JP2013195232A - Apparatus and method for calculating internal resistance of secondary battery, apparatus and method for detecting abnormality of secondary battery, and apparatus and method for estimating deterioration in secondary battery - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an internal resistance calculation apparatus for secondary batteries capable of improving calculation accuracy of an internal resistance value.SOLUTION: An internal resistance calculation apparatus 10 for secondary batteries includes: a voltage measuring unit 3 that is connected between a positive electrode terminal and a negative electrode terminal of each battery cell 1 constituting a secondary battery and measures an interterminal voltage; a current measuring unit 4 that is inserted into a current route and measures charge/discharge currents with respect to the battery cell 1; a control unit 6 that controls measurements in the voltage measuring unit 3 and the current measuring unit 4; and an arithmetic unit 8 that is connected to the control unit 6, performs wavelet transformation from each measurement value of the voltage measuring unit 3 and the current measuring unit 4, obtains wavelet coefficients of the current and the voltage for every frequency and calculates the internal resistance of the secondary battery from a ratio thereof.

Description

  Embodiments described herein relate generally to a secondary battery internal resistance calculation device and its internal resistance calculation method, a secondary battery abnormality detection device and abnormality detection method, a secondary battery deterioration estimation device, and a deterioration estimation method thereof.

  One index for grasping the internal state of the secondary battery is the internal resistance of the battery. Various proposals have been made as a method for calculating the internal resistance.

  For example, firstly, there is a method of linearly approximating a parameter of the current change amount and a parameter based on the voltage change amount and the impedance change amount, and calculating the battery impedance from the slope of the approximated straight line.

  Second, there is a method of calculating an internal resistance by separating current and voltage changes into a high frequency component and a low frequency component using a frequency filter, and extracting only the high frequency component excluding the low frequency component corresponding to the diffusion resistance. is there.

  Third, there is a method in which the current resistance and the voltage change are each subjected to Fourier transform, the signal intensity for each frequency is obtained, and the internal resistance is calculated from the ratio of these frequency components. This method calculates the internal resistance from a plurality of different frequency components, and estimates the circuit constant when the secondary battery is represented by an equivalent circuit.

JP 2006-250905 A JP 2005-106616 A JP 2005-221487 A

  In the first to third methods described above, the internal resistance is estimated from the change in charge / discharge current and battery voltage. However, in the first method, frequency components of current and voltage change are not taken into consideration.

  On the other hand, in the second and third methods, the internal resistance is calculated from a specific frequency using frequency analysis or the like. However, in these methods, it is possible to accurately calculate the internal resistance value at a specific frequency where the correlation between the current and voltage change during charging and discharging is high, but the accuracy is poor at other frequencies. There was a possibility.

  This is considered to be because the correlation between the current and voltage during charging / discharging may be low depending on the frequency. In addition, if a measurement timing shift occurs due to various sensors of the storage battery system, it also affects the correlation between changes in current and voltage, so that the influence of the high-frequency component is prominent.

  Embodiments of the present invention have been made in view of the above problems, and an object thereof is to provide an internal resistance calculation device for a secondary battery and an internal resistance calculation method thereof that can improve the calculation accuracy of the internal resistance value.

  Another object of the present invention is to provide an abnormality detection device for a secondary battery and an abnormality detection method thereof that can detect the abnormality of the secondary battery based on the calculation of the internal resistance value.

  Another object of the present invention is to provide a secondary battery deterioration estimation device and a deterioration estimation method thereof that can estimate the deterioration of the secondary battery based on the calculation of the internal resistance value.

  In order to achieve the above-mentioned object, the internal resistance calculation device according to the first embodiment of the present invention is connected between the positive terminal and the negative terminal of each battery cell constituting the secondary battery, and measures the inter-terminal voltage. A voltage measurement unit, a current measurement unit that is inserted into a current path and measures charge / discharge current to the battery cell, a control unit that controls measurement in the voltage measurement unit and the current measurement unit, and a connection to the control unit A wavelet transform from each measurement value from the voltage measurement unit and the current measurement unit, to obtain a wavelet coefficient of current and voltage for each frequency, and a calculation unit to calculate the internal resistance of the secondary battery from these ratios; It is characterized by providing.

  Further, in the secondary battery abnormality detection device according to the second embodiment of the present invention, each internal resistance of the secondary battery calculated by the calculation unit is added to the control unit of the internal resistance calculation device of the secondary battery. An abnormal cell determination unit that determines an abnormality of the battery cell using a predetermined statistical process from the value is connected.

