JP2018146343A - Battery management device and battery management method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a technique that improves the accuracy of estimating a SOC and makes it possible to correct the SOC even in an SOC region where a change amount of OCV is small.SOLUTION: A battery management device of the present invention includes estimation means for estimating, by a prescribed scheme, the charge rate of a battery that includes a region where the fluctuation width of a battery voltage per unit charge rate transitions within a prescribed range in a correlation between the battery voltage in an unloaded state and a charge rate. The battery management device comprises: acquisition means for acquiring information for estimating the operation state of the battery including at least the terminal voltage of the battery; means for defining the resistance and electrostatic capacity on the low frequency response side of an equivalent circuit model that simulates the operation state of the battery, or the resistance or the electronic capacity as a function of the charge rate; and means for correcting the charge rate of the battery estimated by the estimation means using a charge rate derived from the defined function and the acquired information for estimating the operation state.SELECTED DRAWING: Figure 7

Description

  The present invention relates to a battery management apparatus and a battery management method.

  2. Description of the Related Art In recent years, a battery that can be charged / discharged (hereinafter also referred to as a “secondary battery”) has been reduced in size, weight, and energy density. With the above progress, transportation equipment using a battery as an energy source for driving is becoming widespread. Examples of such transport equipment include vehicles such as electric transport equipment (EV: Electric Vehicle), plug-in hybrid transport equipment (PHV: Plug-in Hybrid Electric Vehicle, PHEV: Plug-in Hybrid Electric Vehicle). .

In the above vehicle, for example, in order to grasp the travelable distance corresponding to the amount of energy stored in the battery, estimation of the state of charge (SOC) of the battery is performed. The estimation of the SOC is performed by a battery management device connected to the battery.

  As a method for estimating the SOC, it is known that the SOC is estimated from the terminal voltage of the battery in operation using the characteristic that the electromotive voltage of the battery changes according to the SOC. In the above estimation method, for example, a table (also referred to as “OCV-SOC characteristics”) that associates the relationship between the open circuit voltage (OCV) and the SOC, in which the inside of the battery is in an equilibrium state when there is no load. Is used. For example, the vehicle obtains the OCV based on the relationship between the terminal voltage and the OCV obtained experimentally in advance, and estimates the SOC corresponding to the OCV using the table. It is known that the OCV-SOC characteristic of a battery depends on, for example, elements (electrode material, electrolyte, active material, etc.) constituting the battery, and does not depend on the temperature of the battery or the degree of deterioration.

  Note that the internal state of the battery in operation varies mainly depending on the usage environment such as the temperature environment and the amount of charge / discharge current. For this reason, in a vehicle, for example, the charge / discharge current of the battery is time-integrated using the Coulomb count method or the like to obtain the current charge, and the table correction (hereinafter referred to as “SOC correction”) is performed to match the OCV-SOC characteristics. Also called).

  In addition, the following patent documents exist as prior art documents in which technologies related to the technologies described in this specification are described.

JP 2005-201743 A JP 2006-30080 A JP 2011-38857 A JP 2012-220199 A

  In the conventional SOC correction method, when the battery state indicated by the observed terminal voltage and the accumulated charge / discharge current value shifts from a flat part to an inclined part on the OCV-SOC characteristic, for example, the OCV-SOC The SOC correction is performed so as to match the characteristics. Here, the inclined portion is a region where the amount of change in the OCV per unit charge rate that appears on the high SOC side and the low SOC side in the OCV-SOC characteristic of the battery is large.

  For example, in a lithium ion battery, the inclined portion on the high SOC side often appears in an SOC region of 95% or more, and the inclined portion on the low SOC side appears in an SOC region of less than 45%. In a lithium ion battery, the change amount of OCV per unit charge rate is small in the flat region where the change amount of OCV per unit charge rate is small and the OCV-SOC characteristic is between 45% and less than 95%. In some cases, SOC correction cannot be performed.

  Early SOC correction is required in order to improve the estimation accuracy of the SOC of the battery in operation and accurately grasp the remaining amount of energy stored in the battery. An object of the present invention is to provide a technique that improves SOC estimation accuracy and enables SOC correction even in an SOC region where the amount of change in OCV is small.

  One aspect of the disclosed technology is exemplified by a battery management device. The battery management device estimates the charging rate of a battery including a region where the fluctuation range of the battery voltage per unit charging rate changes within a predetermined range in a correlation between the battery voltage and the charging rate in a no-load state by a predetermined method. Estimating means for performing The battery management device includes at least a terminal voltage of a battery, acquisition means for acquiring information for estimating the operating state of the battery, resistance and electrostatic on the low frequency response side of an equivalent circuit model that simulates the operating state of the battery Estimated by the estimator using the means to define the capacity or resistance or capacitance as a function of the charge rate, and the charge rate derived from the defined function and the information to estimate the obtained operating state Means for correcting the charging rate of the battery.

  According to one aspect of the disclosed technology, there is provided a technology that improves SOC estimation accuracy and enables SOC correction even in an SOC region where the amount of change in OCV is small.

