Online evaluation method for health degree of battery

Abstract

The invention discloses a battery health degree on-line evaluation method, which specifically comprises the following steps of: according to battery parameter information, obtaining a conversion coefficient between SOC (State of Charge) and open-circuit voltage by adopting linear fitting; calculating the SOH natural attenuation coefficient of the battery based on the conversion coefficient; and in combination with the conversion coefficient and the SOH natural attenuation coefficient, iterative correction is adopted to obtain the real SOH of the battery. The accuracy of battery performance evaluation can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of battery detection, and in particular to an online battery health assessment method. Background Art

[0002] In actual operation, lithium iron phosphate batteries are affected by many factors such as temperature, number of cycles, depth of discharge, discharge rate, capacity attenuation, battery aging, etc. Accurate identification of various state parameters of lithium batteries is a necessary condition for the safe and reliable operation of battery systems. The premise for achieving these functions is to accurately identify battery parameters and establish a high-precision battery model, which is also a key technology that needs to be improved in battery management systems. The terminal voltage, current, temperature and other parameters of lithium iron phosphate power batteries can be directly measured by sensors, but the open circuit voltage, internal resistance, state of charge (SOC) and other parameters of power batteries are constantly changing and cannot be directly measured. The relevant parameters or the relationship between relevant parameters and state of charge (SOC) must be measured in advance through experiments. This kind of test and maintenance operation is relatively complicated. In a laboratory environment, although the identification of SOH (state of health) is relatively accurate, the feasibility is low, the equipment requirements are high, and the real-time performance is poor.

[0003] At the same time, traditional SOH evaluation methods are usually established under laboratory conditions. The disadvantage is that the experimental environment is fixed and the battery's SOH status cannot be obtained in real time. However, in actual use, the external environment of the battery changes unstably and cannot be tested under laboratory conditions. At the same time, the data collected by the sensor will also have certain deviations.

[0004] Therefore, it is urgent to propose an online battery health assessment method to solve the above problems. Summary of the invention

[0005] The purpose of the present invention is to provide a battery health online evaluation method, which can improve the accuracy of battery performance evaluation.

[0006] In order to solve the above technical problems, the present invention provides a battery health online evaluation method, which specifically includes the following:

[0007] According to the battery parameter information, linear fitting is used to obtain the conversion coefficient between SOC and open circuit voltage;

[0008] Calculate the SOH natural decay coefficient of the battery based on the conversion coefficient;

[0009] The conversion coefficient and the SOH natural attenuation coefficient are combined and iterative correction is used to obtain the real SOH of the battery.

[0010] Furthermore, before obtaining the conversion coefficient, the method further includes: evaluating SOH by using battery capacity attenuation, wherein the first formula is: C aged is the existing capacity, C rated is the rated maximum capacity.

[0011] Furthermore, the calculation of the SOH specifically includes:

[0012] During any discharge process, monitor the battery from high load SOC h Discharge to low load SOC l process;

[0013] Integrate the power of the discharge process to obtain the consumed electricity Q;

[0014] Calculate the battery's SOH according to the second formula: Where, ΔSOC = SOC h -SOC l .

[0015] Furthermore, the acquisition of the conversion coefficient includes:

[0016] A linear relationship between SOC and open circuit voltage is established, such as the third formula: SOC = a × V + b; where a is the slope coefficient of the linear approximation, b is the intercept, a and b are both conversion coefficients, and V is the open circuit voltage obtained by sampling;

[0017] According to the linear relationship, the fourth formula is obtained: ΔSOC=a×ΔV;

[0018] Substituting the fourth formula into the second formula: The fifth formula is obtained to calculate the SOH of each battery charge and discharge: Among them, C rated is the rated maximum capacity, ΔV=V h -V l , V h For high voltage value, V l is the low voltage value, and Q is the power consumed.

[0019] Furthermore, the acquisition of the conversion coefficient also includes:

[0020] Obtain the battery capacity change value corresponding to each battery SOC percentage change;

[0021] In actual battery operation, the power integration method is used to calculate the power change information;

[0022] Synchronous sampling to obtain the corresponding open circuit voltage change data during the battery power change process;

[0023] According to the relationship between the battery power change and SOC, and the relationship between the battery power change and the open circuit voltage, it is converted into a relationship between SOC and open circuit voltage.