  Furthermore, the secondary battery deterioration estimation device according to the third embodiment of the present invention is provided with a temperature measurement unit that measures the temperature of the battery cell in the internal resistance calculation device of the secondary battery, and the control unit In addition, a deterioration estimation unit that estimates a degree of deterioration by comparing a reference internal resistance value recorded in advance with a measurement result of the internal resistance of the secondary battery calculated by the calculation unit is connected. .

  The method executed in the first to third embodiments as described above is also one embodiment of the present invention.

Schematic which shows an example of the equivalent circuit model of a secondary battery. The graph which shows the distribution state of the internal resistance value using the linear regression method of the conventional VI characteristic. The graph which shows an example of the relationship between ( ^Wv ) (a, b) and ( ^Wi ) (a, b) in the frequency component (level j = 1) with a low correlation of an electric current and a voltage. The graph which shows an example of the relationship between ( ^Wv ) (a, b) and ( ^Wi ) (a, b) in the frequency component (level j = 5) with a high correlation of an electric current and a voltage. Graph showing an example of the relationship between the frequency correlation (level j) the sampling data (coefficient of determination R ^2). Graph showing another example of the relationship between the frequency correlation (level j) the sampling data (coefficient of determination R ^2). Schematic which shows the structure of the internal resistance calculating apparatus which concerns on the 1st Embodiment of this invention. The flowchart which shows the process sequence which calculates the internal resistance in 1st Embodiment. The graph which shows the distribution state of the internal resistance value computed by the internal resistance calculation method in 1st Embodiment. Schematic which shows the structure of the abnormality detection apparatus of the secondary battery which concerns on the 2nd Embodiment of this invention. The flowchart which shows the process sequence which detects abnormality of the secondary battery in 2nd Embodiment. Schematic which shows the structure of the degradation estimation apparatus of the secondary battery which concerns on the 3rd Embodiment of this invention. The graph which shows the example which plotted the internal resistance of the normal battery and the deterioration battery according to the level.

  In each embodiment of the present invention described below, a method is used in which a current waveform and a voltage waveform are analyzed by wavelet transform, and an internal resistance is calculated at a frequency where the correlation between the signal strengths is high. Hereinafter, the principle of the method used in each embodiment will be described.

(Conventional linear regression method of VI characteristics)
The linear regression method of the VI characteristics of the battery voltage V and the battery current I has already been publicly known by the first method of the conventional example. In this method, for example, the equivalent circuit of the secondary battery shown in FIG. 1 is used. _{Wherein, R Ω, R 1~n, C} 1~n respectively, electrolyte resistance, charge transfer resistance, an electric double layer capacitor. In this case, as for the internal resistance of the secondary battery, the same subscript numbers are combined in parallel for each of the charge transfer resistances R 1 to _n and the electric double layer capacities C 1 to _n , and further combined in series with the electrolyte resistance _RΩ . Expressed as a thing.

Figure 2 is a plot at various combination ratio of the combined resistance of the direct current component R _Omega, the AC component (impedance due C _n and R _n) in the equivalent circuit of the battery shown in FIG. In the VI characteristic graph shown in FIG. 2, a straight line is approximated from each plot, and the internal resistance is obtained from the slope of the straight line. However, in this method, since the slope of the regression line differs depending on the charge / discharge frequency at the time of measurement, the distribution region of the internal resistance value becomes wide as shown by the region A, and the estimation accuracy of the internal resistance value decreases. .

  On the other hand, in the method of linear regression of the relationship between the wavelet coefficient of the voltage value and the current value, the internal resistance value for each frequency can be calculated for the charge / discharge waveform. Hereinafter, an internal resistance calculation method using wavelet transform will be described.

また、Ψ_ａ，ｂ（ｔ）はアナライジング・ウェーブレットと呼ばれ、ダイレーション（拡大縮小）のパラメータを実数ａ、ｔ軸上でのシフトのパラメータを実数ｂとし、式（２）のように定義される。

Ψ（ｔ）としては、さまざまなものが提案されており、適宜選択可能である。式（３）に例としてガボールウェーブレットの定義を示す。

ここで、ω0は振動の中心周波数である。
計測された電流波形をｉ（ｔ）、電圧波形をｖ（ｔ）とすると、それぞれのウェーブレット変換は式（４）、式（５）のようになり、この変換結果はウェーブレット係数と呼ばれる。