It is a figure which shows an example of the block configuration of the battery management system which concerns on this embodiment. It is a figure which shows an example of the hardware constitutions of a battery management apparatus. It is a figure explaining an OCV-SOC characteristic. It is a figure which shows an example of an equivalent circuit model. It is a figure which shows the relationship between each resistance at the time of charging / discharging, and SOC. It is a figure which shows the relationship between each capacitor | condenser and SOC at the time of charging / discharging. It is a flowchart which shows an example of a SOC estimation process. It is a figure which shows an example of the curve approximation of the total resistance-SOC characteristic in a modification. It is a figure which shows an example of the terminal voltage-SOC characteristic at the time of charging / discharging in the equivalent circuit model of the modification 2.

  Hereinafter, a battery management device according to an embodiment will be described with reference to the drawings. The configuration of the following embodiment is an exemplification, and the battery management apparatus is not limited to the configuration of the embodiment.

<Embodiment>
[1. System configuration〕
FIG. 1 is a diagram illustrating an example of a block configuration of a battery management system according to the present embodiment. A battery management system (BMS: Battery Management System) 1 shown in FIG. 1 includes a battery 2, a sensor 3, and a battery management device 10. Battery management device 10 includes an SOC estimation unit 101.

In the battery management system 1 of FIG. 1, a battery state such as a charge / discharge state or a deterioration state (SOH: State Of Health) of the battery 2 in operation is managed. The battery management system 1 is mounted on a vehicle such as an EV, a PHV, or a PHEV that uses the battery 2 as an energy source for driving. In the following description, a battery management system 1 provided in a vehicle such as an EV will be described. However, the battery management system 1 is used in an information device such as a smartphone or a PC (Personal Computer) that uses a chargeable / dischargeable battery 2 as an energy source. It may be provided, or may be provided in a power generation facility that stores electric power generated using natural energy such as sunlight or wind power.

  The battery 2 is a chargeable / dischargeable secondary battery such as a lithium ion battery or a nickel metal hydride battery. A sensor 3 is attached to the battery 2. The sensor 3 is an acquisition unit that acquires information for estimating the state of the battery 2. The sensor 3 includes a plurality of types of sensor devices. For example, the sensor 3 includes a current sensor that detects a charge / discharge current of the battery 2, a voltage sensor that detects a terminal voltage of the battery 2 in operation, and a temperature sensor that detects the temperature of the battery 2.

  The sensor 3 detects information for estimating the state of the battery 2 at a predetermined periodic interval such as 4 ms. Various detection results for estimating the state of the battery 2 detected via the sensor 3 are output to the battery management device 10. Various detection results are output to the battery management apparatus 10 with the time information which shows the time which detected the sensor information of the battery 2, for example.

  The battery management device 10 acquires various detection results detected via the sensor 3. The battery management device 10 manages the battery state such as the current charge / discharge state or deterioration state of the battery 2 based on the acquired various detection results. For example, the battery management apparatus 10 estimates the SOC based on various detection results detected via the sensor 3 and grasps the current battery state of the battery 2. The SOC estimation is provided by the SOC estimation unit 101. The SOC is expressed as a ratio of the remaining capacity to the full charge capacity of the battery 2, for example.

[2. Device configuration〕
FIG. 2 is a diagram illustrating an example of a hardware configuration of the battery management apparatus 10 according to the present embodiment. As illustrated in FIG. 2, the battery management device 10 includes a CPU (Central Processing Unit) 11, a main storage device 12, and an auxiliary storage device 13 that are connected to each other via a connection bus 16.
, And a communication IF (Interface) 14 and an input / output IF 15.

  The CPU 11 is also called an MPU (Microprocessor) or a processor. However, the CPU 11 is not limited to a single processor and may have a multiprocessor configuration. A single CPU connected by a single socket may have a multi-core configuration. The CPU 11 is a central processing unit that controls the battery management apparatus 10 as a whole. For example, the CPU 11 develops a program stored in the auxiliary storage device 13 to be executable in the work area of the main storage device 12, and provides a function that matches a predetermined purpose by controlling peripheral devices through the execution of the program. To do.

The main storage device 12 is a storage medium in which the CPU 11 caches programs and data and develops a work area. The main storage device 12 includes, for example, a flash memory, a RAM (Random Access Memory), and a ROM (Read Only Memory). The auxiliary storage device 13 is a storage medium that stores programs executed by the CPU 11, operation setting information, and the like. The auxiliary storage device 13 is, for example, an HDD (Hard-disk Drive) or an SSD (Solid State D).
rive), EPROM (Erasable Programmable ROM), flash memory, USB memory, SD (Secure Digital) memory card, and the like. The communication IF 14 is an interface with a network connected to the battery management apparatus 10. The network connected to the battery management device 10 includes a CAN (Controller Area Network), a LIN (Local Interconnect Network), and the like. The input / output IF 15 is an interface for inputting / outputting data to / from a sensor or device connected to the battery management apparatus 10.

In the battery management system 1, information indicating the state of the battery 2 detected by the sensor 3 is acquired via the input / output IF 15. In addition, data is input / output to / from various ECUs (Electronic Control Units) connected to the network via the communication IF 14. A plurality of the above components may be provided, or some of the components may not be provided. Further, the above-described components may constitute a part of various ECUs.

