SOH diagnosis method using instantaneous resistance extraction during real-time battery use

Abstract

An SOH diagnosis method using instantaneous resistance extraction during real-time battery use according to the present invention comprises the steps of: monitoring a voltage and a current over time during real-time battery use; selecting at least one section for instantaneous resistance measurement from a result of the monitoring; measuring an instantaneous resistance by using the voltage and the current of the selected section; and diagnosing an SOH by dividing an initial instantaneous resistance by the measured instantaneous resistance.

Description

SOH diagnosis method using instantaneous resistance extraction during real-time battery use

[0001] The present invention relates to a method for diagnosing SOH using instantaneous resistance extraction during real-time battery use, and more specifically, to a method for diagnosing SOH by extracting instantaneous resistance from monitoring results of the actual use of an automobile or an ESS (Energy Storage System).

[0002] Battery State of Health (SOH) is a measure indicating how well a battery's current state is maintained compared to its new state, and it is used as an important indicator to evaluate battery life, performance, and safety.

[0003] The SOH of a battery is lowered by the number of cycles of repeated charging and discharging, exposure to high or low temperature environments, improper use such as overcharging or over-discharging.

[0004] If the battery's SOH decreases, not only does the battery's performance decline, but the risk of fire and explosion also increases.

[0005] In particular, electric vehicles require instantaneous high power output; if the battery's State of Health (SOH) drops below a certain level and performance deteriorates, they cannot deliver the output required for safe operation.

[0006] Diagnosing the battery's State of Health (SOH) is crucial for the safe use and efficient recycling management of batteries.

[0007] SOH is determined based on how much capacity remains in the nth cycle relative to the initial cycle capacity, as shown in Equation 1 below. In this case, the capacity is generally measured during the charging / discharging process with a constant current.

[0008]

[0009] [Equation 1]

[0010]

[0011]

[0012] However, measuring SOH using cycle capacity is practically impossible for batteries in use.

[0013] This is because there are almost no instances of the battery being completely discharged or fully charged during actual use.

[0014] Therefore, there is a need for a new method to diagnose SOH even during real-time battery use.

[0015] One objective of the present invention is to provide a method for diagnosing SOH even during real-time battery use.

[0016] Meanwhile, other unspecified objects of the present invention will be further considered to the extent that they can be easily inferred from the following detailed description and effects.

[0017] To solve the problem described above, the following solution is proposed.

[0018] A method for diagnosing SOH using instantaneous resistance extraction during real-time battery use according to an embodiment of the present invention comprises: a step of monitoring voltage and current over time during real-time battery use; a step of selecting at least one interval for measuring instantaneous resistance from the monitoring results; a step of measuring instantaneous resistance using the voltage and current of the selected at least one interval; and a step of diagnosing SOH by dividing the initial instantaneous resistance by the measured instantaneous resistance.

[0019] In one embodiment, the section selected in the step of selecting at least one section for instantaneous resistance measurement from the monitoring result may be a section where the voltage changes rapidly.

[0020] In one embodiment, the section selected in the step of selecting at least one section for measuring instantaneous resistance from the monitoring result may be a section changing from a charged state to a discharged state or a section changing from a discharged state to a charged state.

[0021] In one embodiment, a step of filtering the selected data may be performed in the step of selecting at least one interval for instantaneous resistance measurement from the monitoring result.

[0022] In one embodiment, the filtering step may be performed by excluding data corresponding to a preset current range among the data selected in the step of selecting at least one interval for instantaneous resistance measurement.

[0023] In one embodiment, the determination of the preset current range may include: a step of performing a pulse test while varying the current for the battery; a step of configuring the amount of change in current and the ohmic resistance from the results of the pulse test into a first dataset; and a step of determining the current range containing an outlier of the ohmic resistance among the first dataset as a current range to be excluded when filtering.

[0024] In one embodiment, the filtering step may be performed by selecting data corresponding to a preset voltage range among the data selected in the step of selecting at least one interval for instantaneous resistance measurement.

[0025] In one embodiment, the determination of the preset voltage range may include: a step of performing a pulse experiment while varying the current for the battery; a step of configuring the voltage and ohmic resistance from the results of the pulse experiment into a second dataset; a step of plotting the second dataset on a graph and dividing the plotted graph into a maintaining portion where the ohmic resistance is constant regardless of voltage, a decreasing portion where the ohmic resistance decreases according to voltage, and an increasing portion where the ohmic resistance increases according to voltage; and a step of determining the voltage range corresponding to the maintaining portion as the preset voltage range.