[0024] Furthermore, the acquisition of the conversion coefficient also includes:

[0025] Repeat the SOC and open circuit voltage relationship conversion process in multiple charge and discharge cycles;

[0026] Mean filtering is used to obtain a stable SOC-open circuit voltage numerical correspondence;

[0027] The values ​​of the slope coefficient a and the intercept b are obtained by linear fitting.

[0028] Furthermore, the calculation of the natural attenuation coefficient includes:

[0029] Establish a historical database based on the SOH calculated for each charge and discharge cycle;

[0030] The SOH ratio of two adjacent cycles is calculated as the natural attenuation coefficient, as shown in the sixth formula: Among them SOH k is the battery SOH of the kth charge and discharge cycle, SOH k-1 Battery SOH at the k-1th battery charge and discharge cycle;

[0031] Mean filtering is used to obtain a stable natural attenuation coefficient.

[0032] Further, the iterative correction includes:

[0033] The SOH after the kth charge and discharge cycle is defined as the seventh formula:

[0034] SOH best =A k ×SOH k +B k ×SOH' k Among them, SOH k SOH' is the value obtained by multiplying the SOH calculated in k-1 charge and discharge cycles by the battery's SOH natural attenuation coefficient e. k is the value obtained by calculating the integral of the k-th operating power, A and B are proportional coefficients, and A+B=1;

[0035] The error is corrected by the proportional coefficient A and the proportional coefficient B, as shown in the eighth and ninth formulas below:

[0036] e best =e×A best

[0037] a best =a×B best

[0038] Among them, e bestis the optimal SOH natural attenuation coefficient, a best is the optimal SOC-open circuit voltage proportional coefficient.

[0039] Furthermore, the iterative correction also includes:

[0040] When the SOH at the k-1th moment is available, the SOH value at the kth time is calculated as follows:

[0041] SOH k =e×SOH k-1

[0042] Where, e is the natural attenuation coefficient;

[0043] Update the SOH variance, which is the following eleventh formula:

[0044] P k =e 2 ×P k-1 +Qq

[0045] Among them, P k is the fluctuation variance of the SOH obtained in history within a predetermined range, and q is the white noise data.

[0046] Furthermore, the iterative correction further includes: according to the power integral and the fifth formula: Calculate the kth SOH' k , and define the actual SOH k and calculated value SOH' k The relationship between; such as the twelfth formula:

[0047] SOH k =SOH k +U

[0048] Among them, U is the numerical fluctuation when the power is used to calculate the SOH in actual battery operation.

[0049] Furthermore, the iterative correction also includes:

[0050] Calculate the difference between the power integral SOH' and the natural attenuation SOH, as shown in the thirteenth formula: k =SOH k '-SOH k ;

[0051] Calculate the variance update base, as shown in the fourteenth formula: Where P is the fluctuation variance calculated by natural attenuation, and R is the fluctuation variance calculated by power integration.

[0052] Furthermore, the iterative correction also includes:

[0053] Calculate the weight distribution coefficient, as shown in the fifteenth formula:

[0054] Update the optimal SOH, as in the sixteenth formula: SOH best =SOH k +T×y k ;

[0055] Update the natural attenuation variance, as shown in Formula 17: P k+1 =P k ×(1-T);

[0056] Add the new power integration result to the historical data to update the R value.

[0057] Through the above technical solution, the present invention has the following beneficial effects:

[0058] The present invention can improve the accuracy of battery performance evaluation by using linear fitting to obtain the conversion coefficient between SOC and open circuit voltage according to battery parameter information; calculating the battery's SOH natural attenuation coefficient based on the conversion coefficient; and combining the conversion coefficient and the SOH natural attenuation coefficient to obtain the battery's true SOH through iterative correction. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 Flow chart of a method for online battery health assessment in one embodiment of the present invention;

[0060] Figure 2 It is a line chart of the first group of test data in the online battery health assessment method in one embodiment of the present invention;

[0061] Figure 3 It is a line chart of the second group of test data in the online battery health assessment method in one embodiment of the present invention;