すると、式（６）によって、同一のダイレーションａ、シフトｂの電流、電圧のウェーブレット係数の比から、内部抵抗を計算することができる。

このとき、ダイレーションａが周波数に相当し、ダイレーションａが定まれば内部抵抗はシフトｂによらず一定であると考えられる。そこで、特定のダイレーションａに対してシフトｂを変化させて、（Ｗ_Ψ ^ｉ）（ａ，ｂ）と（Ｗ_Ψ ^ｖ）（ａ，ｂ）の関係を最小二乗法を用いて直線近似すると、その傾きから周波数毎の内部抵抗値Ｒ（ａ）が算出できる。ここで、直線近似の精度を表す決定係数Ｒ^２を算出する。（Ｗ_Ψ ^ｉ）（ａ，ｂ）と（Ｗ_Ψ ^ｖ）（ａ，ｂ）の分散および共分散をそれぞれν_ｗｉ、ν_ｗｖ、ν_ｗｖｉとすると、ν_ｗｉ、ν_ｗｖ、ν_ｗｖｉは、それぞれ式（７）〜（９）で求められる。

これより、（Ｗ_Ψ ^ｉ）（ａ，ｂ）と（Ｗ_Ψ ^ｖ）（ａ，ｂ）の関係を直線近似した時の決定係数Ｒ^２は以下の式で求められる。

(Internal resistance calculation method using wavelet transform)
The wavelet transform W _Ψ f of the waveform f (t) can be obtained by Expression (1).

Also, Ψ _{a, b} (t) is called an analyzing wavelet, the dilation (enlargement / reduction) parameter is a real number a, the shift parameter on the t-axis is a real number b, and the equation (2) Defined.

Various Ψ (t) have been proposed and can be selected as appropriate. Formula (3) shows the definition of Gabor wavelet as an example.

Here, ω0 is the center frequency of vibration.
Assuming that the measured current waveform is i (t) and the voltage waveform is v (t), the respective wavelet transforms are as shown in Equations (4) and (5), and the transformation results are called wavelet coefficients.

Then, the internal resistance can be calculated from the ratio of the current and voltage wavelet coefficients of the same dilation a, shift b, and voltage according to equation (6).

At this time, the dilation a corresponds to the frequency, and if the dilation a is determined, the internal resistance is considered to be constant regardless of the shift b. Therefore, when the shift b is changed with respect to a specific dilation a, the relationship between (W _Ψ ⁱ ) (a, b) and (W _Ψ ^v ) (a, b) is linearly approximated using the least square method. The internal resistance value R (a) for each frequency can be calculated from the inclination. Here, to calculate the coefficient of determination R ² representing the accuracy of the linear approximation. When the dispersion and covariance of (W _Ψ ⁱ ) (a, b) and (W _Ψ ^v ) (a, b) are respectively ν _wi , ν _wv , and ν _wvi , ν _wi , ν _wv , and ν _wvi are respectively It calculates | requires by Formula (7)-(9).

From this, the determination coefficient R ² when the relationship between (W _Ψ ⁱ ) (a, b) and (W _Ψ ^v ) (a, b) is linearly approximated is obtained by the following expression.

(Correlation of wavelet coefficient of voltage value and current value by frequency)
Calculating the coefficient of determination R ² from the wavelet coefficients of the voltage value and the current value calculated for each frequency, it is possible to determine these correlations.
For example, the coefficient of determination R ² is lower frequency (level j = 1), as shown in FIG. 3, the low correlation of the wavelet coefficient of the voltage and current values. For this reason, if the internal resistance value is calculated from these ratios, the accuracy decreases.

In contrast, in the coefficient of determination R ² is high frequency (level j = 5), as shown in FIG. 4, the correlation of the wavelet coefficient of the voltage and current values higher, to accurately calculate the internal resistance value Can do.

FIG. 5 shows an example of the relationship between the frequency (level j) subjected to wavelet transform and the correlation (determination coefficient R ² ) between sampling data. From this result, it can be seen that when the level j is 5, the correlation (determination coefficient R ² ) is maximum. As described above, by specifying a frequency having a high correlation between current and voltage and calculating the internal resistance value using the frequency component, the calculation accuracy of the internal resistance can be improved.

FIG. 6 shows another example of the relationship between the frequency (level j) subjected to wavelet transform and the correlation (determination coefficient R ² ) between sampling data. Thus, the current and voltage waveforms to be analyzed is changed, the coefficient of determination R ² depending on the frequency components in some cases change the frequency with the maximum. In addition, since the internal resistance value varies depending on the frequency at which the internal resistance is calculated, the calculation is performed for each frequency.

