Lithium iron phosphate battery health state estimation method
By combining the IC curve method and the SOC-V curve method, the feature point calculation is optimized, and the accuracy of lithium battery capacity estimation is solved, and a fast and accurate battery health status evaluation is achieved, which is suitable for the normal working conditions of lithium batteries.
Patent Information
- Application Number
- CN202411192445.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-28
- Publication Date
- 2024-11-19
AI Technical Summary
In the prior art, the accuracy and reliability of lithium battery capacity estimation are limited by the IC curve method and the SOC-V curve method. Especially when the SOC is higher than 20%, the capacity estimation corresponding to the third peak is inaccurate. Moreover, single-point characteristics lead to large errors, affecting the evaluation of battery health status.
Combining the IC curve method and the SOC-V curve method, through optimization and fusion, the characteristic points are obtained using reference data fitting, the current capacity and health status of the battery are calculated, and the power difference is calculated by using the A/H integrator method or the direct power method. Improve the speed and accuracy of the estimation.
Achieve rapid and accurate estimation of the capacity and health status of lithium iron phosphate batteries, with a maximum error of less than 2.5%, and no additional charge and discharge test is required. It is suitable for battery management systems for household cars and energy storage power plants under normal working conditions. .
Smart Images

Figure CN118980939A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of power battery detection, and in particular to a method for estimating the health status of a lithium iron phosphate battery. Background Art
[0002] Lithium batteries are widely used in electronic devices, electric vehicles, energy storage systems and other fields due to their high energy density, long cycle life and environmental friendliness. With the popularity of lithium batteries in various application scenarios. The state of health (SOH) of a lithium battery refers to a measure of the performance retention or remaining life of a lithium battery under current conditions of use relative to its new state. It reflects the degree of aging and remaining performance of the battery under the influence of various factors (such as temperature, charge and discharge cycles, charge and discharge rates, depth of discharge, etc.) during long-term use. It is extremely important to accurately evaluate the health status of lithium batteries during use. Generally, the health status is calculated by comparing the ratio of the current battery capacity to the rated capacity. Therefore, it is extremely important to quickly estimate the battery capacity.
[0003] The capacity of a lithium battery refers to the maximum amount of electricity that a battery can store and release under certain conditions, which directly affects the battery life and performance of the device. In the prior art, the IC (Incremental Capacity) curve method can reveal the details of the internal reaction of the battery. However, in the case of pure peak estimation using the IC curve, only the second and third peaks can generally be used for estimation. However, the SOC (State of Charge) value corresponding to the third peak is less than 20%, or even 15%. In the actual charging and discharging point scenario, the SOC of electric vehicles is often greater than 30%. Therefore, the integrity of the data is often affected, and the practicality of this solution is affected, which in turn leads to a decrease in the accuracy of capacity estimation. In addition, the SOC-V (Voltage) curve method commonly uses only one characteristic point (the front inflection point), which can be used for battery SOC estimation and correction, etc. However, for the estimation of the health state, the error of using only a single point is relatively large, and during the battery aging process, the model parameters may drift, which further increases the estimation error. These problems have an adverse effect on the accuracy and reliability of battery capacity estimation in practical applications. Summary of the invention
[0004] The present invention provides a health status estimation method for a lithium iron phosphate battery, which combines the advantages of an IC curve method and a SOC-V curve method, and effectively improves the speed and accuracy of battery capacity and health status estimation by optimizing and integrating the two methods.
[0005] The present invention can be achieved through the following technical solutions:
[0006] A method for estimating the health status of a lithium iron phosphate battery comprises the following steps:
[0007] Step 1: Based on the charge and discharge data provided by the battery manufacturer, obtain the benchmark SOC-V curve and benchmark IC curve by data fitting;
[0008] Step 2: Calculate the SOC value corresponding to the maximum curvature in the reference SOC-V curve, and record the corresponding point as the front inflection point, which is located in the 80% SOC-100% SOC range. At the same time, calculate the SOC value corresponding to the second peak of the reference IC curve, and record the corresponding point as the second peak point. Calculate the difference between the two SOC values, and record it as ΔSOC*.
