Battery state estimation system, battery state estimation method, and battery state estimation program

Abstract

A data acquisition unit 111 according to the present invention acquires time series battery data including voltage and current of a lithium ion battery in which a mixed material of graphite and silicon is used as a negative electrode active material. An SOC estimation unit 112 estimates, on the basis of a state of charge (SOC)-open circuit voltage (OCV) curve of the lithium ion battery and the voltage of the lithium ion battery included in the battery data, the SOC of the lithium ion battery. A determination unit 113 determines whether to validate or invalidate the SOC estimated using the SOC-OCV curve on the basis of the crystallization state of the silicon.

Description

Battery state estimation system, battery state estimation method, and battery state estimation program

[0001] This disclosure relates to a battery state estimation system, a battery state estimation method, and a battery state estimation program for estimating the internal state of a secondary battery.

[0002] To accurately estimate the internal states of a secondary battery, such as its State of Charge (SOC), Full Charge Capacity (FCC), and State of Health (SOH), it is necessary to accurately estimate its Open Circuit Voltage (OCV) as a prerequisite. The SOC can be estimated from the OCV of the secondary battery using an SOC-OCV curve.

[0003] In recent years, lithium-ion batteries using a mixed material of graphite (C) and silicon (Si) as the negative electrode active material have been developed to increase battery capacity. It is known that when silicon is charged to a deep depth, the silicon crystallizes, and the SOC-CCV (Closed Circuit Voltage) curve deforms during discharge immediately afterward. When silicon crystallizes, a plateau region occurs in the SOC-CCV curve around 10-30% of the SOC during discharge. When silicon is not crystallized, no plateau region occurs in the SOC-CCV curve during discharge. When a plateau region occurs in the SOC-CCV curve during discharge due to silicon crystallization, a plateau region also occurs in the SOC-OCV curve, and estimation errors of SOC are more likely to occur in the plateau region.

[0004] Patent Document 1 discloses a method for estimating the State of Charge (SOC) of a lithium-ion battery containing silicon as the negative electrode active material, using an SOC-OCV curve if there is no effect of silicon crystallization, and estimating the SOC by a current integration method if there is an effect of silicon crystallization, based on the charge-discharge history of the lithium-ion battery.

[0005] However, when using an estimated SOC value to estimate SOH, estimation errors occur in SOH if there is an effect of silicon crystallization. In the current integration method, measurement errors of the current sensor accumulate, so the longer the current integration period, the greater the effect of measurement errors. Also, FCC is required when converting the current integration value to ΔSOC, but using initial or past FCC will not adequately reflect the effects of lithium-ion battery degradation. To estimate the current FCC, it is basically necessary to use the two-point OCV method, and in the two-point OCV method, it is necessary to estimate SOC from OCV using an SOC-OCV curve.

[0006] International Publication No. 22 / 202318

[0007] This disclosure is made in light of these circumstances, and its purpose is to provide a technique for accurately estimating the state of care (SOC) of lithium-ion batteries using a mixed graphite-silicon anode.

[0008] To solve the above problems, a battery state estimation system in one aspect of the present disclosure includes: a data acquisition unit that acquires time-series battery data including voltage and current of a lithium-ion battery using a mixed material of graphite and silicon as the negative electrode active material; an SOC (State of Charge) - OCV (Open Circuit Voltage) curve of the lithium-ion battery and an SOC estimation unit that estimates the SOC of the lithium-ion battery based on the voltage of the lithium-ion battery included in the battery data; and a determination unit that determines whether to enable or disable the SOC estimated using the SOC - OCV curve based on the crystallization state of the silicon.

[0009] Furthermore, any combination of the above components, as well as any conversion of the expressions of this disclosure between devices, systems, methods, computer programs, etc., are also valid forms of this disclosure.

[0010] According to this disclosure, the state of care (SOC) of a lithium-ion battery using a mixed negative electrode of graphite and silicon can be estimated with high accuracy.

[0011] This is a diagram illustrating a battery-equipped device according to an embodiment. This is a diagram illustrating an example configuration of a battery state estimation system according to an embodiment. This is a diagram illustrating a concrete image of the FCC estimation method. This is a diagram showing an example of a lithium-ion battery degradation prediction formula in graph form. This is a diagram illustrating a specific example of the SOC-CCV curve during discharge. This is a diagram illustrating a specific example of the SOC-CCV curve during discharge when multiple charge-discharge cycles are performed between 23% and 100% SOC. This is a diagram illustrating an image of the data structure of the crystallinity-OCV offset amount map. This is a flowchart illustrating a first correction method for the SOC-OCV curve. This is a flowchart illustrating a first determination method for eliminating unreliable SOCs. This is a flowchart illustrating a second determination method for eliminating unreliable SOCs. This is a diagram illustrating an example configuration of a battery state estimation system according to an embodiment. This is a flowchart illustrating a second correction method for the SOC-OCV curve. This is a diagram illustrating a reference SOC-CCV curve and an example of an SOC-CCV curve during actual use generated from the voltage transition of the discharge interval of the battery data. This figure shows an image of the data structure of the deviation-OCV offset amount map. This is a flowchart for explaining a third determination method for eliminating unreliable SOCs. This is a flowchart for explaining a fourth determination method for eliminating unreliable SOCs. This figure shows an example of configuration 3 of the battery state estimation system according to the embodiment. This figure explains the reaction path of silicon. This figure shows a concrete image of the synthesis of the OCP function of graphite and the OCP function of silicon, and the synthesis of the OCP function of the negative electrode and the OCP function of the positive electrode.

[0012] Figure 1 is a diagram illustrating a battery-equipped device 2 according to an embodiment. The battery-equipped device 2 according to the embodiment is a device equipped with a rechargeable battery pack 30. Examples of battery-equipped devices 2 include consumer information devices (e.g., PCs, tablets, smartphones), home appliances (e.g., robotic vacuum cleaners), electric cars, electric motorcycles, electric bicycles, electric kick scooters, and multicopters (drones).

[0013] The battery-powered device 2 comprises a control unit 21, a load unit 22, a charging unit 23, a communication unit 24, and a battery pack 30. The control unit 21 comprehensively controls the entire battery-powered device 2. The functions of the control unit 21 can be realized through the cooperation of hardware and software resources, or solely through hardware resources. Hardware resources that can be used include a CPU, ROM, RAM, GPU (Graphics Processing Unit), NPU (Neural Network Processing Unit), ASIC (Application Specific Integrated Circuit), FPGA (Field Programmable Gate Array), and other LSIs. Software resources that can be used include operating systems and applications.

[0014] The load unit 22 is a general term for the components that consume power in the battery-equipped device 2 (excluding the control unit 21, charging unit 23, and communication unit 24). The charging unit 23 is connected to the commercial power grid 4 and converts the AC power input from the commercial power grid 4 into DC power of a predetermined voltage or current and outputs it.

[0015] The charging unit 23 includes an AC / DC converter and a DC / DC converter. The DC / DC converter (e.g., a switching regulator) controls the voltage or current of the DC power supplied from the AC / DC converter according to the voltage command value or current command value supplied from the control unit 21, and outputs it to at least one of the load unit 22 or the battery pack 30. When charging from an externally installed rapid charger, it is charged with DC power.

[0016] The battery pack 30 includes a battery pack 31 and a battery management device 32. The battery pack 31 includes a plurality of cells E1-En connected in series. The number of cells in series is determined by the specifications of the load unit 22. In this specification, a lithium-ion battery using a mixed material of graphite (C) and silicon (Si) as the negative electrode active material is used. For the positive electrode active material, lithium nickelate (LiNixCoyAlzO2) or lithium cobaltate (LiCoO2) can be used. In this specification, when simply referred to as "lithium-ion battery," it means a lithium-ion battery cell using a mixed negative electrode of graphite (C) and silicon (Si).