As described above, the wavelet coefficients of the charge / discharge current and voltage calculated by the wavelet transform are linearly approximated by the least square method, and the determination coefficient R ² at this time (the residual in the least square method, It is possible to improve the accuracy of calculation of the internal resistance by specifying a frequency having a high (applicable degree of approximation line) and calculating the internal resistance value with the frequency component.

  Hereinafter, each embodiment using the wavelet transform method described above will be specifically described with reference to the drawings.

[First Embodiment]
(Configuration of internal resistance calculation device)
FIG. 7 shows the configuration of the internal resistance arithmetic unit according to the first embodiment of the present invention. The internal resistance calculation device 10 includes a battery pack 2 in which a plurality of battery cells 1 constituting a secondary battery are connected in series, and a voltage that is connected between a positive terminal and a negative terminal of each battery cell 1 and measures a voltage between terminals. A measuring unit 3; a current measuring unit 4 that is inserted in the current path to measure a charging / discharging current to the battery cell 1; and a temperature measuring unit 5 that is disposed in the vicinity of the battery cell 1 and measures the temperature of the battery cell 1. I have.

  Further, the internal resistance calculation device 10 controls the measurement conditions of the voltage measurement unit 3, the current measurement unit 4, and the temperature measurement unit 5, and controls the control unit 6 that processes each measurement value from these parts, and the control unit 6. A recording unit 7 connected to record each measurement value, etc., a calculation unit 8 connected to the control unit 6 to perform wavelet transformation from each measurement value and calculate an internal resistance, and connected to the control unit 6 to communicate with the outside of the secondary battery And a communication interface 9 for performing the above.

An equivalent circuit of the battery cell 1 in the internal resistance calculation device 10 can be shown as shown in FIG. That is, the internal resistance of the secondary battery is a combination of the same subscript numbers in parallel for each of the charge transfer resistances R 1 to _n and the electric double layer capacities C 1 to _n , and further combined in series with the electrolyte resistance _RΩ. Is represented as

  The procedure for calculating the internal resistance by wavelet transformation using the internal resistance computing device 10 will be described by taking the case of using discrete wavelet transformation as an example.

(Processing procedure to calculate internal resistance)
FIG. 8 is a flowchart showing a processing procedure for calculating the internal resistance by wavelet transformation. First, at the start of processing, parameters necessary for calculation by the calculation unit 8 are initialized (step S101). As initial setting parameters in step S101, voltage and current sampling intervals Δt, internal resistance calculation intervals (charging, discharging, charging / discharging), number of samplings N, information defining wavelets used for calculation, etc. There is.

  Next, the current and voltage are measured by the voltage measuring unit 3 and the current measuring unit 4, respectively (step S102). The control unit 6 acquires a voltage value and a current value from the voltage measurement unit 3 and the current measurement unit 4 for each sampling time Δt. Further, the control unit 6 determines whether or not the charge / discharge state is a state where the internal resistance is calculated in the initial setting (step S103). If it is a target section (Yes in step S103), the control unit 6 adds a time stamp (measurement time data) to the measured data and records it in the recording unit 7 (step S104). If it is outside the target section (No in step S103), the current and voltage are measured again (step S102).

  Next, the control unit 6 determines whether or not the measured number of data has reached the number N of data necessary for calculating the internal resistance (step S105). If not reached (No in step S105), the process is repeated from step S102. On the other hand, when the necessary number of data can be acquired (Yes in step S105), the process proceeds to the wavelet transform process from S106.

ここで、Ψ（ｔ）は式（１２）で定義され、式（１２）に含まれるφ（ｔ）は式（１３）で定義される。

上記式（１２）、式（１３）で記載されたｐ_ｎ、ｑ_ｎは、ウェーブレットを表す数列であり、式（１４）の関係がある。

ここで、ウェーブレット数列としては、ハール（Haar）やドベシィ（Daubechies）のウェーブレットなどを利用することができる。 In the wavelet transform process, first, the calculation unit 8 sets the initial value of the parameter j corresponding to the frequency called level to 1 (step S106). Next, the computing unit 8 performs wavelet transform of current and voltage based on the data recorded in the recording unit 7 (step S107). In the discrete wavelet transform, in Equation (2), a = 2 ^j and b = 2 ^j k are used to discretize as in Equation (11).

Here, Ψ (t) is defined by Expression (12), and φ (t) included in Expression (12) is defined by Expression (13).