[0026] The SOH diagnosis method according to one embodiment of the present invention has the advantage of being able to accurately diagnose SOH in real time because it extracts instantaneous resistance during real-time battery use, and allows for easy and convenient diagnosis of SOH as no special conditions for SOH diagnosis are required as in the past.

[0027] It should be added that even if an effect is not explicitly mentioned herein, the effects described in the following specification and the potential effects expected by the technical features of the present invention are treated as described in the specification of the present invention.

[0028] FIG. 1 is a schematic flowchart of a SOH diagnostic method according to one embodiment of the present invention.

[0029] Figure 2 illustrates conventional methods for measuring the resistance of a battery, including (a) the DCIR (Direct Current Internal Resistance) measurement method and (b) the EIS (Electrochemical Impedance Spectroscopy) measurement method.

[0030] FIG. 3 is a diagram illustrating the process of a battery aging experiment conducted to explain a SOH diagnosis method according to one embodiment of the present invention, and includes (a) a voltage graph, (b) a current graph, and (c) a block diagram.

[0031] Figure 4 illustrates the IV experimental process for determining the current range to be excluded and the voltage range to be selected during preprocessing. During the entire process, the two preceding constant current charging / discharging sections are for measuring the battery capacity, and the next section is the resistance measurement section. The resistance measurement section proceeds from SOC 0% to 100%, and data was collected by conducting a pulse test while increasing the SOC by 5% based on the previously measured capacity.

[0032] Figure 5 shows the change in current and R according to voltage among the pulse experiment results of Figure 4. oIt is a city.

[0033] Figure 6 shows the change in current and R according to voltage among the pulse experiment results of Figure 4. o This is illustrated by excluding data corresponding to the current range to be excluded determined in Figure 5.

[0034] Figure 7 shows the changes in current and voltage during the section where the voltage changes rapidly among the instantaneous resistances of the 433rd WLTP in Battery #1 data, with (a) the result before filtering and (b) the result after filtering. After filtering, representative Ros were extracted from hundreds of Ros extracted from WLTP.

[0035] Figure 8 shows the changes in current and voltage during the section where the voltage changes rapidly among the instantaneous resistances of the 241st WLTP in Battery #2 data, with (a) the result before filtering and (b) the result after filtering. After filtering, representative Ros were extracted from hundreds of Ros extracted from WLTP.

[0036] Figure 9 shows the representative R extracted by performing filtering. o This is the result of measuring the correlation between and the capacity measured in the Q-check part of Fig. 3. The capacity measurement was extracted from the second discharge section of the Q-check in Fig. 3. Representative R o It was taken from the first WLTP C3 among the 16 WLTP C3s included in Drive of Fig. 3.

[0037] Figure 10 shows the SOH calculated from data obtained from (a) Battery #1 and (b) Battery #2. Q and SOH R As, SOH according to the cycle number of WLTP C3 Q and SOH R It represents.

[0038] It should be noted that the attached drawings are provided as examples for reference to help understand the technical concept of the present invention, and the scope of the rights of the present invention is not limited by them.

[0039] Hereinafter, with reference to the drawings, we will examine the configuration of the present invention as guided by various embodiments thereof and the effects derived therefrom. In describing the present invention, detailed descriptions of related known functions are omitted if they are deemed obvious to a person skilled in the art and could unnecessarily obscure the essence of the invention.

[0040] FIG. 1 is a schematic flowchart of a SOH diagnostic method according to one embodiment of the present invention.

[0041] Hereinafter, a method for diagnosing SOH according to an embodiment of the present invention will be described with reference to the drawings.

[0042] When a battery is used, discharge and charging form a cycle. A battery SOH diagnosis method according to one embodiment of the present invention calculates the instantaneous resistance of the corresponding cycle using the current and voltage of each cycle, and diagnoses the SOH using the calculated instantaneous resistance.

[0043] More specifically, a method for diagnosing SOH of a battery according to one embodiment of the present invention includes the steps of monitoring voltage and current over time during real-time battery use, selecting at least one interval for measuring instantaneous resistance from the monitoring results, measuring instantaneous resistance using the voltage and current of the selected interval, and diagnosing SOH by dividing the initial instantaneous resistance by the measured instantaneous resistance. It may further include the step of removing noise through filtering before diagnosing SOH.