[0017] A switch SW1 is inserted into the power line connecting the battery pack 31 to the load unit 22 or the charging unit 23, to switch between continuity and non-continuity with the load unit 22 or the charging unit 23. A semiconductor switch or relay can be used for switch SW1.

[0018] The battery management device 32 includes a measurement unit 33 and a control unit 34. The measurement unit 33 is composed of an AFE (Analog Front End) IC or an ASIC (Application Specific Integrated Circuit). The control unit 34 is composed of a microcontroller.

[0019] The measurement unit 33 is connected to each node of the multiple series-connected cells E1-En by multiple voltage measurement lines, and measures the voltage of each cell E1-En by measuring the voltage between two adjacent voltage measurement lines.

[0020] The measurement unit 33 includes a multiplexer and an A / D converter. The multiplexer outputs the voltages of multiple cells E1-En in a predetermined order to the A / D converter. The A / D converter converts the analog voltages input from the multiplexer into digital values. The measurement unit 33 transmits the voltage values ​​of each cell E1-En, which have been converted into digital values, to the control unit 34 via a serial communication interface.

[0021] The measurement unit 33 measures the current flowing through the battery pack 31. A shunt resistor Rs is connected to the power line connecting the battery pack 31 to the load unit 22 or the charging unit 23. A differential amplifier (not shown) amplifies the voltage across the shunt resistor Rs and outputs it to the A / D converter in the measurement unit 33. The A / D converter converts the analog voltage indicating the current flowing through the battery pack 31, which is input from the differential amplifier, into a digital value. The measurement unit 33 transmits the digitally converted current value to the control unit 34 via a serial communication interface.

[0022] A temperature sensor T1 (for example, a thermistor) is installed on the surface of the battery pack 31. The divided voltage of the temperature sensor T1 and a voltage divider resistor (not shown) is input to the measurement unit 33. The A / D converter in the measurement unit 33 converts the input analog voltage representing the temperature into a digital value. The measurement unit 33 transmits the converted digital temperature value to the control unit 34 via a serial communication interface.

[0023] The control unit 34 manages the state of each cell E1-En based on the voltage value of each cell E1-En, the current value flowing through the battery pack 31, and the temperature value of the battery pack 31, all of which are received from the measurement unit 33. When the control unit 34 detects any of the following: overcharge, overdischarge, overcurrent, high temperature abnormality, or low temperature abnormality, it sends a cutoff signal for switch SW1 to the measurement unit 33, causing switch SW1 to turn off.

[0024] The control unit 34 transmits battery data, including the voltage, current, and temperature of each cell of the battery pack 30, to the control unit 21 via the internal network of the device at predetermined intervals (for example, every 10 seconds, every 30 seconds, or every 1 minute). If there are many batteries in series in the battery pack 31, the control unit 34 may transmit only the cell voltage of the cell with the highest voltage and the cell voltage of the cell with the lowest voltage, rather than all of the cell voltages of the battery pack 30.

[0025] The communication unit 24 is an external communication interface (e.g., NIC: Network Interface Card) for connecting to an external network 5 via wired or wireless connection. The control unit 21 accesses the network 5 via the communication unit 24 and transmits the battery data received from the control unit 34 of the battery pack 30 to the battery state estimation system 1 at predetermined transmission intervals (e.g., 10-second, 30-second, or 1-minute intervals). The control unit 21 may also store the received battery data in its internal memory and transmit the stored battery data in a batch at a predetermined timing.

[0026] Network 5 is a general term for communication channels such as the Internet, dedicated lines, and VPNs (Virtual Private Networks), and does not specify the communication medium or protocol. Examples of communication mediums include wired LANs, wireless LANs, mobile phone networks, fiber optic networks, ADSL networks, and CATV networks. Examples of communication protocols include TCP (Transmission Control Protocol) / IP (Internet Protocol), UDP (User Datagram Protocol) / IP, and Ethernet (registered trademark).

[0027] The battery state estimation system 1 is a system for estimating and analyzing the state of a battery pack 30 installed in a battery-equipped device 2. In this embodiment, the battery state estimation system 1 is built on a cloud server located in a data center managed by a cloud service provider. A battery analysis service provider that provides analysis services for the battery pack 30 uses the cloud server by contracting with the cloud service provider. The battery state estimation system 1 may also be built on a server located in the battery analysis service provider's own facility or data center.

[0028] Figure 2 shows an example configuration of a battery state estimation system 1 according to an embodiment. The battery state estimation system 1 comprises a control unit 11, a storage unit 12, and a communication unit 13. The communication unit 13 is an external communication interface for connecting to the network 5 by wire or wireless.

[0029] The control unit 11 includes a data acquisition unit 111, an SOC estimation unit 112, a determination unit 113, a correction unit 114, an SOH estimation unit 115, a degradation prediction formula generation unit 116, and a life prediction unit 117. The functions of the control unit 11 can be realized through the cooperation of hardware resources and software resources, or solely through hardware resources. Hardware resources that can be used include a CPU, ROM, RAM, GPU, NPU, ASIC, FPGA, and other LSIs. Software resources that can be used include operating systems and applications.

[0030] The storage unit 12 includes a non-volatile recording medium such as an HDD or SSD, and stores various types of data. The storage unit 12 includes a battery data storage unit 121, an SOC-OCV curve storage unit 122, a crystallinity map storage unit 123, and a crystallinity-OCV offset amount map storage unit 124.

[0031] The data acquisition unit 111 acquires battery data from the battery pack 30 installed in the battery-equipped device 2 via the network 5. As described above, the battery data includes time-series data of the voltage, current, and temperature of the lithium-ion battery. The data acquisition unit 111 stores the acquired battery data in the battery data storage unit 121.

[0032] The SOC-OCV curve holding unit 122 holds the SOC-OCV curve of the lithium-ion battery contained in the battery pack 30. The SOC-OCV curve of the lithium-ion battery is created in advance based on characteristic tests conducted by the battery manufacturer and registered in the SOC-OCV curve holding unit 122.

[0033] The SOC estimation unit 112 reads the battery data of the lithium-ion battery to be analyzed from the battery data holding unit 121 and reads the SOC-OCV curve of a lithium-ion battery of the same type as the lithium-ion battery from the SOC-OCV curve holding unit 122. Based on the SOC-OCV curve of the lithium-ion battery and the voltage of the lithium-ion battery to be analyzed, the SOC estimation unit 112 estimates the SOC of the lithium-ion battery. The operation of the determination unit 113 and the correction unit 114 will be described later.

[0034] The SOH estimation unit 115 calculates the State of Health (SOH) of a lithium-ion battery over a predetermined period in a time series using a two-point OCV method that utilizes the SOC difference between two points and the integrated current value obtained by referring to battery data of the lithium-ion battery under analysis for a predetermined period. Specifically, the SOH estimation unit 115 determines the SOC difference (ΔSOC) between the SOC of the first resting state and the SOC of the second resting state based on the voltage of the first resting state and the voltage of the second resting state of the lithium-ion battery and the SOC-OCV curve. The SOH estimation unit 115 determines the integrated current value Q for the period from the first resting state to the second resting state based on the current value included in the battery data. The SOH estimation unit 115 determines the current FCC of the lithium-ion battery based on the integrated current value Q and ΔSOC. The SOH estimation unit 115 estimates the SOH of the lithium-ion battery based on the ratio of the current FCC to the initial FCC of the lithium-ion battery.

[0035] Figure 3 shows a concrete image of the FCC estimation method. The SOH estimation unit 115 identifies two points, OCV1 and OCV2, for the first and second resting states, and uses them as the two OCVs. The SOH estimation unit 115 refers to the SOC-OCV curve to identify two points, SOC1 and SOC2, which correspond to the two OCV1 and OCV2 points, and calculates the ΔSOC between the two SOCs.

