JP2018100851A - Secondary battery state detection device and secondary battery state detection method - Google Patents

Abstract

A correction error is reduced even when secondary batteries having different characteristics are used. In a secondary battery state detection device 1 for detecting a state of a secondary battery 14, a calculation means (control unit 10) for calculating an element value of an element constituting an equivalent circuit model of the secondary battery, and a calculation means Correction means (control unit 10) for correcting the element value of the equivalent circuit model calculated by the above to the element value when the secondary battery is in the reference state, and the element of the equivalent circuit model in the reference state corrected by the correction means A coefficient that the correction means has based on the estimation means (control unit 10) for estimating the state of the secondary battery based on the value, the element value at that time calculated by the calculation means, and the state value indicating the reference state Adjusting means (control unit 10) for adjusting [Selection] Figure 1

Description

  The present invention relates to a secondary battery state detection device and a secondary battery state detection method.

  As a technique for coping with individual differences and characteristic changes of the secondary battery, for example, there are techniques disclosed in Patent Documents 1 and 2.

  Patent Document 1 discloses a technique for accurately determining a temperature characteristic of an internal impedance of a secondary battery and determining a deterioration state of the secondary battery with high accuracy.

  In Patent Document 2, the internal resistance ratio Zr (= Z (resistance at that time) / Z0 (reference resistance at the time of new article)) is obtained, and the relative SOC value in each Zr is obtained. A technique is disclosed in which a substantially constant SOC is obtained regardless of the type (for example, manufacturer or capacity) or individual of the secondary battery by creating an associated table.

Japanese Patent Laid-Open No. 2005-091217 JP 2012-189373 A

  By the way, in the techniques disclosed in Patent Documents 1 and 2 described above, since the magnitude of the correction error is not evaluated, the correction error is large in the secondary battery having characteristics different from the secondary battery obtained in advance. There is a problem of becoming.

  The present invention has been made in view of the above situation, and a secondary battery state detection device and a secondary battery state detection capable of reducing correction errors even when secondary batteries having different characteristics are used. It aims to provide a method.

In order to solve the above problems, the present invention provides a secondary battery state detection device for detecting a state of a secondary battery, a calculation means for calculating an element value of an element constituting an equivalent circuit model of the secondary battery, Correction means for correcting the element value of the equivalent circuit model calculated by the calculation means to an element value when the secondary battery is in a reference state; and the equivalent circuit in the reference state corrected by the correction means The estimating means for estimating the state of the secondary battery based on the element value of the model, the current element value calculated by the calculating means, and the correction means are adjusted based on the state value indicating the reference state Adjusting means.
According to such a configuration, correction errors can be reduced even when secondary batteries having different characteristics are used.

Further, the present invention is characterized in that the adjustment means adjusts the correction means when it is estimated that an error occurring during correction by the correction means is less than a predetermined threshold value.
According to such a configuration, the correction unit can be adjusted more accurately.

Further, according to the present invention, the element constituting the equivalent circuit model has at least one of a conductor / liquid resistance, a reaction resistance, and an electric double layer capacity, and the state value indicating the reference state is a liquid value It has at least one of temperature, SOC, and OCV.
According to such a configuration, it is possible to detect the state of the secondary battery using an optimum parameter according to the purpose.

In the invention, it is preferable that the correction unit has a correction formula for correcting the element value of the equivalent circuit model to an element value when the secondary battery is in the reference state, and the adjustment unit includes the correction The coefficient of the correction equation when the value obtained by multiplying the gradient obtained by partial differentiation of the equation with the predetermined state value and the estimated error of the state value at that time is less than the predetermined threshold value It is characterized by adjusting.
According to such a configuration, the coefficient can be appropriately adjusted according to the characteristic of the correction formula.

In the invention, it is preferable that the correction unit has a correction formula for correcting the element value of the equivalent circuit model to an element value when the secondary battery is in the reference state, and the adjustment unit The estimated error when correcting the element value in the reference state to the element value in the reference state, the slope obtained by partial differentiation of the correction formula with the predetermined state value, and the estimated error of the state value at that time When the value obtained by multiplying the value of 未 満 is less than a predetermined threshold, the coefficient of the correction equation is adjusted.
According to such a configuration, the coefficient can be appropriately adjusted according to the characteristic of the correction formula and the estimated error value.

