyzing the aging mechanism of battery cells, this invention reveals that the overall shape of the OCV-SOC curve remains largely unchanged during battery aging, with only a shift in the SOC coordinates. Furthermore, capacity degradation is primarily concentrated above the SOC degradation threshold. Based on this pattern, this invention establishes a dynamic correlation mechanism between the OCV curve and SOH. Specifically, by segmenting, scaling, and stitching the SOC coordinates of the initial (BOL) SOC-OCV baseline curve, an estimated SOC-OCV curve equivalent to the current aging state can be generated. This avoids the high cost and difficulty of real-time measurement of the complete aging curve in an automotive environment while ensuring that the curve can be dynamically updated with changes in SOH.

[0010] (2) Fast convergence of SOH is achieved through voltage feedback iterative correction. This invention first calculates the current SOC estimate based on the discharge capacity and the current SOH estimate; then, it substitutes the SOH estimate into the SOC-OCV scaling and splicing formula to generate the equivalent SOC-OCV curve under the current aging state; finally, it obtains the theoretical expected open-circuit voltage from this curve based on the current SOC estimate. Finally, the actual voltage after resting is used. and By making a comparison, the estimated SOH value is corrected in reverse.

[0011] Furthermore, considering the physical characteristics of SOH as a slow variable, this invention introduces a single-step amplitude hard-limiting constraint in the feedback correction stage. This effectively suppresses instantaneous jumps and miscalibrations caused by disturbances such as voltage acquisition noise, residual polarization, and temperature drift. This makes the convergence trajectory of the SOH estimate smoother and more closely reflects the actual aging process of the battery, further improving the engineering robustness of the estimation. Through multiple iterations, the SOH estimate gradually approaches the true value. Experiments show that under test conditions of a rated capacity of 102 Ah and a current SOH of 82.4%, convergence can be achieved in 20 iterations, with the maximum error controlled within 3.4%, significantly improving the accuracy of SOH calculation.

[0012] (3) Balancing calculation accuracy and engineering feasibility Compared to the calculation deviations caused by fixed curves or simplified models in existing technologies, this invention dynamically adapts to curve offsets through coordinate transformation, which not only eliminates long-term accumulated errors but also ensures that SOH can reflect the real health status of the battery in real time, achieving a balance between the accuracy of SOH calculation and engineering feasibility throughout the entire life cycle. Attached Figure Description

[0013] Figure 1 This is a flowchart illustrating a battery health status calculation method according to the present invention. Figure 2 To obtain the original SOC-OCV curves for the BOL and EOL states of the battery cells; Figure 3 The SOC error diagram is obtained by looking up the original SOC-OCV curve of the BOL state for EOL state cells; Figure 4 Error diagram of SOH value calculated using the original SOC-OCV curve of BOL state for EOL state cells; Figure 5 This is a SOC-OCV comparison curve generated using a scaling and splicing formula in a battery health state calculation method of the present invention. Figure 6 This is a comparison chart of the SOC-OCV curve after 200 iterations in the battery health state calculation method of the present invention and the existing SOC-OCV curve. Figure 7 This describes the convergence process of the SOH estimate with the number of iterations in a battery health state calculation method of the present invention. Detailed Implementation

[0014] The following detailed description illustrates the specific implementation method: The battery health state calculation method in this embodiment is based on the analysis of the battery aging mechanism. By establishing the dynamic relationship between the OCV curve and SOH, a SOC-OCV curve adapted to the current SOH state is constructed, thereby realizing the accurate calculation of SOH and forming a complete technical closed loop from basic testing to dynamic correction and then to accuracy optimization.

[0015] Provide a method for calculating battery health status, as shown in the appendix. Figure 1 As shown, it includes the following steps: S1. Obtain the original SOC-OCV curves for the BOL and EOL states of the battery cell, respectively, and determine the SOC attenuation boundary point. .

[0016] In this embodiment, the SOC-OCV curve was obtained by performing a discharge test at 25°C with an SOC interval of 5%. For example, the discharge SOC-OCV curve test was performed at 25°C for the battery in its BOL (early life) state, with an SOC interval of 5%, and the data obtained are shown in Table 1 below.

