JP6701936B2 - Battery state detecting device, vehicle, program, and battery state detecting method - Google Patents

Description

  The present invention relates to a battery state detection device, a vehicle, a program, and a battery state detection method.

  In recent years, more and more vehicles driven by an engine have an idling stop function. In these vehicles, when the vehicle stops, the state of the storage battery at the time of cranking (restarting) the engine next time is estimated, and idling stop is not performed if it is estimated that restarting is difficult. It is like this. The state of the storage battery is estimated based on the measured values of the voltage, current, etc. of the storage battery, but what timing measurement value is used for the estimation is a problem.

  As an example thereof, in paragraph 0028 of Patent Document 1 below, “First, the measurement of the current Ib and the voltage Vb of the power storage device 101 is started (201), and it is detected that there is a large current discharge (starter start) (202). At that time, when the current has decreased to a predetermined maximum current value Imax, it is determined that the cranking period has started, and the collection of the pair of current Ib and voltage Vb is started, and this collection is performed by the current having a predetermined minimum current value Imin. It continues for the period until it becomes (cranking period), and the collected voltage/current pair group is memorized.”

JP, 2007-223530, A

In the technique of Patent Document 1 described above, in order to estimate the state of the power storage device 101, the timing of storing the measurement value (voltage/current pair group) based on the current value (that is, what timing measurement value is used) Is to determine. However, rather than determining the timing according to the magnitude of the current value, it is more accurate to obtain the measured value at the timing when the current value changes abruptly or discontinuously, and to use the measured value to determine the state of the storage battery more accurately. It turns out that it can be estimated.
The present invention has been made in view of the above circumstances, and an object thereof is to provide a battery state detection device, a vehicle, a program, and a battery state detection method that can accurately estimate the state of a storage battery.

In the battery state detection device of the present invention for solving the above-mentioned problems, does the current flowing in the storage battery or the voltage of the storage battery satisfy a rapid change condition that there is a change exceeding a predetermined value within a predetermined time period? An abrupt change condition determination unit that determines whether or not, and a resistance function identification unit that identifies a resistance function that is a function of the internal resistance of the storage battery with respect to the current, provided that the abrupt change condition is satisfied; A specific current calculation unit that calculates a specific current that is the current when a specific load is connected to the storage battery using the resistance function R(I) that has been set, and the resistance function identification unit, Scaling current for adjusting vertical/horizontal scaling, scaling voltage for adjusting vertical/horizontal scaling, and scaling resistor for adjusting vertical scaling are parameters of the resistance function, and are calculated by the scaling resistor . And a function for obtaining a deviation between an actual voltage drop and the scaling current , the scaling current is updated with a value proportional to the deviation, and the scaling voltage and the scaling resistor are set so that the sum of squares of the deviation is minimized. And a function for obtaining the scaling current ,
A battery state detection device comprising:

  According to the present invention, the state of the storage battery can be accurately estimated.

1 is a block diagram of a vehicle according to a first exemplary embodiment of the present invention. It is a block diagram of a battery monitoring device. It is a flow chart of an idling stop control routine. It is a flowchart of a battery state monitoring routine. It is an example of a waveform diagram of a current I. It is a correlation diagram of current I and voltage change amount ΔV when (a) SOC is 80% and (b) SOC is 90%. FIG. 7 is a block diagram of a vehicle according to a second exemplary embodiment of the present invention. It is a graph of (a) function F(x) and (b) function F'(x) by the Butler-Volmer equation.

[First Embodiment]
<Structure of First Embodiment>
Next, the details of the vehicle 1 according to the first embodiment of the present invention will be described with reference to the block diagram shown in FIG.
The vehicle 1 is a so-called idling stop vehicle having an idling stop function, and stores an engine 11 that drives the vehicle 1, a generator 12 mechanically coupled to the engine 11, and electric power output from the generator 12. Storage battery 10, auxiliary load 14 that is an electric load such as a light, an air conditioner, a starter (specific load) 14a, and an ECU (Electronic Control Unit, engine control unit, battery state detection device) that controls each unit of vehicle 1. 15 and a battery monitoring device (battery state detection device) 30 that detects the state of the storage battery 10. As the storage battery 10, for example, a lead storage battery, a nickel hydride storage battery, or the like can be applied.

  The ECU 15 includes hardware as a general computer such as a CPU (Central Processing Unit), a RAM (Random Access Memory), and a ROM (Read Only Memory), and the ROM executes programs (executed by the CPU). Details will be described later), various data, and the like are stored. The battery monitoring device 30 measures values such as SOC (State of Charge) and SOF (State of Function). Here, SOC is the charging rate, and SOF is the estimated voltage value of the storage battery 10 during cranking (when the engine 11 is started).

  At the time of idling stop for stopping the engine 11, the electric power to the auxiliary load 14 is supplied from the storage battery 10. Further, when the vehicle 1 is decelerated, the generator 12 may be rotated and operated by the rotational force from the tires, and the electric energy generated by the generator 12 may be supplied to the accessory load 14 as electric power. Further, when the SOC of the storage battery 10 is insufficient, the ECU 15 drives the generator 12 by the engine 11 to charge the storage battery 10.

