JP2003222660A - Remaining capacity measuring device of battery - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method and a device of state analysis of battery capable of accurately analyzing the state of battery such as remaining capacity at low cost. <P>SOLUTION: In the state impressing alternating current signal output from a pseudo random noise generation means 12 to an analysis object battery 11 via an impedance element 13, the alternating voltage VB and alternating current iB of the battery 11 are sampled with an analog/digital converter 14. A transfer function of a discrete system consisting of a battery 11 and an impedance element 13 is calculated for estimation with a transfer function operation means 15 using the obtained data, and pole of the transfer function of the alternating current equivalent circuit of the battery 11 is calculated with a pole calculation means 16 using the transfer function of the discrete system. A state judgment means 17 judges the condition of the battery 11 using a correlation to the pole in the transfer function of the alternating current equivalent circuit. As this treats transfer function of discrete system, analog signal processing is unnecessary, and each means can be realized by a microcomputer, for example. <P>COPYRIGHT: (C)2003,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a battery state analysis method and apparatus for analyzing and detecting the state of the remaining battery level of a mobile device such as an electric vehicle, a personal computer, or a mobile phone. It is a thing.

[0002]

2. Description of the Related Art As a conventional method for analyzing the state of a battery such as the remaining amount, firstly, there is a method by a capacity test for measuring the voltage and current of the battery, and secondly, the AC impedance of the battery. There is a method of determining the state of the battery from the correlation between the measured and previously obtained AC impedance and the state of the remaining amount of the battery.

[0003]

However, according to the first method, although the detection accuracy of the state of the battery is good, it requires a long time measurement, and an analog signal processing circuit is required for measuring the current and voltage of the battery. Therefore, there is a problem in that cost is required when the battery condition analyzer is implemented as a device.

The second method has an advantage that the measurement can be performed in a shorter time than the first method, that is, the capacitance test method. However, since the amplitude and the phase are measured using an analog signal, Similar to the first method, an analog signal processing circuit is required, and there is a problem that the cost is high when the device is used, and there is a problem that it is weak against noise.

Our future aim is to incorporate a battery condition analysis device into a device such as an electric vehicle, a personal computer, or a portable device such as a mobile phone in which a battery condition is incorporated to determine the condition of the installed battery condition. Since it is possible to detect in real time, a low cost and highly accurate battery state analysis device is desired.

In view of the above problems, the present invention provides a battery state analysis method and method capable of accurately analyzing the state of the remaining battery level and the like using digital signal processing at low cost. An object is to provide a device used.

[0007]

In order to solve the above problems, the present invention focuses on the fact that there is a correlation between the transition of the poles in the transfer function of the AC equivalent circuit of a battery and the remaining amount of the battery, First, the transfer function of the discrete system of the system including the battery is obtained by an estimation operation, and then the pole in the transfer function of the AC equivalent circuit of the battery is obtained from this transfer function to detect the state of the battery. Since the estimation operation of the transfer function of the discrete system can be performed by digital signal processing, analog signal processing is unnecessary in the present invention.

The solution means taken by the invention described in the present application is as a battery state analysis method for analyzing the state of the remaining battery level,
The transfer function of the discrete system of the system including the battery is estimated and calculated from the time series data of the AC voltage applied to the battery and the AC current flowing in the battery, and the state of the battery is analyzed using the transfer function of the discrete system. It is a thing.

According to the invention described in the present application, analog signal processing is not necessary for estimating and calculating a transfer function of a discrete system of a system including a battery from time series data of an AC voltage applied to the battery and an AC current flowing through the battery. Since all can be performed by digital signal processing, it becomes possible to form an LSI by a one-chip microcomputer, DSP, etc., and when the device is implemented, the cost is reduced and the noise is strong. Further, from the transfer function of the discrete system, for example, the pole in the transfer function of the AC equivalent circuit of the battery is obtained, and by using the correlation between the pole in the transfer function of the AC equivalent circuit of the battery and the state of the battery,
The state of the battery can be analyzed.

The invention described in the present application is a further embodiment of the invention described in the preceding paragraph, which is a method for analyzing the state of a battery such as the remaining battery level, in which an AC signal is applied to the battery to be analyzed. Then, in a state where an AC signal is applied to the battery, a first step of sampling an AC voltage applied to the battery and an AC current flowing in the battery, and a transfer function of a discrete system of a system including the battery are described as follows. A second step of performing an estimation calculation using the alternating voltage and the alternating current sampled in the first step as time series data, and using the transfer function of the discrete system obtained by the estimation calculation in the second step, It analyzes the state of the battery.

In the invention described in the present application, the first step in the battery state analysis method of the preceding paragraph is to apply the AC signal generated and output from the AC signal source to the battery to be analyzed through the impedance element. To do.

According to the invention described in the present application, it is possible to prevent a DC path from being formed between the battery and the AC signal source by the impedance element.

Further, in the invention described in the present application, in the first step in the battery state analysis method of the preceding paragraph, a noise source for generating and outputting a pseudo random noise signal is used as the AC signal source.

According to the invention described in the present application, by applying a pseudo-random noise signal to a battery, a signal having a band including all pole frequencies in the transfer function of the continuous system of the AC equivalent circuit of the battery is applied to the battery. Therefore, the frequency sweep becomes unnecessary.

Further, in the invention described in the present application, the first step in the battery state analyzing method of the preceding paragraph is to directly apply the AC signal having the DC offset voltage generated and output from the voltage offset AC signal source to the battery to be analyzed. It shall be.