[0017] Table 1. SOC-OCV under discharge of lithium iron phosphate cells

[0018] At 25℃, the discharge SOC-OCV curve was tested for the battery EOL (end of life) state, with the SOC interval set to 5%. The data obtained are shown in Table 2 below.

[0019] The EOL curve is obtained from a pre-established standard aging model / database based on accelerated aging tests of the same battery model. The BMS retrieves this data through a lookup table, rather than through real-time testing. The EOL curve is only used for offline verification of the scaling and splicing formula and is not involved in onboard online calculations. During vehicle BMS operation, only the EOL curve and the current SOH estimate are needed to generate an equivalent aging curve.

[0020] Table 2 SOC-OCV under discharge of lithium iron phosphate cells

[0021] The SOC-OCV curves for the cell's BOL (early life) and EOL (end of life) states are thus obtained as shown in the attached figure. Figure 2 As shown. Simultaneously, the initial SOH value is determined based on the established SOC-OCV original curve.

[0022] If the SOC value of a cell in the EOL state is directly looked up using the BOL SOC-OCV curve, a significant SOC calculation error will occur. This error can be further explained in the appendix. Figure 3 A visual demonstration is provided. Meanwhile, when EOL cells employ the existing "two consecutive OCV calibrations to calculate SOH" scheme (i.e., the two-point capacity method based on OCV calibration, calculating SOH through the capacity difference and SOC difference between the two OCV calibrations), the SOC value still relies on the SOC-OCV curve of the BOL state. However, battery aging has caused the actual SOC-OCV curve to shift, and this curve mismatch will further induce SOH calculation errors. This error can be addressed through the attached... Figure 4 This demonstrates the accuracy deficiencies of existing technologies due to their inability to adapt to changes in the SOC-OCV curve after aging.

[0023] In this embodiment, through observation of test results and analysis of the cell mechanism, it was found that: First, although battery aging (decreased SOH) leads to capacity decay, the shape of the OCV-SOC curve remains basically unchanged; Second, the capacity decay caused by battery aging (decreased SOH) is mainly concentrated above SOC=70%, while the capacity from the same OCV value to the discharge cutoff voltage remains basically unchanged before and after aging. Therefore, the SOC decay boundary point is determined based on the concentration of capacity decay in the SOC-OCV curve. .

[0024] In this embodiment, The value range is 60% to 80%. The lower limit of 60% is because a low boundary point will make the high SOC section covered by formula (2) too wide, resulting in an increase in the nonlinear error of linear compression splicing. The upper limit of 80% is because a high boundary point will make the low SOC pure scaling section applicable to formula (1) too narrow, making it difficult to cover the main calibration range generated by the battery under normal discharge conditions. In this embodiment, for the tested lithium iron phosphate cell, the value is determined based on the measured distribution characteristics of its capacity decay concentration. =70%; for other battery systems (such as ternary lithium, lithium cobalt oxide, solid-state batteries, etc.). The above adjustments can be made according to the respective SOC-OCV curve attenuation concentration characteristics within the above range.

[0025] S2, keeping the OCV value unchanged, establish a dynamic mapping relationship between the OCV curve and the current SOH estimate, and form the SOC-OCV scaling and splicing formula.

[0026] First, during the battery aging process, the overall shape of the OCV-SOC curve remains basically unchanged, with only the SOC coordinate shifting due to capacity decay. Second, capacity degradation is mainly concentrated in the high SOC range (≥ ), while the low SOC range (≤ Within the range of discharge cutoff voltage to the same OCV value, the absolute value of available capacity remains essentially unchanged. Therefore, in the low SOC region, the SOC coordinates of the equivalent curve only need to be scaled up according to the total capacity attenuation ratio; in the high SOC region, the high SOC segment of the BOL curve needs to be compressed and spliced ​​into the equivalent region to reflect the concentrated attenuation in this area.

[0027] The SOC-OCV scaling and splicing formula is as follows: For SOC < The range is defined by scaling the SOC coordinates of the BOL curve according to the total capacity decay ratio, and expressed as: (1) For SOC≥ The high SOC segment of the BOL curve is linearly compressed and spliced ​​to an equivalent interval, represented as: (2) In the formula, The SOC coordinates are the equivalent aging curve generated based on the current SOH estimate. The SOC coordinates of the SOC-OCV curve for BOL state battery cell testing; This is the estimated SOH value for the current iteration step (the initial calibrated SOH is used during the first iteration).