Further, the present invention is characterized in that the adjusting means adjusts the correcting means when the polarization state of the secondary battery is substantially eliminated.
According to such a configuration, the correction means can be adjusted more accurately by reducing the influence of polarization.

Further, the present invention provides a secondary battery state detection method for detecting a state of a secondary battery, a calculation step for calculating an element value of an element constituting an equivalent circuit model of the secondary battery, and a calculation in the calculation step. Further, a correction step of correcting the element value of the equivalent circuit model to an element value when the secondary battery is in the reference state, and an element value of the equivalent circuit model in the reference state corrected in the correction step An estimation step for estimating the state of the secondary battery, an adjustment step for adjusting a coefficient of the correction step based on the current element value calculated by the calculation step and the state value indicating the reference state It is characterized by having.
According to such a configuration, correction errors can be reduced even when secondary batteries having different characteristics are used.

  According to the present invention, it is possible to provide a secondary battery state detection device and a secondary battery state detection method capable of reducing correction errors even when secondary batteries having different characteristics are used.

It is a figure which shows the structural example of the secondary battery state detection apparatus which concerns on embodiment of this invention. It is a block diagram which shows the detailed structural example of the control part of FIG. It is a figure which shows the example of the equivalent circuit model of the secondary battery used in embodiment of this invention. It is a figure which shows the dependence with respect to the liquid temperature and SOC of Cr_Rohm. It is a figure which shows the dependence with respect to the liquid temperature and SOC of Cr_Rct1. It is a figure which shows the dependence with respect to the liquid temperature and OCV of Cr_C1. It is a flowchart for demonstrating operation | movement of embodiment of this invention. It is a flowchart for demonstrating the learning process of an equivalent circuit model.

  Next, an embodiment of the present invention will be described.

(A) Description of Configuration of Embodiment of the Present Invention FIG. 1 is a diagram showing a power supply system of a vehicle having a secondary battery state detection device according to an embodiment of the present invention. In this figure, the secondary battery state detection device 1 includes a control unit 10, a voltage sensor 11, a current sensor 12, a temperature sensor 13, and a discharge circuit 15 as main components, and indicates the charge state of the secondary battery 14. Control. Here, the control unit 10 refers to the outputs from the voltage sensor 11, the current sensor 12, and the temperature sensor 13, detects the state of the secondary battery 14, and controls the power generation voltage of the alternator 16 to control the power generation voltage. The charging state of the secondary battery 14 is controlled. The voltage sensor 11 detects the terminal voltage of the secondary battery 14 and notifies the control unit 10 of it. The current sensor 12 detects the current flowing through the secondary battery 14 and notifies the control unit 10 of the current. The temperature sensor 13 detects the electrolyte solution of the secondary battery 14 or the ambient environmental temperature and notifies the control unit 10 of it. The discharge circuit 15 is configured by, for example, a semiconductor switch and a resistance element connected in series, and the secondary battery 14 is intermittently discharged when the control unit 10 performs on / off control of the semiconductor switch.

  The secondary battery 14 is constituted by a secondary battery having an electrolytic solution, for example, a lead storage battery, a nickel cadmium battery, or a nickel metal hydride battery. The secondary battery 14 is charged by the alternator 16 and drives the starter motor 18 to start the engine. At the same time, power is supplied to the load 19. The alternator 16 is driven by the engine 17 to generate AC power, convert it into DC power by a rectifier circuit, and charge the secondary battery 14. The alternator 16 is controlled by the control unit 10 and can adjust the generated voltage.

  The engine 17 is constituted by, for example, a reciprocating engine such as a gasoline engine and a diesel engine, a rotary engine, or the like, and is started by a starter motor 18 to drive driving wheels through a transmission to provide propulsive force to the vehicle. To generate electric power. The starter motor 18 is constituted by, for example, a DC motor, generates a rotational force by the electric power supplied from the secondary battery 14, and starts the engine 17. The load 19 is constituted by, for example, an electric steering motor, a defogger, a seat heater, an ignition coil, a car audio, a car navigation, and the like, and operates with electric power from the secondary battery 14.