  Next, the detailed configuration of the battery monitoring device 30 will be described with reference to the block diagram shown in FIG. The battery monitoring device 30 includes a voltage sensor 31 that measures the output voltage of the storage battery 10, a current sensor 32 that measures the current flowing through the storage battery 10, a CPU 33, a RAM 34 that stores various data, and a control program executed by the CPU 33. It has a ROM 35 that stores (details will be described later), a communication driver 36 that mediates communication between the ECU 15 and the CPU 33, and a temperature sensor 37 that measures the temperature T of the storage battery 10. Each of the sensors 31, 32, 37 includes an AD converter and supplies the measured value to the CPU 33 as a digital value.

  In FIG. 2, the inside of the CPU 33 shows main functions realized by the control program. The abrupt change condition determination unit 301 determines whether or not the current I flowing through the storage battery 10 has changed abruptly. The resistance function identification unit 302 identifies the resistance function R(I) (details will be described later) that is the internal resistance of the storage battery 10. The cranking current calculator (specific current calculator) 304 estimates the current I during cranking (when the engine 11 is started), that is, the cranking current (specific current) Ic. Further, the voltage estimation unit 306 estimates the SOF based on the cranking current Ic.

  The ECU 15 and the battery monitoring device 30 are connected via a communication line 16. Here, the communication line 16 may be configured using a CAN (Controller Area Network) or the like. Further, the current sensor 32 can be configured using a hall element, a shunt resistor, or the like.

<Operation of First Embodiment>
(Operation of ECU 15)
Next, the operation of this embodiment will be described.
First, FIG. 3 is a flowchart of an idling stop control routine executed by the ECU 15. This routine is started when the key is inserted into the vehicle 1 and the ignition is turned on, and is continued until the key is turned off.

  When the process proceeds to step S2 in FIG. 3, it is determined whether the vehicle speed of the vehicle 1 has changed from a positive value to a zero value. When the vehicle 1 is continuously traveling, or when the vehicle 1 is continuously stopped, it is determined as "No", and the loop of step S2 is repeated. On the other hand, when the vehicle speed changes from the positive value to the zero value, it is determined as "Yes" and the process proceeds to step S4. Here, the ECU 15 receives the SOF from the battery monitoring device 30.

  Next, when the process proceeds to step S6, the ECU 15 determines whether SOF exceeds a predetermined minimum voltage. The lowest voltage is the lowest voltage that allows idling stop. If it is determined "No" here, the process returns to step S2, and step S2 is repeatedly executed while continuing idling. On the other hand, if “Yes” is determined in step S6, the process proceeds to step S8, and the ECU 15 stops idling. That is, the engine 11 is stopped.

  Next, when the process proceeds to step S10, the ECU 15 receives the SOF from the battery monitoring device 30 again. Next, when the process proceeds to step S12, it is determined again whether or not SOF exceeds the minimum voltage. If it is determined to be "Yes" here, the process proceeds to step S14, and it is determined whether or not the brake pedal of the vehicle 1 is in the off state. If the determination is "No" here, the process returns to step S10. After that, as long as the SOF exceeds the minimum voltage and the brake pedal is in the on state (state in which the driver is stepping on), the loop of steps S10 to S14 is repeated in the idling stop state.

  Here, when SOF becomes equal to or lower than the minimum voltage, or when the brake pedal is in the off state, the process proceeds to step S16, and the idling stop state is released. That is, the engine 11 is restarted. After that, the process returns to step S2, and the above-described operation is repeated.

(Operation of the battery monitoring device 30)
Next, FIG. 4 is a flowchart of a battery state monitoring routine executed by the battery monitoring device 30. This routine is also started when the key is inserted into the vehicle 1 and the ignition is turned on, and is continued until the key is turned off.
When the process proceeds to step S101 in FIG. 4, the steep change condition determination unit 301 (see FIG. 2) determines whether or not the current time has reached the measurement timing. The “measurement timing” is a timing for each sampling cycle of, for example, several milliseconds. The sampling period is preferably 0.1 seconds or less, more preferably 0.01 seconds or less. If “No” is determined here, the process waits in step S101 until the current time reaches the measurement timing.

  If “Yes” is determined in step S101, the process proceeds to step S102, and the abrupt change condition determination unit 301 measures various values. That is, the voltage V, the current I, and the temperature T of the storage battery 10 are acquired via the voltage sensor 31, the current sensor 32, and the temperature sensor 37. Next, when the process proceeds to step S103, the abrupt change condition determination unit 301 determines that the absolute value of the current change amount ΔI (the value obtained by subtracting the current I at the previous measurement timing from the current I at the current measurement timing) is the current. It is determined whether or not the difference threshold ΔIth is exceeded. That is, it is determined whether or not the current I changes abruptly. Hereinafter, this condition (|ΔI|>ΔIth) is referred to as “a steep change condition (|ΔI|>ΔIth)”. The threshold value ΔIth may be set to 20 [A], for example. The threshold value ΔIth is more preferably set to a value of 50 [A] or more, and further preferably set to a value of 100 [A] or more.

  If the abrupt change condition determination unit 301 determines “Yes” in step S103, the process proceeds to step S104. In step S104, the resistance function identifying unit 302 stores the data of the voltage V, the current I, and the temperature T at the current measurement timing and the previous measurement timing in a predetermined buffer area in the RAM 34. When the buffer area becomes full, the resistance function identifying unit 302 discards the accumulated data in the oldest order.

Next, when the process proceeds to step S106, the resistance function identifying unit 302 updates the resistance function R(I). Here, the resistance function R(I) is the internal resistance of the storage battery 10, but since the internal resistance changes according to the current I, it is called the “resistance function”.
Specifically, the resistance function R(I) is expressed by the following equation (1).