The formula (12), _p n described in Formula (13), _{q n} is a sequence representing wavelet, a relationship of Equation (14).

Here, as the wavelet sequence, Haar, Daubechies wavelets, and the like can be used.

Similar to Equations (4) and (5), when the current waveform is i (t) and the voltage waveform v (t) is discrete wavelet transformed, Equations (15) and (16) are obtained.

Next, the calculation unit 8 linearly approximates the relationship between the current and voltage wavelet coefficients calculated by the equations (15) and (16) at the level j by the least square method, and determines the coefficient R ² representing the correlation at that time. Is calculated (step S108). Further, the computing unit 8 calculates the internal resistance at level j from the slope of the linear approximation obtained in step 108 (step S109).

Subsequently, the arithmetic unit 8, in step S110, the level is incremented by one, wavelet transform (step S107), the calculation of the coefficient of determination ^{R 2} (step S108), and carrying out the processes of the internal resistance calculation (step S109) . Next, it is determined whether or not the level j satisfies the condition of Expression (17) (step S111). If not satisfied (No in step S111), the loop processing is continued and if satisfied (step S111). Yes) ends the loop processing.

Further, the control unit 6 records the internal resistance value obtained for each level in the recording unit 7 (step S112). As described above, the calculation accuracy of the internal resistance value is high where the correlation between the current and the voltage (determination coefficient R ² ) at each level is high. Although there is no clear standard for the coefficient of determination indicating the correlation, it is generally said that the accuracy is good if the coefficient of determination is 0.8 or more. Thus, the internal resistance value calculated with high accuracy is distinguished for each frequency and recorded in the recording unit 7.

  Further, for the purpose of monitoring the state of the secondary battery, the control unit 6 may notify the internal resistance value recorded in the recording unit 7 to the outside via the communication interface 9.

(effect)
(1) Until now, the internal resistance could only be measured by disconnecting the secondary battery from the storage system in actual operation. However, according to this embodiment, the charge / discharge current waveform and voltage waveform in actual operation are By converting and calculating the internal resistance from the ratio of the wavelet coefficients, it can be performed without disconnecting from the power storage system.

(2) By using the wavelet transform, time-frequency analysis is possible, calculation of internal resistance at a frequency where the correlation between the charge / discharge current and the cell voltage is high, and analysis of the internal resistance for each frequency are possible.

(3) The charge and discharge current and voltage each wavelet coefficient calculated by the wavelet transform linearly approximated by the least square method, by identifying the internal resistance value in the coefficient of determination R ² is high frequency upon this, the internal resistance The calculation accuracy can be improved.

  That is, the distribution of the internal resistance value calculated by the conventional linear regression method of the VI characteristic has a large variation as shown in the region A of FIG. 2, but the distribution of the internal resistance value calculated by the present embodiment is As shown in the region B of FIG. 9, the variation is small (the area of the region B <the area of the region A).

(4) Further, according to the present embodiment, there is no need to calculate the internal resistance of the secondary battery cell constituting the storage battery system by a specific charge / discharge operation, and a special analysis such as a frequency analyzer as in the AC impedance method is performed. There is no need to use a measuring device.

[Second Embodiment]
(Detection of secondary battery anomalies using internal resistance calculation results)
In order to detect abnormality of a secondary battery, the internal resistance value for every battery cell is continuously monitored and the method of processing the increase rate etc. statistically can be considered. For this reason, it is necessary to accurately calculate the internal resistance value indicating the internal state of the battery. The secondary battery abnormality detection apparatus using the internal resistance value calculation method described in the first embodiment will be described below.

(Configuration of secondary battery abnormality detection device)
FIG. 10 shows the configuration of an abnormality detection device for a secondary battery according to the second embodiment of the present invention. Note that the same parts as those in the first embodiment are denoted by the same reference numerals and description thereof is omitted.

  The abnormality detection device 20 is configured by connecting an abnormal cell determination unit 11 that determines a battery cell abnormality using a predetermined statistical process to the control unit 6 of the internal resistance calculation device 10 of the first embodiment.

(Processing procedure for detecting secondary battery abnormality)
FIG. 11 is a flowchart showing a processing procedure for detecting an abnormality of the secondary battery using the abnormality detection device 20.

  First, in order to calculate the internal resistance for each battery cell 1 in the assembled battery 2, the voltage value and the charge / discharge current value of each battery cell 1 are measured by the voltage measuring unit 3, the current measuring unit 4, and the temperature measuring unit 5. The battery temperature is measured at regular intervals (step S201). The control unit 6 adds time information to the measured measurement data and records it in the recording unit 7 (step S202).