[0036] The SOH estimation unit 115 calculates the integrated current (= charge / discharge capacity) Q between two points from which OCVs are obtained. The SOH estimation unit 115 estimates the FCC by calculating the following (Equation 1): FCC = Q / ΔSOC ... (Equation 1)

[0037] The SOH estimation unit 115 estimates the SOH by calculating the following (Equation 2). SOH is defined as the ratio of the current FCC to the initial FCC, and a lower value (closer to 0%) indicates that degradation is progressing. SOH = Current FCC / Initial FCC × 100 ... (Equation 2)

[0038] The SOH estimation unit 115 uses battery data of the lithium-ion battery over a predetermined period and calculates the SOH at regular intervals (for example, every month, every two weeks, every week) using the two-point OCV method to generate a time-series SOH.

[0039] Furthermore, in the case of a battery pack 30 that undergoes periodic maintenance charging to check its capacity, the SOH estimation unit 115 can estimate the SOH based on the battery data from the maintenance charging. During maintenance charging, the battery is charged at a constant current from SOC 0% to SOC 100%, so the FCC can be estimated by integrating the current during the maintenance charging period.

[0040] The degradation prediction formula generation unit 116 generates a degradation prediction formula by performing a curve regression on the time-series SOH of the lithium-ion battery over a predetermined period. For example, the least squares method can be used for the curve regression. Generally, it is known that the SOH of a lithium-ion battery decreases in proportion to the square root of the elapsed time (root law, 0.5 power law), so the degradation prediction formula can be defined as follows (Equation 3): SOH = w0 + w1√t ... (Equation 3) w0 is the initial value, and w1 is the degradation coefficient.

[0041] Figure 4 is a graph illustrating an example of a lithium-ion battery degradation prediction formula. The horizontal axis represents the operating period, and the vertical axis represents the State of Health (SOH). The degradation prediction formula generation unit 116 calculates the degradation coefficient (slope) w1 in (Equation 3) above by exponential regression to the power of 0.5, with the operating period t as the independent variable and SOH as the dependent variable. w0 is the initial SOH and is usually set in the range of 1.0 to 1.1. If the actual initial capacity matches the nominal value, w0 is set to 1.0. If the nominal value is set to the minimum guaranteed amount and is lower than the actual initial capacity, a value greater than 1.0 is set.

[0042] The life prediction unit 117 predicts the remaining lifespan of the lithium-ion battery based on the degradation prediction formula generated by the degradation prediction formula generation unit 116. For example, the life prediction unit 117 determines that the timing when the State of Health (SOH) of the lithium-ion battery reaches 70% is the timing for ending the use of the lithium-ion battery.

[0043] As described above, when silicon crystallizes, the SOC-CCV curve deforms during the subsequent discharge. The inventors conducted experiments to investigate how silicon crystallization affects the SOC-CCV curve during discharge. The experiments revealed that the presence or absence of silicon crystallization depends on the SOC value at the start of charging and the SOC value at the end of charging.

[0044] Figure 5 shows a specific example of the SOC-CCV curve during discharge. Figure 5 shows the measured SOC-CCV curve when a lithium-ion battery is charged from SOC 0% to 100% and then discharged, and the measured SOC-CCV curve when it is charged from SOC 20% to 100% and then discharged. In the SOC-CCV curve when it is charged from SOC 20% to 100% and then discharged, a plateau region occurs, and compared to the SOC-CCV curve when it is charged from SOC 0% to 100% and then discharged, the CCV drops by 14 mV around SOC 23%.

[0045] In lithium-ion batteries using a silicon-containing mixed negative electrode, lithium is inserted into the silicon of the negative electrode during charging, and lithium is released from the silicon during discharge. When charging from a state of charge (SOC) of 0%, silicon crystallization does not generally occur. However, when charging is performed before the SOC has sufficiently decreased, silicon crystallization is more likely to occur. Experiments have shown that repeated charging and discharging from a state before the SOC has sufficiently decreased increases the amount of silicon that crystallizes, and the degree of OCV decrease increases.

[0046] Figure 6 shows a specific example of the SOC-CCV curve during discharge when multiple charge-discharge cycles are performed between 23% and 100% SOC. The waveform within the square frame in Figure 6 shows the change in CCV as the lithium-ion battery charges and discharges. In this waveform, there is a section in which (a) discharge from 100% SOC to 23% SOC, (b) charge from 23% SOC to 100% SOC, (c) discharge from 100% SOC to 23% SOC, (d) charge from 23% SOC to 100% SOC, and (e) discharge from 100% SOC to 23% SOC are repeated.

[0047] Comparing the discharge curves in (a), (c), and (e), a difference in CCV appears from around 30% SOC. The discharge curve in (c) shows a decrease in CCV from around 30% SOC compared to the discharge curve in (a), and the discharge curve in (e) shows an even greater decrease in CCV from around 30% SOC.

[0048] When the OCV at the 23% SOC point was measured, the OCV of the discharge curve in (c) was 12 mV lower than the OCV of the discharge curve in (a), and the OCV of the discharge curve in (e) was 25 mV lower than the OCV of the discharge curve in (a). A decrease of -12 mV in OCV corresponds to an SOC error of -0.7%, and a decrease of -25 mV in OCV corresponds to an SOC error of -1.3%.

[0049] As described above, the SOH estimation unit 115 generates a degradation prediction formula for lithium-ion batteries by performing regression on multiple SOH estimation points (see Figure 4). In order to generate a highly accurate degradation prediction formula, it is necessary to increase the number of SOH estimation points and improve the accuracy of each SOH estimation point.

[0050] In this embodiment, as a method to increase the number of SOH estimation points while maintaining the estimation accuracy of SOH, a method is used in which the SOC-OCV curve is corrected to estimate a highly accurate SOC without using the current integration method even when silicon crystallization is present, and the SOH is estimated from the highly accurate SOC. Furthermore, as a method to improve the estimation accuracy of each SOH, a method is used in which SOC estimation is skipped when silicon crystallization is present.

[0051] Returning to Figure 2, the correction unit 114 corrects the SOC-OCV curve of the lithium-ion battery based on the silicon crystallization state. The silicon crystallization state can be estimated by the degree of silicon crystallinity or the degree of deviation from the reference SOC-CCV curve, which will be described later.

[0052] The crystallinity map holding unit 123 holds a crystallinity map that describes the relationship between the charge / discharge pattern of the lithium-ion battery and the crystallinity of silicon, which was generated in advance by the designer based on the results of experiments or simulations. The crystallinity of silicon is defined in the range of 0 to 100, where 0 refers to a state in which the silicon in the negative electrode is not crystallized at all, and 100 refers to a state in which the silicon in the negative electrode is completely crystallized.

[0053] The charge and discharge patterns of lithium-ion batteries are based on the SOC at the start of charging and the SOC at the end of charging, or the SOC at the start of discharging and the SOC at the end of discharging. For example, rules can be established and associated with actions such as "charge once from SOC 20% to 100%" and "add 5% to the crystallinity," or "discharge once to SOC 10%" and "reset the crystallinity to 0%."

[0054] The designer can add various charge-discharge conditions as parameters to estimate the degree of crystallinity more precisely. For example, the number of repetitions of the basic pattern can be used as a parameter. As mentioned above, if charge-discharge is repeated without the crystallinity being resolved, the degree of crystallinity will grow. At least one of the charge-discharge conditions, such as current rate, temperature, and SOH, may also be used as a parameter. The designer derives the degree of crystallinity in charge-discharge patterns under various charge-discharge conditions through experimentation or simulation and generates a degree of crystallinity map.

[0055] The crystallinity-OCV offset amount map holding unit 124 holds a crystallinity-OCV offset amount map, which is pre-generated by the designer based on experimental or simulation results, describing the relationship between the silicon crystallinity and the OCV offset amount for each SOC.

[0056] Figure 7 shows an image of the data structure of the crystallinity-OCV offset map. The crystallinity-OCV offset map is described in a two-dimensional parameter space of SOC and crystallinity, and the OCV offset amount of the SOC-OCV curve is defined for each combination of SOC and crystallinity. In Figure 7, a map is drawn where the unit width of SOC and the unit width of crystallinity are set to 10%, but the unit width of SOC and the unit width of crystallinity may be set to 1%, 3%, or 5%, respectively.