R(I)=(V0/I0)×F(I/I0;α)×I0/I+R0 Equation (1)

In Equation (1), α is a constant, V0 is a vertical scaling voltage , I0 is a horizontal scaling current (scaling current), and R0 is a height scaling resistor . The current I has a positive value in the charging direction. Further, the function F(x;α) is a function that satisfies “x=exp{-(1-α)F}-exp{-αF}”. The function F(x;α) is called the Butler-Volmer equation and is an expression indicating the voltage of the liquid resistance.

In equation (1), vertical and horizontal scaling is adjusted by the vertical scaling voltage V0 and horizontal scaling current I0, and height scaling is adjusted by the height scaling resistor R0.
Here, since the function F′(x;α), which is the differential value of the function F(x;α), is expressed by the following expression (2), the function F(x;α) is converted into the following expression (3). It can be calculated by the recurrence formula based on the Newton method shown.

F′=1/{(1-α)x+exp(−αF)} Equation (2)

F[n+1]=F[n]-{exp((1-α)F[n])-exp(-αF[n])-x}/{(1-α)exp((1-α)F [N])+αexp(−αF[n])} Equation (3)

However, in the equation (3), when x is equal to or less than the inflection point of the function F(x;α), F[0]=2arcsinh(x/2) is set, and x is the function F(x;α). When exceeding the inflection point, F[0]=ln(|x|)/(1-α) may be set. Here, the inflection point of the function F(x;α) is (2-1/α)×(1/α-1) ^(2α-1) .

The reason why the data when the current I sharply changes is used for updating the resistance function R(I) is that accurate data can be acquired while avoiding the influence of polarization. Here, assuming that the current I and the voltage V before the current I sharply change are Ia and Va, and the current I and the voltage V after the current I sharply change are Ib and Vb, the voltage error is E. The squared error E ² at this time is obtained by the following equation (4).

^{E 2 = | V0 × {F} (Ia / I0; α) - F (Ib / I0; α)} + R0 × (Ia-Ib) - (Va-Vb) | 2 ... Equation (4)

  “Updating the resistance function R(I)” in step S106 is preferably “identifying (optimizing) V0, I0, R0, α”. However, if the calculation time is insufficient when the constant α is optimized, α may be fixed to a predetermined value, for example, 0.69 (or any value from 0.5 to 0.9). If the constant α is fixed, the unknowns in the equation (4) will be the parameters V0, I0, and R0. To minimize the sum of squared errors in equation (4), including past data, use the data obtained when the steep change condition has been satisfied for the past n times (n is a natural number of 3 or more). It is advisable to identify V0, I0, R0 that minimizes the total error Em shown in equation (5). Here, in the past n times, data in which the time of the data that is a sudden change condition of current discontinuity is continuous is counted as one group, and indicates data of n groups or more. An example of this data group will be described later.

Em=Σ|V0×{F(Ia(j)/I0;α)−F(Ib(j)/I0;α)}+R0×(Ia(j)−Ib(j))−(Va(j) -Vb(j))| ² ... Formula (5)
(However, j=1, 2,..., N)

In other words, the step S106 is based on the parameters V0, I0, R0 of the resistance function R(I), the voltage drop obtained by the height scaling resistor R0 and the deviation from the actual voltage drop (R0×(Ia−Ib)− (Va-Vb)), the lateral scaling current I0 is updated with a value (small value) proportional to the deviation, and the parameters V0, I0, R0 are calculated so that the sum of squares of deviation (Em) is minimized. It will be.

  The resistance function identifying unit 302 updates V0, I0, and R0 each time step S106 is executed. When step S106 is executed for the first time, the data at the time of the immediately preceding cranking or the key-off last time is performed. It is advisable to use the value of as the initial value. A predetermined value (for example, 20 [A]) may be used as the initial value of the lateral scaling current I0. At the time of factory shipment, a predetermined value (for example, 20 [A]) is set as the lateral scaling current I0, and V0, I0, and R0 may be set by repeatedly calculating the equation (5).

  When step S106 ends, the process returns to step S101. After that, as long as the abrupt change condition (|ΔI|>ΔIth) is satisfied, the resistance function R(I) is updated at each measurement timing. After that, when the abrupt change condition is not satisfied, it is determined as “No” in step S103, and the process proceeds to step S108. In step S108, the cranking current calculator 304 (see FIG. 2) calculates the unsteady open circuit voltage OCV. Here, the open circuit voltage refers to the voltage V when the storage battery 10 is in the open state, and the unsteady open circuit voltage OCV refers to the open circuit voltage including the polarization voltage.

The unsteady open circuit voltage OCV is obtained by the following equation (6).

OCV=VR(I)×I... Formula (6)

Here, the voltage V and the current I are measured values obtained by the voltage sensor 31 and the current sensor 32. The resistance function R(I) is as shown in equation (1).

Next, when the process proceeds to step S110, the cranking current calculation unit 304 estimates the current I in cranking (hereinafter referred to as the cranking current Ic) based on the following equation (7).

R(Ic)×Ic+r0×Ic=OCV (7)

Here, r0 is the resistance value of the starter 14a included in the auxiliary load 14 (hereinafter, referred to as starter resistance value (specific load resistance value)). The starter resistance value r0 is a value including the wiring resistance. Since the starter resistance value r0 changes depending on the temperature T, it is desirable to calculate it at each cranking time. For example, if the minimum value of the voltage V at the time of immediately preceding cranking is the voltage Vst and the current I when the voltage Vst appears is the current Ist, the starter resistance value r0 may be obtained by "r0=Vst/Ist". ..