[0028] It should be noted that the above formula only relies on the SOC-OCV baseline curve at the BOL state and the current estimated SOH value to generate the equivalent aging curve, without the need to obtain the measured OCV data after battery aging in real time. The EOL state cells and their SOC-OCV test curves mentioned in this embodiment are attached. Figure 2 The data shown is offline verification data obtained through accelerated aging tests of the same type of battery. It is only used to verify the accuracy and rationality of the above scaling and splicing formula and is not a necessary input for the vehicle operation of this method.

[0029] S3: After the battery system is fully charged and begins to discharge, monitor the cumulative discharge capacity in real time. When the cumulative discharge capacity reaches 70% of the current available capacity, record the discharge capacity and calculate the current SOC value. Substitute the SOH estimate (based on the SOH value output from the previous iteration or the initial calibration SOH value) into the SOC-OCV scaling and splicing formula to generate an SOC-OCV comparison curve adapted to the current aging level.

[0030] In this embodiment, the battery management system monitors the battery status in real time during vehicle operation and utilizes daily driving conditions to naturally trigger online SOH correction. When the system detects that the battery has reached a fully charged state (e.g., confirmed by the constant voltage stage characteristics at the end of charging or OCV calibration), it uses this moment as the integration starting point. The current sensor collects the charging and discharging current in real time, and accumulates the discharge capacity since the most recent full charge through ampere-hour integration. When the cumulative discharge capacity reaches 70% or more of the current available capacity (corresponding to a remaining SOC ≤ 30%), the system records this discharge capacity and calculates the current estimated SOC value in real time based on the current discharge capacity, expressed as: (3) In the formula, The cumulative discharge capacity is obtained by real-time ampere-hour integration by the BMS since the most recent full charge; This refers to the rated capacity of the battery system. This is the estimated SOH value for the current iteration step (the initial calibrated SOH is used during the first iteration).

[0031] It should be noted that this step sets the discharge capacity threshold to 70% of the current available capacity (corresponding to a remaining SOC ≤ 30%). The primary technical basis for this is not the capacity decay distribution characteristics, but the physical requirements of OCV calibration on the local slope of the SOC-OCV curve.

[0032] The SOC-OCV curves obtained from S1 (Table 1) show that the SOC-OCV curves of lithium iron phosphate batteries exhibit an electrochemical plateau region in the SOC range of approximately 30% to 95%, with a very small local slope of dOCV / dSOC (typically approximately 0.2 to 1.4 mV / %). However, in the low SOC region (SOC ≤ 30%) and the high SOC region (SOC ≥ 95%), the curves show a high slope (typically 5 to 70 mV / %). Considering that the sampling accuracy of BMS cell voltage is typically ±2 to 5 mV, only when the SOC at the time of acquisition is in the aforementioned high-slope region can the SOC inversion error corresponding to the voltage acquisition error be controlled within ±1%, allowing for effective OCV calibration. If calibration occurs in the plateau region, the same voltage acquisition error will cause a SOC inversion error of ±3% to ±10%, rendering the calibration results meaningless.

[0033] While the high SOC steep slope zone (≥95%) meets the slope condition, it requires the battery to be close to full charge, which is difficult to reach stably under normal vehicle usage conditions. In contrast, the low SOC zone (≤30%) is naturally achievable under normal discharge conditions, making it more feasible for engineering. Therefore, this embodiment selects SOC ≤30% (i.e., discharge capacity ≥70%) as the OCV calibration trigger condition.

[0034] Therefore, the 70% in the step "cumulative discharge capacity ≥ 70%" is the same as the SOC decay boundary point in S1. =70% (preferred value) has the same numerical value, but its physical meaning is completely different: the former is the calibration window threshold determined by the engineering boundary of the steep slope region at the lower end of the SOC-OCV curve; the latter is the curve segment splicing point determined by the concentration of battery aging capacity decay. When When taking other values ​​within the range of 60% to 80%, the discharge capacity threshold of this step remains unchanged at 70%. At this time, the calibration point SOC≤30% is always located in the low SOC steep slope area, and the pure scaling assumption of formula (1) also holds.