[0009] Step 3: According to the current charge and discharge data of the battery, calculate the corresponding front inflection point and the second peak point, as well as the capacity difference between the front inflection point and the second peak point, recorded as Q t , and then calculate the current battery capacity C′=Q t / ΔSOC*;
[0010] Step 4: Calculate the battery's current state of health (SOH) using the following formula:
[0011]
[0012] Among them, C 额 Indicates the rated capacity of the battery.
[0013] Further, according to the reference SOC-V curve, the abscissa of the reference IC curve, ie, the voltage value, is replaced with the corresponding SOC value, and then the corresponding SOC value is found according to the second peak point of the reference IC curve.
[0014] Furthermore, according to the charge and discharge data provided by the battery manufacturer, the corresponding benchmark SOC-V curves at different currents and temperatures are obtained by data fitting, and the front inflection points corresponding to the maximum curvature in these benchmark SOC-V curves are calculated respectively, and a lookup table of the voltage values corresponding to the front inflection points at different currents and temperatures is made.
[0015] In step 3, according to the current charge and discharge data of the battery and the corresponding working conditions, the corresponding voltage value is found from the query table and recorded as V 拐 , use data fitting to obtain the IC curve, find the voltage value corresponding to the second peak point of the IC curve, and record it as V IC , according to the battery management system, find out whether the current voltage of the battery reaches voltage V 拐 and voltage V IC The time points corresponding to the 拐 and t IC , use the following formula to calculate the power difference Q using the ampere-hour integration method t;
[0016]
[0017] Or according to the battery management system, find out whether the current voltage of the battery reaches the voltage V 拐 and voltage V IC The corresponding power values are recorded as Q 拐 and Q IC , use the following formula to calculate the power difference Q using the direct power method t ;
[0018] Q t =|Q 拐 -Q IC |.
[0019] Furthermore, when the collected current charge and discharge data of the battery is continuous and includes data corresponding to the front inflection point and the second peak point, the power difference Q is calculated using the ampere-hour integration method or the direct power method. t ;
[0020] When the collected battery current charge and discharge data is discontinuous and includes data corresponding to the previous inflection point or the second peak point, the direct power method is used to calculate the power difference Q t .
[0021] Further, use steps 1 to 3 to calculate the power difference Q corresponding to each cell in the battery pack t , and then calculate the current single battery capacity C' of each battery cell, and take the minimum value of the current single battery capacity to calculate the current health state SOH of the entire battery pack. The specific calculation formula is as follows
[0022]
[0023] Furthermore, when the step calculates the SOC value corresponding to the front inflection point, the curvature corresponding to each point on the SOC-V curve is first calculated, and the curvature change curve, that is, the relationship between the curvature and the voltage, is fitted, and then the horizontal coordinate of the curvature change curve, that is, the voltage value, is replaced with the SOC value according to the SOC-V curve to obtain the maximum curvature at this time, and the corresponding SOC value is found. The point corresponding to the SOC value on the SOC-V curve is the front inflection point.
[0024] The beneficial technical effects of the present invention are:
[0025] 1. Combining the advantages of high precision of IC curve method and fast response of SOC-V curve method, the method makes full use of the extraction of key feature points and the simplified difference calculation model, and can make accurate and fast estimates under normal working conditions. This method requires small amount of calculation in real-time estimation, simple process, high accuracy, and the maximum error is less than 2.5%.
[0026] 2. When the method of the present invention is used to estimate the capacity and SOH of a lithium iron phosphate battery, it can be completed only when the battery is in normal working condition, and no additional charge and discharge test is required.
