JP5307908B2 - Battery state estimation device - Google Patents

Description

  The present invention relates to a battery state estimation device capable of accurately estimating an internal state of a battery.

  Among the batteries, a rechargeable secondary battery is employed in, for example, an electric vehicle. In this case, it is necessary to know the distance that can be traveled by the battery, the current value that can be charged and discharged, etc. In order to grasp these, the battery state of charge (SOC), which is the internal state quantity of the battery, It is necessary to detect the state of health (SOH). However, since these internal state quantities cannot be directly detected, a current integration method (also called a coulomb count method or book keeping method) that detects charge / discharge current values in time series to estimate the internal state, a battery, Establish the battery model's sequential parameters and estimate the open-circuit voltage of the battery by constructing a model and comparing the input / output with the actual battery and reducing the difference with an adaptive filter such as Kalman filter. The open-circuit voltage estimation method for estimating the charging rate is often used.

Although the current integration method is excellent for estimating the charging rate in a short time, there are drawbacks that errors are accumulated and it is difficult to return to the base, and that constant observation is required. On the other hand, the sequential parameter method does not require constant observation because both input and output are observed and does not accumulate errors, but has a drawback that the estimation accuracy of the charging rate in a short time is not good.
Therefore, the charging rate is estimated by combining these two methods.
As such a prior art, the thing of patent document 1 is known.

  That is, the secondary battery charging rate estimation device described in Patent Literature 1 constructs a battery model, performs sequential parameter estimation using an adaptive digital filter, and performs a first charging rate estimation calculation. Means, a second charge rate estimating means for estimating and calculating a second charge rate using a current integration method in a current state in which it is difficult to estimate the charge rate using an adaptive digital filter, a first charge rate and a second charge rate And a final charge rate estimated value selection unit that appropriately selects one of the charge rates. In this case, the final charge rate estimated value selection means selects the first charge rate when the sign of the current is inverted, and when only the charge rate or only the discharge continues from the point in time for a preset first predetermined time. The second charging rate is selected.

JP 2008-164417 A

However, the above-described conventional charging rate estimation device has the problems described below.
That is, when adopting the sequential parameter method, an equivalent circuit model of the battery expressed by the impedance at the interface of the battery, the impedance at each part of the electrolyte, or the like is used.
In this case, the battery has a fast response part at the interface where the charge transfer process takes place (for example, a time constant of several microseconds to several hundred milliseconds) and diffusion in a diffusion layer between the electrolyte interface and the bulk region. Since there is a slow response part (for example, a time constant of 1 second to several hours) that becomes a process, an equivalent circuit model of the battery uses a mathematical model that represents them.

In this case, for the fast response part of the battery, a parameter representing the internal state of the battery can be easily estimated by the sequential parameter method from the viewpoint of S / N ratio and observability.
On the other hand, for the slow response part, the S / N ratio is small, and it is difficult to accurately estimate the parameters by the sequential parameter method from the viewpoint of observability.

In an environment where the fast response of the battery is used as in a hybrid vehicle (HEV: Hybrid Electric Vehicle), the overvoltage component can be accurately calculated even when successive parameter estimation is performed. It is possible to accurately calculate and estimate the charging rate.
On the other hand, in an environment where the battery's slow response part is used, such as an electric vehicle (EV), when parameter estimation is performed sequentially, the parameter estimation accuracy of the battery's slow response part deteriorates, resulting in overvoltage. As a result, there is a problem that the estimation accuracy of the state quantity of the battery such as the open circuit voltage and the charging rate is deteriorated.

  In the above case, in order to obtain the slow response portion of the battery, it is possible to input an arbitrary waveform, and if the conditions for accurately obtaining the open circuit voltage of the battery are aligned, for example, the following This method makes it possible to accurately estimate the parameters of the slow part of the battery.

  In other words, the terminal voltage value Vt (k) of the battery is measured using an accurate voltage sensor, while the charge / discharge current flowing into and out of the battery is measured using an accurate current sensor of the battery shunt resistance type to measure the coulomb count. The charge rate SOC (k) is calculated using the method, and it corresponds to the charge rate SOC (k) using a look-up table that shows the relationship data between the charge rate and the open circuit voltage obtained by experimentation in advance. An open circuit voltage value OCV (k) is obtained. Next, an overvoltage η (k) is obtained by subtracting the open circuit voltage value OCV (k) from the terminal voltage value Vt (k) by a subtractor.