[0062] Figure 4 It is a line chart of the third group of test data in the online battery health assessment method in one embodiment of the present invention;

[0063] Figure 5 It is a line chart of the fourth group of test data in the online battery health assessment method in one embodiment of the present invention;

[0064] Figure 6 It is a line chart of the fifth group of test data in the online battery health assessment method in one embodiment of the present invention;

[0065] Figure 7 It is a line chart of the sixth group of test data in the online battery health assessment method in one embodiment of the present invention;

[0066] Figure 8This is a graph showing the results of the test program console in the online battery health evaluation method according to an embodiment of the present invention;

[0067] Fig. 9 A line graph of the first set of test data results in the online battery health assessment method according to an embodiment of the present invention;

[0068] Fig.10 A line chart of the second set of test data results in the online battery health assessment method according to an embodiment of the present invention;

[0069] Fig.11 A line chart of the third set of test data results in the online battery health assessment method according to an embodiment of the present invention;

[0070] Fig.12 A line chart of the fourth group of test data results in the online battery health assessment method according to an embodiment of the present invention;

[0071] Fig.13 A line graph of the fifth set of test data results in the online battery health assessment method according to an embodiment of the present invention;

[0072] Fig.14 It is a line chart of the sixth set of test data results in the online battery health assessment method in one embodiment of the present invention. DETAILED DESCRIPTION

[0073] A battery health online evaluation method of the present invention will be described in more detail below in conjunction with the accompanying drawings, wherein preferred embodiments of the present invention are shown. It should be understood that those skilled in the art may modify the present invention described herein while still achieving the beneficial effects of the present invention. Therefore, the following description should be understood as being widely known to those skilled in the art and not as a limitation of the present invention.

[0074] The present invention is described in more detail in the following paragraphs by way of example with reference to the accompanying drawings. The advantages and features of the present invention will become more apparent from the following description. It should be noted that the accompanying drawings are in very simplified form and are not in exact proportions, and are only used to facilitate and clearly assist in illustrating the purpose of the embodiments of the present invention.

[0075] like Figure 1-Figure 2 As shown, the embodiment of the present invention proposes a battery health online evaluation method, which specifically includes the following steps:

[0076] S1. According to the battery parameter information, a linear fitting is used to obtain the conversion coefficient between SOC and open circuit voltage;

[0077] S2. Calculating the SOH natural attenuation coefficient of the battery based on the conversion coefficient;

[0078] S3. Combining the conversion coefficient and the SOH natural attenuation coefficient, an iterative correction is used to obtain the real SOH of the battery.

[0079] Before step S1, before the conversion coefficient is obtained, step S0 is also included: Generally, there are many evaluation indicators of SOH. For example, for 32650 batteries, since the rated data of 32650 batteries are calibrated, the battery capacity attenuation can be used to evaluate SOH, wherein the first formula is:

[0080]

[0081] Among them, C aged is the existing capacity, C rated is the rated maximum capacity.

[0082] In this embodiment, the calculation of the SOH specifically includes the following steps:

[0083] S001. During any discharge process, monitor the battery from high load SOC h Discharge to low load SOC l process;

[0084] S002, integrating the power of the discharge process to obtain the consumed power Q;

[0085] S003. Calculate the battery SOH according to the second formula:

[0086]

[0087] Where, ΔSOC = SOC h -SOC l , Q can be obtained by power integration.

[0088] In order to solve the problem of the conversion coefficient between the open circuit voltage and the SOC, and the problem of the natural attenuation of the SOH, this embodiment solves the problem in two stages.

[0089] In the first stage, namely the initial parameter approximation stage, a linear approximation method is used for SOC and open circuit voltage to obtain an initial conversion coefficient, and the approximate result is used to calculate the natural attenuation coefficient of the battery's SOH.

[0090] In this stage, the rated capacity of the battery, the open circuit voltage obtained by sampling, and the battery operating current data are mainly used to approximate the relationship between the battery's SOC and open circuit voltage. Due to the error between the rated capacity and the actual capacity of the battery, the open circuit voltage and SOC of the battery are not a perfect linear relationship, etc., the SOH attenuation ratio and SOC-open circuit voltage corresponding coefficient calculated in the initial parameter approximation stage are not accurate. However, these approximate information provides the necessary basic information for the accurate calculation of parameters in the subsequent stages. The specific steps can be shown as follows.