[0057] Figure 8 is a flowchart illustrating the first correction method for the SOC-OCV curve. The determination unit 113 reads the battery data of the lithium-ion battery to be analyzed from the battery data holding unit 121 and reads the crystallinity map of a lithium-ion battery of the same type as the lithium-ion battery from the crystallinity map holding unit 123. The determination unit 113 estimates the crystallinity of silicon based on the read crystallinity map and the charge / discharge transition based on the read battery data (S10).

[0058] The determination unit 113 compares the estimated degree of silicon crystallinity with a first threshold (S11). The first threshold may be set to 0% or to a few percent. If the estimated degree of silicon crystallinity is greater than or equal to the first threshold (Y in S11), the determination unit 113 determines that silicon crystallinity has occurred, and if it is less than the first threshold (N in S11), it determines that silicon crystallinity has been resolved.

[0059] If the crystallinity of silicon is above the first threshold (Y in S11), the correction unit 114 reads the crystallinity-OCV offset amount map of a lithium-ion battery of the same type as the lithium-ion battery being analyzed from the crystallinity-OCV offset amount map holding unit 124. The correction unit 114 estimates the OCV offset amount based on the crystallinity-OCV offset amount map and the estimated crystallinity of silicon (S12), and corrects the default SOC-OCV curve using the estimated OCV offset amount (S13).

[0060] If the crystallinity of silicon is below the first threshold (N in S11), steps S12 and S13 are skipped. That is, the correction of the SOC-OCV curve by the correction unit 114 is skipped. Alternatively, step S11 may be omitted, and the OCV offset amount may always be estimated by referring to the crystallinity-OCV offset amount map, and the SOC-OCV curve may be corrected.

[0061] As a method to maintain the accuracy of SOC estimation, it is also conceivable that the determination unit 113, without correcting the SOC-OCV curve, determine whether to validate or invalidate the SOC estimated using the SOC-OCV curve based on the crystallization state of silicon, and select the SOC accordingly.

[0062] Figure 9 is a flowchart illustrating a first determination method for eliminating unreliable SOCs. The determination unit 113 reads the battery data of the lithium-ion battery to be analyzed from the battery data holding unit 121 and reads the crystallinity map of a lithium-ion battery of the same type as the lithium-ion battery from the crystallinity map holding unit 123. The determination unit 113 estimates the crystallinity of silicon based on the read crystallinity map and the charge / discharge transition based on the read battery data (S20).

[0063] The determination unit 113 compares the estimated degree of silicon crystallinity with a first threshold (S21). If the estimated degree of silicon crystallinity is greater than or equal to the first threshold (Y in S21), the determination unit 113 determines that silicon crystallinity has occurred, and if it is less than the first threshold (N in S21), it determines that silicon crystallinity has been resolved.

[0064] If the crystallinity of silicon is below the first threshold (N in S21), the SOC estimation unit 112 reads the SOC-OCV curve of a lithium-ion battery of the same type as the lithium-ion battery to be analyzed from the SOC-OCV curve holding unit 122 and estimates the SOC of the lithium-ion battery to be analyzed (S22). If the crystallinity of silicon is equal to or greater than the first threshold (Y in S21), the process in step S22 is skipped. That is, the determination unit 113 controls the system so that SOC estimation using the SOC-OCV curve is skipped.

[0065] Figure 10 is a flowchart illustrating a second determination method for eliminating unreliable SOCs. The SOC estimation unit 112 reads the battery data of the lithium-ion battery to be analyzed from the battery data holding unit 121 and reads the SOC-OCV curve of a lithium-ion battery of the same type as the lithium-ion battery in question from the SOC-OCV curve holding unit 122 (S30).

[0066] The determination unit 113 reads the crystallinity map of a lithium-ion battery of the same type as the lithium-ion battery in question from the crystallinity map holding unit 123, and reads the crystallinity-OCV offset map of a lithium-ion battery of the same type from the crystallinity-OCV offset map holding unit 124.

[0067] The determination unit 113 estimates the crystallinity of silicon based on the read crystallinity map and the charge / discharge transition based on the read battery data (S31). The determination unit 113 estimates the OCV offset amount based on the read crystallinity-OCV offset amount map and the estimated silicon crystallinity (S32). The correction unit 114 corrects the read SOC-OCV curve using the estimated OCV offset amount (S33). The determination unit 113 estimates the SOC from the corrected SOC-OCV curve and the OCV estimated from the voltage value included in the battery data (S34), and obtains the OCV offset amount corresponding to the estimated SOC and estimated crystallinity by referring to the crystallinity-OCV offset amount map (S35). The determination unit 113 compares the obtained OCV offset amount with a second threshold (S36). The second threshold is set by the designer to the value of the allowable OCV offset amount.

[0068] If the estimated OCV offset amount is greater than or equal to the second threshold (Y in S36), the determination unit 113 invalidates the estimated SOC (S37). If the estimated OCV offset amount is less than the second threshold (N in S36), the process in step S37 is skipped. In other words, the determination unit 113 validates the estimated SOC.

[0069] As mentioned above, the effect of silicon crystallization on OCV reduction is significant around 10-30% SOC, and becomes smaller in the high SOC region. For example, even when the silicon crystallinity is 50% or higher, if the SOC is high enough to around 80%, the OCV reduction becomes almost negligible.

[0070] Figure 11 shows an example configuration 2 of the battery state estimation system 1 according to an embodiment. In example configuration 2 of the battery state estimation system 1, a reference SOC-CCV curve holding unit 125 and a deviation-OCV offset amount map holding unit 126 are used instead of the crystallinity map holding unit 123 and crystallinity-OCV offset amount map holding unit 124 used in example configuration 1 of the battery state estimation system 1 shown in Figure 2.

[0071] The reference SOC-CCV curve holding unit 125 holds the SOC-CCV curve of a lithium-ion battery during discharge in a state where silicon crystallization has not occurred as the reference SOC-CCV curve during discharge. The reference SOC-CCV curve during discharge is created based on data obtained from a prior discharge test and registered in the reference SOC-CCV curve holding unit 125.

[0072] Figure 12 is a flowchart illustrating a second correction method for the SOC-OCV curve. The determination unit 113 reads the battery data of the lithium-ion battery to be analyzed from the battery data holding unit 121 and reads the reference SOC-CCV curve of the lithium-ion battery from the reference SOC-CCV curve holding unit 125. The determination unit 113 generates an SOC-CCV curve for the discharge section from the voltage transition of the discharge section of the read battery data (S40). The determination unit 113 calculates the degree of deviation between the read reference SOC-CCV curve and the generated SOC-CCV curve for the discharge section (S41).

[0073] Figure 13 shows a reference SOC-CCV curve and an example of an SOC-CCV curve generated from the voltage transition during the discharge interval of battery data during actual use. The degree of deviation between the reference SOC-CCV curve and the SOC-CCV curve during actual use may be defined, for example, by the difference in CCV between the two at a specific SOC (e.g., 25%). Alternatively, this degree of deviation may be defined by the cumulative value or average value of the difference in CCV between the two in a specific SOC range (e.g., 10-30%).

[0074] Furthermore, the degree of deviation may be defined as the value obtained by dividing the area of ​​the difference between the reference SOC-CCV curve and the SOC-CCV curve during actual use by the ΔSOC of the difference interval between the two curves, as shown in (Equation 4) below. Degree of deviation = Area of ​​the difference between SOC-CCV curves ÷ ΔSOC of the difference interval between SOC-CCV curves ... (Equation 4)

[0075] The degree of deviation may be normalized to a range of 0-100. 0 refers to a state where the reference SOC-CCV curve and the SOC-CCV curve during actual use perfectly match. 100 refers to a state where the silicon in the negative electrode of the lithium-ion battery is fully crystallized, and the maximum deformation SOC-CCV curve when discharged at the rated discharge current of the lithium-ion battery matches the SOC-CCV curve during actual use.