  Next, the control unit 6 calculates the SOC (State Of Charge) of the battery from the integration of the charge / discharge current (step S203). It is conceivable that the battery SOC is calculated from the open circuit voltage of the battery as well as the integration of the charge / discharge current.

  Furthermore, in order to collect sampling data for calculating the internal resistance, measurement of the cell voltage, charge / discharge current, temperature, and calculation of the battery SOC are repeated for a certain period (step S204). When a certain period satisfying the sampling number necessary for calculating the internal resistance is reached (Yes in step S204), the process proceeds to the next step. On the other hand, if it has not been reached (No in step S204), the process is repeated from step S201.

  Next, the control unit 6 confirms that the battery temperature measured by the temperature measurement unit 5 and the change amount of the battery SOC calculated by the control unit 6 are within the specified range within the measurement period of the sampling data. Confirmation (step S205). If the amount of change exceeds the specified range (No in step S205), it is assumed that the internal resistance value has changed significantly during this measurement period, and the calculated internal resistance value is used for battery cell abnormality detection. do not do.

  On the other hand, when the amount of change is within the specified range (Yes in step S205), the calculation unit 8 uses the wavelet to calculate the data string of the measured charge / discharge current value and the voltage value of each battery cell at level j = 1. The wavelet coefficient is calculated by conversion (step S206).

Next, the arithmetic unit 8, and the linear approximation the relation of the wavelet coefficients of the formula (15) and the current and the cell voltage calculated by the equation (16) at level j, and calculates a determination coefficient R ² at that time (step S207 ). Further, the internal resistance value at level j is calculated from the slope of the linear approximation (step S208). Subsequently, the level j approximated by the scaling function is increased by 1 (step S209), and it is further determined whether or not the level j satisfies the condition of expression (17) (step S210). ), Go to the next step. On the other hand, if it has not been reached (No in step S210), the process is repeated from step S206.

  Subsequently, the control unit 6 has a certain coefficient of determination more than a certain value among the battery temperature at the time of measuring the current and the cell voltage, the battery SOC, and the internal resistance value calculated for each level in steps S207 and S208. The internal resistance value is recorded in the recording unit 7 (step S211). Next, the control part 6 repeats the process (step S201-step S211) until calculation of the internal resistance of all the battery cells 1 for a fixed period (step S212) by a fixed period (for example, 1 day unit). When calculation data sufficient for comparison of internal resistance can be acquired (Yes in step S212), an abnormality detection method for the battery cell 1 using statistical processing described below is performed (step S213).

In this step, first, under the same measurement conditions (battery temperature, battery SOC), the calculation unit 8 calculates the internal resistance value of the battery cell 1 at regular intervals, and calculates the internal resistance value with respect to the charge / discharge cycle and the elapsed time. Calculate the rate of increase. This is defined as an internal resistance increase rate X _i (ΔR / Δt, i: cell number).

  Next, the abnormal cell determination unit 11 performs statistical processing using a rejection test on the internal resistance increase rate for each battery cell 1 to determine whether or not the battery cell exhibiting a particularly large increase rate is an abnormal cell. Make a decision. As one of the rejection test methods, the Smirnov-Grubbs test is known, and the rejection test can be performed by the following procedure.

（４）上記Ｔｉと、有意水準αの有意点Ｔｎ（α）を比較することによって、棄却検定を行なう。
Ｔｉ＜Ｔｎ（α）の時、Ｔｉは棄却されない。
Ｔｉ≧Ｔｎ（α）の時、Ｔｉは棄却される。
Ｔｎ(α)は、ｔ_α／ｎを自由度ｎ−２のｔ分布の上側１００α／ｎ（％）とした時、式（１９）により求められる。

(1) the number of samples n, (in this case, the internal resistance increase rate) sample data _{X 1,} X 2, and · · · _{X n.}
(2) Let X ′ be the sample mean and U be the unbiased variance.
(3) for the maximum measured value _{X i} to calculate the Ti according to equation (18).

(4) A rejection test is performed by comparing the Ti with the significance point Tn (α) of the significance level α.
When Ti <Tn (α), Ti is not rejected.
When Ti ≧ Tn (α), Ti is rejected.
Tn (α) is obtained by Expression (19), where t _{α / n} is 100 α / n (%) above the t distribution with n−2 degrees of freedom.