[0044] The step of monitoring voltage and current over time during real-time battery use may utilize a Battery Management System (BMS). However, the present invention is not limited thereto, and it is also possible to use other devices.

[0045] When the voltage and current are monitored in real-time during battery use and the results are collected, the battery management system or a separate SOH diagnostic system performs the process to diagnose SOH.

[0046] First, a step is performed to select at least one interval for measuring instantaneous resistance from the monitoring results.

[0047] As shown in Figure 2, either the DCIR measurement method or the EIS measurement method can be used to measure the resistance of the battery.

[0048] DCIR measures resistance by utilizing the degree of voltage change over time when a direct current is applied, whereas EIS measures impedance according to frequency by applying a very small AC voltage or current. DCIR has the advantage of being easy to measure and does not require significant cost for setting up the experimental environment, and to measure resistance, the change in voltage is calculated by dividing the change in current as shown in Equation 2 below.

[0049]

[0050] [Equation 2]

[0051]

[0052]

[0053] Here, I1 represents the current previously applied to the battery, and V1 represents the voltage previously possessed by the battery. I2 represents the current applied to the battery, and V2 represents the voltage possessed by the battery after the current is applied. DCIR is the resistance value R calculated through Equation 2. o(Ohmic resistance), R ct (Charge transfer resistance), R p It is difficult to separate them because the (polarization resistance) appears superimposed. For example, the resistance appearing 5 seconds after applying a DC current is R o and R ct A mixed resistance value appears. Conversely, for EIS, R depends on the frequency o , R ct , R p Although values ​​can be viewed separately, using EIS requires high-precision equipment and necessitates keeping the battery in an OCV state for an extended period to bring its internal state into equilibrium. Furthermore, both DCIR and EIS require the insertion of an additional step at a specific SOC during battery testing to enable measurement. Due to these drawbacks, DCIR and EIS are not suitable resistance measurement methods for use in environments where batteries are operating.

[0054] A SOH diagnostic method according to one embodiment of the present invention measures instantaneous resistance using a method based on DCIR (Direct Current Internal Resistance). At this time, instantaneous resistance is not measured in all sections, but rather sections satisfying specific conditions are selected. In the SOH diagnostic method of one embodiment of the present invention, the section selected for instantaneous resistance measurement refers to a section (P in FIG. 2) where the voltage changes rapidly over time to the point of being close to vertical. More specifically, a section changing from a charged state to a discharged state, or from a discharged state to a charged state, may be the target section for measuring instantaneous resistance. Systematically, it is also possible to set a reference value for the rate of change of voltage over time and automatically extract sections exceeding the reference value as sections selected for instantaneous resistance measurement.

[0055] After selecting at least one interval for measuring instantaneous resistance, the instantaneous resistance is measured using the voltage and current of the selected interval. The instantaneous resistance in the selected interval refers to the ohmic resistance. The instantaneous resistance can be measured using Equation 2. Alternatively, the representative instantaneous resistance can be measured by calculating the slope of the voltage and current datasets of the selected intervals. The Least Squares Method (LSM) can be used to calculate the slope.

[0056] SOH can be diagnosed by dividing the representative instantaneous resistance value at a specific point in time by the initial representative instantaneous resistance value.

[0057]

[0058] [Equation 3]

[0059]

[0060]

[0061] Meanwhile, for more accurate SOH diagnosis, filtering can be performed before measuring instantaneous resistance. Filtering is performed by excluding data corresponding to a preset current range and selecting data corresponding to a preset voltage range from the selected data during the step of selecting at least one interval for instantaneous resistance measurement.

[0062] The important thing here is to determine the current range to exclude and the voltage range to select for filtering.

[0063] The current range to be excluded and the voltage range to be selected during filtering can be determined in advance for each battery. More specifically, if there is a specific battery product (e.g., Samsung SDI 41J), the current range to be excluded and the voltage range to be selected during filtering applicable to that battery product can be determined in advance.

[0064] The method for determining the current range to exclude and the voltage range to select is as follows.

[0065] First, a pulse test is performed on the battery while varying the current. For example, the pulse test is performed while varying the current to C-rates of 0.01C, 0.05C, 0.1C, 0.2C, 0.3C, 0.4C, and 0.5C. Using the measurement results and Equation 2, the ohmic resistance (R o Measures ).