[0076] The SOC-CCV curve depends on the current rate. Therefore, there is a possibility of confusing the deformation of the SOC-CCV curve due to the current rate with the deformation of the SOC-CCV curve due to silicon crystallization. To address this, the determination unit 113 converts the SOC-CCV curve during actual use to the SOC-CCV curve when discharged at a constant current rate before comparing the reference SOC-CCV curve with the SOC-CCV curve during actual use.

[0077] Alternatively, reference SOC-CCV curves may be created for various current rates, and a map of these reference SOC-CCV curves along the current axis may be prepared in advance. In this case, the determination unit 113 appropriately selects a reference SOC-CCV curve corresponding to the current rate identified from the battery data. The determination unit 113 compares the selected reference SOC-CCV curve with the SOC-CCV curve used in actual operation.

[0078] The deviation-OCV offset amount map holding unit 126 holds a deviation-OCV offset amount map for each SOC, which is generated in advance by the designer based on experimental or simulation results and describes the relationship between the deviation degree from the reference SOC-CCV curve and the OCV offset amount.

[0079] Figure 14 shows an image of the data structure of the deviation-OCV offset map. The deviation-OCV offset map is described in a two-dimensional parameter space of SOC and deviation, and the OCV offset amount of the SOC-OCV curve is defined for each combination of SOC and deviation. In Figure 14, a map is drawn where the unit width of SOC and the unit width of deviation are set to 10%, but the unit width of SOC and the unit width of deviation may be set to 1%, 3%, or 5%, respectively.

[0080] Returning to Figure 12, the determination unit 113 compares the degree of deviation from the calculated reference SOC-CCV curve with a third threshold (S42). The third threshold may be set to 0% or to a few percent. The determination unit 113 determines that silicon crystallization has occurred if the degree of deviation from the calculated reference SOC-CCV curve is greater than or equal to the third threshold (Y in S42), and determines that silicon crystallization has been resolved if it is less than the third threshold (N in S42).

[0081] If the degree of deviation from the reference SOC-CCV curve is greater than or equal to the third threshold (Y in S42), the correction unit 114 reads the degree of deviation-OCV offset amount map of a lithium-ion battery of the same type as the lithium-ion battery being analyzed from the degree of deviation-OCV offset amount map holding unit 126. The correction unit 114 estimates the OCV offset amount based on the degree of deviation between the degree of deviation-OCV offset amount map read from the degree of deviation-OCV offset amount map holding unit 126 and the calculated reference SOC-CCV curve (S43), and corrects the default SOC-OCV curve using the estimated OCV offset amount (S44).

[0082] If the deviation from the reference SOC-OCV curve is less than the third threshold (N in S42), steps S43 and S44 are skipped. That is, the correction of the SOC-OCV curve by the correction unit 114 is skipped. Alternatively, step S42 may be omitted, and the OCV offset amount may be estimated by always referring to the deviation-OCV offset amount map, and the SOC-OCV curve may be corrected.

[0083] As a method to maintain the accuracy of SOC estimation, the determination unit 113 may select whether to validate or invalidate the SOC estimated using the SOC-OCV curve based on the crystallization state of silicon.

[0084] Figure 15 is a flowchart illustrating a third determination method for eliminating unreliable SOCs. The determination unit 113 reads the battery data of the lithium-ion battery to be analyzed from the battery data holding unit 121 and reads the reference SOC-CCV curve of the lithium-ion battery from the reference SOC-CCV curve holding unit 125. The determination unit 113 generates an SOC-CCV curve for the discharge section from the voltage transition of the discharge section of the read battery data (S50). The determination unit 113 calculates the degree of deviation between the read reference SOC-CCV curve and the generated SOC-CCV curve for the discharge section (S51).

[0085] The determination unit 113 compares the degree of deviation from the calculated reference SOC-CCV curve with a third threshold (S52). If the degree of deviation from the calculated reference SOC-CCV curve is greater than or equal to the third threshold (Y in S51), the determination unit 113 determines that silicon crystallization has occurred, and if it is less than the third threshold (N in S51), it determines that silicon crystallization has been resolved.

[0086] If the degree of deviation from the calculated reference SOC-OCV curve is less than the third threshold (N in S51), the SOC estimation unit 112 reads the SOC-OCV curve of a lithium-ion battery of the same type as the lithium-ion battery to be analyzed from the SOC-OCV curve holding unit 122 and estimates the SOC of the lithium-ion battery to be analyzed (S53).

[0087] If the degree of deviation from the calculated reference SOC-OCV curve is greater than or equal to the third threshold (Y in S51), the process in step S53 is skipped. In other words, the determination unit 113 controls the process so that SOC estimation using the SOC-OCV curve is skipped.

[0088] Figure 16 is a flowchart illustrating a fourth determination method for eliminating unreliable SOCs. The SOC estimation unit 112 reads the battery data of the lithium-ion battery to be analyzed from the battery data holding unit 121 and reads the SOC-OCV curve of a lithium-ion battery of the same type as the lithium-ion battery in question from the SOC-OCV curve holding unit 122 (S60).

[0089] The determination unit 113 reads the reference SOC-CCV curve of the lithium-ion battery from the reference SOC-CCV curve holding unit 125 and reads the deviation-OCV offset amount map of the lithium-ion battery of the said type from the deviation-OCV offset amount map holding unit 126. The determination unit 113 generates an SOC-CCV curve for the discharge section from the voltage transition of the discharge section of the read battery data (S61). The determination unit 113 calculates the deviation between the read reference SOC-CCV curve and the generated SOC-CCV curve for the discharge section (S62). The determination unit 113 estimates the OCV offset amount based on the deviation between the read deviation-OCV offset amount map and the calculated reference SOC-CCV curve (S63). The correction unit 114 corrects the read SOC-OCV curve using the estimated OCV offset amount (S64). The determination unit 113 estimates the SOC from the corrected SOC-OCV curve and the OCV estimated from the voltage value included in the battery data (S65), and obtains the OCV offset amount corresponding to the estimated SOC and the calculated deviation amount by referring to the deviation-OCV offset amount map (S66).

[0090] The determination unit 113 compares the acquired OCV offset amount with the fourth threshold (S67). The fourth threshold is set by the designer to the value of an acceptable OCV offset amount. If the estimated OCV offset amount is greater than or equal to the fourth threshold (Y in S67), the determination unit 113 invalidates the estimated SOC (S68). If the estimated OCV offset amount is less than the fourth threshold (N in S67), the process in step S68 is skipped. In other words, the determination unit 113 validates the estimated SOC.

[0091] Figure 17 shows an example of configuration 3 of the battery state estimation system 1 according to an embodiment. In configuration example 3 of the battery state estimation system 1, compared to configuration example 1 of the battery state estimation system 1 shown in Figure 2, the crystallinity map holding unit 123 and the crystallinity-OCV offset amount map holding unit 124 are omitted, and an internal state model estimation unit 118 is added.

[0092] The internal state model estimation unit 118 estimates the amount of lithium inserted into silicon, the amount of lithium desorbed from silicon, and the OCP (Open Circuit Potential) function of silicon based on the internal model of the lithium-ion battery and the current included in the battery data. The internal state model estimation unit 118 generates the OCP function of the negative electrode by combining the estimated silicon OCP function and the graphite OCP function, and generates the SOC-OCV curve by combining the generated negative electrode OCP function and the positive electrode OCP function. The correction unit 114 uses the SOC-OCV curve estimated by the internal state model estimation unit 118 as a corrected SOC-OCV curve.

[0093] As a method to maintain the accuracy of SOC estimation, it is also conceivable that the determination unit 113, without correcting the SOC-OCV curve, determine whether to validate or invalidate the SOC estimated using the SOC-OCV curve based on the crystallization state of silicon, and select the SOC accordingly.