Then, an equivalent circuit model of the overvoltage part is constructed using the current as input and the overvoltage as output. The equivalent circuit model of the overvoltage portion may be a mathematical model representing the inside of the battery, such as a diffusion equation such as a Foster-type equivalent circuit model.
In this way, it is possible to estimate the parameter of the slow response part of the battery by experiment etc., but considering the environment where the battery is actually used, for example, in the case of EV, an arbitrary waveform can be obtained. There is almost no input, and it is almost the condition / situation where it is difficult to obtain the open-circuit voltage with high accuracy.
Therefore, under the situation where the battery is actually used, it is very difficult to estimate the parameters of the slow response part of the battery, and as a result, the internal state of the battery such as the open circuit voltage and the charging rate of the battery is estimated with high accuracy. There is a problem that it is difficult to estimate.

  The present invention has been made paying attention to the above-mentioned problems, and the object of the present invention is to improve the estimation accuracy of the battery overvoltage in consideration of the slow response part of the battery, thereby accurately determining the internal state of the battery. It is an object of the present invention to provide a battery state estimation device that can be well estimated.

For this purpose, the battery state estimation device according to the present invention as set forth in claim 1 comprises:
Charge / discharge current detection means for detecting the charge / discharge current value of the battery;
A terminal voltage detecting means for detecting a terminal voltage value of the battery;
An equivalent circuit model having a fast response portion and a slow response portion of the battery;
Sequential parameter estimation means for performing sequential parameter estimation using only the fast response portion of the equivalent circuit model based on the charge / discharge current value input from the charge / discharge current detection means and the terminal voltage value input from the terminal voltage detection means When,
A state quantity that estimates the overvoltage value of the slow response part using the slow response part of the equivalent circuit model based on the charge / discharge current value input from the charge / discharge current detection means and the terminal voltage value input from the terminal voltage detection means An estimation means;
Multiplication means for obtaining the overvoltage value of the fast response part by multiplying the parameter estimated by the sequential parameter estimation means by the charge / discharge current value;
Overvoltage detection means for adding the overvoltage value of the fast response part obtained by the multiplication means and the overvoltage value of the slow response part obtained by the state quantity estimation means to obtain the battery overvoltage value;
It is provided with.

The battery state estimation device according to the present invention as set forth in claim 2 comprises:
The battery state estimation device according to claim 1,
Subtracting means for subtracting the overvoltage value obtained by the terminal voltage detecting means from the terminal voltage value obtained by the terminal voltage detecting means to obtain the open-circuit voltage value of the battery;
An open-circuit voltage-charge rate estimating means for obtaining a charge rate of the battery based on the open-circuit voltage value obtained by the subtracting means;
It is characterized by having

The battery state estimation apparatus according to the present invention as set forth in claim 3 comprises:
In the battery state estimation device according to claim 1 or 2,
It has a filter processing means for removing the slow response portion of the terminal voltage obtained by the terminal voltage detection means and sequentially inputting it to the parameter estimation means.
It is characterized by that.

The battery state estimation apparatus according to the present invention as set forth in claim 4 comprises:
In the battery state estimation device according to claim 3,
The filter processing means removes the slow response portion from the charge / discharge current obtained by the charge / discharge current detection means and sequentially inputs it to the parameter estimation means.
It is characterized by that.

  In the battery state estimation apparatus according to the first aspect of the present invention, sequential parameter estimation is performed using a fast response portion of an equivalent circuit model of the battery, and state quantity estimation is performed for a slow response portion of the battery. Therefore, the estimation accuracy of the battery overvoltage can be improved, and therefore the internal state of the battery can be estimated with high accuracy.

  In the battery state estimating apparatus according to the second aspect of the present invention, the overvoltage value can be subtracted from the terminal voltage value to accurately obtain the open circuit voltage value of the battery. Therefore, the charging rate, which is one of the internal states of the battery, can be accurately estimated.