By the way, since the above equation (7) can be transformed into the following equation (8), the cranking current Ic can be obtained by solving the nonlinear equation of the equation (8).

V0×F(Ic/I0; α)+(R0+r0)×Ic=OCV Equation (8)

The cranking current Ic in the equation (8) can be calculated by the recurrence equation by the Newton method using the function x(n) shown in the equation (9).

x(n+1)=x(n)-{exp(x(n))-1+(Ax(n)-B)exp(αx(n))}/{(1-α)exp(x(n)) +α+Aexp(αx(n))} Equation (9)

However, in equation (9),
A=V0/(R0+r0),
B=OCV/(R0+r0).
When B is less than or equal to the inflection point of the function F(x;α), x(0)=arcsinh(B/2)/B is set, and B is the inflection point of the function F(x;α). When it exceeds, x(0)=ln(|B|)/(1-α) is preferable. As described above, the inflection point of the function F(x;α) is (2-1/α)×(1/α-1) ^(2α-1) .

The cranking current calculator 304 estimates the cranking current Ic by "Ic={OCV-V0*x(n+1)}/(R0+r0)". Note that “x(n+1)” is “n+1” when the equation (9) converges in an appropriate range. For example, it is preferable to use the function x(n+1) when the convergence condition of |x(n+1)−x(n)|<1e ⁻⁶ is satisfied.

When the cranking current Ic is calculated in step S110, the process next proceeds to step S112, and the voltage estimating unit 306 calculates the cranking current Ic and the starter resistance value r0 based on the following equation (10). SOF, that is, the estimated value of the voltage V during cranking is calculated.

SOF=Ic×r0 Equation (10)

  Next, when the process proceeds to step S114, the voltage estimation unit 306 outputs (transmits) the calculated SOF. The voltage estimation unit 306 stores the SOF in the communication driver 36 so that the ECU 15 can read the SOF at any time. When the above processing is completed, the processing returns to step S101, and the processing after step S101 is repeated.

  Here, an example of a waveform diagram of the current I in the present embodiment is shown in FIG. In FIG. 5, the section before time t2 is in the idling stop state. In this section, since a small current is supplied to the auxiliary load 14 except the starter 14a, the current I is a value (discharge state) slightly smaller than zero. At time t2, the idling stop is canceled, that is, the engine 11 is started. Since a large current flows through the starter 14a, the current I drops sharply, and the current I reaches the minimum value at time t4. Here, the above-mentioned "data group" means one group consisting of t2 to t4, one group consisting of t6 to t8, and one group consisting of t10 to t12. All the data of the past three or more groups are used for the optimization of the resistance function mentioned above.

  After the engine 11 is cranked, the current I returns from a negative value to a positive value (charge state) at time t6, and a maximum value appears in the current I at time t8. That is, the charging current becomes maximum. After time t8, the SOC of the storage battery 10 gradually rises, so that the current I also gradually drops. At time t10, the vehicle 1 is stopped, so idling stop is started. That is, since the engine 11 is stopped, the current I sharply drops and reaches the zero value at the time t12. After that, as before time t2, a small current is supplied to the auxiliary load 14 other than the starter 14a.

  In FIG. 5, during the period from time t2 to t8 and time t10 to t12, since the change in the current I is relatively large, the steep change condition (|ΔI|>ΔIth) in step S103 is likely to be satisfied, and step S106. There is a strong tendency for the resistance function R(I) to be updated. On the contrary, in the other sections, the change in the current I is relatively small, and therefore there is a high possibility that the steep change condition is not satisfied, and the SOF is more likely to be calculated in step S112.

<Measured data>
The data according to the present embodiment is shown in FIGS. 6(a) and 6(b) show that when the SOC is 80% and 90%, respectively, a predetermined current I (-300 to 100 [A in the illustrated example) is applied to the lead storage battery for a light vehicle from 0 [A]. ]] is a correlation diagram in which the voltage change amount when the step response is up to ΔV is ΔV, the vertical axis is the voltage change amount ΔV, and the horizontal axis is the current I.

  The black squares (♦) are the measured data, and the solid curves are the theoretical values calculated based on the Butler-Volmer equation (function F(x;α)). 6(a) and 6(b), it can be seen that the actually measured data substantially match the theoretical value. It is considered that the reason for this is that the resistance due to sulfuric acid is a liquid and follows the Butler-Volmer equation, and linear resistance (mainly metallic lead and lead dioxide) is connected in series to it. Further, the constant α becomes larger than 0.5, which is considered to be because it is difficult to charge the lead storage battery.

  As described above, according to the present embodiment, when the steep change condition is satisfied (when “Yes” is determined in step S103), the resistance function R(I) is updated in steps S104 to S106. In other cases, since SOF is calculated through steps S108 to S114, a precise resistance function (R(I)) can be obtained and the state of the storage battery 10 can be accurately estimated.

[Second Embodiment]
Next, the details of the vehicle 2 according to the second embodiment of the present invention will be described with reference to the block diagram shown in FIG. 7.
The vehicle 2 of the present embodiment is different in that a measuring unit 40 is provided instead of the battery monitoring device 30 of the vehicle 1 of the first embodiment. The measuring unit 40 has sensors (see FIG. 1) such as a voltage sensor 31, a current sensor 32, and a temperature sensor 37, and detection signals of these sensors are supplied to the ECU 15.