[0027] 3. The estimation method of the present invention has low requirements for the selection of battery charging and discharging data segments. It only needs to select the data segment containing the front inflection point of the SOC-V curve and the second peak point of the IC curve to quickly calculate the current battery capacity, which is generally between 40% SOC-100% SOC. Therefore, the starting SOC of charging is low and does not need to be below 40%. It can be well applied to the normal charging range of household cars and can also be used in the battery management system of energy storage power stations. BRIEF DESCRIPTION OF THE DRAWINGS
[0028] Figure 1 , 2 Schematic diagrams of the change rules of the SOC value and the difference corresponding to the front inflection point of the SOC-V curve and the second peak point of the IC curve in 1500 cycles of the present invention;
[0029] Figure 3 It is a schematic diagram of the overall process of the present invention;
[0030] Figure 4 It is a schematic diagram of the benchmark SOC-V curve and the corresponding curvature curve and IC curve of the present invention;
[0031] Figure 5 It is a schematic diagram of the error variation between the estimated value and the accurate value of the current capacity of all cells in a battery pack of the present invention. DETAILED DESCRIPTION
[0032] The specific implementation of the present invention is described in detail below with reference to the accompanying drawings and preferred embodiments.
[0033] For lithium iron phosphate batteries, the points with the largest curvature at both ends of the capacity-voltage SOC-V curve are called inflection points. The curvature is the rotation rate of the tangent direction angle at a point on the curve to the arc length, indicating the degree to which the curve deviates from a straight line. Mathematically, it is a numerical value that indicates the degree of curvature of the curve at a certain point.
[0034] In order to accurately obtain the position of the inflection point, a mathematical model can be established using the charge and discharge data provided by the battery manufacturer, and the least squares polynomial fitting can be used to obtain the actual SOC-V curve; then, the curvature of each point at both ends of the curve is calculated, where the two end areas are approximately above 80% and below 20% of the SOC value. Since the general use of the battery will not reach SOC=20% or less, the front inflection point, that is, the inflection point above SOC=80%, is selected as a characteristic point of the present invention. The IC curve represents the relationship between the capacity increment dQ / dV and the voltage V of the battery during the charge and discharge process. The IC curve is obtained by differentiating the voltage and capacity changes during the charge and discharge process of the battery.
[0035] Through a series of experimental comparisons of lithium iron phosphate batteries, it can be found that although the remaining life of the battery is gradually decreasing, the curves have different changes in shape and characteristics, but in the same charge and discharge cycle test, the SOC value corresponding to the front inflection point of the SOC-V curve and the SOC value corresponding to the second peak point of the IC curve only decrease very slightly with the increase in the number of cycles, but the SOC difference between the two has almost no change. In addition, by comparing the charge and discharge cycle experiments under different temperatures and currents, it is found that the temperature and current have little effect on it and can be ignored. Figure 1 , 2 This is the changing pattern of the SOC value and the difference corresponding to the front inflection point of the SOC-V curve and the second peak point of the IC curve in 1500 cycles.
[0036] Based on this law of change, Figure 3 As shown, the present invention uses the front inflection point and the second peak point as two characteristic points to provide a method for estimating the health state of a lithium iron phosphate battery. First, based on the charge and discharge data provided by the battery manufacturer, the SOC value corresponding to the front inflection point of the benchmark SOC-V curve and the SOC value corresponding to the second peak point of the benchmark IC curve are obtained by data fitting, and the difference between the two SOC values is calculated; then, according to the current charge and discharge data of the battery, the corresponding front inflection point and the second peak point and their corresponding capacity difference are calculated, thereby estimating the current capacity of the battery and the current health state SOH.
[0037] The details are as follows:
[0038] S1. Construct the benchmark SOC-V curve and IC curve, and calculate the corresponding front inflection point and second peak point
[0039] For the target lithium iron phosphate battery, obtain the battery benchmark curve data. This data is the detailed pre-factory test data obtained from the manufacturer, mainly including SOC, voltage, temperature, current, charge / discharge power and rated capacity, and the SOC value interval is ≤1%.
[0040] First, the benchmark SOC-V curve of the target lithium-ion battery is obtained according to the SOC and corresponding voltage in the benchmark data, and then the corresponding curvature change curve is obtained according to the benchmark SOC-V curve, such as Figure 3 shown.
[0041] For a curve in a two-dimensional plane, the curvature can be defined as:
[0042]
[0043] Among them, x and y are the parameter equations of the curve. Substituting into the SOC-V curve, with SOC as the independent variable and voltage V as the dependent variable, we can deduce:
[0044]
[0045] The curvature of each point of the SOC-V curve is used as the ordinate and SOC is used as the abscissa to fit the curvature change curve corresponding to the benchmark SOC-V curve, and the SOC value corresponding to the maximum peak point of the curvature change curve is found, which is recorded as SOC 拐* .