  In the battery state estimation device according to the third aspect of the present invention, since the filter processing unit is provided to remove the slow response portion of the terminal voltage and input to the sequential parameter estimation unit, the terminal It is possible to easily and reliably remove the overvoltage value overlap calculation in the slow response portion and the fast response portion based on the voltage.

  In the battery state estimation apparatus according to the fourth aspect of the present invention, the filter processing unit removes the slow response portion from the charge / discharge current and inputs it to the sequential parameter estimation unit. Calculation of minutes becomes easy.

It is a block diagram which shows the relationship of the functional block which comprises the battery state estimation apparatus of Example 1 of this invention connected to the actual battery. It is a block diagram which shows the relationship of the functional block which comprises the state quantity estimation part used with the battery state estimation apparatus of FIG. It is a figure showing the battery equivalent circuit model of the slow response part of the real battery used in the state quantity estimation part of FIG. FIG. 3 is a diagram for explaining a sampling method for separating a fast response portion and a slow response portion of the battery in the equivalent circuit model of the battery used in the battery state estimation device of FIG. 1. FIG. 5 is a Bode diagram used in an example in which a boundary between a fast response portion and a slow response portion of the battery is determined using the sampling method of FIG. 4. It is the figure which compared the estimation result of the charging rate of the battery between a prior art and the battery state estimation apparatus of Example 1. FIG. It is a block diagram which shows the relationship of the functional block which comprises the battery state estimation apparatus of Example 2 of this invention connected to the actual battery. It is a figure which shows the relationship of the functional block which comprises the filter process part used with the battery state estimation apparatus of Example 2. FIG.

  Hereinafter, embodiments of the present invention will be described in detail based on examples shown in the accompanying drawings.

First, the overall configuration of the battery state estimation device of Example 1 will be described.
The battery state estimation apparatus according to the first embodiment is connected to an actual battery (secondary battery such as a lithium ion battery) 1 that is mounted on, for example, an electric vehicle and can supply power to a drive motor (not shown). , Current sensor 2, voltage sensor 3, sequential parameter estimation unit 4, multiplier 5, first subtractor 6, state quantity estimation unit 7, adder 8, second subtractor 9, open circuit voltage -The charging rate conversion part 10 is provided.

The current sensor 2 collects a part of braking energy by causing the electric motor to function as a generator at the time of vehicle braking, or to collect a part of braking energy when supplying power from the actual battery 1 to the drive motor or the like. The charging / discharging current value Ia detected here detects the magnitude of the charging current in the case of charging from the equipment, and the sequential parameter estimation unit 4 as an input signal in which charging is + and discharging is −, It is output to the multiplier 5 and the state quantity estimation unit 7, respectively.
In addition, what has various structures and forms can be employ | adopted for the current sensor 22 suitably, and is equivalent to the charging / discharging electric current detection means of this invention.

The voltage sensor 3 detects a voltage value between the terminals of the battery 1, and the detected terminal voltage value Va is output to the sequential parameter estimation unit 4, the first subtractor 6, and the second subtracter 9, respectively. The
As the voltage sensor 3, ones having various structures and formats can be adopted as appropriate and correspond to the terminal voltage detection means of the present invention.

The sequential parameter estimator 4 has a fast response portion (R ₀ , R ₁ and C ₁ , R ₂ and C ₂ , primary and secondary resistance-capacitors in the equivalent circuit model of the battery shown in FIG. The charge / discharge current value Ia obtained from the current sensor 2 and the terminal voltage value Va obtained from the voltage sensor 3 as input signals using, for example, a Kalman filter. Compare the value (terminal voltage value Va obtained from the voltage sensor 3) with the output value (estimated terminal voltage value) of the fast response part of the battery equivalent circuit model so that the difference between these output values is reduced. The parameters of the fast response part are estimated by sequentially adjusting the parameters of the state equation. Details of parameter estimation by the Kalman filter are described in Japanese Patent Application No. 2011-007874 of the present applicant.
The resistance values (R ₀ , R ₁ , R ₂ ) and the capacitor capacities (C ₁ , C ₂ ), which are parameters estimated by the sequential parameter estimation unit 4, are output to the multiplier 5.
The sequential parameter estimation unit 4 corresponds to sequential parameter estimation means of the present invention.