[0035] Substituting the calculated current SOC estimate and SOH estimate into the SOC-OCV scaling and stitching formula constructed in S2, a SOC-OCV comparison curve adapted to the current aging level is generated in real time, denoted as... This curve dynamically reflects the open-circuit voltage characteristics of the battery under its current aging state, and is used for subsequent theoretical OCV lookup comparison. The entire S3 step is automatically completed by the battery management system during normal vehicle operation, without the need for offline testing or manual intervention, making full use of the deep discharge conditions that naturally occur during daily driving to achieve online SOH estimation.

[0036] S4, based on the current SOC estimate obtained in S3, in the value generated in S3 By looking up the table on the curve, the theoretical open-circuit voltage value corresponding to this SOC can be obtained. .

[0037] In this embodiment, after completing the real-time SOC calculation and curve generation in S3, the battery management system uses the current SOC estimate as input and compares the newly generated SOC-OCV curve (i.e., Linear interpolation is performed on the SOC point to look up the theoretical open-circuit voltage. .

[0038] Since the comparison curve has been adapted to the aging state corresponding to the current SOH estimate using the scaling and splicing formula of S2, the result obtained from the table is... It reflects the open-circuit voltage level that the battery should have at its current state of aging and SOC level, rather than directly applying the theoretical value from the BOL state curve. As the benchmark expected value compared with the actual resting voltage in subsequent S5 steps, it is a key intermediate quantity in the connection curve generation and voltage feedback process throughout the entire closed-loop correction process. The entire table lookup process is completed in real time by the battery management system without offline intervention.

[0039] S5: After the vehicle is parked, the battery management system automatically monitors the battery's resting conditions. When the stability criteria are met, it collects the individual cell voltage and compares it with the theoretical open-circuit voltage obtained in S4. The SOH value is adaptively adjusted according to the direction and magnitude of the voltage deviation.

[0040] In this embodiment, after the vehicle completes driving or charging, it is parked. The battery management system continuously monitors the battery circuit current I(t) and the individual battery cell terminal voltage U(t). The battery is determined to have entered a stable open-circuit voltage state when both of the following conditions are met simultaneously: Condition 1: The absolute value of the loop current is continuously lower than the preset current threshold I_th, that is, |I(t)| < I_th. In this embodiment, I_th = 0.005C, where C is the rated capacity of the battery cell. Condition 2: The duration of the above current condition is not less than the preset resting time T_rest. In this embodiment, T_rest = 2 hours.

[0041] In this embodiment, the current threshold I_th is set to exclude the minor disturbances to the single-cell terminal voltage caused by vehicle static power consumption (such as BMS self-consumption and 12V low-voltage load leakage) and battery self-discharge. That is, when the value of I_th is too large, the residual current will generate a non-negligible voltage drop across the battery internal resistance, contaminating the open-circuit voltage acquisition; when the value of I_th is too small, it will be lower than the background noise level of the BMS current sampling circuit, causing the static criterion to be consistently unsatisfactory and the SOH calibration to fail to be triggered.

[0042] The resting time T_rest is set to ensure that the polarization voltage (including ohmic polarization, electrochemical polarization, and concentration polarization) generated by the previous charge-discharge process inside the cell has sufficiently decayed to a negligible level. If the T_rest value is too short, the polarization will not be fully released, and the measured terminal voltage will deviate from the true OCV. If the T_rest value is too long, the SOH calibration time will be delayed, reducing the timeliness of online estimation. Considering the above engineering factors, this embodiment determines that I_th=0.005C and T_rest=2h is the optimal combination under typical operating conditions. It can also be adjusted within the range of 0.001C≤I_th≤0.02C and 1h≤T_rest≤4h according to the specific battery system (such as ternary, lithium iron phosphate, solid-state battery, etc.) and the vehicle's overall power consumption characteristics.