According to the invention described in the present application, it is possible to prevent the formation of a DC path between the battery and the voltage offset AC signal source by adjusting the DC offset voltage to the DC voltage of the battery. It becomes unnecessary.

Further, in the invention described in the present application, in the first step in the battery state analysis method of the preceding paragraph, a noise source for generating and outputting a pseudo random noise signal having a DC offset voltage is used as the voltage offset AC signal source. And

According to the invention described in the present application, by applying the pseudo random noise signal to the battery, a signal having a band including all the frequencies of the poles in the transfer function of the continuous system of the AC equivalent circuit of the battery is applied to the battery. Therefore, the frequency sweep becomes unnecessary.

In the invention described in the present application, the noise source in the above-mentioned battery state analysis method is assumed to be a noise source using the M-sequence code or the Gold code.

In the invention described in the present application, the second step in the above-mentioned battery state analysis method is that the order of poles in the transfer function of the discrete system of the battery is the continuous system of the AC equivalent circuit of the battery. It is assumed that the transfer function of the discrete system of the system including the battery is estimated as a higher order than the order of the poles in the transfer function.

Further, in the invention described in the present application, in the second step in the above-mentioned battery state analysis method, the coefficient parameter of the transfer function of the discrete system of the system including the battery is determined from the AC equivalent circuit of the battery. An initial value is given to improve the convergence of the estimation calculation.

Further, in the invention described in the present application, in the above-described battery state analysis method, the poles in the discrete system transfer function are obtained from the discrete system transfer function obtained in the second step and then obtained. A third step of converting the poles into the poles in the transfer function of the continuous system of the AC equivalent circuit of the battery, and the poles in the transfer function of the continuous system of the AC equivalent circuit of the battery obtained in the third step. It shall be used to analyze the state of the battery.

In the invention described in the present application, in the battery state analyzing method of the preceding paragraph, in the first step, the AC signal generated and output from the AC signal source is applied to the battery to be analyzed through the impedance element. And the third
The step of is, from the polynomial of the denominator of the transfer function of the discrete system of the system including the battery, the sum of the order of the pole in the transfer function of the continuous system of the AC equivalent circuit of the battery and the order of the pole in the impedance of the impedance element. After truncating higher order terms, the poles in the transfer function of the discrete system are obtained.

Further, in the invention described in the present application, in the above-mentioned battery state analysis method, in the first step, the AC signal generated and output from the AC signal source is directly applied to the battery to be analyzed. In the third step, a term having a higher order than a pole order in the transfer function of the continuous system of the AC equivalent circuit of the battery is truncated from the polynomial of the denominator of the transfer function of the discrete system of the system including the battery. After that, the poles in the transfer function of the discrete system are obtained.

In the invention described in the present application, in the above-described battery state analysis method, in the third step, after obtaining the pole in the transfer function of the continuous system of the AC equivalent circuit of the battery, the obtained pole is obtained. Correction shall be made according to the ambient temperature of the battery.

Further, in the invention described in the present application, in the battery state analysis method described above, the correlation between the pole in the transfer function of the continuous system of the AC equivalent circuit of the battery and the remaining amount of the battery, which is obtained in advance. Accordingly, a fourth step of determining the remaining amount of the battery using the pole in the transfer function of the continuous system of the AC equivalent circuit of the battery obtained in the third step is provided.

Further, the solution means taken by the invention described in the present application is a time-series data of an AC voltage applied to the battery and an AC current flowing in the battery as a battery state analysis device for analyzing the state of the battery such as the remaining amount. From the above, a means for estimating and calculating the transfer function of the discrete system of the system including the battery is provided, and the state of the battery is analyzed based on the transfer function of the discrete system estimated and calculated by this means.

According to the invention described in the present application, the analog signal processing is not necessary for the estimation calculation of the transfer function of the discrete system, and all can be performed by the digital signal processing. Therefore, the AC voltage applied to the battery and the AC current flowing in the battery are The means for estimating and calculating the transfer function of the discrete system of the system including the battery from the time series data can be realized by a one-chip microcomputer, DSP, etc., which reduces the cost when implemented as a device and is resistant to noise. Further, from the transfer function of the discrete system obtained by the means, for example, the pole in the transfer function of the AC equivalent circuit of the battery is obtained, and the correlation between the pole in the transfer function of the AC equivalent circuit of the battery and the state of the battery is calculated. It can be used to analyze the state of the battery.

The invention described in the present application embodies the invention of the preceding paragraph, and applies an AC signal to a battery to be analyzed as a battery condition analyzer for analyzing the condition of the remaining battery level and the like. AC signal applying means, sampling means for sampling the AC voltage applied to the battery and the AC current flowing in the battery when the AC signal is applied to the battery by the AC signal applying means, and the sampling means. And a transfer function calculating means for estimating and calculating a transfer function of a discrete system including the battery from time series data of AC voltage and AC current.

In the invention described in the present application, the AC signal applying means in the battery state analyzing device of the preceding paragraph has an AC signal source for generating and outputting an AC signal and an impedance element, and the AC signal source The generated and output AC signal is applied to the battery to be analyzed via the impedance element.

According to the invention described in the present application, it is possible to prevent a DC path from being formed between the battery and the AC signal source by the impedance element.

Further, in the invention described in the present application, the AC signal source in the battery state analysis device of the preceding paragraph is a noise source for generating and outputting a pseudo random noise signal.

According to the invention described in the present application, by applying the pseudo random noise signal to the battery, a signal having a band including all the frequencies of the poles in the transfer function of the continuous system of the AC equivalent circuit of the battery is applied to the battery. Therefore, the frequency sweep becomes unnecessary.