The multiplier 5 calculates the charge / discharge current value Ia detected by the current sensor 2, the resistance values (R ₀ , R ₁ , R ₂ ) and the capacitor capacity (C ₁ , C ₂ ) estimated by the sequential parameter estimation unit 4. Are multiplied to obtain a first overvoltage value V ₀₁ . The first overvoltage value V ₀₁ is output to the first subtracter 6 and the adder 8, respectively.
The multiplier 5 corresponds to the multiplication means of the present invention.

The first subtracter 6 subtracts the first overvoltage value V ₀₁ obtained by the multiplier 5 from the terminal voltage value Va detected by the voltage sensor 3 to obtain a first open-circuit voltage value OCV1, which is obtained as a state quantity. Output to the estimation unit 7.

The state quantity estimation unit 7 estimates the state quantity based on the charge / discharge current value Ia detected by the current sensor 2 and the first open-circuit voltage value OCV1 obtained by the first subtractor 6 to calculate the second overvoltage value. V ₀₂ is obtained and output to the adder 8. The estimation of the second overvoltage value V _{02 in} the state quantity estimation unit 7 will be described in detail later.
The state quantity estimation unit 7 corresponds to the state quantity estimation of the present invention.

The adder 8 adds the first overvoltage value V _{01 for} the fast response of the battery obtained by the multiplier 5 and the second overvoltage value V _{02 for} the slow response portion of the battery obtained by the state quantity estimation unit 7. Thus, the battery overvoltage value V ₀ is obtained and output to the second subtractor 9.
The adder 8 corresponds to the adding means of the present invention.

Second subtractor 9, to obtain a second open-circuit voltage value OCV2 battery by subtracting the overvoltage value V ₀ obtained by the adder 8 from the terminal voltage Va detected by the voltage sensor 3, which open circuit voltage - the charging rate conversion To the unit 10.

The open-circuit voltage-charge rate conversion unit 10 stores data representing a relationship between the open-circuit voltage and the charge rate obtained in advance as a look-up table, and the second open circuit obtained by the second subtracter 9 The voltage value OCV2 is input, and the corresponding charge rate SOC _OCV is output.
The open-circuit voltage-charge rate conversion unit 10 corresponds to the open-circuit voltage-charge rate conversion means of the present invention.

Next, the configuration of the state quantity estimation unit 7 that estimates the state quantity of the slow response portion of the battery will be described with reference to FIGS. 2 and 3.
The state quantity estimation unit 7 is a mechanism for estimating the state x from the input u and the output y when the state x cannot be observed directly, that is, an observer. In this embodiment, the state quantity estimation unit 7 is equivalent to FIG. A circuit model and the Kalman filter shown in FIG. 2 are used.

  In the Kalman filter, a model of a target system (in this embodiment, an equivalent circuit model of a slow response part of a battery) is designed, and the same input signal (in this embodiment, an actual battery 1) is input to the same system (the actual battery 1 in this embodiment). In this embodiment, the charging / discharging current Ia) detected by the current sensor 2 is input, and the outputs (in this embodiment, the overvoltage) in both cases are compared.・ By multiplying the gain K and feeding back to the model, the above model is corrected so that the error between them is minimized. By repeating this, the true internal state quantity is estimated.

In FIG. 2 showing the state quantity estimation unit 7, the part BT surrounded by the upper broken line is a part corresponding to the actual battery 1, and the part BM surrounded by the lower broken line is a part corresponding to the slow response part of the battery of FIG. . A battery equivalent circuit model corresponding to the slow response portion of the battery will be described in detail later.
The real part BT includes a first coefficient unit 71, a first adder 72, a first integrator 73, a second coefficient unit 74, and a third coefficient unit 75.
On the other hand, the model part BM has the same configuration as the real part BT, that is, the fourth coefficient unit 81, the second adder 82, the second integrator 83, the fifth coefficient unit 84, and the sixth coefficient unit 85. It is equipped with.
The state quantity estimation unit 7 further includes a subtractor 86 for obtaining a subtraction value obtained by subtracting the output value w (t) of the model part BM from the output value y (t) of the real part BT, and the subtraction value obtained by the subtractor 86. And a multiplier 87 that multiplies them by Kalman gain K and inputs the result to the second adder 82.