  FIG. 2 is a diagram illustrating a detailed configuration example of the control unit 10 illustrated in FIG. 1. As shown in this figure, the control unit 10 includes a CPU (Central Processing Unit) 10a, a ROM (Read Only Memory) 10b, a RAM (Random Access Memory) 10c, a communication unit 10d, and an I / F (Interface) 10e. ing. Here, the CPU 10a controls each unit based on the program 10ba stored in the ROM 10b. The ROM 10b is configured by a semiconductor memory or the like, and stores a program 10ba or the like. The RAM 10c is configured by a semiconductor memory or the like, and stores data generated when the program 10ba is executed, and parameters 10ca such as mathematical formulas or tables described later. The communication unit 10d communicates with an upper device such as an ECU (Electronic Control Unit) and notifies the detected information or control information to the upper device. The I / F 10e converts the signals supplied from the voltage sensor 11, the current sensor 12, and the temperature sensor 13 into digital signals and takes them in, and supplies drive current to the discharge circuit 15, the alternator 16, the starter motor 18, and the like. Supply them to control them.

(B) Description of Operation of the Embodiment of the Present Invention Next, the operation of the embodiment of the present invention will be described. In the following, after describing the operation principle of the embodiment of the present invention, the detailed operation will be described.

  First, the operation principle of the embodiment of the present invention will be described. In the embodiment of the present invention, the liquid temperature of the secondary battery 14 is Temp, the charging rate is SOC, and the open circuit voltage is OCV. Then, dependency correction curved surfaces (or dependency correction curves) relating to the liquid temperature, SOC, and OCV of the elements constituting the equivalent circuit model (equivalent circuit model acquired in advance) showing the same tendency even in different types of secondary batteries. When the secondary battery 14 is used (when the control unit 10 knows the state of the secondary battery 14 (liquid temperature, SOC, OCV, etc.)), it matches the characteristics of each secondary battery. Correct (learn) as follows. More specifically, the liquid temperature at that time, the magnitude of the slope of the tangent surface of the dependency correction curved surface (or the tangent line of the dependency correction curve) in the SOC and OCV, and the estimated error relating to each of the liquid temperature, SOC, and OCV. Based on this, the dependency correction curved surface (or dependency) for each of the liquid temperature, SOC, and OCV is taken into consideration while considering the magnitude of the influence on the dependency correction curved surface (or dependency correction curve) of the liquid temperature, SOC, and OCV. Correction curve). By such an operation, an accurate state can be detected regardless of the type or individual difference of the secondary battery 14.

  FIG. 3A is an example of an equivalent circuit model of the secondary battery 14 used in the embodiment of the present invention. In the example of FIG. 3A, the equivalent circuit model has conductor resistance and liquid resistance (hereinafter referred to as “conductor / liquid resistance”) R_ohm, reaction resistance Rct1, and electric double layer capacitance C1. ing. In the example of FIG. 3A, the reaction resistance Rct1 and the electric double layer capacitance C1 are connected in parallel, and the conductor / liquid resistance R_ohm is connected in series to the reaction resistance Rct1 and electric double layer capacitance C1 connected in parallel. Has been.

  First, in the present embodiment, the specified SOC (for example, 100%), the specified liquid temperature (for example, 25 ° C.), and the specified OCV (for example, 12.0 V) are set as the reference state of the secondary battery 14. Here, SOC, OCV, and liquid temperature are referred to as state values indicating the reference state. Needless to say, the value of the exemplified reference state is an example, and other values may be used. Then, the correction coefficient to the reference state of R_ohm which is the conductor / liquid resistance constituting the equivalent circuit model is Cr_Rohm, the correction coefficient to the reference state of Rct1 which is the reaction resistance is Cr_Rct1, and the electric double layer capacitance of C1 The correction coefficient for the reference state is Cr_C1. Then, calculation formulas for obtaining these correction coefficients Cr_Rohm, Cr_Rct1, and Cr_C1 are defined as follows, and these calculation formulas are obtained in advance by actual measurement.

Cr_Rohm = f (SOC, Temp, OCV) (1)
Cr_Rct1 = g (SOC, Temp, OCV) (2)
Cr_C1 = h (SOC, Temp, OCV) (3)

  The expressions (1) to (3) thus obtained by actual measurement are stored as parameters 10ca in the RAM 10c shown in FIG. The operation of storing the expressions (1) to (3) in the RAM 10c is performed when the vehicle is assembled or when the control unit 10 is assembled. In that case, as for the formulas (1) to (3), a formula corresponding to the type of the secondary battery 14 mounted on the vehicle may be selected and stored.

  When the vehicle in which the expressions (1) to (3) are stored is delivered to the user and the user starts using the vehicle, the timing (except immediately after charging / discharging in which the state of the secondary battery 14 is unstable ( For example, the following processing is executed at a timing when a predetermined time has elapsed since the vehicle stopped.