[0094] The determination unit 113 compares the amount of lithium inserted into silicon estimated by the internal state model estimation unit 118 with a fifth threshold value. The fifth threshold value is set by the designer based on the results of experiments or simulations. When the amount of lithium inserted into silicon is greater than or equal to the fifth threshold value, the determination unit 113 determines that silicon crystallization has occurred and controls so that the SOC estimation using the SOC-OCV curve is skipped. When the amount of lithium inserted into silicon is less than the fifth threshold value, the determination unit 113 determines that the silicon crystallization has disappeared. In this case, the SOC estimated using the SOC-OCV curve is valid.

[0095] Next, a specific method for estimating the amount of lithium inserted into silicon, the amount of lithium desorbed from silicon, and the OCP function of silicon using theoretical formulas will be described. First, the variables used in the theoretical formulas described below will be defined. m: Object (Si (silicon) / g (graphite) / n (negative electrode) / p (positive electrode)) r: Phase (s (solid phase) / e (electrolyte phase)) k: Type of reaction (1 / 2 / f3 / f4 / b2) (see FIG. 18 described later) F: Faraday constant [Cmol -2 , -1 , cryst , m,0 , -1 , n,tot , -3 , Li , -3 , -3 , Si(k) , m , cryst , Li , (k) R: Gas constant [Jmol -1 K -1 T: Absolute temperature [K] t: Time [s] φ m,r : Potential of r of m [V] φ m,ocp : Open circuit potential of m [V] a m : Specific surface area of m [m -1 i m,0 : Exchange current density of m [Am -2 j m Li : Reaction current density per unit volume of m [Am -3 j Si(k) Li : Reaction current density per unit volume of reaction k of Si [Am -3 Q: Cell capacity [C] Q n,tot : Capacity per unit volume of the negative electrode [Cm -3 k cryst : Rate constant of crystallization [s -1 x (k) : Ratio of sites filled by reaction k x cryst : c-Li15 Mole fraction of Si4 Δx (k) : Percentage of maximum lithium insertion amount due to reaction k w j : Reaction coefficient φ between adjacent ions in the host lattice (k),0 : Standard equilibrium potential [V] of reaction k φ (k) : Equilibrium potential of reaction k [V] θ m : Stoichiometric ratio of m I: Applied current [A]

[0096] The internal state model estimation unit 118 estimates the potential φ of the solid phase of the negative electrode using an equivalent circuit model or a DFN (Doyle-Fuller-Newman) model. n,s , the potential φ of the electrolyte phase of the negative electrode n,e And the current density j into silicon Si Li To estimate the current density j in silicon. Si Li Three methods can be considered for estimating this.

[0097] The internal state model estimation unit 118, as a first method, directly estimates the current density j to silicon using the DFN model as shown in (Equation 5) below. Si Li We seek j Si Li = 2a si i si,0 ・sinh[0.5F / RT(φ n,s -φ n,e -φ si,ocp )] ... (Formula 5)

[0098] Open-circuit potential φ of silicon si,ocp φ can be defined as shown in (Equation 6) below. si,ocp =sigmoid(-100I / Q)φ si,ocp de +sigmoid(-100I / Q)φ si,ocp li　 ...(Formula 6) sigmoid(x)=[1+sinh(x)] / 2

[0099] The coefficient 100 is introduced to rescale the applied current I, which is normalized by the cell capacitance Q. si,ocp de This indicates the open-circuit potential during lithium desorption, and φsi,ocp li This indicates the open-circuit potential when lithium is inserted. The applied current I is the current value included in the battery data.

[0100] The internal state model estimation unit 118 uses the DFN model as a second method, as shown in (Equation 7) below, to estimate the current density j to the negative electrode. n Li Current density j from to graphite g Li By subtracting this, the current density j to silicon is obtained. Si Li We seek j g Li = 2a g i g,0 ・sinh[0.5F / RT(φ n,s -φ n,e -φ g,ocp ) ] j Si Li = j n Li -j g Li ...(Formula 7)

[0101] As a third method, the internal state model estimation unit 118 configures an equivalent circuit model including circuit elements representing the positive electrode, graphite negative electrode, and silicon negative electrode, and determines the potential φ of the solid phase of the negative electrode. n,s , the potential φ of the electrolyte phase of the negative electrode n,e And the current density j into silicon Si Li It is also possible to estimate this.

[0102] The determination unit 113 determines the current density j to silicon as shown in (Equation 8) below. Si Li When the cumulative amount of lithium inserted into silicon exceeds the fifth threshold, it is determined that silicon crystallization has occurred. Si Li ≥ Fifth threshold... (Equation 8)

[0103] Furthermore, the determination unit 113 determines the potential φ of the solid phase of the negative electrode, as shown in (Equation 9) below. n,sWhen it becomes less than the sixth threshold value, it is determined that silicon crystallization has occurred. The sixth threshold value is set by the designer based on the results of experiments or simulations. φ n,s <Sixth threshold value... (Equation 9)

[0104] Below, the method for obtaining the current density j to silicon by the first method Si Li will be specifically described. The internal state model estimation unit 118 is φ (k) , x (k) , x cryst obtained from the initial value or the calculation result at the time one step before, and φ (k) , x (k) , x cryst is updated. The current density of each reaction is expressed by the Butler-Volmer equation as shown in the following (Equation 10). j Si(k) Li = 2a si i si,0 ·sinh[0.5F / RT(φ n,s - φ n,e - φ (k) )] ··· (Equation 10)

[0105] The time evolution of each reaction amount of silicon is defined as in the following (Equation 11), (Equation 12). The internal state model estimation unit 118 updates x (k) , x cryst . dx (k) / dt = j Si(k) Li / Q n,tot (k = 1, 2, f3, b2) ··· (Equation 11) dx cryst / dt = k cryst x cryst (x (f3) / Δx (f3) - x cryst ) (f4) ··· (Equation 12)

[0106] FIG. 18 is a diagram for explaining the reaction path of silicon. When the depth of charge of silicon is not so deep (> 0.05 V), Reaction 1 and Reaction 2 occur. Reaction 1 and Reaction 2 occur when discharging at a high DOD (Depth Of Discharge) and then charging.

[0107] Reaction 1 involves the reaction of amorphous silicon a-Si with lithium ions Li ＋ a-Li is inserted. x This is a reversible reaction that forms Si. Reaction 2 is a-Li x Si and Li ＋ a-Li is inserted. x This is a reversible reaction that forms Si4.

[0108] When the silicon charge depth is deep (<0.05V), reactions f3, f4, and b2 occur. Reactions f3, f4, and b2 occur when the silicon is discharged and then charged in a medium DOD state. Reaction f3 is a-Li 15 Si4 with Li ＋ a-Li is inserted. (15+δ) This is an irreversible reaction that forms Si4 (formation of nuclei in the crystal structure). Reaction f4 is a-Li (15+δ) Si4 crystallizes, c-Li 15 This is an irreversible reaction that forms Si4 (growth of crystal structure). Reaction b2 involves c-Li 15 Si4 to Li ＋ It detaches, a-Li x This is an irreversible reaction that forms silicon.