[0091] In step S1, the acquisition of the conversion coefficient includes:

[0092] S101, establish a linear relationship between SOC and open circuit voltage, such as the third formula:

[0093] SOC=a×V+b (3)

[0094] Wherein, a is the slope coefficient of the linear approximation, b is the intercept, a and b are both conversion coefficients, and V is the open circuit voltage obtained by sampling.

[0095] According to the linear relationship, the fourth formula is obtained:

[0096] ΔSOC=a×ΔV (4)

[0097] Substituting the fourth formula into the second formula, the fifth formula is obtained to calculate the SOH of each battery charge and discharge:

[0098]

[0099] Among them, C rated is the rated maximum capacity, ΔV=V h -V l , V h For high voltage value, V l is the low voltage value, and Q is the power consumed.

[0100] In step S1, the acquisition of the conversion coefficient further includes the following steps:

[0101] S102, dividing the rated capacity of the battery by 100 to obtain a battery capacity change value corresponding to a percentage change in the SOC of each battery;

[0102] S103, using the power integration method to calculate the power change information in actual battery operation;

[0103] S104, synchronously sampling to obtain corresponding open circuit voltage change data during the battery power change process;

[0104] S105 . Convert the relationship between the battery power change and the SOC and the relationship between the battery power change and the open circuit voltage into a relationship between the SOC and the open circuit voltage.

[0105] Furthermore, in step S1, the acquisition of the conversion coefficient also includes: S106, repeating the conversion process of the relationship between SOC and open circuit voltage in multiple charge and discharge cycles; using mean filtering to obtain a stable SOC-open circuit voltage numerical correspondence; and obtaining the values ​​of the slope coefficient a and the intercept b by linear fitting.

[0106] In step S2, the calculation of the natural attenuation coefficient includes:

[0107] S201, establishing a historical database based on the SOH calculated for each charge and discharge cycle;

[0108] S202, calculating the SOH ratio of two adjacent cycles as the natural attenuation coefficient, such as the sixth formula:

[0109]

[0110] Among them SOH k is the battery SOH of the kth charge and discharge cycle, SOH k-1 Battery SOH at the k-1th battery charge and discharge cycle;

[0111] S203, using mean filtering to obtain a stable natural attenuation coefficient.

[0112] In summary, this embodiment solves the problem of the conversion coefficient between open circuit voltage and SOC, as well as the natural attenuation problem of SOH through the following steps. Specifically: Step 1. At this stage, SOC and open circuit voltage are linearly fitted, that is, formula (3); wherein the proportional coefficient a is the slope of the linear approximation, b is the intercept, and V is the sampled open circuit voltage. At the same time, according to the factory data of the battery, it can be calculated that when the battery is at the maximum output voltage, the SOC is 100%. Step 2. Since the rated capacity of the battery is known, the rated capacity of the battery is divided by 100 to obtain the battery capacity change value when the SOC percentage of each battery changes. Based on these data, the relationship between battery power change and SOC can be calculated. Step 3. In the actual operation of the battery, the power integration method is used to calculate the change information of the power, and the corresponding open circuit voltage change data is sampled to obtain the battery power change process. Based on these data, the relationship between battery power change and open circuit voltage is calculated. Step 4: According to the battery power change-SOC relationship obtained in step 2 and the battery power change-open circuit voltage relationship obtained in step 3, the relationship is converted into the SOC-open circuit voltage relationship. Step 5: Repeat steps 3 and 4 in multiple battery charge and discharge cycles, use the mean filtering method to obtain a stable SOC-open circuit voltage value change correspondence, and substitute it into formula (3) to obtain the slope coefficient a and intercept b.