[0046] Similarly, the reference capacity increment IC curve is obtained based on the voltage V and charge / discharge quantity Q in the reference data. The specific calculation formula is:
[0047]
[0048] in, represents the incremental charge, Q represents the charge of the battery, V represents the voltage of the battery, dQ is the differential of the capacity, dV is the differential of the voltage, ΔQ is the charge difference between adjacent sampling data, and ΔV is the voltage difference between adjacent sampling data.
[0049] First, based on the benchmark data, the curve L is fitted with the voltage V as the x-axis and the charge / discharge quantity Q as the Y-axis, and the differential value of this curve L is obtained. Let voltage V be the X-axis. The IC curve is obtained by fitting the Y-axis, filtering and smoothing it, and finally the IC curve abscissa is replaced by the voltage to the corresponding SOC value according to the reference SOC-V curve. Generally, the IC curve of lithium iron phosphate battery will show 2-3 characteristic peaks. In this invention, the second peak point is taken and the corresponding SOC value is recorded, which is recorded as SOC IC* .
[0050] S2. Calculate the benchmark parameter ΔSOC*
[0051] Obtain the SOC difference corresponding to the characteristic points of the above two reference curves:
[0052] ΔSOC*=|SOC 拐* -SOCIC* |
[0053] Taking ΔSOC* as the reference parameter, the SOC difference between the two characteristic points mentioned above remains unchanged, which means that the ratio of the charging / discharging capacity value between the two characteristic points to the total capacity remains unchanged. In practical applications, it means the relationship coefficient between the charging / discharging capacity value between the two characteristic points and the total capacity.
[0054] Since the ΔSOC* value is little affected by temperature and current, it can be obtained by simply using the corresponding data at any temperature and current in the charge and discharge data provided by the manufacturer.
[0055] S3. Estimate the current capacity of the battery and its current health status
[0056] For the target lithium iron phosphate battery during normal operation, the data in the SOC range where the front inflection point appears is selected, which is generally below 100% and above 80%; and the SOC curve where the second peak point of the capacity increment curve IC curve appears is generally between 40% and 60%, so generally only the data range of approximately 40% to 100% SOC is required.
[0057] According to the current charge and discharge data of the battery, only the data segment corresponding to the data interval of 40% SOC-100% SOC can be used to calculate the corresponding front inflection point and the second peak point, as well as the capacity difference corresponding to the front inflection point and the second peak point, recorded as Q t , and then calculate the current capacity of the battery C′=Q t / ΔSOC*;
[0058] We can use the battery management system BMS to extract the voltage V, current I, charge / discharge quantity Q, etc. at every change time Δt=1s in the data segment, and then obtain the voltage and quantity data The IC curve is fitted, and the voltage value corresponding to the second peak point of the IC curve is extracted and recorded as V IC .
[0059] However, for the calculation of the charge or discharge amount corresponding to the front inflection point, it is necessary to fit the SOC-V curve first, then calculate the curvature corresponding to each point, and find the inflection point position corresponding to the maximum curvature value. The calculation process is relatively complicated. After testing, the voltage value of the inflection point of the SOC-V curve will change due to temperature and charge / discharge current. Therefore, in step S1, we can use the above method of solving the front inflection point to obtain the voltage value corresponding to the front inflection point at different currents and temperatures based on the charge and discharge data at different currents and temperatures in the benchmark data in advance, so as to establish a query table. In this way, when calculating the front inflection point based on the current data fragment of the battery, there is no need to follow the above solution method, and the corresponding voltage value can be directly found from the query table according to the current operating parameters of the battery charge and discharge, such as current and temperature, and recorded as V 拐, and then find the operating voltage V=V according to the data recorded by BMS 拐 and V = V IC The corresponding charge or discharge amount and time point are recorded as Q 拐 and Q IC , t 拐 and t IC , which can effectively save calculation time and improve reaction efficiency.