Further, in the invention described in the present application, the AC signal applying means in the above-mentioned battery state analyzing apparatus has a voltage offset AC signal source for generating and outputting an AC signal having a DC offset voltage, The AC signal generated and output from the offset AC signal source is applied to the battery to be analyzed.

According to the invention described in the present application, it is possible to prevent the formation of a DC path between the battery and the voltage offset AC signal source by adjusting the DC offset voltage to the DC voltage of the battery. It becomes unnecessary.

Further, in the invention described in the present application, the voltage offset AC signal source in the battery state analysis device of the preceding paragraph is:
It shall be a noise source that generates and outputs a pseudo random noise signal having a DC offset voltage.

According to the invention described in the present application, by applying the pseudo random noise signal to the battery, a signal having a band including all the frequencies of the poles in the transfer function of the continuous system of the AC equivalent circuit of the battery is applied to the battery. Therefore, the frequency sweep becomes unnecessary.

Further, in the invention described in the present application, the noise source in the above-mentioned battery state analyzing apparatus is assumed to be a noise source using the M-sequence code or the Gold code.

Further, in the invention described in the present application, in the above-mentioned battery state analyzing apparatus, the transfer function of the discrete system of the system including the battery, which is obtained by the transfer function calculating means, is converted into the pole in the transfer function of the discrete system. Along with determining the pole, the obtained pole is provided with a pole calculating means for converting into a pole in the transfer function of the continuous system of the AC equivalent circuit of the battery, based on the pole in the transfer function of the continuous system obtained by the pole calculating means, The state of the battery shall be analyzed.

Further, in the invention described in the present application, the pole calculating means in the above-mentioned battery state analyzing apparatus factorizes the denominator polynomial of the transfer function of the discrete system of the system including the battery,
Factoring operation means for obtaining a complex solution of poles in the transfer function of the discrete system, and truncating an imaginary part from the complex solution of poles obtained by the factoring operation means, from the remaining real number part,
Pole calculating means for specifying poles in the transfer function of the discrete system corresponding to the positive electrode and the negative electrode of the battery, and poles in the transfer function of the discrete system corresponding to the positive electrode and the negative electrode of the battery specified by the pole calculating means. Pole conversion calculation means for converting into a pole in the transfer function of the continuous system of the AC equivalent circuit of the battery based on the sampling cycle by the sampling means.

Further, in the invention described in the present application, in the battery state analyzing apparatus of the preceding paragraph, the AC signal applying means has an AC signal source for generating and outputting an AC signal, and an impedance element, and the AC signal An AC signal generated and output from a source is applied to the battery to be analyzed through the impedance element, and the factorization calculation means is
The polynomial of the denominator of the transfer function of the discrete system of the system including the battery, the order higher than the sum of the order of the pole in the transfer function of the continuous system of the AC equivalent circuit of the battery and the order of the pole in the impedance of the impedance element It is assumed that the term is truncated and the factorization is performed to find the poles in the transfer function of the discrete system.

Further, in the invention described in the present application, in the above-described battery state analyzing apparatus, the AC signal applying means has an AC signal source for generating and outputting an AC signal having a DC offset voltage. The AC signal generated and output from the signal source is directly applied to the battery to be analyzed,
The factorization calculation means discards the polynomial of the denominator of the transfer function of the discrete system of the system including the battery, rounding down terms having a higher order than the order of the pole in the transfer function of the continuous system of the AC equivalent circuit of the battery. It is assumed that the factorization is performed to find the pole in the transfer function of the discrete system.

Further, in the invention described in the present application, in the above-mentioned battery state analyzing apparatus, a temperature measuring means for measuring the ambient temperature of the battery is provided, and the pole calculating means is provided for the obtained AC equivalent circuit of the battery. It is assumed that a pole in the transfer function of the continuous system is provided with a pole temperature correction means for correcting the pole according to the ambient temperature of the battery measured by the temperature measurement means.

Further, in the invention described in the present application, the correlation between the pole in the transfer function of the continuous system of the AC equivalent circuit of the battery and the remaining amount of the battery, which is obtained in advance in the battery state analysis device described above. According to the above, it is assumed that the state determination means for determining the remaining capacity of the battery is provided based on the pole in the transfer function of the continuous system of the AC equivalent circuit of the battery obtained by the pole calculation means.

Here, the AC signal includes an AC signal containing a plurality of frequency components. Further, the AC signal includes a pulse wave and a composite wave of pulse waves having different duty ratios. Further, the pole here means a singular point, and the pole and the zero point having the reciprocal relation thereof are essentially synonymous. Further, here, pseudo random noise generating means is used as one means of the AC signal source.

[0046]

BEST MODE FOR CARRYING OUT THE INVENTION First, the basic principle of the present invention will be described.

(1) Definition of Battery Transfer Function In measuring the AC impedance of the battery, if the interface between the electrode and the electrolyte of the battery is polarized by a minute AC signal having an amplitude of about 10 mV or less superimposed on the DC voltage, the interface is hardly disturbed. Therefore, the obtained AC signal can be regarded as a change in the vicinity of the DC potential.