[0066] When the change in current and ohmic resistance are plotted as a dataset, data containing a significant amount of noise is also plotted. To define data containing a significant amount of noise, the standard deviation of ohmic resistance values ​​within the same change in current was used. More specifically, based on the standard deviation of ohmic resistance values ​​with the largest absolute value of the change in current, groups of change in current that have a standard deviation greater than 10% are considered to contain a significant amount of noise and are filtered out.

[0067] Next, voltage and ohmic resistance are plotted as a dataset. At this stage, the group of current variations considered as noise in the ohmic resistance dataset is excluded. The plotted voltage-ohmic resistance graph exhibits a "stagnant" section where the ohmic resistance value remains constant regardless of voltage, a "decreasing" section where the ohmic resistance decreases with voltage, and / or an "increasing" section where the ohmic resistance increases with voltage. Further filtering is performed here to retain only the data corresponding to the stable section.

[0068] When filtering is performed in this manner, the error is reduced, improving the accuracy of SOH, and there is an advantage that the instantaneous resistance of the filtered data remains constant regardless of the data's position.

[0069]

[0070] Examples

[0071] A battery aging experiment was performed to examine the accuracy of the SOH diagnosis method according to one embodiment of the present invention. A Samsung SDI 41J battery was used for real-time SOH diagnosis. The battery aging experiment was performed in environments of 30 degrees (Battery #1) and 50 degrees (Battery #2).

[0072] FIG. 3 is a diagram illustrating the process of a battery aging experiment conducted to explain a SOH diagnosis method according to an embodiment of the present invention, and includes (a) a voltage graph, (b) a current graph, and (c) a block diagram. In addition, Table 1 below shows detailed information on each step of the battery aging experiment.

[0073]

[0074] [Table 1]

[0075]

[0076]

[0077] The battery aging experiment consists of four parts.

[0078] First, the battery was maintained in the OCV state for 3 hours, and then the Capacity Check (Q check) interval was set. The Q check consists of 3 charge and 2 discharge cycles. The charge was conducted in CC-CV mode with a current of 0.5C, a voltage of 4.2V, and a cut-off current of 0.05C; the final charge, charge*, was charged to only 50% of the SOC in CC mode. The discharge was conducted in CC mode with a current of 0.5C and a cut-off voltage of 3V, and the capacity was extracted during the second discharge. The reference for the SOC in each test part is the capacity extracted during the Q check interval. By calculating the capacity from the Q check, the SOH Q After calculating, the SOH of the present invention RIt was used for verification. This will be discussed later.

[0079] Finally, the Drive, which is the main objective of the present invention, was performed. The Drive is intended to simulate charging and discharging during the actual use of the battery. The Drive consists of two parts: charge and discharge. To simulate the high-speed charging conditions of a vehicle, charging was performed in CC-CV mode with a current of 1C, a voltage of 4.2V, and a cut-off SOC of 90%. When charging is performed in this manner, the charging ends at a voltage of 4.2V and a current of 1.2A (0.3C). Next, WLTP C3 (Worldwide Harmonized Light Vehicles Test Cycle Class 3) was used as the discharge profile, and the cut-off voltage was set to 2.3V to accelerate battery degradation.

[0080] Prior to SOH diagnosis, the current range to be excluded from filtering and the voltage range to be selected were determined for the battery used in the example.

[0081] Figure 4 illustrates the IV experimental process for selecting pretreatment criteria, and to show in detail how the pulse experiment proceeds, the current and voltage graphs over the same time range were enlarged and inserted as insets in Figures 4(a) and (b). Table 2 shows detailed information on each step.

[0082]

[0083] [Table 2]

[0084]

[0085]

[0086] First, two full charges and two full discharges were performed. The charge was conducted in CC-CV mode with a current of 0.5C, a voltage of 4.15V, and a cut-off current of 0.05C, while the discharge was conducted in CC mode with a current of 0.2C and a cut-off voltage of 3V. After the second discharge was completed, pulse experiments were performed while increasing the battery SOC in 5% increments up to 100%. The pulse experiments were conducted by applying currents of 0.01C, 0.05C, 0.1C, 0.2C, 0.3C, 0.4C, and 0.5C. Sixty data points were accumulated for each current during the pulse experiments. Data were taken from the pulse experiments when the current changed from charge to discharge or vice versa, and R was calculated using Equation 2. o Calculated.