The battery management device 10 is provided with a processing function including the SOC estimation unit 101 by executing a program of the CPU 11. However, at least a part of the processing functions may be provided by a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), or the like. In addition, at least a part of each of the processing functions may be a dedicated LSI (large scale integration) such as a field-programmable gate array (FPGA) or other digital circuit. In addition, an analog circuit may be included in at least a part of each processing function.

  The battery management apparatus 10 includes an auxiliary storage device 13 as a storage destination of data to be referred to or managed by the processing function. The auxiliary storage device 13 is defined using OCV-SOC characteristic map information described in FIG. 3, the resistance of the low-frequency RC circuit of the equivalent model described in FIGS. Contains functions. The information stored in the auxiliary storage device 13 is read together with the execution of the program, and is temporarily stored in a predetermined work area of the main storage device 12.

[3. Processing function)
FIG. 3 is a diagram for explaining the OCV-SOC characteristic of the battery 2. The OCV-SOC characteristic data shown in FIG. 3 is an example of a lithium ion battery. In FIG. 3, the vertical axis represents OCV (unit: V), and the horizontal axis represents SOC (unit: percent). In FIG. 3, a curve connected with “♦” represents a characteristic on the discharge side of the electric energy stored in the battery 2, and a curve connected with “■” represents a charge for storing electric energy in the battery 2. Represents the characteristics on the side. It is known that the OCV-SOC characteristic of a battery depends on, for example, elements (electrode material, electrolyte, active material, etc.) constituting the battery, and does not depend on the temperature of the battery or the degree of deterioration.

  As shown in FIG. 3, in the OCV-SOC characteristic, the characteristic curve on the discharge side and the characteristic curve on the charge side tend to change approximately, and each is a region where the change amount of the OCV per unit charging rate is large. It has an inclined part. In addition, it has a flat portion that is a region in which the OCV per unit charging rate is roughly constant without changing (or fluctuates within a predetermined minute range). In FIG. 3, the inclined portion appears in a high SOC region of 95% or more and a low SOC region of less than 45%. A region where the SOC is 45% or more and less than 95% is a flat portion. Such OCV-SOC characteristics are mapped and held in the main storage device 12 of the battery management device 10.

In the management of the battery state, when charging / discharging is repeated in the flat part, since the OCV with respect to the SOC changes roughly and constant, it is difficult to estimate the current SOC based on the terminal voltage of the battery 2, for example. It is. For this reason, a method of estimating the SOC by the Coulomb count method or the like is used.

  In the case of the coulomb counting method, the charge / discharge current is integrated over time, and the SOC is estimated from the integrated value of the electric charge (current) that enters and exits the battery 2 at the present time. However, the SOC estimation by the Coulomb count method or the like includes an error, and there is a possibility that the estimated SOC is deviated from the true value due to this error. For this reason, when shifting from the flat portion to the inclined portion, SOC correction has been performed in which the SOC value estimated by the Coulomb count method or the like is updated to the SOC estimated by the OCV-SOC characteristic. In the following description, the Coulomb count method will be described as an example of SOC estimation by the Coulomb count method or the like.

  As shown in FIG. 3, the SOC at the boundary between the flat portion and the inclined portion in the high SOC region is 95% or more. For this reason, at the time of charge, there is a possibility that overcharge in which the charged state continues despite the full charge capacity of the battery 2 has occurred. Further, in the low SOC region of less than 45%, when the ratio is 10% or less, the change slope of the OCV per unit charge rate becomes steep. For this reason, at the time of discharging, the SOC becomes 0%, and there is a possibility that the vehicle 2 or the like using the battery 2 as an energy source for driving unexpectedly stops or the battery 2 is overdischarged.

  Overcharge and overdischarge promote deterioration of the battery 2 and shorten the battery life. From the viewpoint of battery state management, it is preferable to perform the above SOC correction at an early stage in the flat SOC region before reaching the high SOC region of 95% or more in order to suppress overcharge.

  The battery management device 10 according to the present embodiment determines that the current battery state is a flat portion on the OCV-SOC characteristic based on the terminal voltage and charge / discharge current of the battery 2 observed through the sensor 3. To do. When the current battery state is in the flat SOC region, the battery management device 10 sets each parameter such as resistance and capacity using an equivalent circuit model that simulates the battery 2 in the operating state. Estimate the terminal voltage. The terminal voltage is estimated using the equivalent circuit model as a predicted value of the battery state observed next time. Then, the battery management device 10 obtains the SOC using the error (deviation) between the observed battery state and the predicted battery state, and performs the above-described SOC correction, so that the early SOC correction in the flat SOC region is performed. Enable.