[0113] Through steps 1 to 5, we can get relatively stable values ​​of slope coefficient a and intercept b. Since we only need the difference of SOC in the determination formula (2) of SOH, we can get formula (4), where a in formula (4) is the slope coefficient in formula (3), which can be calculated in step 5. Then, substitute formula (4) into formula (2) to get: formula (5), where C rated is the rated maximum capacity, which can be obtained from the factory instructions, and a is the slope coefficient in formula (3). These two parameters are known. In any discharge process, the battery is discharged from a high voltage V h Discharge operation to low load V l During the operation, the power of the battery during operation is integrated to obtain the power consumption data, that is, the consumed power Q, and ΔV = V h -V l Then, the SOH of each battery charge and discharge can be calculated using formula (5).

[0114] Next, the natural attenuation rate of the battery's SOH can be calculated based on the results of each battery charge and discharge cycle. Generally speaking, the charge and discharge cycle of a lithium battery is about 5,000 times, and after each battery charge and discharge cycle, part of the battery's material will undergo irreversible chemical changes, resulting in a slight decrease in the battery's maximum capacity. According to formula (1), since the maximum capacity of the battery has decreased, the battery's SOH will also have a slight decrease. Therefore, in order to improve the accuracy of the SOH system's estimation of the battery's SOH, the SOH system will consider the natural attenuation of the battery and comprehensively evaluate the battery's SOH. Specifically, the battery's natural attenuation coefficient e is defined as the ratio of the battery's SOH at the kth charge and discharge cycle to the battery's SOH at the k-1th charge and discharge cycle, that is, formula (6). In the initial parameter approximation stage, based on the SOH calculated for each charge and discharge cycle, an approximate battery natural attenuation coefficient e can be obtained by using the mean filtering method. The entire initial parameter approximation stage lasts about 20 battery charge and discharge cycles.

[0115] In the second stage, i.e. the parameter correction stage, the conversion coefficient and the natural attenuation coefficient of SOH are corrected, and secondary calculations are performed based on actual sampling to continuously approach the true value.

[0116] After about 20 battery charge and discharge cycles, the second stage, the parameter correction stage, will begin. In this stage, we iteratively update the battery SOH in each charge and discharge cycle based on the initial parameters, gradually approaching the true value. The specific method is described as follows.

[0117] First, in step S3, the iterative correction includes:

[0118] The SOH after the kth charge and discharge cycle is defined as the seventh formula:

[0119] SOH best =A k ×SOH k +B k ×SOH' k (7)

[0120] Among them, SOH k SOH' is the value obtained by multiplying the SOH calculated in k-1 charge and discharge cycles by the battery's SOH natural attenuation coefficient e. k is the value obtained by integrating the k-th operating power, A and B are proportional coefficients, and A+B=1.

[0121] The error is corrected by the proportional coefficient A and the proportional coefficient B, as shown in the eighth and ninth formulas below:

[0122] e best=e×A best (8)

[0123] a best =a×B best (9)

[0124] Among them, e best is the optimal SOH natural attenuation coefficient, a best is the optimal SOC-open circuit voltage proportional coefficient.

[0125] In this embodiment, the iterative correction further includes:

[0126] When the SOH at the k-1th moment is available, the SOH value at the kth time is calculated as follows:

[0127] SOH k =e×SOH k-1 (10)

[0128] Where e is the natural attenuation coefficient.

[0129] Update the SOH variance, which is the following eleventh formula:

[0130] P k =e 2 ×P k-1 +Qq (11)

[0131] Among them, P k is the fluctuation variance of the SOH obtained in history within a predetermined range, and q is the white noise data.

[0132] Furthermore, the iterative correction further includes: according to the power integral and the fifth formula: Calculate the kth SOH' k , and define the actual SOH k and calculated value SOH' k The relationship between; such as the twelfth formula:

[0133] SOH k =SOH k +U (12)

[0134] Among them, U is the numerical fluctuation when the power is used to calculate the SOH in actual battery operation.

[0135] Furthermore, the iterative correction also includes:

[0136] Calculate the difference between the power integrated SOH' and the natural attenuation SOH, as shown in the thirteenth formula:

[0137] y k =SOH k '-SOHk (13)

[0138] Calculate the variance update base, as shown in the fourteenth formula:

[0139]

[0140] Where P is the fluctuation variance calculated by natural attenuation, and R is the fluctuation variance calculated by power integration.