  That is, the CPU 10a of the control unit 10 performs a process of estimating the liquid temperature, SOC, and OCV of the secondary battery 14. More specifically, the CPU 10 a estimates the liquid temperature Temp of the secondary battery 14 by referring to the output of the temperature sensor 13. Moreover, CPU10a detects the terminal voltage when the secondary battery 14 will be in the stable state by the voltage sensor 11, and estimates OCV. Furthermore, the CPU 10a estimates the SOC by applying the previously estimated OCV to the relational expression between the OCV and the SOC. The secondary battery 14 is pulse-discharged by controlling the discharge circuit 15, and the internal impedance of the secondary battery 14 is measured by detecting the voltage and current at that time with the voltage sensor 11 and the current sensor 12. The SOC may be detected based on the above.

  Next, the CPU 10a calculates the element values of the elements constituting the equivalent circuit model of FIG. Details of this process will be described later with reference to FIG.

  Next, the CPU 10a obtains estimated liquid crystal temperature, SOC, and OCV estimated error values corresponding to the previously obtained liquid temperature, SOC, and OCV estimation results from the parameter 10ca of the RAM 10c, for example. More specifically, when the liquid temperature is estimated, an error corresponding to the liquid temperature occurs. For example, an error of about 3 ° C. may occur when the liquid temperature is 20 ° C., and an error of about 5 ° C. may occur when the liquid temperature is 30 ° C. Similarly, in the case of SOC and OCV, an error corresponding to the value occurs. The estimated estimated error value is stored in the parameter 10ca of the RAM 10c as a table associated with the liquid temperature, SOC, and OCV values.

  Subsequently, the CPU 10a calculates the slopes of the equations (1) to (3) in the liquid temperature, the SOC, and the OCV at that time. Specifically, regarding the formula (1), the inclination is obtained by the following formulas (4) to (6). Note that ∂ is a partial differential symbol.

  ∂f / ∂Temp = {f (SOC, Temp + dTemp, OCV) −f (SOC, Temp, OCV)} / dTemp (4)

  ∂f / ∂SOC = {f (SOC + dSOC, Temp, OCV) −f (SOC, Temp, OCV)} / dSOC (5)

  ∂f / ∂OCV = {f (SOC, Temp, OCV + dOCV) −f (SOC, Temp, OCV)} / dOCV (6)

  For equation (2), the slope is determined by the following equations (7) to (9).

  ∂g / ∂Temp = {g (SOC, Temp + dTemp, OCV) −g (SOC, Temp, OCV)} / dTemp (7)

  ∂g / ∂SOC = {g (SOC + dSOC, Temp, OCV) −g (SOC, Temp, OCV)} / dSOC (8)

  ∂g / ∂OCV = {g (SOC, Temp, OCV + dOCV) −g (SOC, Temp, OCV)} / dOCV (9)

  In addition, regarding the expression (3), the inclination is obtained by the following expressions (10) to (12).

  ∂h / ∂Temp = {h (SOC, Temp + dTemp, OCV) −h (SOC, Temp, OCV)} / dTemp (10)

  ∂h / ∂SOC = {h (SOC + dSOC, Temp, OCV) −h (SOC, Temp, OCV)} / dSOC (11)

  ∂h / ∂OCV = {h (SOC, Temp, OCV + dOCV) −h (SOC, Temp, OCV)} / dOCV (12)

  FIG. 4 is a graph showing the dependence of Cr_Rohm on the liquid temperature and SOC, corresponding to equation (4). In FIG. 4, the horizontal axis indicates the liquid temperature, and the vertical axis indicates the value of Cr_Rohm. Moreover, each curve has shown the relationship between the liquid temperature in case SOC differs, and Cr_Rohm. In FIG. 4, for example, ∂f / ∂Temp when the SOC is 100% and the liquid temperature is 10 ° C. is as shown by a broken line.

  FIG. 5 is a graph showing the dependence of Cr_Rct1 on the liquid temperature and SOC corresponding to equation (8). In FIG. 5, the horizontal axis indicates the SOC, and the vertical axis indicates the value of Cr_Rct1. Moreover, each curve has shown the relationship between SOC and Cr_Rct1 in case liquid temperature differs. In FIG. 5, for example, when the liquid temperature is 0 ° C. and the SOC is 40%, ＳＯg / ∂SOC is as shown by a broken line.