[0060] Using the following formula, the difference in power between the two peaks, Q, can be obtained: t .
[0061]
[0062] Among them, t IC is the time to reach the second peak of the IC curve, t 拐 It is the time to reach the inflection point of the curvature curve, I represents the working current of the charge and discharge condition. If it is constant current charge and discharge, I is a constant value. If it is not constant current charge and discharge, I is a function of time t.
[0063] Note: If the selected charge and discharge data segment contains only one characteristic value, you can select the most recent data segment containing another characteristic value. Note that the working current of the two data segments must be consistent. When the data segments are continuous and contain two characteristic points, Q t It can be obtained directly by the ampere-hour integration method (current integration over time), or by taking the difference between the charge / discharge quantity at two characteristic points. If the data fragments are not continuous, Q t It can only be obtained by using the direct charge method and subtracting the charge / discharge corresponding to the two characteristic points.
[0064] The current capacity of the battery can then be found:
[0065]
[0066] Where C' is the current capacity of the battery.
[0067] S4. Estimate the battery's current state of health (SOH)
[0068] Considering that the battery pack may be composed of multiple cells connected in series or parallel, we can calculate the current capacity of all cell monomers one by one according to the above steps S1-S3, and take the capacity of the cell with the smallest current capacity as the overall capacity of the battery pack, so as to calculate the current health status of the battery pack:
[0069]
[0070] Among them, C 额 is the rated capacity of the battery pack.
[0071] In order to verify the feasibility of the estimation method of the present invention, we conducted the following experiments:
[0072] The object to be estimated is the battery pack in the Tengfei Building Energy Storage Power Station, which contains 240 batteries in series. Battery No. 1 is used as an example to explain. The battery in the power station is a lithium iron phosphate battery with a rated capacity of 180Ah. Based on the initial battery benchmark data stored in the power station, the benchmark SOC-V curve is fitted, such as Figure 4 As shown, the benchmark curvature curve is obtained according to the benchmark SOC-V curve, and then the dQ / dV value is obtained according to the capacity and corresponding voltage in the benchmark data, and then the IC curve is obtained. Finally, according to the two benchmark curves, the first characteristic value, that is, the front inflection point SOC of the benchmark curvature curve, is 95.41%, and the second characteristic value, that is, the second peak point SOC of the IC curve, is 50.3%. Figure 4 As shown, the corresponding difference can be obtained by calculation, namely
[0073] ΔSOC*=|SOC 拐* -SOC IC* |=|95.41%-50.3%|=0.4511
[0074] At this time, the current is 0.25C, i.e. 45A, and the temperature is 25°C. According to the established lookup table, the voltage of the front inflection point is extracted to be 3.28V.
[0075] Extract the battery discharge data of a certain time in the power station BMS system, extract the data every 1s, the data includes the voltage and the corresponding discharge amount Q, find V 拐 =3.28V position corresponds to a discharge capacity of 8.06Ah, and then the IC curve of the discharge data is obtained. According to the curve, the discharge capacity corresponding to the second peak point of the IC curve is 87.27Ah. At this time, it can be obtained
[0076] C t =|C 拐 -C IC |=|8.06-87.27|=79.21Ah
[0077] Substituting into the formula, we can calculate C′=Q t / ΔSOC*=79.21Ah / 0.4511=175.59Ah.
[0078] To verify the accuracy, the battery was calibrated using standard operating conditions, and the result was 174.82Ah with an error of less than 0.5%.
[0079] Use the same method to calculate the capacity of the remaining monomers and obtain the current capacity C' of the minimum monomer minThe error between the estimated current capacity of all cells and the accurate value is about 2.2% at most, which is the No. 133 battery. Figure 5 As shown, the estimation errors of the remaining battery cells are all within 1.5%.
[0080] The current health status SOH of the battery pack at this time:
[0081] Although specific embodiments of the present invention are described above, those skilled in the art should understand that these are merely examples and that various changes or modifications may be made to these embodiments without departing from the principles and essence of the present invention. Therefore, the scope of protection of the present invention is limited by the appended claims.