The AC equivalent circuit of the battery when a minute AC signal is applied is as shown in FIG. The impedance due to the contact between the electrode and the electrolytic solution can be represented by the parallel connection of the capacitance of the electric double layer generated between the electrode interface and the electrolytic solution interface and the Faraday impedance. Assuming that the charge transfer process is rate-determining in the electrode reaction, the Faraday impedance is expressed by the polarization resistance (charge transfer resistance) of the electrode and the capacitance component of the electrode connected in series. In FIG. 1, 31p and 31n are polarization resistances of positive and negative electrodes, and 32p
Is an electric double layer capacity in the positive electrode, 32n is an electric double layer capacity in the negative electrode, 33p and 33n are capacity components of the positive electrode and the negative electrode, 34 is resistance of the electrolytic solution, and impedance of the positive electrode is composed of polarization resistance 31p and capacity component 33p The Faraday impedance and the electric double layer capacitance 32p are connected in parallel, and the negative impedance is represented by the Faraday impedance composed of the polarization resistance 31n and the capacitance component 33n and the electric double layer capacitance 32n.

Here, the resistance value of the polarization resistor 31p is Rp,
If the resistance value of the polarization resistor 31n is Rn,

[0050]

[Equation 1]

It is expressed as In equations (1) and (2),
R is the Boltzmann constant, T is the absolute temperature, n is the number of charges in the electrode reaction, F is the Faraday constant, iop is the exchange current of the positive electrode, i
on indicates the exchange current of the negative electrode. In FIG. 1, Cdp is the capacitance value of the electric double layer capacitance 32p in the positive electrode, Cdn is the capacitance value of the electric double layer capacitance 32n in the negative electrode, and Rel is the resistance value of the resistor 34 of the electrolytic solution.

When the frequency of the AC signal is relatively high and the effect of the electrode capacitance components 33p and 33n on the battery impedance can be ignored, the AC equivalent circuit of the battery eliminates the capacitance components 33p and 33n from FIG. Then, it can be regarded as shown in FIG.

Assuming that the battery during AC polarization is a time-invariant linear system, the battery transfer function HB (s) in the AC equivalent circuit shown in FIG.

[0054]

[Equation 2]

It becomes In addition, the impedance of the impedance element used for measuring the AC impedance of the battery is set to H
Let I (s) be the total transfer function G (s) of the system consisting of the battery and the impedance element is defined as follows.

[0056]

[Equation 3]

Here, vB is the voltage applied to the battery during the AC impedance measurement, and iB is the current flowing through the battery during the AC impedance measurement. The pole order of the total transfer function G (s) is determined by the pole order in the impedance HI (s) of the impedance element and the battery transfer function HB.
It is the sum of the order of the poles in (s).

FIG. 3 is a complex impedance plot of a battery having a battery transfer function HB (s) as shown in equation (3). As shown in FIG. 3, the complex impedance of the battery goes through the semicircular locus of the negative electrode and the semicircular locus of the positive electrode as the angular frequency ω of the AC signal increases, and the resistance of the electrolyte resistance 34 at ω = ∞. The value becomes Rel. When the angular frequencies at the first pole s1 and the second pole s2 are ω1 and ω2, respectively, ω1 = -1 / Rn Cdn [rad / s] (5) ω2 = -1 / Rp Cdp [rad / s] ( 6) is given in.

(2) Estimating calculation of transfer function In the present invention, assuming that the battery is a time-invariant linear system, the total transfer function G (s) expressed by the equation (4) is calculated as follows. It is given as a general transfer function in the discrete system of shape.

[0060]

[Equation 4]

The transfer function G (z, θ) is estimated by determining the coefficient of the equation (7) based on the input / output signal of the battery. Generally, a method of estimating a transfer function of a system based on input / output data is called system identification. FIG. 4 is a diagram showing a model used for system identification. When the noise transfer function Hn (z, θ) is represented by Hn (z, θ) = 1 / A (z) (8), this model is Autoregressive with external input (A
RX) model. As shown in FIG. 4, the output signal y
Given that (t) is given by a linear combination of the input signal u (t) and the noise signal e (t), the output response is: A (z) y (t) = B (z) u (t) + e ( t) (9)

Coefficient parameter θ of transfer function G (z, θ) to be estimated, regression vector φ based on input / output data series
(T)

[0063]

[Equation 5]

Then, the equation (9) is expressed as y (t) = φT θ + e (t) (12). When the matrix operation of the equation (12) is solved by, for example, the RLS iterative method, the coefficient parameter θ ( t) is

[0065]

[Equation 6]

Is given by This coefficient parameter θ
(T) converges after a predetermined number of repeated calculations.

The initial value of the calculation is generally θ (0) = 0, P (0) = αId (14), but for example Rp = 0.001 to 10 [Ω] Rn = 0 .001 to 10 [Ω] Cdp = 10 −6 to 10 −2 [F] Cdn = 10 −6 to 10 −2 [F] (15) If an initial value within the range is used, iterative calculation until convergence is performed. Can be reduced.

(3) Calculation of Discrete System Poles The case where the number of discrete system poles and the number of continuous system poles are made equal will be described.

(3-1) When the impedance element is a pure resistance When the impedance element is a pure resistance, the order of the poles in the total transfer function G (s) shown in the expression (4) is quadratic, so that the expression (7) The system identification is performed on the assumption that the pole order in the discrete system transfer function G (z, θ) is also quadratic. That is, the equation (7) is

[0070]

[Equation 7]

It becomes Here, assuming that each parameter θ is converged in Expression (16), the pole in the transfer function G (z, θ) of the discrete system shown in Expression (16) is

[0072]

[Equation 8]

It is given by the solution of the quadratic equation of z as shown in. Here, the coefficient of equation (17) is

[0074]

[Equation 9]

If we put that, two poles can be obtained from the solution formula.