  FIG. 4 is a diagram illustrating an example of an equivalent circuit model that simulates the battery 2 in operation. In general, an equivalent circuit model that simulates the battery 2 in operation can be represented by two elements, that is, an open circuit voltage (OCV) of the battery 2 and an internal impedance. The internal impedance is further expressed by three elements: an ion migration process in the electrolyte, a charge transfer process for transferring electrons at the electrode-electrolyte interface, and an ion diffusion process at the interface. Here, in the battery 2 in operation, the charge transfer process for transferring electrons at the interface between the electrode and the electrolyte, and the ion diffusion process at the interface are transient phenomena approximated as RC parallel circuits. The former has a time constant. A transient response with a time constant on the order of seconds to thousands of milliseconds, from a few milliseconds to hundreds of milliseconds. The iontophoretic process in the electrolyte is expressed as ohmic resistance.

As shown in FIG. 4, in the equivalent circuit model according to the present embodiment, the open circuit voltage of the battery 2 is a battery voltage (Vovv) in an unloaded state, and the iontophoretic process in the electrolyte of the battery 2 in operation is a resistance (R0). ). Further, the transient phenomenon in the operating battery 2 is expressed as a low frequency side RC circuit (R1, C1) and a high frequency side RC circuit (R2, C2). The low-frequency RC circuit (R1, C1) represents the ion diffusion process at the interface, and the high-frequency RC circuit (R2, C2) represents the charge transfer process for transferring electrons at the electrode-electrolyte interface. In FIG. 4, the current (I) represents the current of the battery 2 observed through the sensor 3, and the terminal voltage (V) represents the terminal voltage of the battery 2 observed through the sensor 3.

  Based on the equivalent circuit model shown in FIG. 4, the battery 2 was charged and discharged. As a result, the results shown in FIGS. 5 and 6 were obtained. FIG. 5 is a diagram showing the relationship between each resistance (R0, R1, R2) of the battery 2 and the SOC at the time of charging / discharging, the vertical axis represents the resistance value (mΩ), and the horizontal axis represents the SOC. FIG. 6 is a diagram showing the relationship between each capacitor (C1, C2) and SOC of the battery 2 at the time of charging / discharging. The vertical axis represents the capacitor (F), and the horizontal axis represents the SOC.

  In FIG. 5, the curve connected with “◯” represents the SOC characteristic of the resistor R2 during discharging, and the curve connected with “▲” represents the SOC characteristic of the resistor R2 during charging. Similarly, a curve connected with “x” represents the SOC characteristic of the resistor R0 during discharging, and a curve connected with “♦” represents the SOC characteristic of the resistor R0 during charging. A curve connected with “*” represents the SOC characteristic of the resistor R1 during discharging, and a curve connected with “■” represents the SOC characteristic of the resistor R1 during charging.

  As shown in FIG. 5, the resistance R2 at the time of charging / discharging has a change tendency which decreases within the range of 0.9 mΩ to 0.8 mΩ as the SOC increases. Similarly, the resistance R0 at the time of charging / discharging has a change tendency which decreases within the range of 0.8 mΩ to 0.7 mΩ as the SOC increases. Further, the resistance R1 at the time of charging / discharging has a changing tendency to decrease within the range of 0.8 mΩ to 0.3 mΩ as the SOC increases.

  Comparing the reduction characteristics of each resistance with increasing SOC, it can be seen that the rate of change of the resistance R1 is relatively largest. It can also be seen that the decreasing change tendency of the resistance R1 when the SOC is around 40 percent to 90 percent changes depending on the increase / decrease change in the SOC.

  Next, in FIG. 6, the curve connected with “♦” represents the SOC characteristic of the capacitor C1 during charging, and the curve connected with “▲” represents the SOC characteristic of the capacitor C1 during discharging. Similarly, the curve connected with “■” represents the SOC characteristic of the capacitor C2 during charging, and the curve connected with “x” represents the SOC characteristic of the capacitor C2 during discharging. In FIG. 6, the SOC characteristic of the capacitor C2 during charging and the SOC characteristic of the capacitor C2 during discharging overlap.

  As shown in FIG. 6, the capacitor C1 at the time of charging / discharging has a change tendency which increases in the range from about 120F to about 280F as the SOC increases. On the other hand, the capacitor C2 at the time of charging / discharging is constant at rough 0F, and does not have a change tendency with an increase in SOC. As the change tendency of the capacitor C1 with the increase in the SOC, it can be seen that the change rate in the region where the SOC is 70% or more is larger than the change rate in the region where the SOC is less than 70%.

  As shown in FIGS. 5 and 6, even if the current battery state is a flat region on the OCV-SOC characteristic, the low frequency side RC circuit (R1, C1) shown in the equivalent circuit model is It turns out that it depends on an increase change and a decrease change.

  When the current battery state is in the flat SOC region, the SOC estimation unit 101 of the battery management apparatus 10 according to the present embodiment has constants (R1, R1) of the low frequency side RC circuit of the equivalent circuit model shown in FIG. C1) is defined as a function that depends on the SOC. That is, for the constants (R1, C1) of the low frequency side RC circuit, a function using the SOC as a parameter so as to have the characteristics shown in FIG. 5 and FIG. When the battery state is in the flat SOC region, constants (R1, C1) of the low frequency side RC circuit are estimated based on this function.