In FIG. 2, the input u (t) is the charge / discharge current value Ia in this embodiment, and this input signal is supplied to the first coefficient unit 71 of the real part BT and the fourth coefficient unit 81 of the model part BM, respectively. Entered.
The first coefficient unit 71 multiplies the input u (t) by the coefficient B and inputs this value to the first adder 72. In the first adder 72, the output value from the first coefficient unit 71 and the output value from the third coefficient unit 75 are added and this added value is input to the first integrator 73. In the first integrator 73, the inputted addition value is time-integrated, and this integrated value is input to the second coefficient unit 74 and the third coefficient unit 75. The second coefficient unit 74 multiplies the integral value obtained by the first integrator 73 by the coefficient C to obtain an output value y (t). In this embodiment, the output value y (t) is the overvoltage value of the slow response part of the real battery. The current is + when charging and-when discharging.
On the other hand, the third coefficient unit 75 multiplies the integral value obtained by the integrator 73 by the coefficient A and inputs this value to the first adder 72.

The fourth coefficient unit 81 multiplies the input value u (t) by the coefficient B and inputs this value to the second adder 82. In the second adder 82, the output value from the fourth coefficient unit 81, the output value from the sixth coefficient unit 85, and the output value obtained by the multiplier 87 are added, and this added value is added to the second integrator 83. To enter. In the second integrator 83, the inputted addition value is time-integrated, and this integrated value is input to the fifth coefficient unit 84 and the sixth coefficient unit 85. The fourth coefficient unit 84 multiplies the integral value obtained by the second integrator 83 by the coefficient C to obtain an output value w (t). In this embodiment, the output value w (t) is an overvoltage value of a slow response portion of the battery equivalent circuit model.
On the other hand, the sixth coefficient unit 85 multiplies the integral value obtained by the second integrator 83 by the coefficient A and inputs this value to the second adder 82.

Here, the slow response part of the battery equivalent circuit model is slightly different from the battery equivalent circuit model shown in FIG. 4, and the circuit model shown in FIG. 3 is used.
That is, as shown in FIG. 3, this model is equivalent to a resistor R ₀ , which is an equivalent circuit model portion corresponding to the fast answer portion of the battery in FIG. 4, and a primary-secondary resistor-capacitor parallel circuit (R _1). , C ₁ , R ₂ , C ₂ ) are removed, and the remaining third to fifth resistor-capacitor parallel circuits (R ₃ , C ₃ , R ₄ , C ₄ , R ₅ , C ₅ ) and this And the open circuit voltage (C _OCV ) of the battery connected in series.

Then, in the state quantity estimation unit 7, the open circuit voltage value C _OCV , the voltage value V3 at R ₃ and C ₃ , the voltage value V4 at R ₄ and C ₄ , and the voltage value V5 at R ₅ and C ₅ Then, the state quantity is estimated by the observer using the state quantity x to estimate the overvoltage w (t) (= V3 + V4 + V5) in the slow response part of the battery.

In the equivalent circuit model of the slow response part of the battery in FIG. 3, it is necessary to perform sequential parameter estimation so that the overvoltage value does not overlap with the fast response part of the battery. For this reason, in this embodiment, the equivalent circuit model is divided into a fast response portion and a slow response portion by changing the sampling period in the sequential parameter estimation.
In this embodiment, as shown in FIG. 4, when parameters are estimated at different sampling periods (10 seconds and 0.1 seconds) for an overvoltage battery equivalent circuit model, which frequency band parameters can be obtained is examined. . The Bode diagram obtained at this time is shown in FIG.