Claims
1. A method for estimating the health status of a lithium iron phosphate battery, characterized in that The following steps are involved: Step 1: Based on the charge and discharge data provided by the battery manufacturer, obtain the benchmark SOC-V curve and benchmark IC curve by data fitting; Step 2: Calculate the SOC value corresponding to the maximum curvature in the reference SOC-V curve, and record the corresponding point as the front inflection point, which is located in the 80% SOC-100% SOC range. At the same time, calculate the SOC value corresponding to the second peak of the reference IC curve, and record the corresponding point as the second peak point. Calculate the difference between the two SOC values, and record it as ΔSOC*. Step 3: According to the current charge and discharge data of the battery, calculate the corresponding front inflection point and the second peak point, as well as the capacity difference between the front inflection point and the second peak point, recorded as Q t , and then calculate the current battery capacity C′=Q t / ΔSOC*; Step 4: Calculate the battery's current state of health (SOH) using the following formula: Among them, C 额 Indicates the rated capacity of the battery.
2. The method for estimating the health status of a lithium iron phosphate battery according to claim 1, characterized in that: According to the reference SOC-V curve, the horizontal axis of the reference IC curve, that is, the voltage value, is replaced with the corresponding SOC value, and then the corresponding SOC value is found according to the second peak point of the reference IC curve.
3. The health status estimation method of a lithium iron phosphate battery according to claim 2, characterized in that: According to the charge and discharge data provided by the battery manufacturer, the corresponding benchmark SOC-V curves at different currents and temperatures are obtained by data fitting. The front inflection points corresponding to the maximum curvature in these benchmark SOC-V curves are calculated respectively, and a lookup table of the voltage values corresponding to the front inflection points at different currents and temperatures is made. In step 3, according to the current charge and discharge data of the battery and the corresponding working conditions, the corresponding voltage value is found from the query table and recorded as V 拐 , use data fitting to obtain the IC curve, find the voltage value corresponding to the second peak point of the IC curve, and record it as V IC , according to the battery management system, find out whether the current voltage of the battery reaches voltage V 拐 and voltage V IC The time points corresponding to the 拐 and t IC , use the following formula to calculate the power difference Q using the ampere-hour integration method t ; Or according to the battery management system, find out whether the current voltage of the battery reaches the voltage V 拐 and voltage V IC The corresponding power values are recorded as Q 拐 and Q IC , use the following formula to calculate the power difference Q using the direct power method t ; Q t =|Q 拐 -Q IC |。 4. The method for estimating the health status of a lithium iron phosphate battery according to claim 3, characterized in that: When the collected battery current charge and discharge data is continuous and includes data corresponding to the front inflection point and the second peak point, the ampere-hour integration method or the direct power method is used to calculate the power difference Q t ; When the collected battery current charge and discharge data is discontinuous and includes data corresponding to the previous inflection point or the second peak point, the direct power method is used to calculate the power difference Q t .
5. The method for estimating the health status of a lithium iron phosphate battery according to claim 1, characterized in that: Use steps 1 to 3 to calculate the power difference Q corresponding to each cell in the battery pack t , and then calculate the current single battery capacity C' of each battery cell, and take the minimum value of the current single battery capacity to calculate the current health state SOH of the entire battery pack. The specific calculation formula is as follows 6. The health status estimation method of a lithium iron phosphate battery according to claim 1, characterized in that: When the step calculates the SOC value corresponding to the front inflection point, the curvature corresponding to each point on the SOC-V curve is first calculated, and the curvature change curve, that is, the relationship between the curvature and the voltage is fitted, and then the horizontal coordinate of the curvature change curve, that is, the voltage value, is replaced with the SOC value according to the SOC-V curve, the maximum curvature at this time is obtained, and the corresponding SOC value is found. The point corresponding to the SOC value on the SOC-V curve is the front inflection point.
Citation Information
Cited By
Lithium iron phosphate battery capacity change prediction method and device, computer equipment, storage medium and program product
CN121254116A
Methods, devices, computer equipment, storage media, and software products for predicting capacity changes in lithium iron phosphate batteries.
CN121254116B