[0076]

[Equation 10]

This solution is generally given by the following complex solution. z1 = σ1 + jβ1 z2 = σ2 + jβ2 (20) Further, since the two poles in the total transfer function G (s) of the continuous system to be originally obtained exist on the real axis on the s plane, the transfer function G (z of the discrete system is obtained. , Θ) also has two poles on the real axis of the z-plane, the imaginary number component of the equation (20) is truncated, and z1 = σ1 z2 = σ2 (21) is a battery transfer function in a discrete system. The pole is G (z, θ).

The absolute values of the poles z1 and z2 are compared according to | z2 | <| z1 | (22), and when the expression (22) is true, the pole z1 corresponds to the negative pole of the battery, z2 becomes the pole corresponding to the positive electrode of the battery, and
When 2) is false, the pole z1 corresponds to the positive electrode of the battery and the pole z2 corresponds to the negative electrode of the battery.

(3-2) When the impedance element includes a pure capacitance When the impedance element includes a pure capacitance, the order of the poles in the total transfer function G (s) shown in the equation (4) becomes the third order. System identification is performed assuming that the pole order in the discrete system transfer function G (z, θ) shown in (7) is also cubic.
That is, the equation (7) is

[0080]

[Equation 11]

It becomes Transfer function G shown in equation (23)
The pole of (z, θ) is

[0082]

[Equation 12]

It is given by the solution of the cubic equation of z as shown in. And the three poles z1, z2, z3 are

[0084]

[Equation 13]

This solution is generally given by the following complex solution. z0 = σ0 + jβ0 z1 = σ1 + jβ1 z2 = σ2 + jβ2 (26) Furthermore, 3 of the total transfer function G (s) of the continuous system originally desired to be obtained.
Since the two poles are on the real axis on the s-plane, the poles are also on the real axis on the z-plane in the transfer function G (z, θ) of the discrete system. Truncate.
Z0, z1, and z2 are calculated as z0 = σ0 z1 = σ1 z2 = σ2 (27) in descending order of their absolute values. Of these, the pole z0 with the largest absolute value is z
The remaining poles z1 and z2 are excluded as the poles of the impedance element that should be = 1 and their absolute values are compared according to | z2 | <| z1 | (28). When formula (28) is true, the pole z1 corresponds to the negative electrode of the battery, and pole z2 corresponds to the positive electrode of the battery. When formula (28) is false, the pole z1 corresponds to the positive electrode of the battery. The pole and the pole z2 are the poles corresponding to the negative electrode of the battery.

(4) Transformation of poles from discrete system to continuous system FIG. 5 is a diagram showing mapping from the z plane to the s plane. By the mapping as shown in FIG. 5, the pole z1, in the discrete system is
From z2, the poles s1 and s2 in the continuous system as shown in equations (5) and (6) are obtained. If the sampling time in a discrete system is T [s], the mapping of poles from the z plane to the s plane is

[0087]

[Equation 14]

It is performed according to. (5) Correlation between battery state and pole transition The battery state can be evaluated by, for example, the remaining amount. The remaining amount means the remaining amount of electricity charged in the battery. There is a correlation between the transition of the pole in the battery transfer function and the remaining battery level.

FIG. 6 is a diagram showing a complex impedance plot of a lithium ion battery, and is a diagram showing a change in the complex impedance plot with a change in the remaining amount of the battery.
In the figure, (a), (b), (c), and (d) are cases where the remaining amount is 100%, 50%, 10%, and 0%, respectively. The resistance values Rp and Rn of the polarization resistors 31p and 31n shown in FIG. 2 change according to changes in the exchange current densities iop and ion caused by the chemical reaction of the positive electrode and the negative electrode due to charging and discharging of the battery. For this reason, the resistance values Rp and Rn of the polarization resistors 31p and 31n change with the change of the remaining amount of the battery, so that the complex impedance plot also changes with the change of the remaining amount of the battery as shown in FIG. To do. Therefore,
The angular frequency of the pole also changes as the battery level changes.

FIG. 7 is a graph showing the correlation between the remaining amount of the battery and the angular frequency ω of the pole in the battery transfer function.
(A) and (b) are a 1st pole and a 2nd pole, respectively. As shown in FIG. 7, the higher the angular frequency ω of the pole, the larger the remaining amount of the battery, and the lower the angular frequency ω of the pole, the smaller the remaining amount of the battery.

(Embodiment) An embodiment of the present invention will be described below with reference to the drawings. FIG. 8 is a diagram showing a configuration of a battery state analysis device according to an embodiment of the present invention, in which FIG. 8A is a diagram showing a system for applying an AC signal to the battery, and FIG. 8B is a state of the battery. It is a block diagram which shows the structure of the part which analyzes. The battery state analyzing apparatus according to the present embodiment shown in FIG. 8 estimates and evaluates the state of the battery 11 to be analyzed, for example, the degree of remaining capacity. In FIG. 8, 11
Is a battery to be analyzed, 12 is pseudo random noise generating means for outputting an AC signal applied to the battery 11, and 13 is the battery 11
, An impedance element provided between the pseudo random noise generating means 12 and the pseudo random noise generating means 12, and 14 is an AC voltage v applied to the battery 11.
It is an analog / digital converter as a sampling means for sampling the AC current iB flowing in B and the battery 11 and converting the analog value into a digital value.