  FIG. 6 is a graph showing the dependence of Cr_C1 on the liquid temperature and OCV corresponding to Expression (12). In FIG. 6, the horizontal axis indicates OCV, and the vertical axis indicates the value of Cr_C1. Moreover, each curve has shown the relationship between OCV and Cr_C1 in case liquid temperature differs. In FIG. 6, for example, when the liquid temperature is 50 ° C. and the OCV is 12.4, ∂h / ∂OCV is as shown by a broken line.

  Note that with respect to the equation (1), according to the equations (4) to (6), for example, ∂f / ∂Temp = 0.03, ∂f / ∂SOC = 0.01, ∂f / ∂OCV = 0. A value like 005 is obtained.

  Next, CPU10a acquires the correction | amendment error estimated value of Formula (1)-Formula (3) in liquid temperature, SOC, and OCV calculated | required previously from the parameter 10ca of RAM10c, for example. More specifically, for example, when obtaining Cr_Rohm by f (SOC, Temp, OCV), the error of Cr_Rohm varies depending on the values of SOC, Temp, OCV. Similarly, when obtaining Cr_Rct1 and Cr_C1 by g (SOC, Temp, OCV) and h (SOC, Temp, OCV), the errors of Cr_Rohm and Cr_C1 differ depending on the values of SOC, Temp, and OCV. Therefore, the expected correction error values of Cr_Rohm, Cr_Rct1, and Cr_C1 based on the values of SOC, Temp, and OCV are stored as a table in the parameter 10ca of the RAM 10c, and the correction errors corresponding to the SOC, Temp, and OCV values at that time are stored. Expected values can be obtained from a table. For example, 10% is obtained as the expected correction error value of Cr_Rohm corresponding to SOC, Temp, and OCV at that time.

  Next, the CPU 10a calculates a temperature-dependent learning error estimated value, an SOC-dependent learning error estimated value, and an OCV-dependent learning error estimated value for each of the equations (1) to (3). . For example, for the equation (1), based on the following equations (13) to (15), the temperature-dependent learning error expected value E_Rohm_Temp, the SOC-dependent learning error estimated value E_Rohm_SOC, and the OCV-dependent A predicted learning error value E_Rohm_OCV is calculated.

  E_Rohm_Temp = ∂f / ∂Temp × Temp Estimated Error Expected Value × Cr_Rohm Corrected Error Expected Value (13)

  E_Rohm_SOC = ∂f / ∂SOC × SOC estimated error expected value × Cr_Rohm expected correction error value (14)

  E_Rohm_OCV = ∂f / ∂OCV × OCV estimated error expected value × Cr_Rohm expected correction error value (15)

  Note that also for the expression (2), based on the following expressions (16) to (18), the temperature-dependent learning error estimated value E_Rct1_Temp, the SOC-dependent learning error estimated value E_Rct1_SOC, and the OCV-dependent learning The estimated error value E_Rct1_OCV is calculated.

  E_Rct1_Temp = ∂g / ∂Temp × Temp estimated error expected value × Cr_Rct1 expected correction error value (16)

  E_Rct1_SOC = ∂g / ∂SOC × SOC estimated error expected value × Cr_Rct1 expected correction error value (17)

  E_Rct1_OCV = ∂g / ∂OCV × OCV estimated error expected value × Cr_Rct1 expected correction error value (18)

  Further, with regard to the expression (3), based on the following expressions (19) to (21), the temperature-dependent learning error estimated value E_C1_Temp, the SOC-dependent learning error estimated value E_C1_SOC, and the OCV-dependent learning The estimated error value E_C1_OCV is calculated.

  E_C1_Temp = ∂h / ∂Temp × Temp estimated error value × Cr_C1 expected correction error value (19)

  E_C1_SOC = ∂h / ∂SOC × SOC estimated error expected value × Cr_C1 expected correction error value (20)

  E_C1_OCV = ∂h / ∂OCV × OCV estimated error expected value × Cr_C1 expected correction error value (21)

  Then, when the nine kinds of estimated learning error values obtained by the above formulas (13) to (21) are all below a predetermined threshold value, the CPU 10a determines the Rohm, Rct1, and C1 at that time. Based on the measured values and Rohm_0, Rct1_0, and C1_0 that are Rohm, Rct1, and C1 in the reference state, Cr_Rohm, Cr_Rct1, and Cr_C1 are calculated by the following formulas (22) to (24).