[0141] Furthermore, the iterative correction also includes:

[0142] Calculate the weight distribution coefficient, as shown in the fifteenth formula:

[0143]

[0144] Update the optimal SOH, as shown in the sixteenth formula:

[0145] SOH best =SOH k +T×y k (16)

[0146] Update the natural attenuation variance, as shown in the seventeenth formula:

[0147] P k+1 =P k ×(1-T) (17)

[0148] Add the new power integration result to the historical data to update the R value.

[0149] Therefore, in summary, in the second stage, i.e., the parameter correction stage, the following steps can be followed: Step 1, first define the SOH of the battery after the kth charge and discharge cycle of the battery as formula (7). The SOH here contains two parts. The first part is obtained by multiplying the SOH calculated in the k-1th charge and discharge cycle by the natural attenuation coefficient e of the battery SOH, i.e., obtained according to formula (6). The second part SOH' is calculated based on the kth operation power integral, i.e., calculated according to formula (5). Here, A and B are proportional coefficients. Since in formulas (5) and (6), the SOC-open circuit voltage conversion coefficient a and the SOH natural attenuation coefficient e may have errors, the errors can be proportionally corrected by A and B, so that the calculated SOH is closer to the true value, i.e., formula (8) and formula (9). Although there are errors in the e and a estimated approximately in the first stage, these errors can be corrected by A and B, and finally the optimal natural attenuation coefficient e and open circuit voltage-SOC proportional coefficient a are obtained.

[0150] Since this correction coefficient needs to be updated with each charge and discharge cycle iteration, both coefficients have a subscript k, indicating that this coefficient is the result obtained after the kth iteration; and the sum of these two coefficients must be 1. Both the natural attenuation calculation method and the power integration calculation method are an estimate of SOH. The two are redistributed through the weight coefficient, and the actual SOH value can be obtained by adding them together.

[0151] In the first charge and discharge cycle of the parameter correction phase, these two coefficients can be initialized to 0.5. Although any value does not affect the subsequent iterations, a more stable initialization value can reduce the number of iterations. The following is a step-by-step introduction to how these parameters are obtained.

[0152] Step 2: Regardless of the data sampling results, SOH itself has natural attenuation. So if we calculate it according to natural attenuation, if we have the SOH at the k-1th moment, we can infer that the kth SOH should be formula (10). Since there may be some fluctuations in the calculation of SOH, the SOH obtained in history varies within a range, and this fluctuating data has a mean and variance. The initial variance can be obtained in the first stage. In each charge and discharge cycle, the SOH can be calculated. The result of the mean filter is the SOH we use here. By fitting the distribution of all samples, we can obtain the initial variance P of SOH. It must be recognized that after SOH is multiplied by the attenuation coefficient e, the variance of SOH needs to be multiplied by the square of e, that is, formula (11) and Q in formula (11) is white noise data.

[0153] Step 3: If the natural attenuation of SOH is not considered, the power integral data is used for each battery SOH, and it is calculated by formula (5), then SOH has the relationship as shown in formula (12); and SOH' in formula (12) is the SOH calculated by formula (5), and the SOH here is the actual SOH. Because when the power is used to calculate SOH in the actual operation of the battery, there will be a certain fluctuation in the value. This fluctuation is recorded as U, the mean of U is 0, and the variance of U is R. Similar to the calculation of the variance P in the natural attenuation calculation method, the initial value of the variance R can be calculated in the parameter approximation stage.

[0154] Step 4. Combining Step 2 and Step 3, it can be seen that in each charge and discharge cycle, the SOH' calculated by power integration and the SOH calculated by SOH natural decay can be obtained. Due to data fluctuations between the two, or inaccurate SOC-open circuit voltage coefficient, natural decay coefficient, etc., there will be a difference between the power-integrated SOH' and the natural decay SOH. The size of this difference y is shown in formula (13).

[0155] Step 5. Combining steps 2 and 3, it can be seen that whether it is the SOH calculated by natural attenuation or the SOH' calculated by power integration, these two values ​​fluctuate within a range, and the fluctuation variances are P and R respectively. Here, the variance update base S is defined, and the calculation formula is formula (14).