Reference numeral 15 is an analog / digital converter 14
Transfer function calculating means for estimating and calculating a transfer function of a discrete system representing the characteristics of the system composed of the battery 11 and the impedance element 13 based on the output data of the transfer function calculating means 1
From the transfer function of the discrete system estimated by 5 to the battery 1
1, a pole calculating means for calculating the pole in the transfer function of the continuous system 1, 17 is state determining means for determining the state of the battery based on the pole calculated by the pole calculating means 16, and 18 is the battery 1
1 is a temperature sensor as a temperature measuring unit that measures the ambient temperature of No. 1 and outputs it to the pole calculating unit 16. The transfer function calculating means 15, the pole calculating means 16, and the state determining means 17 are realized by the microcomputer 20 in this embodiment.

FIG. 9 is a flow chart showing the operation of the battery state analyzing apparatus according to this embodiment shown in FIG. The operation of the battery state analysis device according to the present embodiment shown in FIG. 8 will be described below with reference to FIG.

First, in step S1, an AC signal is applied to the battery 11 to be analyzed. At this time, the frequency may be swept so as to include the frequency of the pole in the transfer function of the battery 11 using the pseudo random noise generating means, but in the present embodiment, the analysis can be performed efficiently. The pseudo random noise generating means 12 is a battery 11
It is assumed that an AC signal in a frequency band sufficiently including the frequency of the pole in the transfer function of is output.

FIG. 10 is a graph showing an example of frequency characteristics of the pseudo random noise generating means 12, where the horizontal axis represents frequency [Hz] and the vertical axis represents amplitude [dB]. As shown in FIG. 10, the frequency band of the AC signal output from the pseudo random noise generating means 12 is the first pole S of the transfer function of the battery 11.
It is necessary to include the components of the angular frequency ω1 at 1 and the angular frequency ω2 at the second pole S2. By using such a pseudo random noise generating means 12, frequency sweep becomes unnecessary. Also, the pseudo random noise generating means 12
The amplitude of the AC signal output from the device is below a level where the influence of the disturbance at the electrode interface can be ignored, and is below about 10 mV.

The pseudo random noise generating means 12 may be constituted by using a noise source which generates a noise signal such as white noise. Further, a pseudo noise signal generating means using a pseudo random code such as a noise source using an M sequence code as shown in FIG. 11 may be used. In FIG. 11, 41
Is a flip-flop as a delay operator having a predetermined delay time, and 42 is an exclusive OR calculator. Figure 1
In 1, the pseudo noise signal generating means is realized by hardware using a flip-flop and an exclusive OR calculator, but software such as storing and outputting the pseudo random code pattern in the memory of the microprocessor is used. There is also a method to realize it.

The impedance element 13 is the battery 11
It is intended to prevent a DC path from being formed between the battery 11 and the pseudo random noise generating means 12 when an AC signal is applied to the battery, and the battery state analyzing apparatus according to the present embodiment shown in FIG. A capacitor having a sufficiently large capacitance and a known capacitance value is used as the impedance element 13.

Next, in step S2, the AC voltage vB applied to the battery 11 and the AC current iB flowing in the battery 11 are sampled to obtain the time series input / output data used for the estimation calculation of the transfer function of the discrete system. Step S2 is performed by the analog / digital converter 14. AC voltage vB applied to the battery 11 and AC current iB flowing in the battery 11
Respectively correspond to the input signal u (t) and the output signal y (t) in the ARX model shown in FIG. 4, and thus from the digital values sampled by the analog-to-digital converter 14 at predetermined time intervals, equation (11) ), The regression vector φ (t) based on the input / output data series is obtained. In other words, the AC voltage vB applied to the battery 11 and the AC current iB flowing in the battery 11 are time-series input / output data used for the estimation calculation of the transfer function of the discrete system via the analog / digital converter 14 as the sampling means. Becomes

Next, in step S3, step S2
The transfer function of the discrete system is estimated and calculated using the time-series input / output data obtained in. Step S3 is the transfer function calculation means 15.
According to the procedure described in (2) Battery transfer function estimation calculation in the basic principle of the present invention. That is, the transfer function calculating means 15 is
AC voltage vB and AC current iB sampled by the digital converter 14 are converted into time series input / output data u (t),
The regression vector ψ (t) of the equation (11) is obtained as y (t), and the transfer function estimation calculation shown in the equation (13) is performed using the initial value given by the equation (14) or (15). The coefficient parameter θ (t) of the equation (10) is calculated to obtain the equation (7)
Find the transfer function of the discrete system as shown in.

Next, in step S4, step S3
The poles in the transfer function of the continuous system are calculated from the transfer function of the discrete system obtained in. Step S4 is performed by the pole calculating means 16 according to the procedure described in the paragraphs (3) calculation of the poles of the discrete system and (4) conversion of the poles from the discrete system to the continuous system in the basic principle of the present invention. Done.

FIG. 12 is a block diagram showing the structure of the pole calculating means 16. First, the factorization calculation means 16a factorizes the rational polynomial that is the denominator of the transfer function of the discrete system estimated by the transfer function calculation means 15 in step S3. In this embodiment, since a pure capacitance is given as the impedance element 13, as described in (3-2) in the basic principle of the present invention, the transfer function of the estimated expression (7) is obtained. The polynomial A (z) that is the denominator is 3
After truncating higher-order terms, the expressions (23) to (2
Factor according to 5).

As shown in the equation (27), the pole calculating means 16b truncates the imaginary part from the complex solution obtained from the equation (25) to leave only the real number part, and further compares the magnitude according to the equation (28) to determine the battery. Determine the poles corresponding to 11 positive and negative electrodes.