  In the present embodiment, the ECU 15 executes the idling stop control routine (FIG. 3) and the battery state monitoring routine (FIG. 4) in parallel in different processes. According to the present embodiment, since the processing contents executed by both the ECU 15 and the battery monitoring device 30 in the first embodiment can be executed only by the ECU 15, in addition to the effects of the first embodiment, a mounting space for the device can be saved. It has the effect of saving money.

[Modification]
The present invention is not limited to the above-described embodiment, and various modifications can be made. The above-described embodiment has been described as an example in order to facilitate understanding of the present invention, and is not necessarily limited to one including all the configurations described. Further, a part of the configuration of one embodiment can be replaced with the configuration of another embodiment, and the configuration of another embodiment can be added to the configuration of one embodiment. Further, it is possible to delete a part of the configuration of each embodiment or add/replace another configuration. The possible modifications to the above embodiment are, for example, as follows.

In each of the above-described embodiments, the abrupt change condition for updating the resistance function R(I) was “a condition of |ΔI|>ΔIth is satisfied” (step S103).
However, the change in the voltage V may be measured, and when the absolute value of the voltage change amount ΔV exceeds the voltage threshold ΔVth, the steep change condition may be satisfied.

Further, as the abrupt change condition, for example, "|dI/dt|>|I|/(R0*C) is satisfied" may be applied. Where R0 is the height scaling resistance and C is the polarization capacitance. The polarization capacitance C may be a preset value, for example, “100 farads”.

The “abrupt change condition” may be “within a predetermined period”. Here, the “predetermined period” is a period in which the current change amount ΔI is predicted to have a strong tendency to exceed the threshold value ΔIth. More specifically, "period during cranking" (time t2 to t4 in Fig. 5), "period after cranking until charging" (time t4 to t6), and "start of idling stop". The resistance function R(I) may be updated with the “period” (at the same time t10 to t12) as the “predetermined period”.
In addition, when both "there is each period (time t2 to t4, t4 to t6, t10 to t12)" and "|ΔI|>ΔIth is satisfied", the abrupt change condition is satisfied. It may be satisfied.

Further, in order to identify the resistance function R(I), it is particularly preferable to attach importance to the data in the period in which the negative current I having a large absolute value is obtained. You may make it use only the data of the period which is a negative value. Alternatively, only the data during the “during cranking period” (at the same time t2 to t4) may be used. In particular, during cranking, the sharp change condition is often satisfied continuously at a plurality of measurement timings. In this case as well, steps S104 and S106 may be executed at each measurement timing to update the resistance function R(I), or data of a plurality of continuous measurement timings may be treated as one set of data (set data). Good.

○ In addition, when the charged state is discharged or the current becomes zero, the resistance may increase due to gassing. In such a case, the "idling stop start period" (at the same time t10 to t12) may be excluded from the "predetermined period".

The minimization of the total error Em based on the above equation (5) increases the amount of calculation. Therefore, if it is desired to reduce the amount of calculation, the content of calculation may be simplified. For example, the following formula (11) is used by using the set data of the latest three periods of “period during cranking”, “period after cranking until charging”, and “starting period of idling stop”. May update V0, I0, R0.
e(j)=V0×{F(Ia(j)/I0;α)−F(Ib(j)/I0;α)}+R0×(Ia(j)−Ib(j))−(Va(j )-Vb(j)),

V0(n+1)=V0(n)−ε×Σe(j)×{F(Ia(j)/I0;α)−F(Ib(j)/I0;α)},

R0(n+1)=R0(n)-ε×Σe(j)×{ Ia(j)-Ib(j)},

1/I0(n+1)=1/I0(n)−ε×V0(n)×Σe(j)×{F′(Ia(j)/I0; α)−F′(Ib(j)/I0; α)}
... Formula (11)

Expression (11) is repeated a plurality of times until Σ|e(j)| ² becomes, for example, 1e ⁻⁶ or less. ε may be a predetermined minute constant (for example, 1e ⁻³ ) or ε may be the reciprocal of the update count n. As the initial values of V0, I0, and R0, the values at the time of the previous update may be used as they are. In addition, when α is also changed, the equation (5) may be partially differentiated by α and I0 instead of the equation (11), and the steepest descent or the Newton method may be used for the calculation. Here, F and the partial differential value of F by α and I0 may be numerically calculated in advance and held in a table or the like.

The equation (11) described above slightly changes the three values of V0, I0, and R0. Instead, the lateral scaling current I0 is changed slightly to change I0 to V0, R0. You may ask. An example is shown in formula (12). In equation (12), R0 and V0 are obtained by the least squares method with I0 as a fixed element. Since the equation (12) is a two-dimensional simultaneous equation of R0 and V0, R0 and V0 are obtained by solving the simultaneous equations.

In each of the above embodiments, the height scaling resistance R0 is variable, but for example, when the storage battery 10 is a lead storage battery, it is considered that the height scaling resistance R0 is mainly due to metallic lead and lead dioxide. You can In this case, the height scaling resistor R0 may be identified as a function of temperature T, and V0, I0 may be identified based on a recurrence formula. More specifically, this means that the height scaling resistor R0 is not updated in the above equation (11), so the following equation (13) may be applied instead of the above equation (12).