  Then, the SOC estimation unit 101 of the battery management device 10 derives an estimation formula representing the terminal voltage of the battery 2 in operation based on R1 and C1 of the low frequency side RC circuit defined as a function of the SOC. The SOC estimation unit 101 of the battery management apparatus 10 according to the present embodiment obtains the SOC for the current battery state based on the error between the estimation result obtained by the estimation expression representing the terminal voltage and the observation result, and performs the SOC correction.

[4. Estimation function based on equivalent circuit model]
An estimation formula indicating the battery state of the battery 2 in operation based on the equivalent circuit model shown in FIG. 4 is expressed by the following formula (1).

  In equation (1), the state function: x ′ (k) represents the battery state of the battery 2 that is operating at the time of observation, and the argument: (k) estimates the battery state detected via the sensor 3. Represents a sampling sequence of information (for example, terminal voltage, current, temperature, etc.). The information for estimating the battery state is acquired, for example, at a predetermined constant cycle (Δt) interval. In equation (1), the variable: v1 (k) represents the voltage across the low frequency side RC circuit (R1, C1), and the variable: v2 (k) represents the voltage across the high frequency side RC circuit (R2, C2). The variable (Sca) represents the full charge capacity (rated capacity) of the battery 2, and the variable i represents the current (I) observed via the sensor 3 at regular intervals.

As described with reference to FIGS. 5 and 6, the low-frequency RC circuit (R1, C1) is defined as a function that depends on the SOC. For example, if the provider of the battery management apparatus 10 experimentally measures the characteristics illustrated in FIGS. 5 and 6 and defines a function depending on the SOC based on the experimental result, as shown in Equation (2), Good.

In the definition of the function based on the experimental result, the provider of the battery management device 10 may define a plurality of functions corresponding to the temperature of the battery 2. As described with reference to FIGS. 5 and 6, it is estimated that the high frequency side RC circuit (R2, C2) has a low dependency on the SOC. For this reason, the value of the high frequency side RC circuit (R2, C2) may be a constant obtained experimentally in advance. Similarly to the low frequency side RC circuit (R1, C1), the provider of the battery management device 10 has a plurality of constants corresponding to the temperature of the battery 2 for the value of the high frequency side RC circuit (R2, C2). It is good.

Moreover, the observation prediction formula of the battery 2 in operation based on the equivalent circuit model shown in FIG. 4 is expressed by the following formula (3).

  In Expression (3), the predicted terminal voltage: Vp (k) represents the predicted value of the terminal voltage of the battery 2 at the observation time. Further, the resistance R0 in the equation (2) is expressed as a constant obtained experimentally in advance. The value of the resistance R0 is specified by elements (electrode material, electrolyte, active material, etc.) constituting the battery. In addition, each of the variables: v1 (k), v2 (k), and SOC (k) in equation (3) is estimated by equation (1).

  The SOC estimation unit 101 of the battery management apparatus 10 predicts the OCV (k) shown in the equation (3) from the estimated SOC estimated by the equation (1), and the OCV (k) and the variable: v1 (k), Predicted terminal voltage: Vp (k) is predicted from v2 (k). Here, OCV (k) is obtained from the mapped OCV-SOC characteristic described with reference to FIG. 3 (charge / discharge curve fitting).

As shown in equation (4), the SOC estimation unit 101 of the battery management apparatus 10 predicts the predicted terminal voltage: Vp (k) predicted by equation (3) and the terminal voltage measured at the observation time: Vt (k). And compare. When the predicted terminal voltage: Vp (k) predicted is different from the terminal voltage: Vt (k) measured at the time of observation, the SOC estimation unit 101 obtains an estimation error: y (k) Estimation error: The estimated SOC estimated by Expression (1) is corrected based on y (k). Here, the estimated SOC estimated by Expression (1) is also referred to as “preliminarily estimated SOC”, and the SOC corrected by Expression (4) is also referred to as “post-mortem estimated SOC”. In the battery management device 10, the SOC estimated by the coulomb counting method or the like is corrected using the post-estimation SOC.

In Equation (4), Kalman gain: G (k) is derived by an extended Kalman filter. The SOC estimation unit 101 of the battery management apparatus 10 includes the prior error covariance shown in Expression (5): P￣ (k), the Kalman gain derivation expression shown in Expression (6), and the posterior error covariance shown in Expression (7). A Kalman gain: G (k) shown in Expression (4) is derived using P (k).

The SOC estimation unit 101 of the battery management apparatus 10 repeats the above processing every time an input update of information (current, voltage, temperature, etc.) for estimating the battery state observed via the sensor 3 is performed. As a result, in the battery management apparatus 10 of the present embodiment, as described with reference to FIGS. 5 and 6, for example, in the region where the SOC is 70% or more, the predicted terminal voltage accompanying the change tendency of C1 and R1 : The amount of change in Vp (k) is captured to estimate the SOC, and the SOC estimated by the Coulomb count method can be corrected.
Similarly, in the battery management apparatus 10, for example, in the region where the SOC is less than 70%, the SOC is estimated by capturing the amount of change of the predicted terminal voltage: Vp (k) mainly accompanying the change tendency of R1, and the coulomb count This makes it possible to correct the estimated SOC.
In the battery management apparatus 10 of the present embodiment, based on the SOC dependency of the low-frequency RC circuit (R1, C1) shown in the equivalent circuit model, it is possible to perform SOC correction in the flat region on the OCV-SOC characteristic. Become. The above embodiment has shown the example which correct | amends SOC estimated by the Coulomb count method with SOC estimated using the Kalman filter.