  With the method described above, it is possible to determine whether or not the battery cell 1 that exhibits a particularly large rate of increase in internal resistance is an abnormal cell.

(effect)
(1) According to this embodiment, it is possible to detect abnormality of the secondary battery by continuously monitoring the internal resistance of each battery cell 1 and statistically processing the calculated individual cell internal resistance values. Thereby, in the storage battery system using a secondary battery, the abnormality of a secondary battery by the secular deterioration etc. which cannot be detected at the time of battery manufacture can be detected at an early stage.

(2) By detecting the battery cell 1 exhibiting a specific internal resistance, it is possible to grasp the tendency of battery deterioration and predict charge / discharge power. This makes it possible to decide on a policy for efficient operation, maintenance, reuse, and recycling of the secondary battery.

(3) By using the communication interface 9 to notify the higher-order monitoring device of the internal resistance and abnormality information of the secondary battery, it is possible to monitor abnormality or performance degradation of the storage battery system.

[Third Embodiment]
The secondary battery deterioration estimation apparatus according to the third embodiment using the internal resistance value calculation method described in the first embodiment will be described below.

(Configuration of secondary battery deterioration estimation device)
FIG. 12 shows the configuration of a secondary battery deterioration estimation apparatus according to the third embodiment of the present invention. Note that the same parts as those in the first embodiment are denoted by the same reference numerals and description thereof is omitted.

  The degradation estimation apparatus 30 of the present embodiment compares the measurement result of the internal resistance of the secondary battery with the internal resistance value recorded in advance in the control unit 6 of the internal resistance calculation apparatus 10 of the first embodiment. A deterioration estimation unit 21 that estimates the degree is connected.

  As shown in FIG. 13, a BOL (Begin Of Life) secondary battery (normal battery) and a deteriorated secondary battery have different internal resistance values, and the difference also varies depending on the frequency (level). For this reason, in this embodiment, the internal resistance according to the frequency is measured using a secondary battery whose degree of deterioration is known in advance at the reference battery temperature and SOC (charged state), and the recording unit 7 is deteriorated. Record as an internal resistance value table for each degree and frequency.

  Next, the calculation unit 8 calculates the internal resistance of each battery cell 1 for each frequency by the secondary battery internal resistance calculation method according to the first embodiment, and the deterioration estimation unit 21 stores the calculation result in the recording unit 7. Compare with the recorded internal resistance value table.

  Specifically, first, in the recording unit 7, for the internal resistance value of the BOL (Begin Of Life) secondary battery, the reference battery temperature and SOC (charged state) characteristic data are classified by frequency. Record as an internal resistance value table. Next, the internal resistance value calculated by the calculation unit 8 is converted (corrected) from the current battery temperature and SOC (charged state) to the internal resistance value of the reference battery state (standard temperature, standard charged state). Furthermore, the deterioration state of the secondary battery is estimated from the ratio between the converted internal resistance value and the initial internal resistance value recorded as the internal resistance value table.

(effect)
According to the present embodiment, the result of the internal resistance of each battery cell 1 calculated for each frequency by the method for calculating the internal resistance of the secondary battery according to the first embodiment, and recorded in advance in the recording unit 7 for each degree of deterioration and frequency. By comparing with the internal resistance value table thus made, it becomes possible to accurately grasp the deterioration state of the secondary battery.

[Other Embodiments]
(1) In each of the embodiments described above, the communication interface 9 for transmitting the internal resistance value to the outside is provided for the purpose of monitoring the state of the secondary battery, but may be omitted.

(2) In the second embodiment, as a method for detecting abnormal cells, a method of comparing the rate of increase in internal resistance for each battery cell 1 is used. The internal resistance value itself may be compared.

(3) In the third embodiment, the calculated internal resistance value is converted (corrected) into the reference battery state (standard temperature, standard charge state), and compared with the value in the internal resistance value table to estimate the deterioration state. However, the deterioration state of the secondary battery can also be estimated from the increase ratio of the internal resistance value after conversion (correction).

(4) You may combine the abnormality detection apparatus 20 which concerns on 2nd Embodiment, and the degradation estimation apparatus 30 which concerns on 3rd Embodiment. In this case, the abnormal cell determination unit 11 of the abnormality detection device 20 and the deterioration estimation unit 21 of the deterioration estimation device 30 can coexist in one device. The abnormal cell determination unit 11 may have the function of the deterioration estimation unit 21, or conversely, the deterioration estimation unit 21 may have the function of the abnormal cell determination unit 11.