In each of the above embodiments, the polarization resistance value was ignored when calculating the cranking current Ic based on the equations (7) and (8). However, in some cases, it is desirable to add the polarization resistance value. Therefore, the calculation method in which the polarization resistance value is added will be described. At the time of cranking, the discharge current is about 300 [A], although it depends on the types of vehicles 1 and 2. At that time, the impedance generated by the polarization can be approximately considered to be only the resistance component.

Assuming that the polarization resistance value at that time is r, “r×cranking current Ic” becomes an error of the unsteady open circuit voltage OCV at the next cranking. When the actual current Ir, the estimated current Ie, and the resistance function R(I) are used when the voltage V becomes the minimum voltage by cranking, the polarization resistance value r can be expressed by the following equation (14).

r=R(I)×(Ie/Ir−1) Equation (14)

Therefore, the polarization resistance value r is obtained based on the equation (14) after cranking, and the cranking current Ic is estimated based on the following equation (15) instead of the above equation (7) at the next cranking. Good.

R(Ic)×Ic+(r+r0)×Ic=OCV Equation (15)

In equation (15), r0 is the starter resistance value, r is the polarization resistance value during cranking, and OCV is the unsteady open circuit voltage. Alternatively, “minimum voltage correction voltage during cranking=OCV estimated value×constant 1+charge during cranking×constant 2+constant 3” may be used. The constant 1, the constant 2, and the constant 3 may be estimated by least squares from the data of past cranking. The past actual value of the minimum voltage correction voltage during cranking is the cranking start voltage-the minimum voltage during cranking, and the charge during cranking is the current integration from the start of cranking to the application of the minimum cranking voltage. In this case, the estimated SOF is output as the equation (10) to which the minimum voltage correction voltage during cranking is added.

In each of the above embodiments, the battery monitoring device 30 or the ECU 15 calculates SOC, SOF, etc., but in addition to these, SOH (State Of Health) may be calculated. Here, SOH is an index indicating the deterioration state of the storage battery 10, and is defined by “(full charge capacity/initial full charge capacity)×100%”. As a method of calculating SOH, for example, the method disclosed in JP-A-2005-259624 may be applied.

In each of the above embodiments, the function F(x;α) and the derivative thereof, the function F′(x;α), are obtained by calculation. However, the functions F(x;α) and F′(x;α) are obtained. May be stored in the ROM 35 (see FIG. 2) as a table. Graphs of the function F(x) and the function F'(x) when the constant α is fixed to "0.69" are shown in FIGS. It should be noted that discrete values may be stored in the tables representing these graphs, and the function F(x) or the function F'(x) may be obtained by using linear interpolation, spline interpolation, or the like.

In this case, the cranking current Ic may be obtained by the recurrence formula of the following formula (16).

Ic(n+1)=Ic(n)-{V0*F(Ic(n)/I0)+(r0+R0)*Ic(n)-OCV}/{V0*F'(Ic(n)/I0)/I0+ (R0+R0)}
... Formula (16)

Further, “satisfying |ε(I)/ΔI|+|ε(V)/ΔV|<δ/2” may be adopted as the abrupt change condition for updating the resistance function R(I). Here, ΔI is the deviation between the current I at the current measurement timing and the previous current I (one sampling cycle before). Further, ε(I) is an error of the current sensor 32 and is, for example, 0.1 [A]. Further, ε(V) is an error of the voltage sensor 31 and is, for example, 30 [mV]. The errors ε(I) and ε(V) include quantization error, gain error, white noise, and the like. However, since the offset error is canceled by the difference between the measured values, it need not be particularly considered.

  Further, δ is a target error of |ΔV/ΔI|, which may be set to 5%, for example. For example, assuming that ε(I)=0.1 [A], ε(V)=30 [mV], and target error δ=0.05, “|0.1/ΔI|+|0.03/ΔV| When <0.025”, it is determined that the sharp change condition is satisfied. In other words, if the data in the case of “|0.1/ΔI|+|0.03/ΔV|≧0.025” is used, the error of the resistance function R(I) becomes large. However, the target error δ may be 10% or 20% depending on the situation.

Further, when the error ε(V) of the voltage sensor 31 is small, the error ε(V) is ignored, and attention is paid to ε(I) of the current sensor 32 to set the update condition of the resistance function R(I). May be set. That is, “meeting |ε(I)/ΔI|<δ/2” may be adopted as the abrupt change condition. For example, when the error ε(I) is 0.1 [A] and the target error δ is 0.01, “|ΔI|>20 [A]” is a sharp change condition. In other words, if the data in the case of “|ΔI|≦20 [A]” is used, the error of the resistance function R(I) becomes large. Also in this case, the target error δ may be 10% or 20% depending on the situation.

In each of the above-described embodiments, it is determined whether or not the steep change condition is satisfied using the current change amount ΔI of the current I between the measurement timings. However, a differentiating circuit is provided to output the differential value of the current I. The condition that the differential value has exceeded the predetermined value may be used as the abrupt change condition.

Since the hardware of the ECU 15 and the battery monitoring device 30 in each of the above-described embodiments can be realized by a general computer, the programs according to the flowcharts shown in FIG. 3 and FIG. You may distribute it.

The processing shown in FIGS. 3 and 4 has been described as software processing using a program in the above embodiment, but a part or all of the processing is performed by an ASIC (Application Specific Integrated Circuit), or It may be replaced with hardware-like processing using an FPGA (field-programmable gate array) or the like.