Cr_Rohm = Rohm_0 / Rohm (22)
Cr_Rct1 = Rct1_0 / Rct1 (23)
Cr_C1 = C1_0 / C1 (24)

  Then, the CPU 10a associates the calculated Cr_Rohm, Cr_Rct1, and Cr_C1 and the SOC, Temp, and OCV at that time and stores them in the RAM 10c as the parameter 10ca.

  When the above-described processing is repeated and a predetermined number of Cr_Rohm, Cr_Rct1, and Cr_C1 are stored in the RAM 10c, the coefficients of the expressions (1) to (3) are optimized based on these values. Execute the process to convert.

  According to the above processing, the correction formula for correcting the element values of the elements constituting the equivalent circuit model to the reference state can be optimized according to the secondary battery 14 being used. For this reason, when a secondary battery of a type different from that specified by the manufacturer is mounted as the secondary battery 14, there is an individual difference of the secondary battery 14, or the secondary battery 14 deteriorates and the characteristics are new. Even in the case of a change, the element values of the elements constituting the equivalent circuit model can be accurately calculated.

  Next, processing executed in the control unit 10 shown in FIG. 1 will be described with reference to FIG. The processing of the flowchart shown in FIG. 7 is performed when, for example, a predetermined time has elapsed since the charging / discharging of the secondary battery 14 has ended (for example, a certain time has elapsed since the vehicle on which the secondary battery 14 is mounted is stopped). And when the polarization of the secondary battery 14 is substantially eliminated (for example, when the voltage generated by the polarization becomes less than a predetermined threshold). When the processing of the flowchart of FIG. 7 is started, the following steps are executed.

  In step S10, the CPU 10a of the control unit 10 estimates the liquid temperature, SOC, and OCV of the secondary battery 14. More specifically, the CPU 10a estimates the liquid temperature by referring to the output of the temperature sensor 13, detects the terminal voltage of the secondary battery 14 by the voltage sensor 11, and estimates the OCV. Moreover, CPU10a estimates SOC by applying OCV estimated previously with respect to the relational expression of OCV and SOC. In addition, as described above, the discharge circuit 15 is controlled to cause the secondary battery 14 to pulse discharge, and the voltage and current at that time are detected by the voltage sensor 11 and the current sensor 12, whereby the internal impedance of the secondary battery 14 is determined. The SOC may be estimated based on the internal impedance.

  In step S11, the CPU 10a learns the element values of the equivalent circuit model shown in FIG. 3A of the secondary battery 14 at that time. Details of this process will be described later with reference to FIG.

  In step S12, the CPU 10a calculates estimated liquid error, SOC, and OCV estimated error values. More specifically, for example, the CPU 10a acquires an estimated error expected value corresponding to the liquid temperature, the SOC, and the OCV at that time, which is stored in the RAM 10c as the parameter 10ca.

  In step S13, the CPU 10a calculates the inclination of the dependency correction formula (or correction curved surface) for each of the liquid temperature, the SOC, and the OCV. More specifically, the inclination of the dependency correction formula (or correction curved surface) is calculated based on the above-described formulas (4) to (12).

  In step S14, the CPU 10a acquires a predicted correction error value corresponding to the SOC, Temp, and OCV values at that time from the parameter 10ca of the RAM 10c. More specifically, the CPU 10a acquires the expected correction error values of Cr_Rohm, Cr_Rct1, and Cr_C1 based on the SOC, Temp, and OCV values stored in advance in the RAM 10c as the parameters 10ca.

  In step S15, the CPU 10a, based on the above-described equations (13) to (21), predicts the temperature-dependent learning error value E_Rohm_Temp, E_Rct1_Temp, E_C1_Temp, and the SOC-dependent learning error probability value E_Rohm_SOC, E_Rct1_SOC, E_C1, E_C1 Then, OCV-dependent learning error estimated values E_Rohm_OCV, E_Rct1_OCV, and E_C1_OCV are calculated.

  In step S <b> 16, the CPU 10 a determines whether all nine types of estimated learning error values obtained by the equations (13) to (21) are all below a predetermined threshold, and all are below the predetermined threshold. Is determined (step S16: Y), the process proceeds to step S17. Otherwise (step S16: N), the process ends.