[0043] At the moment when the above criteria are met, the BMS sampling circuit acquires the voltage of each battery cell and records it as follows. This is the current actual open-circuit voltage. (The remaining text appears to be incomplete and requires further context.) Obtained by looking up table with S4 Compare and calculate voltage deviation The SOH estimate is adaptively adjusted based on the direction and magnitude of the deviation. The adjusted SOH value is denoted as SOH_Cal, and the adjustment method is expressed as follows: (4) (5) (6) In the formula, This is the original SOH correction amount calculated according to the formula in this round; To preset the maximum correction range per step (in this embodiment) = 1%) The sign function (takes +1 when the input is positive, and takes -1 when the input is negative) 1. When the input is 0, take 0). This is a function that takes the minimum value. This is the correction amount that actually affects the estimated SOH value after amplitude limiting, and its absolute value does not exceed .

[0044] k is the correction gain coefficient from voltage deviation to SOH deviation, and its calculation method is expressed as follows: (7) The local slope of the original SOC-OCV curve of the BOL state established for S1 at the current SOC estimation point can be approximated in engineering by the difference between two adjacent data points of the curve; This is a coordinate transformation factor used to convert the "SOC coordinate deviation on the equivalent SOC-OCV curve" into the "SOH percentage deviation". The preset damping coefficient has a value range that satisfies 0 < ≤1, in this embodiment =0.7.

[0045] The calculation formula for k in the above formula (7), the first term To achieve adaptive correction magnitude to the nonlinearity of the SOC-OCV curve: In the steep range of the curve (such as when SOC is close to 0% or 100%), the same voltage deviation corresponds to a smaller SOC error, and k is automatically set to a smaller value to avoid over-adjustment; in the relatively flat plateau range of the curve, the same voltage deviation corresponds to a larger SOC error, and k is automatically set to a larger value to accelerate convergence.

[0046] Second item The SOC-OCV scaling and splicing formula is derived from S2.

[0047] Depend on It can be deduced that the deviation of the SOC coordinates of the equivalent curve is... Deviation from SOH Between Therefore, this factor realizes the coordinate transformation from SOC deviation to SOH deviation.

[0048] Third item It is the damping coefficient, used to suppress over-adjustment oscillation of the SOH estimate caused by voltage acquisition noise, temperature drift, or operating condition disturbances. The larger the value, the larger the magnitude of a single correction and the faster the convergence speed, but the risk of oscillation increases; The smaller the value, the smaller the magnitude of a single correction and the smoother the convergence trajectory, but the more iterations are required. In this embodiment... =0.7 is a typical fixed value.

[0049] As a preferred embodiment. It can be dynamically adjusted during the iteration process: set the iteration switching point. When the current iteration number n≤ Time to take (The damping coefficient is taken as a large value in the initial stage of iteration to quickly approximate the true state of harmonic equilibrium (SOH) when it is far from the actual state of harmonic equilibrium (SOH); when n > Time to take (The damping coefficient in the later stages of iteration is chosen to be smaller for steady-state refinement near the actual SOH, avoiding over-adjustment oscillations.) The typical value range is 0.5 ≤ ≤1.0, 0.3≤ ≤0.7, The time is the preset switching point, in this embodiment =10, corresponding to =1.0、 =0.5).

[0050] In this embodiment, the dynamic adjustment strategy is compared with the fixed one. The advantage of this approach is that the SOH estimation error is relatively large in the early stages of iteration. Cooperate Amplitude limiting allows the SOH to quickly approach the true value with a step size close to the maximum allowable amplitude; in the later stages of iteration, the SOH estimation error is significantly reduced, and the smaller... This allows the single-step correction amount to be naturally smaller than (At this time, the amplitude limiting mechanism is not triggered), and the corrected trajectory is smooth and the steady-state fluctuation amplitude is small, thus balancing convergence speed and steady-state accuracy without increasing the total number of iterations.

[0051] Given that the battery state of OH (SOH) is a state quantity reflecting the long-term aging of the battery, the actual change between two adjacent OCV calibration events is usually very small (under normal operating conditions, the actual SOH change within a single calibration interval generally does not exceed 0.5%). In order to prevent overcorrection and miscalibration caused by abnormal single voltage deviations (such as voltage sampling noise, insufficient polarization release, temperature compensation residuals, etc.), this embodiment applies a hard limit constraint on the single-step SOH correction amplitude.