[5. Processing flow
Hereinafter, the SOC estimation processing according to the present embodiment will be described with reference to FIG. FIG. 7 is a flowchart illustrating an example of the SOC estimation process provided by the battery management apparatus 10. The battery management device 10 according to the present embodiment provides the SOC estimation process illustrated in FIG. 7 by, for example, the CPU 11 and the like reading and executing various programs and various data stored in the auxiliary storage device 13. Note that the SOC estimation process is mainly provided by the SOC estimation unit 101. Further, the CPU 11 or the like of the battery management device 10 develops the OCV-SOC characteristics, various parameters, and the like stored in the auxiliary storage device 13 in a predetermined area of the main storage device 12 so as to be readable, and executes processing.

  In the battery management apparatus 10, the SOC estimation process shown in FIG. 7 is repeated for each input update of information (current, voltage, temperature) indicating the battery state of the battery 2 observed via the sensor 3, for example.

  In the flowchart shown in FIG. 7, the start of the process is exemplified when the input of sensor information (current, terminal voltage, temperature, etc.) for estimating the battery state of the battery 2 is updated. The sensor 3 detects information such as the current, terminal voltage, and temperature of the battery 2 at a predetermined cycle interval of 4 ms, for example. Various detection results for estimating the state of the battery 2 detected via the sensor 3 are output to the battery management apparatus 10 together with time information indicating the detection time. For example, the battery management device 10 temporarily stores various detection results output from the sensor 3 together with time information in a predetermined area of the main storage device 12.

  The battery management device 10 refers to the main storage device 12, and determines the current change of the battery 2 during a predetermined period (S1). Here, the predetermined period is exemplified by the unit time (for example, 100 ms) in the latest 10 minutes. The current change in unit time can be expressed as, for example, a current integration value in unit time. The battery management device 10 determines the polarization generated in the electrode of the battery 2 (a charge transfer process for transferring electrons at the interface between the electrode and the electrolyte, an ion diffusion process at the interface) based on the current integral value per unit time in a predetermined period. ) Is resolved.

  When the current integrated value per unit time in the latest 10 minutes is equal to or less than a predetermined value (for example, 300 mA), it can be determined that the polarization generated in the electrode of the battery 2 has been eliminated. When it can be determined that the polarization of the battery 2 has been eliminated, the battery management device 10 compares the terminal voltage of the battery 2 detected via the sensor 3 with the OCV-SOC characteristics described with reference to FIG. It can be regarded as a possible OCV.

  On the other hand, when the current integrated value per unit time in the latest 10 minutes is equal to or greater than a predetermined value (for example, 300 mA), it can be determined that the polarization generated in the electrode of the battery 2 has not been eliminated. If it can be determined that the polarization of the battery 2 has not been eliminated, the battery management device 10 estimates the SOC of the battery 2 at the current time by the Coulomb count method.

  If the current integrated value of the unit time in the latest 10 minutes is equal to or smaller than a predetermined value (S1, yes), the battery management device 10 proceeds to the process of S2. On the other hand, when the current integrated value of the unit time in the latest 10 minutes is not equal to or less than a predetermined value (S1, no), the battery management device 10 proceeds to the process of S4.

  In the process of S2, the battery management apparatus 10 employs the terminal voltage of the battery 2 detected via the sensor 3 as the OCV. Then, the battery management device 10 obtains the SOC value based on the mapped OCV-SOC characteristic held in a predetermined area of the main storage device 12, and uses the obtained SOC value (also referred to as SOCm) as a control value. (S3).

In the process of S4, the battery management device 10 time-integrates the observed charging current by the Coulomb count method, and obtains the integrated value of the charge (current) charged in the battery 2 to obtain the current state of the battery 2. Estimate the corresponding SOC. Note that the SOC estimated by the coulomb counting method is also referred to as “SOCc”.

  It should be noted that “yes” in the process of S1 is when the battery voltage is stable, such as when the vehicle is started, and the battery 2 is repeatedly charged and discharged during operation of the vehicle. There are few opportunities to become “yes”. Therefore, the SOCc estimated by the Coulomb method of the process of S4 is used for control at almost all occasions during driving of the vehicle.

  In the process of S5, the battery management device 10 refers to the main storage device 12, and acquires the current value (i) sampled in the latest 100 ms period. The battery management device 10 integrates the acquired current value, and determines that the integrated value is “0” or more. The battery management device 10 determines that the battery 2 is in the charged state when the current integral value in the latest 100 ms period is “0” or more. The battery management device 10 determines that the battery 2 is in a discharged state when the current integrated value in the latest 100 ms period is not “0” or more.