[0087] Figure 5 shows R according to the change in current extracted from the pulse experiment during experiment IV of Figure 4. o As illustrated, (a) all data extracted from the current pulse experiment section of experiment IV in Fig. 4 are plotted as changes in current and resistance, and (b) voltage and resistance are plotted. Table 3 below shows the standard deviation of resistance according to the changes in current. The change in current direction in Table 3 refers to the state immediately after changing from charging (C) to discharging (D), or immediately after changing from discharging to charging.

[0088] Based on the standard deviation of the ohmic resistance values ​​with the largest absolute value of current change, |ΔI| = 1C, groups of current change with a standard deviation greater than 10% were considered to contain a significant amount of noise and were displayed in gray. The remaining standard deviation values ​​of less than 10% were displayed in white. Since groups of current change with an error of 10% or more contain many abnormal resistance values, they were designated as groups of current change to be excluded during filtering. Consequently, after filtering, only data with a current change of |ΔI| = 0.3C or greater were selected. In Figure 5(a), the groups of current change to be excluded are also indicated as shaded areas.

[0089]

[0090] [Table 3]

[0091]

[0092]

[0093] Figure 6 is the result of removing the data corresponding to the shaded area from the data of Figure 5, and is shown as (a) the change in current and resistance and (b) voltage and resistance.

[0094] When comparing the data in Fig. 6(b) with the data in Fig. 5(b), it can be seen that a significant amount of noise has been removed by excluding data corresponding to a specific range of current change using the standard deviation as described above.

[0095] Voltage-R in Fig. 6(b) o In the graph, R is independent of voltage o R depending on the voltage and the maintaining part with a constant value o This decreasing part, R depending on the voltage o An increasing section with increasing values ​​appears. Here, additional filtering is performed to retain only the data corresponding to the voltage range of the maintaining section (shaded area in Fig. 6(b)).

[0096] As a result, through this process, data with an absolute value of |ΔI| = 0.3C for the change in current during filtering of Samsung SDI 41J is primarily excluded, and subsequently, only data with a voltage range of 3.7 to 4.0 V remains.

[0097] Next, to see how performing filtering changes the actual data, the 433rd WLTP C3 from Battery #1 data and the 241st WLTP C3 from Battery #2 data were selected. To verify the performance of the filtering, sets were randomly selected from various WLTP datasets.

[0098] Figure 7 shows the change in current and the change in voltage during the rapid voltage change interval of the 433rd WLTP instantaneous resistance, with (a) the result before filtering and (b) the result after filtering. Also, Figure 8 shows the change in current and the change in voltage during the rapid voltage change interval of the 241st WLTP instantaneous resistance, with (a) the result before filtering and (b) the result after filtering.

[0099] The straight line plotted on the graphs in Figs. 7(a) and 8(a) is the line obtained using LSM, and the slope of the function is the representative R of the corresponding cycle of WLTP C3 according to Equation 2. o It means. R 2 The closer this is to 1, the more the data converges to a straight line. R after filtering 2 R before proceeding 2 You can see that it is closer to 1. In other words, after performing filtering, the error range of the data is reduced, and accuracy is improved.

[0100] Verification of the accuracy improvement due to filtering was also conducted through numerical comparison. In the inset of Fig. 7(a), the points indicated by the left and right arrows, respectively, represent the minimum and maximum values ​​of the resistance, having values ​​of -20.01 mΩ and 51.89 mΩ, and the representative R obtained by LSM in Fig. 7(b) o is 30.10 mΩ. In Fig. 8(a), the points indicated by the left and right arrows are -60.01 mΩ and 37.72 mΩ, respectively, and the representative R obtained by LSM in Fig. 8(b) o is 31.11 mΩ. Figures 7(a) and 8(a), which were not preprocessed, contain a significant amount of noise, so some R o The value is the representative R in Figs. 7(b) and 8(b). o It shows a very large difference in value, and it can be confirmed that negative resistance values, which cannot actually exist, are also observed. In other words, R before and after filtering 2 Even if the values ​​do not show a significant difference, since there is a large amount of noisy data in the dataset before filtering, the noisy data must be removed through appropriate filtering.

[0101] Next, R extracted by performing filtering o Pearson correlation analysis was performed to determine whether it could be used as a diagnostic factor for SOH. Pearson correlation analysis is used to determine how strong a relationship two factors form through correlation coefficients. The results are shown in Figure 9.