  In step S17, the CPU 10a determines Cr_Rohm, Cr_Rct1, and R1 based on the measured values of Rohm, Rct1, and C1, and Rohm_0, Rct1_0, and C1_0 that are Rohm, Rct1, and C1 in the reference state. , Cr_C1 is calculated according to the above-described equations (22) to (24), and the obtained Cr_Rohm, Cr_Rct1, and Cr_C1 and the SOC, Temp, and OCV at that time are associated with each other and stored in the RAM 10c as the parameter 10ca. .

  In step S18, when a predetermined number of Cr_Rohm, Cr_Rct1, and Cr_C1 are stored in the RAM 10c, the CPU 10a uses the values of the above-described equations (1) to (3) based on these values. Execute the process to be optimized.

  Next, with reference to FIG. 8, it is a figure explaining the learning process of the equivalent circuit model shown to FIG. 3 (A). When the process of FIG. 8 is started, the following steps are executed.

In step S50, the CPU 10a of the control unit 10 substitutes a value obtained by adding ΔT to the previous value T _n−1 to a variable T _n indicating time. As ΔT, for example, several msec to several hundred msec can be used.

In step S51, the CPU 10a measures the current I _n , the voltage V _n , and the temperature θ _n based on the detection signals from the voltage sensor 11, the current sensor 12, and the temperature sensor 13.

In step S52, the CPU 10a calculates the voltage drop ΔV _n by applying the voltage V _n measured in step S51 to the following equation (25).

ΔV _n = V _n −OCV (25)

  Here, OCV is an open circuit voltage. As a method for obtaining the OCV, for example, the terminal voltage measured immediately before the secondary battery 14 is started or the open circuit voltage of the secondary battery 14 estimated from the charge / discharge state of the secondary battery 14 is defined as OCV. it can.

In step S52, CPU 10a from the n-th observed value and the previous state vector estimate to update the Jacobian _{F n} based on the following equation (26). Note that diag () indicates a diagonal matrix.

F _n = diag (1-ΔT / Rct1 _n · C1 _n , 1,1,1,1) (26)

In step S54, the CPU 10a sets ΔV _n obtained by calculation in step S52 as the observed observation value Y _n of the extended Kalman filter, as shown by the following equation (27).

Y _n = ΔV _n (27)

In step S55, the CPU 10a obtains a state vector _{Xn + 1} | _n ahead of the time based on the following equation (28).

_{X n + 1 | n = F} n · X n + U n ··· (28)

Here, _Xn and _Un are represented by the following formulas (29) and (30). T represents a transposed matrix.

X _n ^T = (ΔV2, Rct1, Rohm, C1, V0) (29)

U _n ^T = (Δt · I _n / C1, 0, 0, 0, 0) (30)

Note that by defining H _n ^T as in the following equation (31), the observation equation and the predicted observation value Y _{n + 1} | _n can be defined as in equation (32).

H _n ^T = (1, I _n , 0, 0, 0) (31)

Y _n = H _n ^T · X _n (32)

In step S56, CPU 10a, the prediction value one period ahead of the state vector _{X n} + 1 _{| n} and the measured observations _{Y n + 1} and the predicted observation value _{Y n} + 1 _| based on _n, the extended Kalman filter operation using a Kalman gain calculation and filtering calculation, The optimum state vector _Xn is sequentially estimated, and the adjustment parameter (of the equivalent circuit model) is updated to the optimum one from the estimated state vector _Xn .

  Through the above processing, the element values of the elements constituting the equivalent circuit model shown in FIG. 3A can be obtained.

  Rohm_0, Rct1_0, and C1_0, which are the element values of the equivalent circuit model in the corrected reference state, obtained by the processing shown in FIGS. By applying, it is possible to know SOH or SOF as the state of the secondary battery 14.

  SOH = F (Rohm_0, Rct1_0, C1_0) (33)

  SOF = G (Rohm_0, Rct1_0, C1_0) (34)

(C) Description of Modified Embodiment It goes without saying that the above embodiment is merely an example, and the present invention is not limited to the case described above. For example, in the above embodiment, the equivalent circuit model Rohm, Rct1, and C1 are corrected according to the SOC, Temp, and OCV using the expressions (1) to (3). Any one of Rohm, Rct1, and C1, or any two of them may be corrected. Further, in the expressions (1) to (3), SOC, Temp, and OCV are variables, but any one or two of these may be used as variables. Alternatively, in the expressions (1) to (3), SOC, Temp, and OCV are variables, but other variables, for example, the polarization of the secondary battery 14 may be used as a variable.