  When the current integrated value in the latest 100 ms period is “0” or more (S5, yes), the battery management device 10 proceeds to the process of S6. On the other hand, when the current integrated value in the latest 100 ms period is not “0” or more (S5, no), the battery management device 10 proceeds to the process of S8.

  In the process of S6, it is determined whether or not the battery 2 at the time of observation in the charged state is a flat area on the OCV-SOC characteristic. That is, it is determined whether or not the current SOC estimated in S3 or S4 corresponds to a flat area on the OCV-SOC characteristic. When the current SOC does not correspond to the flat area on the OCV-SOC characteristic (S6, no), the battery management apparatus 10 ends the SOC estimation process shown in FIG. In the battery management device 10, the SOCm employed in the process of S3 or the SOCc determined in the process of S4 is employed as the SOC of the battery 2 at the time of observation.

  On the other hand, when the current SOC corresponds to the flat area on the OCV-SOC characteristic (S6, yes), the battery management apparatus 10 proceeds to the process of S7.

  In the process of S7, the battery management device 10 uses the equation (2) that simulates the characteristics shown in FIGS. 5 and 6 for the constants (R1, C1) of the low frequency side RC circuit of the equivalent circuit model shown in FIG. Estimate from the SOC dependent function shown. The battery management apparatus 10 then estimates the estimated error between the predicted terminal voltage: Vp (k) and the terminal voltage: Vt (k) measured at the time of observation based on the equations (1) to (7): y (k ) And the estimated SOC estimated by the equation (1) based on the estimated error: y (k) is corrected to obtain a post-estimated SOC.

  The battery management device 10 updates the SOC estimated in step 3 by the post-estimation SOC, or the SOC by the Coulomb count method estimated in step S4. That is, the battery management device 10 corrects the SOC estimated from the OCV-SOC map by the post-estimation SOC or the SOC estimated by the coulomb count method. The battery management apparatus 10 ends the SOC estimation process shown in FIG. 7 after the process of S7.

In the process of S8, it is determined whether or not the battery 2 at the observation time in the discharged state is a flat area on the OCV-SOC characteristic. That is, the battery management device 10 determines whether or not the current SOC estimated in the processing of S3 or S4 corresponds to a flat area on the OCV-SOC characteristic. When the current SOC does not correspond to the flat area on the OCV-SOC characteristic (S8, no), the battery management apparatus 10 ends the SOC estimation process shown in FIG. In the battery management device 10, the SOCm employed in the process of S3, or
The SOCc obtained in the process of S4 is adopted as the SOC of the battery 2 at the time of observation.

  On the other hand, when the current SOC corresponds to the flat area on the OCV-SOC characteristic (S8, yes), the battery management apparatus 10 proceeds to the process of S9.

  In the process of S9, the battery management device 10 calculates the constants (R1, C1) of the low frequency side RC circuit of the equivalent circuit model shown in FIG. 4 by the equation (2) that simulates the characteristics shown in FIGS. Estimate from the SOC dependent function shown. Then, similarly to the process of S7, the battery management device 10 uses the predicted terminal voltage: Vp (k) and the terminal voltage measured at the observation time: Vt (k) based on the formulas (1) to (7). The estimation error: y (k) is obtained, and the estimated SOC estimated by the equation (1) based on the estimation error: y (k) is corrected to obtain the post-estimation SOC.

  The battery management device 10 updates the SOC estimated in step S3 by the post-estimation SOC or the SOC by the Coulomb count method estimated in step 4. Similar to the processing of S7, the battery management device 10 corrects the SOC estimated from the OCV-SOC map by the post-estimation SOC or the SOC estimated by the coulomb count method. The battery management apparatus 10 ends the SOC estimation process shown in FIG. 7 after the process of S9.

  As described above, in the battery management apparatus 10 according to the present embodiment, the battery 2 at the time of observation is updated according to the input update of information (current, terminal voltage, temperature, etc.) for estimating the battery state of the battery 2. It can be determined whether or not corresponds to a flat area on the OCV-SOC characteristic. When the battery 2 at the time of observation corresponds to the flat area, the battery management device 10 determines Vp (k associated with the change tendency of the low-frequency RC circuit (R1, C1) based on the equivalent circuit model. ) And the SOC is estimated, and the SOC estimated by the Coulomb count method can be corrected.

  As a result, in the battery management apparatus 10 of the present embodiment, it is possible to perform SOC correction in the flat region (the SOC region where the amount of change in OCV is small) on the OCV-SOC characteristic. According to the battery management device 10 of the present embodiment, the SOC estimation accuracy is improved.

(Modification)
As described with reference to FIG. 5, each resistance (R0, R1, R2) in the equivalent circuit model has a tendency to change within a predetermined range as the SOC increases. In the battery management apparatus 10 according to the modified example, a correlation characteristic between the total resistance (R0 + R1 + R2) of the battery 2 during charge / discharge and the SOC is experimentally acquired in advance, and the SOC correction is performed by approximating the change characteristic of the total resistance by a curve. It is good.