[0052] In this embodiment, the above-mentioned limiting mechanism is based on the following three engineering considerations: First, SOH is a slow variable that physically characterizes the cumulative aging of a battery. Its actual rate of change is determined by factors such as the number of cycles, temperature, and depth of discharge, and changes gradually under reasonable operating conditions. Between two adjacent OCV calibration events, the actual change in SOH is usually much less than 1%.

[0053] Secondly, during a single voltage acquisition and table lookup process, there are multiple sources of disturbance, such as mV-level voltage sampling noise, polarization residual voltage, temperature compensation error, and curve interpolation error, which may generate abnormally large disturbances in a certain iteration. If adopted directly, the estimated SOH value will jump, becoming out of sync with the actual aging process of the battery.

[0054] Third, the amplitude limiting constraint does not affect the final convergence of the iteration process. Since S6 iterates multiple times, even if a single correction is truncated by the amplitude limiting, the remaining deviation will be gradually eliminated in subsequent iterations. The amplitude limiting mechanism can effectively suppress the instantaneous jumps triggered by occasional abnormal operating conditions, making the convergence trajectory of the SOH estimate smoother and more in line with the physical characteristics of SOH as a slow variable.

[0055] when At that time, the amplitude limiting mechanism does not work. , Correct normally according to the original formula; when At that time, the amount of this round of correction was truncated to The remaining deviations will be corrected in subsequent iterations.

[0056] S6. Using the adjusted SOH value as the new input, repeat S3 to S5 to continuously iterate and calculate, gradually reducing the SOH error.

[0057] In this embodiment, the corrected SOH estimate is used as the new input, and S3 to S5 are re-executed. The SOH error is gradually reduced through multiple iterations. The comparison chart of the SOC-OCV curve after 200 iterations and the existing SOC-OCV curve is attached. Figure 6 As shown, through multiple iterations of optimization, the estimated SOH value gradually approaches the true value, verifying the accuracy improvement effect of the process.

[0058] In this embodiment, a practical application is used as an example for illustration.

[0059] If a battery cell with a rated capacity of 102Ah and a current SOH of 82.4% is used for testing, and it is randomly discharged from 70% to 100%, the SOH convergence process is calculated as shown in the attached figure. Figure 7 As shown, its maximum error is 3.4%. In this embodiment, the SOH estimate has entered the steady-state fluctuation range of ±1% after 20 iterations (considered as convergence). Subsequent iterations are performed up to 200 times to verify the steady-state stability.

[0060] In this embodiment, it is first clarified that the capacity decay caused by battery aging is mainly concentrated in the case of SOC ≥ SOC < The capacity remains basically unchanged from the same OCV value to the discharge cutoff voltage, and the shape of the OCV-SOC curve remains stable. Based on this, the SOC coordinates are segmented. When SOC < SOC attenuation boundary point, the SOC coordinates of the initial SOC-OCV curve are scaled by formula (1). When SOC ≥ SOC attenuation boundary point, the coordinates are spliced ​​by formula (2) to construct an SOC-OCV curve that adapts to the current SOH state.

[0061] Secondly, the full charge step provides a standardized starting point for subsequent discharge, ensuring consistent initial conditions for each iteration; while the setting of a discharge capacity ≥70% is to ensure that the SOC falls into the low SOC region (≤30%) during calibration. The SOC-OCV curve constructed based on scaling and stitching is the core bridge connecting the battery aging mechanism and SOH calculation, and its accuracy directly affects the reliability of the expected OCV.

[0062] Finally, the voltage acquisition and comparison after resting serves as the feedback loop for SOH correction, dynamically adjusting the SOH based on the deviation between the measured and theoretical values. The iterative process gradually eliminates errors through multiple cycles, ultimately achieving precise convergence of the SOH value. Each step is interconnected, forming a complete technical closed loop from basic testing to dynamic correction and then to accuracy optimization.

[0063] The above descriptions are merely embodiments of the present invention, and common knowledge such as specific technical solutions and / or characteristics are not described in detail here. It should be noted that those skilled in the art can make various modifications and improvements without departing from the technical solutions of the present invention, and these should also be considered within the scope of protection of the present invention. These modifications and improvements will not affect the effectiveness of the implementation of the present invention or the practicality of the patent. The scope of protection claimed in this application should be determined by the content of its claims, and the specific embodiments described in the specification can be used to interpret the content of the claims.