[0156] Step 6: Variance update base S can be used to evaluate the optimal SOH best The weight distribution of SOH and SOH' in the estimation. The calculation formula of the distribution coefficient T is as shown in formula (15).

[0157] Step 7: According to the distribution coefficient T, the optimal SOH can be calculated best Estimate, as shown in formula (16).

[0158] Among them, formula (16) and formula (7) are completely equivalent, and the relationship between A, B and T can be obtained through mathematical conversion.

[0159] Step 8: In this step, the two variance parameters are iteratively updated according to the SOH calculation results. The update formula of P is as shown in formula (17); in addition, R can be iteratively updated by substituting the k-th power integration result into all historical power integration data.

[0160] Through the SOH evaluation process of this embodiment, it is possible to allow a fixed deviation in the battery data sampling circuit. In addition, the deviation ratio can be updated through each measurement calculation. In the case of inherent deviation, the true SOH can still be accurately inferred, thereby being able to well meet the accuracy requirements of online SOH calculation.

[0161] In this embodiment, for example, a 5Ah lithium iron phosphate battery is selected for testing, and the data samples are divided into two columns. The first column is the sampled voltage U, in V (volts), and the second column is the sampled current I, in A (amperes). There are 6611 sets of U and I data in the test data, and 601 sets of data are collected for each round of charge and discharge, so there are a total of 11 rounds of charge and discharge data. In this test, 3 rounds of charge and discharge data were used in the initialization parameter stage; 6 rounds of charge and discharge data were used in the operation stage.

[0162] The test data includes data without noise and data with noise, with a total of 6 test data sets. To make the display more intuitive, this embodiment visualizes the data sets, which are Figure 2-Figure 7 From the data visualization results, it can be clearly seen that the test data has obvious periodicity, and the charge and discharge cycle is relatively obvious, which is consistent with this test. The test results are as follows: The test data is imported into the test program respectively, and the results of each PC test program run are as follows: Figure 8As shown. The running time is 1s, which is fast. Figure 8 As you can see, in the initialization phase, the program calculates the initial SOH parameters, and after the parameters are processed, the program calculates the optimal estimated SOH. best , and called the police.

[0163] Then, for groups 1-6 of data, the program runs the SOH change results and displays them, such as Figure 9-14 As shown in the test results of Table 1 below, it can be seen from the results that with the increase in the number of charge and discharge cycles, SOH has a trend of gradually decreasing, which is in line with the actual situation of the battery, and this also directly proves the feasibility and correctness of the algorithm program.

[0164] Table 1 Test results

[0165]

[0166]

[0167] In summary, the online battery health assessment method proposed in the present invention has the following advantages:

[0168] The present invention can improve the accuracy of battery performance evaluation by using linear fitting to obtain the conversion coefficient between SOC and open circuit voltage according to battery parameter information; calculating the battery's SOH natural attenuation coefficient based on the conversion coefficient; and combining the conversion coefficient and the SOH natural attenuation coefficient to obtain the battery's true SOH through iterative correction.

[0169] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A battery health online evaluation method, characterized in that: The details include: According to the battery parameter information, linear fitting is used to obtain the conversion coefficient between SOC and open circuit voltage; Calculate the SOH natural attenuation coefficient of the battery based on the conversion coefficient; The conversion coefficient and the SOH natural attenuation coefficient are combined and iterative correction is used to obtain the real SOH of the battery.

2. The method for online battery health assessment according to claim 1, characterized in that: Before obtaining the conversion coefficient, the method further includes: evaluating SOH by using battery capacity attenuation, wherein the first formula is: C aged is the existing capacity, C rated is the rated maximum capacity.

3. The method for online battery health assessment according to claim 2, characterized in that: The calculation of the SOH specifically includes: During any discharge process, monitor the battery from high load SOC h Discharge to low load SOC l process; Integrate the power of the discharge process to obtain the consumed electricity Q; Calculate the battery's SOH according to the second formula: Where, ΔSOC = SOC h -SOC l .