[0102] Figure 9 shows the representative R extracted by performing filtering. o This is the result of measuring the correlation between the dose measured in the Q-check part of Fig. 3. To perform correlation analysis, the number of data points for both factors must be identical. Therefore, the representative R used for the correlation analysis is oε is a value extracted from the first WLTP C3 with the least degradation among the 16 WLTP C3 runs in Fig. 3(c), and the capacity was obtained from the second discharge within the Q check interval. Looking at Fig. 9, the correlation coefficient between the two factors is -0.9936, with a representative R o It can be confirmed that the capacity and have a relationship that is close to direct proportionality. In other words, the representative R extracted by performing filtering o It can be seen that it is suitable as a diagnostic factor for SOH.

[0103] Finally, SOH Q and SOH R They compared it.

[0104] Figure 10 shows the capacity obtained from (a) Battery #1 and (b) Battery #2 and the representative R o SOH data in Equation (1) and Equation (3), respectively Q and SOH R This is a graph plotted after calculating. In the entire cycle of Fig. 3, the capacity is measured once during the Q check interval, and the representative R o Since it is measured 16 times in WLTP C3, SOH Q and SOH R It can be calculated as 1 and 16 respectively per cycle.

[0105] The WLTP numbers on the x-axis shown in Fig. 10 represent the 16 WLTPs obtained per cycle arranged in chronological order. To explain in detail, WLTPs 1 through 16 are generated in the first cycle, and WLTPs 17 through 32 are generated in the second cycle. SOH calculated for each WLTP R It is indicated by a solid line. On the other hand, the SOH obtained through the capacity measurement section Q is extracted one per cycle (represented by multiple dots in Fig. 10), SOH R For comparison, it was set to the same number as the first WLTP of each cycle. That is, the SOH measured in the first cycleQ The WLTP number is 1, and the SOH measured in the second cycle Q Its WLTP number is 17. In this way, SOH at every cycle Q The WLTP number increases by 16.

[0106] SOH in Fig. 10 Q and SOH R It can be seen that this is matched. This suggests that SOH can be estimated using the resistance extracted from WLTP without measuring the capacity. To explain in detail, ideally, to calculate SOH, it is desirable to use the ratio of the capacity in a specific state to the initial capacity. However, accurately measuring capacity requires the use of high-performance sensors, which not only increases hardware costs but also necessitates fully discharging and then fully charging the battery. Therefore, measuring the battery capacity every time to determine SOH is nearly impossible.

[0107] It is worth noting that if the resistance extracted from the data used while the battery is in use is utilized, there is no need to measure the capacity every time. For example, when the WLTP number in Fig. 10 is 8, there is no capacity data and only representative Ro data exists, so SOH Q cannot be calculated and SOH R Only can be calculated. SOH Q and SOH R Based on its high consistency, SOH in WLTP number 8 R SOH Q It can be determined that it will be similar to. In other words, instead of calculating SOH using a commonly used capacity, SOH can be estimated using a representative Ro.

[0108] SOH at the point where battery degradation has progressed significantly Q and SOH RTo verify if this shows consistency, the SOH near the end of the time series of Battery #1 and #2 Q and SOH R It was compared in Table 4.

[0109]

[0110] [Table 4]

[0111]

[0112]

[0113] If 16 cycles of WLTP C3 are considered to be 5,000 km, then 513 cycles is approximately 160,000 km. SOH of Battery #1 in 513 cycles. Q is 93.64%, SOH R It is 94.35%, with an error rate of less than 1%. In the case of Battery #2, SOH Q is 89.76%, SOH R It shows an error rate of 90.18%, which is lower than Battery #1. This result indicates that even if capacity measurements are not performed continuously, R o This means that a highly accurate SOH can be calculated through changes in . That is, R o If we continue to measure it, we can know the maximum capacity of the battery at a specific cycle whenever we want.

[0114] In Fig. 10(b), the abnormal resistance values ​​that occurred at the 165th to 175th WLTP C3 of Battery #2 were confirmed to be due to equipment problems at the time, so they are excluded from the result analysis.

[0115] Meanwhile, Tables 5 and 6 below show the SOH of Battery #1 and #2 at their respective WLTP Numbers. Q and SOH R This is the result of comparing.

[0116]

[0117] [Table 5]

[0118]

[0119] [Table 6]

[0120]

[0121]

[0122] As can be seen in Tables 5 and 6, the SOH measured at each WLTP number Q and SOH R When comparing them, it can be seen that the error rate is approximately ±1%. This means that when using the SOH diagnostic method according to one embodiment of the present invention, accurate diagnosis is possible at any point during actual use of the battery.