  In the above embodiment, the equivalent circuit model shown in FIG. 3A is used. However, for example, the equivalent circuit model shown in FIG. 3B may be used. In that case, equations corresponding to Rct2 and C2 are added to equations (1) to (3).

  Further, in step S16 of FIG. 7, when all the dependency correction curved surface learning error expected values are below a predetermined threshold value, the process proceeds to step S17. When the value falls below a predetermined threshold value, the process may proceed to step S17. Or you may make it progress to step S17 in all cases irrespective of the dependence correction | amendment curved surface learning error estimated value.

  Further, in the above embodiment, the value obtained by multiplying the slope of the dependency correction equation, the estimated error estimated value, and the estimated dependency correction error value and the learning error estimated value of the dependency correction curved surface, For example, the process of step S16 may be executed by using a value obtained by multiplying the inclination of the dependency correction formula by the estimated error estimated value as a learning error estimated value of the dependency corrected curved surface.

  Further, in the above embodiment, the process shown in FIG. 7 is performed when a predetermined time has elapsed after the charge / discharge of the secondary battery 14 is completed, but is performed during the charge / discharge. Or may be executed before a predetermined time has elapsed since the end of charging and discharging.

  In the above embodiment, the constants of the equivalent circuit model are calculated according to the flowchart shown in FIG. 8, but the constants of the equivalent circuit model may be calculated by a process other than that in FIG.

DESCRIPTION OF SYMBOLS 1 Secondary battery state detection apparatus 10 Control part (Calculation means, estimation means, correction means, adjustment means)
10a CPU
10b ROM
10c RAM
10d Communication unit 10e I / F
DESCRIPTION OF SYMBOLS 11 Voltage sensor 12 Current sensor 13 Temperature sensor 14 Secondary battery 15 Discharge circuit 16 Alternator 17 Engine 18 Starter motor 19 Load

Claims (7)

In the secondary battery state detection device for detecting the state of the secondary battery,
Calculating means for calculating an element value of an element constituting the equivalent circuit model of the secondary battery;
Correction means for correcting the element value of the equivalent circuit model calculated by the calculation means to an element value when the secondary battery is in a reference state;
Estimating means for estimating a state of the secondary battery based on an element value of the equivalent circuit model in the reference state corrected by the correcting means;
Adjusting means for adjusting the correcting means based on the element value at that time calculated by the calculating means and a state value indicating the reference state;
A secondary battery state detection device comprising:   2. The secondary battery state detection according to claim 1, wherein the adjustment unit adjusts the correction unit when it is estimated that an error occurring during correction by the correction unit is less than a predetermined threshold value. apparatus. The element constituting the equivalent circuit model has at least one of conductor / liquid resistance, reaction resistance, and electric double layer capacitance,
The state value indicating the reference state has at least one of liquid temperature, SOC, and OCV.
The secondary battery state detection device according to claim 2. The correction means has a correction formula for correcting the element value of the equivalent circuit model to the element value when the secondary battery is in the reference state,
When the value obtained by multiplying the gradient obtained by partial differentiation of the correction equation with a predetermined state value by the estimated error of the state value at that time is less than a predetermined threshold The secondary battery state detection device according to claim 3, wherein a coefficient of the correction formula is adjusted. The correction means has a correction formula for correcting the element value of the equivalent circuit model to the element value when the secondary battery is in the reference state,
The adjustment means includes an estimated error value when correcting the element value in the current state to the element value in the reference state, a slope obtained by partial differentiation of the correction formula with a predetermined state value, and the time point 4. The coefficient of the correction formula is adjusted when the value obtained by multiplying the estimated error value of the state value in FIG. 1 is less than a predetermined threshold value. Secondary battery state detection device.   6. The secondary battery state detection device according to claim 2, wherein the adjustment unit adjusts the correction unit when the polarization state of the secondary battery is substantially eliminated. . In the secondary battery state detection method for detecting the state of the secondary battery,
A calculation step of calculating an element value of an element constituting the equivalent circuit model of the secondary battery;
A correction step of correcting the element value of the equivalent circuit model calculated in the calculation step to an element value when the secondary battery is in a reference state;
An estimation step for estimating a state of the secondary battery based on an element value of the equivalent circuit model in the reference state corrected in the correction step;
An adjustment step of adjusting a coefficient of the correction step based on the current element value calculated by the calculation step and a state value indicating the reference state;
A secondary battery state detection method comprising:

Images