The pole transformation calculating means 16d transforms the poles from the z domain to the s domain using the sampling period T according to the equation (29). The factorization calculation means 16a, the pole calculation means 16b, and the pole conversion calculation means 16c determine the poles in the battery transfer function HB (s) of the continuous system given by the equation (3) from the transfer function of the discrete system.

The pole calculating means 16 in this embodiment is further provided with a pole temperature correcting means 16d, and the factorization calculating means 16a, the pole calculating means 16b and the pole conversion calculating means 16c.
Battery transfer function HB (s) of continuous system obtained by
The pole at is corrected by the ambient temperature of the battery measured by the temperature sensor 18.

The pole temperature correction is performed as follows. The temperature sensor 18 measures the surface temperature or the ambient temperature of the battery 11. As shown in equations (1) and (2), the polarization resistance 3 of the positive electrode
The resistance value Rp of 1p and the resistance value R of the polarization resistance 31n of the negative electrode
Since both n are proportional to the absolute temperature T, equations (5) and (6)
The angular frequency of the pole indicated by is inversely proportional to absolute temperature. Here, it is assumed that the temperature inside the battery 11 is equal to the temperature measured by the temperature sensor 18. Let Ta [K] be the reference temperature (experimental temperature that obtains the correlation between the state of the electrode and the battery) and Td [K] be the temperature measured by the temperature sensor 18. Resistance values Rp and Rn of the polarization resistance at the reference temperature Ta.
And the relationship between the resistance values Rp 'and Rn' of the polarization resistance at the temperature Td are as follows from equations (1) and (2)

[0106]

[Equation 15]

It becomes as follows. From equations (30) and (31), the angular frequencies ω1 and ω2 of the poles at the reference temperature Ta are
It is determined by correcting the polar angular frequencies ω1 ′ and ω2 ′ at the temperature Td according to the following equation.

[0108]

[Equation 16]

It becomes: The polar temperature correction means 16d expresses the angular frequency of the pole in the battery transfer function HB (s) of the continuous system obtained by the polar conversion calculation means 16c by the equations (32) and (3).
According to 3), the temperature is corrected and output.

Finally, in step S5, step S
The remaining battery level is determined based on the polar angular frequency obtained in 4. In step S5, the state determination means 17
This is performed according to the procedure described in the section (5) Correlation between battery state and pole transition in the basic principle of the present invention.

When the pseudo random noise generating means 12 has the same DC offset voltage as the battery 11, the impedance element 13 can be omitted. FIG. 13 is a diagram showing a configuration of a battery state analyzing device that applies an AC signal to the battery 11 without passing through an impedance element, and shows only a system that applies an AC signal to the battery 11. In FIG. 13, 12A is a voltage offset pseudo-random noise generating means having a DC offset voltage equal to that of the battery 11. The voltage offset pseudo random noise generating means 12A is the pseudo random noise generating means 12 shown in FIG.
Similarly, a voltage offset noise source that generates a noise signal may be used.

Similarly to the battery state analyzing apparatus shown in FIG. 8, the battery state analyzing apparatus shown in FIG. 13 samples the AC voltage vB applied to the battery 11 and the AC current iB flowing in the battery 11 to obtain a discrete system transfer function. The state of the battery 11 can be determined by calculation. However, in the case of the battery state analysis device of FIG. 13, since the (3-1) impedance element in the basic principle of the present invention corresponds to a pure resistance, the factorization calculation means 16a constituting the pole calculation means 16 is , The rational polynomial A (z) that is the denominator of the transfer function of the discrete system estimated by the transfer function calculating means 15 is 2
After truncating higher-order terms, the equations (17) to (1
Factor 9 according to 9).

The transfer function calculating means 15 may be provided with an initial value calculated in advance based on the polarization resistance and capacitance components of the electrodes of the battery 11 to be analyzed. Thereby, the convergence of the estimation operation of the transfer function of the discrete system can be improved.

In the pole calculating means 16, the pole temperature correcting means 16d may not be an essential component. In this case, the temperature sensor 18 becomes unnecessary.

Further, a battery state analyzing device for judging the state of the battery other than the remaining amount from the transfer function of the discrete system obtained by the transfer function calculating means 15 is also conceivable. For example, it is conceivable to judge the safety and deterioration of the battery from the transfer function of the discrete system.

As a method different from the present embodiment, a method may be considered in which the order of poles of the transfer function of the discrete system to be estimated is arbitrary and the transition of the poles is directly correlated with the remaining battery level. For example, it is assumed that the order of the poles of the transfer function of the discrete system to be estimated and calculated is higher than the order of the poles of the transfer function of the continuous system of the AC equivalent circuit that is predetermined for the battery 11. Here, assuming that the pole order of the transfer function is the 30th order, the transfer function G (z, θ) is expressed by the following equation.

[0117]

[Equation 17]

After the coefficient parameter θ converges, the transfer function G obtained by the calculation using the discharge method algorithm
Find 30 poles of (z, θ). FIG. 14 is a graph showing an example of the position of the 30th pole in the z plane. 14
As shown in, the state of the battery 11 is analyzed by extracting all the poles associated with the battery 11 and the impedance element 12 and analyzing the transition of a specific pole among them. The determination of the remaining amount can be performed by observing a pole that is previously identified as having a correlation with the remaining amount by an experiment or the like.

[0119]

As described above, according to the present invention, the transfer function of the discrete system of the system including the battery is first obtained by the estimation operation,
After obtaining the poles in the transfer function of the continuous system of the AC equivalent circuit of the battery from the transfer function of this discrete system, the correlation between the poles in the transfer function of the AC equivalent circuit of the battery and the states such as the remaining capacity of the battery Since the state of the battery is analyzed by using, it is possible to perform the state of the battery reliably and quantitatively.