[0109] The equilibrium potential of each reaction is defined as shown in (Equation 13) below. The internal state model estimation unit 118 updates x using (Equation 11) above. (k) Using φ (k) Update. φ (k) = φ (k),0 +RTw (k) / F・ln(Δx (k) -x (k) / x (k) ) (k=1, 2, f3, b2) ... (Formula 13)

[0110] The internal state model estimation unit 118 determines the proportion of silicon C that transitions between reaction 2 and reaction 1 paths when switching from charging to discharging. (2,1) , the proportion of silicon C that transitions through the reaction path f4 (f4) The proportion of silicon C that transitions through reaction b2 and reaction 1. (b2,1) This is calculated using the following equations (Equations 14, 15, and 16). C (2,1) = 1 - (x (f3) / Δx(f3) ) ... (Formula 14) C (f4) = (x (f3) / Δx (f3) ) - x cryst ...(Formula 15) C (b2,1) = x cryst ...(Formula 16)

[0111] The determination unit 113 determines the proportion of silicon C that transitions between reaction 2 and reaction 1, as shown in (Equation 17) below. (2,1) If the value exceeds the seventh threshold, it is determined that silicon crystallization has been resolved. The seventh threshold is set by the designer based on experimental or simulation results. (2,1) > Seventh threshold... (Equation 17)

[0112] The internal state model estimation unit 118 transforms the above (Equation 13) as shown below (Equation 18), and x (k) to φ (k) Represented by x (k) (φ (k) ) = Δx (k) / [1+exp((F / RTw (k) )・(φ (k) -φ (k),0 ))] (k=1, 2, f3, b2) ... (Formula 18)

[0113] The internal state model estimation unit 118 divides the discharge into two cases: one where the reaction path transitions from reaction path 2 to reaction path 1, and another where it transitions from reaction path b2 to reaction path 1. It then calculates the inverse function of the OCP of the silicon on each discharge side, as shown in (Equation 19) and (Equation 20) below. θ Si(2,1) (φ Si,OCP ) = Σx (k) (φ Si,OCP )(k=1,2)...(Formula 19) θ Si(b2,1) (φ Si,OCP ) = Σx (k) (φ Si,OCP ) (k=1, b2) ... (Formula 20)

[0114] The internal state model estimation unit 118 uses the ratios calculated in (Equation 14) and (Equation 16) above to derive the inverse function of the OCP of the discharge-side silicon, which is a combination of (Equation 19) and (Equation 20), as shown in (Equation 21) below. θ Si (φ Si,OCP) = C (2,1) θ Si(2,1) (φ Si,OCP ) + C (b2,1) θ Si(b2,1) (φ Si,OCP )...(Formula 21)

[0115] The internal state model estimation unit 118 calculates φ from the above (Equation 21). Si,OCP θ Si By solving for this, we obtain the OCP function of the silicon on the discharge side. If the OCP function is deformed, we obtain the deformed OCP function (see Figure 19 below).

[0116] The internal state model estimation unit 118 updates φ Si,OCP This is reflected in the above (Equation 5). Si,OCP The update process is repeated.

[0117] The internal state model estimation unit 118 generates the OCP function of the negative electrode by combining the OCP function of graphite and the OCP function of silicon at a predetermined timing, and then generates the SOC-OCV curve of the lithium-ion battery by combining the generated OCP function of the negative electrode and the OCP function of the positive electrode.

[0118] A maximum deformation curve of the SOC-OCV curve, where all silicon in the negative electrode has crystallized, is prepared in advance. When the internal state model estimation unit 118 generates the SOC-OCV curve, it limits the generation range of the SOC-OCV curve so that it does not deform beyond the maximum deformation curve.

[0119] Figure 19 shows a concrete image of the combined OCP function of graphite and silicon, and the combined OCP function of the negative electrode and the positive electrode. Figure 19 also shows an image of the case where the OCP function on the discharge side is deformed.

[0120] The internal state model estimation unit 118 can, at a predetermined timing, take the measured current value included in the battery data as input and output the voltage estimated from the SOC-OCV curve generated by (Equation 5) - (Equation 21) above, using a Kalman filter or the like to feed back the error between the estimated voltage and the measured voltage for each reaction amount. Specifically, the internal state model estimation unit 118 uses the error between the estimated voltage value estimated from the SOC-OCV curve and the measured voltage value included in the battery data to calculate x in (Equation 11) above. (k) , x in the above (Equation 12) cryst Correct it.

[0121] As described above, according to this embodiment, it is determined whether or not silicon crystallization is occurring, and if silicon crystallization is occurring, the SOC-OCV curve is corrected. Alternatively, if silicon crystallization is occurring, the impact of silicon crystallization on the SOC-OCV curve is determined, and if the impact is significant, the SOC estimation is skipped or the estimated SOC is invalidated. This makes it possible to estimate the SOC of a lithium-ion battery using a mixed negative electrode of graphite and silicon with high accuracy.

[0122] The present disclosure has been described above based on embodiments. The embodiments are illustrative, and it will be understood by those skilled in the art that various modifications are possible in combinations of their components and processing processes, and that such modifications are also within the scope of the present disclosure.

[0123] In the above embodiment, an example was described in which the State of Charge (SOC) of a lithium-ion battery is estimated using a battery state estimation system 1 built on a cloud server. However, the functions of the battery state estimation system may also be implemented on the edge side. For example, the battery state estimation system may be implemented in the battery management device 32 within the battery pack 30, or in the control unit 21 of the battery-equipped device 2.

[0124] The embodiments may be specified by the following items.

[0125] [Item 1] A battery state estimation system (1) comprising: a data acquisition unit (111) that acquires time-series battery data including voltage and current of a lithium-ion battery (E1) using a mixed material of graphite and silicon as the negative electrode active material; an SOC (State Of Charge) estimation unit (112) that estimates the SOC of the lithium-ion battery (E1) based on the SOC-OCV (Open Circuit Voltage) curve of the lithium-ion battery (E1) and the voltage of the lithium-ion battery (E1) included in the battery data; and a determination unit (113) that determines whether to enable or disable the SOC estimated using the SOC-OCV curve based on the crystallization state of the silicon. This system can improve the accuracy of SOC estimation of the lithium-ion battery (E1). [Item 2] The battery state estimation system (1) described in Item 1, further comprising a crystallinity map (123) describing the relationship between the charge-discharge pattern of the lithium-ion battery (E1) and the crystallinity of the silicon, wherein the determination unit (113) estimates the crystallinity of the silicon based on the crystallinity map (123) and the charge-discharge transition based on the battery data, and if the estimated crystallinity of the silicon is above a threshold, the SOC estimation unit (112) skips the SOC estimation using the SOC-OCV curve.[Item 3] The battery state estimation system (1) according to Item 1, further comprising: a crystallinity map (123) describing the relationship between the charge-discharge pattern of the lithium-ion battery (E1) and the crystallinity of the silicon; and an OCV offset map (124) describing the relationship between the crystallinity of the silicon and the OCV offset amount for each SOC, wherein the determination unit (113) estimates the crystallinity of the silicon based on the crystallinity map (123) and the charge-discharge transition based on the battery data; estimates the SOC based on the SOC-OCV curve corrected based on the estimated crystallinity of the silicon and the OCV estimated from the voltage included in the battery data; estimates the OCV offset amount based on the estimated SOC, the estimated crystallinity of the silicon and the OCV offset map (124); and invalidates the estimated SOC if the estimated OCV offset amount is greater than or equal to a threshold. According to this, if the effect of silicon crystallization is small, the SOC estimated using the SOC-OCV curve can be made valid, and if the effect of silicon crystallization is large, the SOC estimated using the SOC-OCV curve can be made invalid. [Item 4] The determination unit (113) calculates the degree of deviation between the reference SOC-CCV (Closed Circuit Voltage) curve when the lithium-ion battery (E1) is discharged in a state where the silicon has not crystallized and the SOC-CCV curve when the lithium-ion battery (E1) is discharged based on the battery data, and if the calculated degree of deviation is greater than or equal to a threshold, the SOC estimation unit (112) skips the SOC estimation using the SOC-OCV curve, the battery state estimation system (1) according to Item 1. According to this, by referring to the degree of deviation from the reference SOC-CCV curve, it is possible to accurately determine whether or not highly accurate SOC estimation is possible using the SOC-OCV curve.[Item 5] The determination unit (113) calculates the degree of deviation between a reference SOC-CCV (Closed Circuit Voltage) curve when the lithium-ion battery (E1) is discharged in a state where the silicon has not crystallized and the SOC-CCV curve when the lithium-ion battery (E1) is discharged based on the battery data, and the battery state estimation system (1) further includes an OCV offset amount map (126) for each SOC that describes the relationship between the degree of deviation between the reference SOC-CCV curve and the SOC-CCV curve when discharged under multiple charge-discharge conditions and the OCV offset amount, and the determination unit (113) estimates the SOC based on the SOC-OCV curve corrected based on the calculated degree of deviation and the OCV estimated from the voltage included in the battery data. A battery state estimation system (1) according to item 1, wherein the OCV offset amount is estimated based on the estimated SOC, the calculated deviation degree, and the OCV offset amount map (126), and if the estimated OCV offset amount is greater than or equal to a threshold, the estimated SOC is invalidated.According to this, if the effect of silicon crystallization is small, the SOC estimated using the SOC-OCV curve can be made valid, and if the effect of silicon crystallization is large, the SOC estimated using the SOC-OCV curve can be invalidated.[Item 6] A battery state estimation system (1) according to item 1, further comprising an internal state model estimation unit (118) that estimates the amount of lithium inserted into the silicon based on the internal model of the lithium-ion battery (E1) and the current included in the battery data, wherein if the estimated amount of lithium inserted into the silicon is greater than or equal to a threshold, the SOC estimation unit (112) skips the SOC estimation using the SOC-OCV curve. According to this, by estimating the amount of lithium inserted into silicon using the theoretical formula for lithium-ion batteries (E1), it is possible to accurately determine whether or not high-precision SOC estimation is possible using the SOC-OCV curve.[Item 7] A battery state estimation method comprising: acquiring time-series battery data including voltage and current of a lithium-ion battery (E1) using a mixed material of graphite and silicon as the negative electrode active material; estimating the state of charge (SOC) of the lithium-ion battery (E1) based on the SOC-OCV curve of the lithium-ion battery (E1) and the voltage of the lithium-ion battery (E1) included in the battery data; and determining whether to enable or disable the SOC estimated using the SOC-OCV curve based on the crystallization state of the silicon. According to this method, the accuracy of SOC estimation of the lithium-ion battery (E1) can be improved. [Item 8] A battery state estimation program that causes a computer to perform the following steps: acquiring time-series battery data including voltage and current of a lithium-ion battery (E1) using a mixed material of graphite and silicon as the negative electrode active material; estimating the State of Control (SOC) of the lithium-ion battery (E1) based on the SOC-OCV curve of the lithium-ion battery (E1) and the voltage of the lithium-ion battery (E1) included in the battery data; and determining whether to enable or disable the SOC estimated using the SOC-OCV curve based on the crystallization state of the silicon. This program makes it possible to improve the accuracy of SOC estimation for lithium-ion batteries (E1).