For example, the relationship between the terminal voltage Vt and the current i at the time of observation can be expressed by the following formula (8). Vocv represents a battery voltage (OCV) in a no-load state.
Terminal voltage Vt = Vocv + (R0 + R1 + R2) * current i Formula (8)

  FIG. 8 is a diagram showing an example of curve approximation of the change characteristic of the total resistance. In FIG. 8, the vertical axis represents resistance (mΩ), and the horizontal axis represents SOC (percentage). In FIG. 8, a curve connected with “♦” represents the total resistance-SOC characteristic during charging, and a curve connected with “■” represents the total resistance-SOC characteristic during discharging. A curve indicated by a solid line is a curve approximation of the total resistance-SOC characteristic during charging, and a curve indicated by a broken line is a curve approximation of the total resistance-SOC characteristic during discharging.

As shown in FIG. 8, the total resistance-SOC characteristic during charging is approximated by a curve as an SOC polynomial: (Y = 9E￣0.5x ² −0.0154x + 2.5462). Further, the total resistance-SOC characteristic at the time of discharging is approximated by a curve as an SOC polynomial: (Y = 6E￣0.5x ² −0.0127x + 2.446).

  In the battery management apparatus 10 of the modification, for example, the total resistance (R0 + R1 + R2) of the battery 2 in operation is obtained using the terminal voltage Vt and the current i at the time of observation and the equation (8). The battery management apparatus 10 according to the modified example specifies an SOC (corresponding to “post-estimated SOC”) corresponding to the total resistance at the time of observation using the approximate curve shown in FIG. And the battery management apparatus 10 of a modification should just correct | amend SOC estimated by the Coulomb count method etc. based on the specified said SOC.

(Modification 2)
Moreover, the battery management apparatus 10 of the modification 2 may experimentally measure the terminal voltage Vt and the current i of the battery 2 at the time of charging / discharging, and map and hold the SOC characteristics corresponding to the measured values. The terminal voltage Vt and current i relating to charging / discharging of the battery 2 obtained experimentally in advance can be regarded as the overall output value of the equivalent model shown in FIG.

  FIG. 9 is a diagram illustrating an example of terminal voltage-SOC characteristics measured based on an equivalent circuit model during charging. The vertical axis in FIG. 9 represents SOC (percent), and the horizontal axis represents terminal voltage (V). In FIG. 9, a curve g1 represents a terminal voltage-SOC characteristic with respect to a charging current: 0.8A. Similarly, the curves g2, g3, g4, g5, g6, g7, g8, g9, g10 are the charging currents: 4A, 8A, 40A, 80A, 112A, 144A, 176A, 208A, 240A, respectively. Represents SOC characteristics.

  As shown by curves g1 to g6 in FIG. 9, it can be seen that there is a dependency on the terminal voltage in the relationship between the terminal voltage and the SOC at the time of charging until the charging current is around 140A. At 140A or higher, the SOC is 90 percent (curve g7: 144A), 70 percent (curve g8: 176A), 50 percent (curve g9: 208A), and 35 percent (curve g10: 240A). It can be seen that there is a dependency on voltage.

  For example, the battery management device 10 maps the terminal voltage-SOC characteristics shown in FIG. 9 for each current value or terminal voltage and stores the mapping in the auxiliary storage device 13. The current value held in the auxiliary storage device 13 or the mapped characteristics for each terminal voltage is developed so as to be readable in a predetermined area of the main storage device 12 when the processing is executed. For example, the battery management device 10 specifies the SOC according to the terminal voltage Vt, current i measured from the battery 2, and the mapped terminal voltage-SOC characteristic. Then, the battery management device 10 may correct the SOC estimated by the Coulomb count method or the like with the specified SOC.

1 Battery management system 2 Battery (chargeable / dischargeable secondary battery)
3 Sensor 10 Battery management device 11 CPU
12 Main storage device 13 Auxiliary storage device 14 Communication IF
15 I / O IF
16 Connection Bus 101 SOC Estimator

Claims (2)

In the correlation between the battery voltage in the no-load state and the charging rate, a battery management device for a battery including a region where the fluctuation range of the battery voltage per unit charging rate changes within a predetermined range,
Estimating means for estimating a charging rate of the battery by a predetermined method;
Obtaining means for obtaining information for estimating an operating state of the battery, including at least a terminal voltage of the battery;
A resistor and capacitance on the low frequency response side of an equivalent circuit model that simulates the operating state of the battery, or means for defining the resistance or capacitance as a function of the charge rate;
Means for correcting the charging rate of the battery estimated by the estimating means using a charging rate derived from the defined function and information for estimating the acquired operating state;
A battery management device comprising: In the correlation between the battery voltage in the no-load state and the charging rate, a battery management method for a battery including a region where the fluctuation range of the battery voltage per unit charging rate changes within a predetermined range,
While estimating the charging rate of the battery by a predetermined method,
Including at least the terminal voltage of the battery, obtaining information for estimating the operating state of the battery,
Define the resistance and capacitance on the low frequency response side of the equivalent circuit model that simulates the operating state of the battery, or define the resistance or capacitance as a function of the charge rate,
Correcting the charging rate of the battery estimated by the predetermined method using the charging rate derived from the defined function and the information for estimating the acquired operating state;
A battery management method.

Images