(5) Although several embodiments of the present invention have been described above, these embodiments are presented as examples and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, replacements, and changes can be made without departing from the spirit of the invention. These embodiments and their modifications are included in the scope and gist of the invention, and are also included in the invention described in the claims and the equivalents thereof.

DESCRIPTION OF SYMBOLS 1 ... Battery cell 2 ... Assembly battery 3 ... Voltage measurement part 4 ... Current measurement part 5 ... Temperature measurement part 6 ... Control part 7 ... Recording part 8 ... Calculation part 9 ... Communication interface 10 ... Internal resistance calculation apparatus 11 ... Abnormal cell determination 20: Abnormality detection device 21 ... Degradation estimation unit 30 ... Degradation estimation device

Claims (12)

A voltage measuring unit that is connected between the positive electrode terminal and the negative electrode terminal of each battery cell constituting the secondary battery, and measures a voltage between the terminals;
A current measurement unit that is inserted into a current path and measures a charge / discharge current to the battery cell;
A control unit for controlling measurement in the voltage measurement unit and the current measurement unit;
Connected to the control unit, performs wavelet transform from each measurement value from the voltage measurement unit and the current measurement unit, obtains a wavelet coefficient of current and voltage for each frequency, and calculates the internal resistance of the secondary battery from these ratios A computing unit for computing,
An internal resistance calculation device for a secondary battery, comprising:   The arithmetic unit specifies a frequency having a high correlation when the relationship between the current and voltage wavelet coefficients is linearly approximated, and calculates the internal resistance of the secondary battery from the ratio of the wavelet coefficients at the frequency. The internal resistance calculation device of the secondary battery according to claim 1.   The calculation unit uses N measurement values measured at predetermined sampling time intervals in the voltage measurement unit and the current measurement unit, respectively, and uses a discrete wavelet transform as the wavelet transform. The internal resistance calculation apparatus of the secondary battery of Claim 1 or 2.   A predetermined statistical process is used for the control unit of the internal resistance calculation device for a secondary battery according to any one of claims 1 to 3 from each internal resistance value of the secondary battery calculated by the calculation unit. An abnormality detection device for a secondary battery, wherein an abnormality cell determination unit for determining abnormality of a battery cell is connected.   5. The secondary battery abnormality detection device according to claim 4, wherein a rejection test is used as the statistical processing.   The secondary battery abnormality detection device according to claim 4 or 5, wherein a communication interface for notifying an external device of abnormality information of the secondary battery detected by the abnormal cell determination unit is connected to the control unit. .   The internal resistance calculation device for a secondary battery according to any one of claims 1 to 3, further comprising: a temperature measurement unit that measures the temperature of the battery cell; and a reference recorded in advance in the control unit. A deterioration estimation device for a secondary battery, characterized in that a deterioration estimation unit for comparing the internal resistance value and the measurement result of the internal resistance of the secondary battery calculated by the calculation unit to estimate the degree of deterioration is connected.   The deterioration estimation unit includes a value of an internal resistance value table in which characteristic data in a battery temperature and a charging state serving as a reference for an internal resistance value of a secondary battery at an early stage of life, and an internal value calculated by the calculation unit 8. The secondary battery according to claim 7, wherein a resistance value is compared with a correction value converted into an internal resistance value of a battery state as a reference, and a deterioration state of the secondary battery is estimated from these ratios. Degradation estimation device.   The deterioration estimation unit estimates a deterioration state of a secondary battery from an increase rate when the internal resistance value calculated by the calculation unit is converted into an internal resistance value of a battery state serving as a reference. The deterioration estimation device for a secondary battery according to claim 7.   When the current measurement value and the voltage measurement value in each battery cell constituting the secondary battery are respectively wavelet transformed, the current and voltage wavelet coefficients are obtained at each frequency, and the relationship between the current and voltage wavelet coefficients is linearly approximated A method for calculating an internal resistance of a secondary battery, characterized by specifying a highly correlated frequency and calculating the internal resistance of the secondary battery from a ratio of wavelet coefficients at the frequency.   An abnormality detection method for a secondary battery, wherein abnormality of a battery cell is determined from each internal resistance value calculated by the internal resistance calculation method for a secondary battery according to claim 10 using a rejection test.   Comparing the internal resistance value of the secondary battery in the initial stage of life recorded in advance with each internal resistance value calculated by the internal resistance calculation method of the secondary battery according to claim 10 to estimate the degree of deterioration. A method for estimating deterioration of a secondary battery.

Images