  In the Bode diagram of FIG. 5 (frequency on the horizontal axis and amplitude on the vertical axis), the broken line is a filter process for the charge / discharge current value Ia detected by the current sensor 2 and the terminal voltage value Va obtained by the voltage sensor 3. Otherwise, the alternate long and short dash line performs high frequency cut filtering on the charge / discharge current value Ia and the terminal voltage value Va and performs sampling at 10 second intervals, while the solid line performs the same filter processing at 0.1 second intervals. When sampling is performed, each system identification result is shown.

As can be seen from FIG. 5, in the case of the experiment in which the parameter estimation is performed sequentially with a sampling period of 10 seconds, it is shown that the matching is achieved in the band of the slow response part. However, in actuality, when the sequential parameters are performed at a sampling period of 10 seconds, the slow response part of the battery has a small S / N ratio and it is difficult to estimate the parameters sequentially from the viewpoint of observability.
On the other hand, when the parameter estimation is performed sequentially with a sampling period of 0.1 seconds, it is shown that the fast response part bands of the battery are identical, but the slow response part of the battery is not identical.

  That is, in the band of the fast response part of the battery, different from the slow response part of the battery, the sequential parameter estimation can be easily performed from the viewpoint of S / N ratio and observability. Therefore, if parameter estimation is performed with the sampling period set to 0.1 seconds, it is possible to calculate parameters for only the fast response part. As a result, using these parameters, the overvoltage for only the fast response part is calculated. Will be able to.

Next, the operation of the battery internal state estimating device of Example 1 configured as described above will be described.
The current sensor 2 detects a charge / discharge current value Ia charged / discharged to / from the actual battery 1, and inputs this value to the multiplier 5, the sequential parameter estimation unit 4, and the state quantity estimation unit 7.
On the other hand, the voltage sensor 3 detects the terminal voltage value Va of the actual battery 1 and inputs this value to the sequential parameter estimation unit 4, the first subtracter 6, and the second subtracter 9.

The sequential parameter estimation unit 4 is based on the input charge / discharge current Ia and the terminal voltage Va, and is equivalent circuit model of the fast response part of the battery in FIG. 4 (the resistance R ₀ and the first and second order in FIG. 4). Resistance-capacitor parallel circuit, R ₁ , C ₁ , R ₂ , C ₂ ), resistance value (R ₀ , R ₁ , R ₂ ) and capacitor capacitance (C1) , C2). These resistance values are input to the multiplier 5 and multiplied by the charge / discharge current value Ia also input from the current sensor 2 to obtain the first overvoltage value V ₀₁ . The first overvoltage value V ₀₁ is input to the adder 8 and the first subtracter 6.

In the first subtracter 6, the first overvoltage value V ₀₁ is subtracted from the terminal voltage value Va to obtain the first open voltage value OCV 1, and this first open voltage value OCV 1 is input to the state quantity estimation unit 7.
The state quantity estimation unit 7 is based on the input charge / discharge current value Ia and the first open-circuit voltage value OCV1, and is equivalent circuit model (third-order to fifth-order resistance − Capacitor parallel circuit, R ₃ , C ₃ , R ₄ , C ₄ , R ₅ , C ₅ ) using the Kalman filter and the slow response part state quantity (C _OCV , V3, V4, V5) to the second overvoltage value Estimate V ₀₂ (= V3 + V4 + V5). The second overvoltage value V ₀₂ is input to the adder 8.

The adder 8 adds the input first overvoltage value V ₀₁ and the second overvoltage value V ₀₂ to obtain the battery overvoltage value V ₀ . This overvoltage value V ₀ is input to the second subtractor 9. The second subtracter 9 subtracts the overvoltage value V ₀ from the terminal voltage Va to obtain a second open-circuit voltage value OCV 2, which is input to the open-circuit voltage-charge rate conversion unit 10.

The open-circuit voltage-charge rate conversion unit 10 obtains the charge rate SOC _OCV corresponding to the input second open-circuit voltage value OCV2 using the open-circuit voltage-charge rate look-up table, and can drive this value, for example, Output to a necessary calculation unit such as a distance calculation unit (not shown).