[Summary of composition and effects]
As described above, the battery state detection device (15, 30) in each embodiment is
An abrupt change that determines whether or not the current (I) that flows in the storage battery (10) or the voltage (V) of the storage battery (10) satisfies the abrupt change condition that the change exceeds a predetermined value within a predetermined time. A condition determination unit (301),
A resistance function identification unit (302) that identifies a resistance function (R(I)) that is a function of the internal resistance of the storage battery (10) with respect to the current (I), provided that the abrupt change condition is satisfied.
Using the identified resistance function (R(I)), a specific current calculation unit (calculates a specific current (Ic) that is a current (I) when the specific load (14a) is connected to the storage battery (10) ( 304),
It is characterized by having.
With this, the resistance function (R(I)) can be identified when a sharp change condition occurs, and the state of the storage battery can be accurately estimated by obtaining a precise resistance function (R(I)).

The specific load (14a) starts the engine (11) mounted on the vehicle (1, 2),
The specific current (Ic) is a current flowing through the specific load (14a) when the engine (11) is stopped when the vehicle (1, 2) is stopped and then the engine (11) is started,
The battery state detection device (15, 30) is
A voltage estimation unit (306) that calculates an estimated voltage value (SOF) that is an estimated value of the voltage (V) when the engine (11) is started, based on the specific current (Ic),
An engine control unit (15) for determining whether or not the engine (11) being started can be stopped or whether or not the stopped engine (11) needs to be started based on the estimated voltage value (SOF);
Is further included.
This makes it possible to appropriately determine whether or not the engine (11) can be stopped or whether or not the stopped engine (11) needs to be started.

In addition, the abrupt change condition determination unit (301)
The period until the minimum value appears in the current (I) after starting the engine (11) (t2 to t4), and the period until the current (I) becomes zero after the minimum value appears in the current (I). (T4 to t6), or one of a plurality of periods (t10 to t12) until the current (I) becomes zero after the engine (11) has been stopped, satisfying the sharp change condition It is characterized by determining.
Thereby, the resistance function (R(I)) can be identified within an appropriate period.

The resistance function identification unit (302) calculates the resistance function (R(I)) based on the Butler-Volmer equation,
The scaling current (I0), the scaling voltage (V0), and the scaling resistance (R0) are used as parameters of the resistance function (R(I)), and the voltage drop obtained by the scaling resistance (R0) and the actual voltage drop are The function to obtain the deviation (E),
The scaling current (I0) is updated with a value (small value) proportional to the deviation (E), and the scaling voltage (V0) and the scaling resistance (R0) are set so that the sum of squares (Em) of the deviation (E) is minimized. ) And a scaling current (I0),
It is characterized by having.
As a result, the accurate scaling voltage (V0), scaling resistance (R0) and scaling current (I0) can be obtained.

The specific current calculation unit (304) has a function of estimating the unsteady open circuit voltage (OCV) of the storage battery (10) and a function of calculating a specific current (Ic) based on the unsteady open circuit voltage (OCV). A function of calculating an estimated voltage value (SOF) of the storage battery (10) when the specific load (14a) is connected to the storage battery (10) based on the specific current (Ic);
Thus, it is possible to appropriately determine whether or not the engine (11) can be stopped or whether or not the stopped engine (11) needs to be started.

Further, the specific current calculation unit (304) calculates the unsteady open circuit voltage (OCV) based on the current voltage (V), the current (I), the Butler-Volmer equation, and the scaling resistance (R0). It is characterized by estimating it.
This allows the unsteady open circuit voltage (OCV) to be accurately estimated.

In addition, the specific current calculation unit (304), the voltage (V) of the storage battery (10) when the storage battery (10) is connected to the specific load (14a), and the actual current (I) flowing through the specific load (14a). ) And a specific load resistance value (r0) that is a resistance value of the specific load (14a).
As a result, the specific load resistance value (r0) can be accurately calculated, and the state of the storage battery (10) can be detected more accurately.

Further, the specific current calculation unit (304) is based on the current (I) flowing through the storage battery (10) when the storage battery (10) is connected to the specific load (14a) and the specific current (Ic) obtained in advance. The function of obtaining the polarization resistance value (r) of the storage battery (10),
Based on the polarization resistance value (r), the estimated value of the specific current (Ic) when connecting the specific load (14a) to the storage battery (10) next time is updated.
Thereby, the specific current (Ic) can be accurately estimated, and the state of the storage battery (10) can be detected more accurately.

1, 2 vehicle 10 storage battery 11 engine 12 generator 14 supplementary load 14a starter (specific load)
15 ECU (engine control unit, battery state detection device)
30 Battery monitoring device (battery status detection device)
301 abrupt change condition determination unit (abrupt change condition determination means)
302 resistance function identification unit (resistance function identification means)
304 Cranking current calculator (specific current calculator, specific current calculator)
306 Voltage estimating unit (voltage estimating means)
ΔIth (threshold)
I0 Horizontal scaling current (scaling current)
Ic cranking current (specific current)
r0 Starter resistance value (specific load resistance value)

Claims (12)