[0123] The scope of protection of the present invention is not limited to the description and expression of the embodiments explicitly described above. Furthermore, it is added once again that the scope of protection of the present invention cannot be limited by obvious changes or substitutions in the technical field to which the present invention belongs.

[0124]

[0125] [National R&D projects that supported this invention]

[0126] [Project No.] NP2022-0077

[0127] [Ministry Name] Ministry of Trade, Industry and Energy

[0128] [Name of Project Management (Specialized) Agency] Korea Industrial Complex Corporation

[0129] [Research Project Name] Industrial Technology Development Project (Korea Industrial Complex Corporation)

[0130] [Research Project Title] Establishment of a Gwangju Advanced Industrial Complex-type Intelligent Distributed Energy Corporate Joint Research and Utilization Center

[0131] [Name of Project Performing Organization] Korea Institute of Industrial Technology

[0132] [Research Period] 2022.08.01 ~ 2024.12.31

Claims

1. A step of monitoring voltage and current over time during real-time battery use; A step of selecting at least one interval for measuring instantaneous resistance from the monitoring results; A step of measuring instantaneous resistance using the voltage and current of at least one selected interval; and A step of diagnosing SOH by dividing the initial instantaneous resistance by the measured instantaneous resistance; including SOH diagnosis method using instantaneous resistance extraction during real-time battery use.

2. In Paragraph 1, In the step of selecting at least one interval for instantaneous resistance measurement from the above monitoring results, the interval selected is an interval where the voltage changes rapidly, SOH diagnosis method using instantaneous resistance extraction during real-time battery use.

3. In Paragraph 1, A method for diagnosing SOH using instantaneous resistance extraction during real-time battery use, wherein the section selected in the step of selecting at least one section for instantaneous resistance measurement from the above monitoring results is a section changing from a charged state to a discharged state or a section changing from a discharged state to a charged state.

4. In Paragraph 1, In the step of selecting at least one interval for instantaneous resistance measurement from the above monitoring results, a step of filtering the selected data is performed. SOH diagnosis method using instantaneous resistance extraction during real-time battery use.

5. In Paragraph 4, The above filtering step is performed by excluding data corresponding to a preset current range from the data selected in the step of selecting at least one interval for instantaneous resistance measurement. SOH diagnosis method using instantaneous resistance extraction during real-time battery use.

6. In Paragraph 5, The determination of the above-mentioned preset current range is, A step of performing pulse experiments on the above battery while varying the current; A step of configuring the change in current and ohmic resistance from the above pulse experiment results into a first dataset; and A step of determining a current range containing an outlier of ohmic resistance among the first dataset as a current range to be excluded when filtering; comprising SOH diagnosis method using instantaneous resistance extraction during real-time battery use.

7. In Paragraph 4, The above filtering step is performed by selecting data corresponding to a preset voltage range among the data selected in the step of selecting at least one interval for instantaneous resistance measurement. SOH diagnosis method using instantaneous resistance extraction during real-time battery use.

8. In Paragraph 7, The determination of the above-mentioned preset voltage range is, A step of performing pulse experiments on the above battery while varying the current; A step of configuring voltage and ohmic resistance from the above pulse experiment results into a second dataset; A step of plotting the second dataset on a graph, and dividing the plotted graph into a maintaining portion where the ohmic resistance is constant regardless of voltage, a decreasing portion where the ohmic resistance decreases according to voltage, and an increasing portion where the ohmic resistance increases according to voltage; and A step of determining a voltage range corresponding to the above-mentioned maintenance part as a preset voltage range; comprising SOH diagnosis method using instantaneous resistance extraction during real-time battery use.

9. In Paragraph 1, The step of measuring the instantaneous resistance above involves extracting a representative Ro using the Least Square Method (LSM) from the voltage and current range data of a selected section, SOH diagnosis method using instantaneous resistance extraction during real-time battery use.

10. In Paragraph 9, The step of diagnosing the above SOH involves estimating the SOH using the extracted representative Ro according to the following mathematical formula, SOH diagnosis method using instantaneous resistance extraction during real-time battery use. [Mathematical Formula] (where R o,0 represents the instantaneous resistance extracted in the first section, and R o,n represents the instantaneous resistance extracted from the selected interval, the nth interval.)

Images