Here, in order to confirm the effect of the battery state estimation apparatus of Example 1, the result of an experiment performed using an actual battery is shown in FIG. In the figure, the horizontal axis represents the measurement time (seconds), and the vertical axis represents the charging rate. The solid line indicates the true value of the charging rate, the alternate long and short dash line indicates the method according to the prior art, and the broken line indicates the method according to the present embodiment.
As is clear from FIG. 6, the charging rate obtained by the method of the prior art has a larger error than the true value, whereas the charging rate obtained by the method according to the present embodiment is almost equal to the true value. It turns out that it becomes an equivalent value and the estimation accuracy of a charging rate is high.

As can be seen from the above description, the battery state estimation device of Example 1 has the following effects.
In the battery state estimation apparatus according to the first embodiment, the first parameter corresponding to the fast response of the battery is obtained by multiplying the parameter obtained by the successive parameter estimation using the equivalent circuit model of the fast response part of the battery and the current value Ia. The overvoltage value V ₀₁ is calculated, the state quantity is estimated for the slow response part of the battery, the second overvoltage value V ₀₂ of the slow response part of the battery is calculated, and these first and second overvoltage values V ₀₁ , V _Using the overvoltage value V ₀ obtained by adding ₀₂ , the second open-circuit voltage value OCV2 is calculated to estimate the internal state of the battery (charge rate SOC _OCV, etc.). In the environment, it is possible to accurately estimate the internal state of the battery in consideration of the slow response part of the battery, which is difficult with the sequential parameter method.

To determine the charging rate of the battery, determine the second open-circuit voltage value OCV2 from the terminal voltage value Va by subtracting said over-voltage value V _0, open circuit voltage - the second open-circuit voltage value by using the relationship data of the charging rate OCV2 Since the corresponding charge rate SOC _OCV is obtained, the charge rate can be obtained with a simple calculation.

In sequential parameter estimation and state quantity estimation, an equivalent circuit model of a battery such as a Foster type can be used. In this case, the fast response part of the battery is based on the waveform of the input charge / discharge current value Ia and the terminal voltage value Va. In addition, it is possible to use sequential parameter estimation in a portion excluding a slow response portion (current and voltage components generated by parameters obtained by state quantity estimation) in the equivalent circuit model of the battery.
That is, the fast response portion and the slow response portion are separated by changing the sampling period at the sequential parameter. Therefore, it is possible to prevent the overvoltage value of the fast response part and the overvoltage value of the slow part of the battery from being calculated redundantly.

  Next, another embodiment will be described. In the description of the other embodiments, the same components as those of the first embodiment are not shown, or the same reference numerals are given and the description thereof is omitted, and only the differences are described.

  As shown in FIG. 7, the battery internal state estimation apparatus according to the second embodiment is provided with a filter processing unit 11 between the current sensor 2, the voltage sensor 3, and the sequential parameter estimation unit 4. This is different from the internal state estimation apparatus (shown in FIG. 1). The filter processing unit 11 corresponds to the filter processing means of the present invention.

  The filter processing unit 11 removes at least the slow response (diffusion resistance) from the terminal voltage value Va input from the voltage sensor 3, and converts the fast response (connection resistance + electrolyte resistance + charge transfer resistance) to the filter processing voltage. The value Vb is sequentially input to the parameter estimation unit 4.

In this embodiment, the filter processing unit 11 is composed of a low pass filter. An example of this low-pass filter is shown in FIG. 8. Here, the voltage value Vc of the slow response part obtained by calculation using the charge / discharge current value Ia is subtracted from the terminal voltage value Va to obtain the fast response part. A filter processing voltage value Vb which is a voltage value is calculated.
That is, the voltage value Vc of the slow response portion is represented by the transfer function 12 corresponding to the third order R3, C3 and the fourth order R4 in the equivalent circuit model of the slow response portion of the battery of FIG. , C4 and transfer function 14 corresponding to fifth-order R5 and C5 are input with charge / discharge current values Ia to obtain respective overvoltage values, and these overvoltage values are added by adder 15. Add the voltage value Vc of the slow response part.
The subtracter 16 subtracts the slow response portion voltage value Vc from the terminal voltage value Va to obtain the fast response portion voltage value Vb.