A sharp change condition determination unit that determines whether or not the current flowing through the storage battery or the voltage of the storage battery satisfies a sharp change condition indicating that there is a change exceeding a predetermined value within a predetermined time.
A resistance function identification unit that identifies a resistance function that is a function of the internal resistance of the storage battery with respect to the current, provided that the abrupt change condition is satisfied,
Using the identified resistance function, a specific current calculation unit that calculates a specific current that is the current when a specific load is connected to the storage battery,
Have
The resistance function identification unit,
Scaling current for adjusting vertical/horizontal scaling, scaling voltage for adjusting vertical/horizontal scaling, and scaling resistor for adjusting vertical scaling are parameters of the resistance function, and are calculated by the scaling resistor . And the function to find the deviation between the actual voltage drop and
A function of updating the scaling current with a value proportional to the deviation, and obtaining the scaling voltage , the scaling resistor, and the scaling current so that the sum of squares of the deviation is minimized,
A battery state detection device comprising: The specific load is to start an engine mounted on the vehicle,
The specific current is the current that flows into the specific load when the engine is stopped when the vehicle is stopped and the engine is subsequently started,
A voltage estimation unit that calculates a voltage estimation value that is an estimation value of the voltage when the engine is started, based on the specific current,
An engine control unit that determines, based on the estimated voltage value, whether or not the engine that is being started can be stopped, or whether or not the engine that is stopped needs to be started.
The battery state detection device according to claim 1, further comprising: The abrupt change condition determination unit,
A period until a minimum value appears in the current after starting the engine, a period until the current becomes zero after the minimum value appears in the current, or the current after the engine is stopped. The battery state detection device according to claim 2, wherein it is determined that the abrupt change condition is satisfied in one or a plurality of periods until the value becomes zero. The battery state detection device according to claim 3, wherein the resistance function identification unit calculates the resistance function based on a Butler-Volmer equation. The specific current calculation unit, a function of estimating the unsteady open circuit voltage of the storage battery, a function of calculating the specific current based on the unsteady open circuit voltage, based on the specific current, the specific load on the storage battery. A function of calculating an estimated voltage value of the storage battery when is connected,
The battery state detection device according to claim 1, further comprising: The specific current calculation unit estimates the unsteady open circuit voltage based on the current voltage, the current, the Butler-Volmer formula, and the scaling resistor. The battery state detection device described. The specific current calculation unit, based on the voltage of the storage battery when the storage battery is connected to the specific load, and the actual current flowing in the specific load, a specific load that is a resistance value of the specific load. The battery state detecting device according to claim 6, having a function of calculating a resistance value. The specific current calculation unit, a function of obtaining a polarization resistance value of the storage battery, based on the current flowing in the storage battery when the storage battery is connected to the specific load, and the specific current obtained in advance,
The battery state detection device according to claim 6, wherein the estimated value of the specific current when the specific load is connected to the storage battery next time is updated based on the polarization resistance value. The battery state detection device according to claim 1, wherein the abrupt change condition is a condition that a change of 20 amperes or more occurs in the current flowing through the storage battery within a period of 0.1 second or less. .. Storage battery,
An engine that appropriately charges the storage battery,
A current flowing through the storage battery or a voltage of the storage battery, a sharp change condition determination unit that determines whether or not a predetermined sharp change condition is satisfied,
A resistance function identification unit that identifies a resistance function that is a function of the internal resistance of the storage battery with respect to the current, provided that the abrupt change condition is satisfied,
Using the identified resistance function, a specific current calculation unit that calculates a specific current that is the current when a specific load is connected to the storage battery,
Have
The resistance function identification unit,
Scaling current for adjusting vertical/horizontal scaling, scaling voltage for adjusting vertical/horizontal scaling, and scaling resistor for adjusting vertical scaling are parameters of the resistance function, and are calculated by the scaling resistor . And the function to find the deviation between the actual voltage drop and
A function of updating the scaling current with a value proportional to the deviation, and obtaining the scaling voltage , the scaling resistor, and the scaling current so that the sum of squares of the deviation is minimized,
A vehicle having: Computer,
The current flowing through the storage battery or the voltage of the storage battery determines a sharp change condition, which determines whether or not the sharp change condition is determined.
Resistance function identification means for identifying a resistance function that is a function of the internal resistance of the storage battery with respect to the current, provided that the abrupt change condition is satisfied,
Using the identified resistance function, a specific current calculation means for calculating a specific current which is the current when a specific load is connected to the storage battery,
Function as
The resistance function identification means,
Scaling current for adjusting vertical/horizontal scaling, scaling voltage for adjusting vertical/horizontal scaling, and scaling resistor for adjusting vertical scaling are parameters of the resistance function, and are calculated by the scaling resistor . Means for obtaining the deviation between the voltage drop and the actual voltage drop,
Means for updating the scaling current with a value proportional to the deviation, and obtaining the scaling voltage , the scaling resistor, and the scaling current so that the sum of squares of the deviation is minimized,
Program to function as. A process of determining whether or not the current flowing through the storage battery or the voltage of the storage battery satisfies a predetermined sharp change condition,
A resistance function identification step of identifying a resistance function that is a function of the internal resistance of the storage battery with respect to the current, provided that the abrupt change condition is satisfied,
Using the identified resistance function, a step of calculating a specific current that is the current when a specific load is connected to the storage battery,
Have
The resistance function identification process is
Scaling current for adjusting vertical/horizontal scaling, scaling voltage for adjusting vertical/horizontal scaling, and scaling resistor for adjusting vertical scaling are parameters of the resistance function, and are calculated by the scaling resistor . And the process of obtaining the deviation between the actual voltage drop and
Updating the scaling current with a value proportional to the deviation, and obtaining the scaling voltage , the scaling resistor, and the scaling current so that the sum of squares of the deviation is minimized,
A battery state detecting method comprising:

Images