[0126] This disclosure is applicable to lithium-ion batteries using a mixed material of graphite and silicon as the negative electrode active material.

[0127] 2 Battery-equipped equipment, 4 Commercial power grid, 5 Network, 21 Control unit, 22 Load unit, 23 Charging unit, 24 Communication unit, 30 Battery pack, 31 Battery assembly, 32 Battery management device, 33 Measurement unit, 34 Control unit, E1-En Cell, Rs Shunt resistor, SW1 Switch, 1 Battery state estimation system, 111 Data acquisition unit, 112 SOC estimation unit, 113 Judgment unit, 114 Correction unit, 115 SOH estimation unit, 116 Degradation prediction formula generation unit, 117 Life prediction unit, 118 Internal state model estimation unit, 12 Storage unit, 121 Battery data holding unit, 122 SOC-OCV curve holding unit, 123 Crystallinity map holding unit, 124 Crystallinity-OCV offset amount map holding unit, 125 Reference SOC-CCV curve holding unit, 126 Degree of deviation-OCV offset amount map holding unit, 13 Communication unit.

Claims

1. A battery state estimation system comprising: a data acquisition unit that acquires time-series battery data including voltage and current of a lithium-ion battery using a mixed material of graphite and silicon as the negative electrode active material; an SOC (State of Charge) estimation unit that estimates the SOC of the lithium-ion battery based on the SOC-OCV (Open Circuit Voltage) curve of the lithium-ion battery and the voltage of the lithium-ion battery included in the battery data; and a determination unit that determines whether to enable or disable the SOC estimated using the SOC-OCV curve based on the crystallization state of the silicon.

2. The battery state estimation system according to claim 1, further comprising a crystallinity map describing the relationship between the charge-discharge pattern of the lithium-ion battery and the crystallinity of the silicon, wherein the determination unit estimates the crystallinity of the silicon based on the crystallinity map and the charge-discharge transition based on the battery data, and if the estimated crystallinity of the silicon is above a threshold, the SOC estimation unit skips the SOC estimation using the SOC-OCV curve.

3. A battery state estimation system according to claim 1, further comprising: a crystallinity map describing the relationship between the charge / discharge pattern of the lithium-ion battery and the crystallinity of the silicon; an OCV offset amount map describing the relationship between the crystallinity of the silicon and the OCV offset amount for each SOC, wherein the determination unit estimates the crystallinity of the silicon based on the crystallinity map and the charge / discharge transition based on the battery data; estimates the SOC based on the SOC-OCV curve corrected based on the estimated crystallinity of the silicon and the OCV estimated from the voltage included in the battery data; estimates the OCV offset amount based on the estimated SOC, the estimated crystallinity of the silicon and the OCV offset amount map; and invalidates the estimated SOC if the estimated OCV offset amount is greater than or equal to a threshold.

4. The determination unit calculates the degree of deviation between a reference SOC-CCV (Closed Circuit Voltage) curve when the lithium-ion battery is discharged while the silicon is not crystallized and the SOC-CCV curve when the lithium-ion battery is discharged based on the battery data, and if the calculated degree of deviation is greater than or equal to a threshold, the SOC estimation unit skips SOC estimation using the SOC-OCV curve, the battery state estimation system according to claim 1.

5. The determination unit calculates the degree of deviation between a reference SOC-CCV (Closed Circuit Voltage) curve when the lithium-ion battery is discharged while the silicon is not crystallized and an SOC-CCV curve when the lithium-ion battery is discharged based on the battery data; the battery state estimation system further comprises, for each SOC, an OCV offset amount map that describes the relationship between the degree of deviation between the reference SOC-CCV curve and the SOC-CCV curve when discharged under multiple charge / discharge conditions and the OCV offset amount; the determination unit estimates the SOC based on the SOC-OCV curve corrected based on the calculated degree of deviation and the OCV estimated from the voltage included in the battery data; estimates the OCV offset amount based on the estimated SOC, the calculated degree of deviation and the OCV offset amount map; and invalidates the estimated SOC if the estimated OCV offset amount is greater than or equal to a threshold; the battery state estimation system according to claim 1.

6. The battery state estimation system according to claim 1, further comprising an internal state model estimation unit that estimates the amount of lithium inserted into the silicon based on the internal model of the lithium-ion battery and the current included in the battery data, wherein if the estimated amount of lithium inserted into the silicon is greater than or equal to a threshold, the SOC estimation unit skips SOC estimation using the SOC-OCV curve.

7. A battery state estimation method comprising: acquiring time-series battery data including voltage and current of a lithium-ion battery using a mixed material of graphite and silicon as the negative electrode active material; estimating the state of charge (SOC) of the lithium-ion battery based on the SOC-OCV curve of the lithium-ion battery and the voltage of the lithium-ion battery included in the battery data; and determining whether to enable or disable the SOC estimated using the SOC-OCV curve based on the crystallization state of the silicon.

8. A battery state estimation program that causes a computer to perform the following steps: acquire time-series battery data including voltage and current of a lithium-ion battery using a mixed material of graphite and silicon as the negative electrode active material; estimate the state of charge (SOC) of the lithium-ion battery based on the SOC-OCV curve of the lithium-ion battery and the voltage of the lithium-ion battery included in the battery data; and determine whether to enable or disable the SOC estimated using the SOC-OCV curve based on the crystallization state of the silicon.

Images