On the other hand, the current may be input to the parameter estimation unit 4 as it is without being processed by the filter processing unit 11, or the slow response part may be removed using a high pass filter to obtain the filtered current value Ib. You may input into the sequential parameter estimation part 4. FIG.
Other configurations are the same as those of the first embodiment.

  As a result, in the battery internal state estimating device of Example 2, since only the voltage of the fast response part excluding the slow response part of the battery is input to the sequential parameter estimation unit 4, the overvoltage value in the slow response part and It is possible to avoid redundant calculations.

As described above, the battery internal state estimating apparatus according to the second embodiment can accurately estimate the internal state of the battery, similarly to the first embodiment.
Further, by providing the filter processing unit 11, it is possible to easily remove the slow response portion from the input signal to the sequential parameter estimation unit 4.

  The present invention has been described based on the above embodiments. However, the present invention is not limited to these embodiments, and is included in the present invention even when there is a design change or the like without departing from the gist of the present invention. .

For example, the low-pass filter and the high-pass filter used in the filter processing unit 11 are not limited to those in the embodiment, and various other types may be used.
The equivalent circuit model of the battery is not limited to the Foster type, and may be any other mathematical model representing the inside of the battery, such as a diffusion equation.

  Moreover, the battery internal state estimating device of the present invention is not limited to a vehicle such as an electric vehicle, and may be applied to any device as long as it can estimate the internal state of a secondary battery.

1 actual battery 2 current sensor (charge / discharge current detection means)
3 Voltage sensor (terminal voltage detection means)
4 Sequential parameter estimation unit (sequential parameter estimation means)
5 Multiplier (multiplication means)
6 first subtractor 7 state quantity estimation unit (state quantity estimation means)
71 first coefficient unit 72 first adder 73 first integrator 74 second coefficient unit 75 third coefficient unit 81 fourth coefficient unit 82 second adder 83 second integrator 84 fifth coefficient unit 85 sixth coefficient unit 86 Subtractor 87 Multiplier 8 Adder (addition means)
9 Second subtractor (subtraction means)
10 Open-circuit voltage-charge rate converter (open-circuit voltage-charge rate estimation means)
11 Filter processing section (filter processing means)
12, 13, 14 Transfer function 15 Adder 16 Subtractor

Claims (4)

Charge / discharge current detection means for detecting the charge / discharge current value of the battery;
Terminal voltage detecting means for detecting a terminal voltage value of the battery;
An equivalent circuit model having a fast response portion and a slow response portion of the battery;
Based on the charging / discharging current value input from the charging / discharging current detection means and the terminal voltage value input from the terminal voltage detection means, sequential parameter estimation is performed using only the fast response portion of the response portions of the equivalent circuit model. Sequential parameter estimation means for performing
Based on the charge / discharge current value input from the charge / discharge current detection means and the terminal voltage value input from the terminal voltage detection means, the overvoltage value of the slow response portion is estimated using the slow response portion of the equivalent circuit model. State quantity estimating means to perform,
Multiplication means for obtaining an overvoltage value of the fast response portion by multiplying the charge / discharge current value by the parameter estimated by the sequential parameter estimation means;
An overvoltage detection means for adding the overvoltage value of the fast response portion obtained by the multiplication means and the overvoltage value of the slow response portion obtained by the state quantity estimation means to obtain an overvoltage value of the battery;
A battery state estimation device comprising: The battery state estimation device according to claim 1,
Subtracting means for subtracting the overvoltage value obtained by the terminal voltage detecting means from the terminal voltage value obtained by the terminal voltage detecting means to obtain the open-circuit voltage value of the battery;
An open-circuit voltage-charge rate estimating means for obtaining a charge rate of the battery based on the open-circuit voltage value obtained by the subtracting means;
Having
A battery state estimation device. In the battery state estimation device according to claim 1 or 2,
Filter processing means for removing the slow response portion of the terminal voltage obtained by the terminal voltage detection means and inputting the voltage to the sequential parameter estimation means.
A battery state estimation device. In the battery state estimation device according to claim 3,
The filter processing means removes the slow response portion from the charge / discharge current obtained by the charge / discharge current detection means and inputs it to the sequential parameter estimation means.
A battery state estimation device.

Images