4. The method for online battery health assessment according to claim 1, wherein: The acquisition of the conversion coefficient includes: A linear relationship between SOC and open circuit voltage is established, such as the third formula: SOC = a × V + b; where a is the slope coefficient of the linear approximation, b is the intercept, a and b are both conversion coefficients, and V is the open circuit voltage obtained by sampling; According to the linear relationship, the fourth formula is obtained: ΔSOC=a×ΔV; Substituting the fourth formula into the second formula: The fifth formula is obtained to calculate the SOH of each battery charge and discharge: Among them, C rated is the rated maximum capacity, ΔV=V h -V l , V h For high voltage value, V l is the low voltage value, and Q is the power consumed.

5. The method for online battery health assessment according to claim 4, characterized in that: The acquisition of the conversion coefficient also includes: Obtain the battery capacity change value corresponding to each battery SOC percentage change; In actual battery operation, the power integration method is used to calculate the power change information; Synchronous sampling to obtain the corresponding open circuit voltage change data during the battery power change process; According to the relationship between the battery power change and SOC, and the relationship between the battery power change and the open circuit voltage, it is converted into a relationship between SOC and open circuit voltage.

6. The method for online battery health assessment according to claim 5, characterized in that: The acquisition of the conversion coefficient also includes: Repeat the SOC and open circuit voltage relationship conversion process in multiple charge and discharge cycles; Mean filtering is used to obtain a stable SOC-open circuit voltage numerical correspondence; The values ​​of the slope coefficient a and the intercept b are obtained by linear fitting.

7. The method for online battery health assessment according to claim 1, characterized in that: The calculation of the natural attenuation coefficient includes: Establish a historical database based on the SOH calculated for each charge and discharge cycle; The SOH ratio of two adjacent cycles is calculated as the natural attenuation coefficient, as shown in the sixth formula: Among them SOH k is the battery SOH of the kth charge and discharge cycle, SOH k-1 Battery SOH at the k-1th battery charge and discharge cycle; Mean filtering is used to obtain a stable natural attenuation coefficient.

8. The method for online battery health assessment according to claim 1, wherein: The iterative correction includes: The SOH after the kth charge and discharge cycle is defined as the seventh formula: SOH best =A k ×SOH k +B k ×SOH' k Among them, SOH k SOH' is the value obtained by multiplying the SOH calculated in k-1 charge and discharge cycles by the battery's SOH natural attenuation coefficient e. k is the value obtained by calculating the integral of the k-th operating power, A and B are proportional coefficients, and A+B=1; The error is corrected by the proportional coefficient A and the proportional coefficient B, as shown in the eighth and ninth formulas below: and best =e×A best a best =a×B best Among them, e best is the optimal SOH natural attenuation coefficient, a best is the optimal SOC-open circuit voltage proportional coefficient.

9. The method for online battery health assessment according to claim 8, characterized in that: The iterative correction also includes: When the SOH at the k-1th moment is available, the SOH value at the kth time is calculated as follows: SOH k =e×SOH k-1 Where, e is the natural attenuation coefficient; Update the SOH variance, which is the following eleventh formula: P k =e 2 ×P k-1 +Qq Among them, P k is the fluctuation variance of the SOH obtained in history within a predetermined range, and q is the white noise data.

10. The method for online battery health assessment according to claim 9, characterized in that: The iterative correction further comprises: according to the power integral and the fifth formula: Calculate the kth SOH' k , and define the actual SOH k and calculated value SOH' k The relationship between; such as the twelfth formula: SOH k '=SOH k +U Among them, U is the numerical fluctuation when the power is used to calculate the SOH in actual battery operation.

11. The method for online battery health assessment according to claim 10, characterized in that: The iterative correction also includes: Calculate the difference between the power integrated SOH' and the natural attenuation SOH, as shown in the thirteenth formula: y k =SOH k '-SOH k ; Calculate the variance update base, as shown in the fourteenth formula: Where P is the fluctuation variance calculated by natural attenuation, and R is the fluctuation variance calculated by power integration.

12. The method for online battery health assessment according to claim 11, characterized in that: The iterative correction also includes: Calculate the weight distribution coefficient, as shown in the fifteenth formula: Update the optimal SOH, as in the sixteenth formula: SOH best =SOH k +T×y k ; Update the natural attenuation variance, as shown in Formula 17: P k+1 =P k ×(1-T); Add the new power integration result to the historical data to update the R value.

Images