Claims

1. A method for calculating battery health status, characterized in that, Includes the following steps: S1. Obtain the original SOC-OCV curves for the BOL and EOL states of the battery cell, respectively, and determine the SOC attenuation boundary point. ; S2, keeping the OCV value unchanged, establish the relationship between the OCV curve and SOH, and form the SOC-OCV scaling and splicing formula; S3, after the battery system is fully charged and begins to discharge, monitor the cumulative discharge capacity in real time; when the cumulative discharge capacity reaches 70% of the current available capacity, record the discharge capacity and calculate the current SOC value; Substitute the estimated SOH value into the SOC-OCV scaling and splicing formula to generate an SOC-OCV comparison curve adapted to the current aging level; S4. Based on the current SOC value of the battery, obtain the theoretical open-circuit voltage value corresponding to that SOC through the SOC-OCV comparison curve. ; S5: After the vehicle is parked, the battery management system automatically monitors the battery's resting conditions. When the stability criteria are met, it collects the individual cell voltage and compares it with the theoretical open-circuit voltage obtained in S4. In comparison, the SOH value is adaptively adjusted based on the direction and magnitude of the voltage deviation; S6. Using the adjusted SOH value as the new input, repeat S3 to S5 to continuously iterate and calculate, gradually reducing the SOH error.

2. The battery health status calculation method according to claim 1, characterized in that: In S2, the SOC-OCV scaling and splicing formula is: for SOC < The range is defined by scaling the SOC coordinates of the BOL curve according to the total capacity decay ratio, and expressed as: ； For SOC≥ The high SOC segment of the BOL curve is linearly compressed and spliced ​​to an equivalent interval, represented as: ; In the formula, The SOC coordinates are the equivalent aging curve generated based on the current SOH estimate. The SOC coordinates of the SOC-OCV curve for BOL state battery cell testing; This is the estimated SOH value for the current iteration step (the initial calibrated SOH is used during the first iteration).

3. The battery health status calculation method according to claim 1, characterized in that: In S3, the current SOC value is calculated based on the current discharge capacity and expressed as follows: ； In the formula, This represents the current discharge capacity. This refers to the rated capacity of the battery system. The currently calculated SOH value.

4. A method for calculating battery health status according to claim 3, characterized in that: In S5, the adjusted SOH value is recorded as follows. The adjustment methods include: ； in, The SOH correction amount after single-step amplitude limiting is expressed as: ； ； In the formula, The correction gain coefficient is determined based on the local slope of the SOC-OCV curve at the current SOC estimate point; This is the preset maximum correction range per step; It is a symbolic function; This is a function that takes the minimum value.

5. A method for calculating battery health status according to claim 4, characterized in that: The maximum single-step correction limit satisfies 0.5% ≤ ≤ 2%.

6. A method for calculating battery health status according to claim 4, characterized in that: The corrected gain coefficient The calculation method is as follows: ； The slope of the original SOC-OCV curve for the BOL state at the current SOC estimate point is the local slope. The preset damping coefficient has a value range that satisfies 0 < ≤1.

7. The battery health status calculation method according to claim 6, characterized in that: The damping coefficient The settings are dynamically adjusted during the iteration process, with the initial step being... Take in the later stage of iteration ,and < , where 0.5≤ ≤1.0, 0.3 ≤ ≤0.

7.

8. The battery health status calculation method according to claim 1, characterized in that: The The capacity decay concentration is determined based on the SOC-OCV curve, ≤ 60%. ≤80%; preferably, =70%.

9. The method for calculating battery health status according to claim 1, characterized in that: The SOC-OCV curve was obtained by conducting a discharge test at 25°C with the SOC interval set to 5%.

10. The method for calculating battery health status according to claim 1, characterized in that: In S5, the stability criterion is that the following two conditions are met simultaneously: Condition 1: The absolute value of the battery circuit current is continuously lower than the preset current threshold I_th; Condition 2: The duration of satisfying Condition 1 is not less than the preset rest duration T_rest; Where 0.001C≤I_th≤0.02C, 1h≤T_rest≤4h, and C is the rated capacity of the battery cell.

Images