Further, in order to estimate and calculate the transfer function of the discrete system, analog signal processing is not necessary and all can be performed by digital signal processing, so that it is possible to realize an LSI by a one-chip microcomputer or DSP, It is possible to realize a low-cost and high-accuracy battery state analysis device.

[Brief description of drawings]

FIG. 1 is a diagram for explaining the principle of the present invention and is a circuit diagram showing an AC equivalent circuit of a battery.

FIG. 2 is a diagram for explaining the principle of the present invention and is a circuit diagram showing an AC equivalent circuit of a battery when the frequency of an AC signal is relatively high.

FIG. 3 is a diagram for explaining the principle of the present invention, and is a graph showing a complex impedance plot of a battery.

FIG. 4 is a diagram for explaining the principle of the present invention and is a diagram showing an ARX (autoregressive) model used for system identification.

FIG. 5 is a diagram for explaining the principle of the present invention, showing a mapping from the z plane to the s plane for converting the poles obtained from the transfer function of the discrete system of the battery into the poles in the transfer function of the continuous system. Figure

FIG. 6 is a diagram for explaining the principle of the present invention, showing a change in the complex impedance plot with a change in the remaining amount of the battery (a) a diagram showing a case where the remaining amount is 100% (b) a remaining amount Figure showing the case of 50% (c) Figure showing the case of the remaining amount of 10% (d) Figure showing the case of the remaining amount of 0%

FIG. 7 is a diagram for explaining the principle of the present invention, showing the correlation between the remaining amount of the battery and the angular frequency of the pole in the transfer function of the battery (a) showing the case of the first pole (b) ) Graph for the second pole

FIG. 8 is a diagram showing a configuration of a battery state analysis apparatus according to an embodiment of the present invention, FIG. 8A is a diagram showing a system for applying an AC signal to the battery, and FIG. 8B is a configuration of a portion for performing battery state analysis. Block diagram showing

9 is a flowchart showing the operation of the battery state analysis device according to the embodiment of the present invention shown in FIG.

FIG. 10 is a graph showing an example of frequency characteristics of the pseudo random noise generating means used in the embodiment of the present invention.

FIG. 11 is a diagram showing a configuration of a noise source using an M-sequence code, which is used as a pseudo-random noise generating means in the embodiment of the present invention.

FIG. 12 is a block diagram showing a configuration of a pole calculating means in the battery state analyzing device according to the embodiment of the present invention.

FIG. 13 is a diagram showing a configuration of a battery state analyzing device according to an embodiment of the present invention for applying an AC signal to the battery without passing through an impedance element, and only a system for applying the AC signal to the battery. Showing

FIG. 14 is a graph showing an example of the position of the pole in the z plane when the order of the pole in the transfer function of the battery is the 30th order.

[Explanation of symbols]

AC voltage applied to VB battery AC current flowing in iB battery 11 batteries 12 Pseudo-random noise generating means 12A voltage offset pseudo-random noise generating means 13 Impedance element 14 Analog / digital converter (sampling means) 15 Transfer function calculation means 16 pole calculation means 16a Factorization operation means 16b Pole calculation means 16c Pole conversion calculation means 16d extreme temperature correction means 17 State determination means 18 Temperature sensor (temperature measuring means)

   ─────────────────────────────────────────────────── ─── Continued front page    (72) Inventor Taketoshi Nakao             1006 Kadoma, Kadoma-shi, Osaka Matsushita Electric             Sangyo Co., Ltd. (72) Inventor Yoshinori Kumamoto             1006 Kadoma, Kadoma-shi, Osaka Matsushita Electric             Sangyo Co., Ltd. F-term (reference) 2G016 CB12 CC01 CC04 CC09 CC13                       CC16 CC24 CC27 CD14 CF06                 5H030 AS08 AS14 FF41 FF42 FF43                       FF44

Claims (7)

[Claims] 1. A pseudo random noise generating means for outputting an alternating current signal applied to a battery, and an alternating voltage applied to the battery or an alternating current flowing through the battery is sampled and used for estimation calculation of a transfer function of a discrete system. Sampling means for obtaining series data, transfer function calculating means for calculating a pole in a continuous system of a battery from an estimated transfer function of a discrete system, and a state for determining the remaining amount of the battery based on the obtained pole A remaining capacity measuring device for a battery, comprising: a determining means. 2. The battery residual capacity measuring device according to claim 1, wherein the pseudo random noise generating means has a variable frequency. 3. The state-of-charge measuring device according to claim 1, wherein the frequency band of the AC signal output by the pseudo random noise generating means includes a pole component of the transfer function of the battery. 4. The battery residual capacity measuring device according to claim 1, wherein the pseudo-random noise generating means is noise using M-sequence code or Gold code or white noise. 5. The pseudo random noise generating means is voltage offset pseudo random noise.
The battery residual capacity measuring device according to any one of to 4. 6. The battery according to claim 1, wherein the AC signal applying means comprises an AC signal source for generating and outputting an AC signal, and means for applying the AC signal to the battery via an impedance element. Remaining capacity measuring device. 7. The sampling means comprises: time-series data of an alternating voltage and an alternating current flowing through the battery; and an analog / digital converting means for converting an analog value obtained from the sampling means into a digital value. The battery residual capacity measuring device according to any one of claims 1 to 6, which is characterized in that.