JP6277864B2 - Battery internal state estimation device - Google Patents

Description

  The present invention relates to a battery internal state estimation device that calculates internal impedance of a secondary battery.

  In order to operate the secondary battery safely, it is necessary to grasp the internal state of the secondary battery. As an index indicating the internal state of the secondary battery, there are a charge amount (SOC: State of Charge) of the secondary battery, allowable charge / discharge power, internal resistance, a degree of deterioration, and the like.

  In order to grasp the internal state of the secondary battery, a method for capturing electrical and chemical phenomena occurring inside the secondary battery is required. The response between voltage and current varies with electrical and chemical phenomena. The response between voltage and current corresponds to the internal impedance of the secondary battery. A method is known in which a secondary battery is modeled by an equivalent circuit based on the internal impedance of the secondary battery, and the equivalent circuit is used for characteristic measurement and control of the secondary battery.

  It is known that the internal impedance of a secondary battery can be modeled by a combination of a plurality of resistance components and a plurality of capacity components. The absolute values of the real part and the imaginary part of the internal impedance at a certain frequency correspond to the internal resistance of the secondary battery at that frequency. Since the internal impedance changes according to the frequency of the voltage and current input / output to / from the secondary battery, the internal resistance of the secondary battery changes.

  For example, consider a secondary battery mounted on a vehicle. In a vehicle-mounted environment, an electrical load such as a generator and a motor is connected to the secondary battery via a DC-AC converter, and aperiodic charge / discharge is performed between the secondary battery, the generator and the electrical load. Is done. In this non-periodic charge / discharge, since the magnitude of the charge / discharge current and the duration of charge / discharge change, the AC impedance method of measuring impedance by applying a sine wave cannot be applied. In addition, an analysis method for accurately calculating the frequency characteristics is required while the internal state of the secondary battery changes every moment.

  There has been proposed a method for calculating the frequency characteristics of the internal resistance of the secondary battery from the voltage and current of the secondary battery obtained in a vehicle-mounted environment. Applying wavelet transform to the charge / discharge current of the secondary battery and the voltage between terminals, the internal impedance is calculated from the slope of the obtained voltage wavelet transform coefficient and current wavelet transform coefficient, and the size of the real part of the internal impedance is A method of calculating the frequency characteristic of the resistor is proposed (Patent Document 1).

JP2012-83142A

  In the prior art method, the internal resistance at a certain frequency of the secondary battery can be obtained from the voltage and current waveforms of the secondary battery obtained in the operating system, such as when the vehicle is operating. ing. However, the prior art method only calculates the size of the real part of the internal impedance. If the absolute value of both the real part and the imaginary part of the internal impedance or the real part and the imaginary part is not obtained together with the frequency information, the contribution of various phenomena of the secondary battery is related to a plurality of resistance components and the imaginary part related to the real part. A plurality of capacitive components cannot be obtained. Without an imaginary part, a plurality of resistance components corresponding to various phenomena cannot be separated. Further, the capacitance component cannot be obtained from the frequency characteristic of the impedance absolute value not including the imaginary part. From the above, time constants of various phenomena cannot be obtained. That is, in the operating system, the contribution degree of various phenomena of the secondary battery cannot be obtained, and the internal state of the secondary battery cannot be accurately estimated.

  The present invention has been made to solve the above-described problem, and an object of the present invention is to provide a battery internal state estimation device that can accurately estimate the details of the internal state of a secondary battery even in an operating system. .

  The battery internal state estimation device according to the present invention is detected by voltage detection means (31) for detecting the voltage across the terminals of the secondary battery (10) and current detection means (30) for detecting the charge / discharge current flowing in the secondary battery. A storage means (54) for storing values in time series, and a voltage wavelet transform coefficient is calculated by performing wavelet transform on the detected value of the inter-terminal voltage stored in the time series, and is stored in the time series. Wavelet transform means (60) for calculating a current wavelet transform coefficient by performing wavelet transform on the detected charge / discharge current value, and a ratio between the voltage wavelet transform coefficient and the current wavelet transform coefficient for each frequency. Internal impedance calculation means (61) for calculating a real part and an imaginary part of the internal impedance of the secondary battery Characterized in that it obtain.

  Unlike the Fourier transform, the wavelet transform is characterized in that both frequency information can be obtained without losing time information, and the accuracy of both the real part and the imaginary part can be expected. In addition, the wavelet transform can accurately calculate each frequency component even if the object to be transformed is an aperiodic waveform. That is, the real part and the imaginary part of the internal impedance of the secondary battery at each frequency can be accurately calculated as the internal state of the secondary battery while performing aperiodic and complicated charge / discharge. In addition, when the wavelet transform is used, charging / discharging of the secondary battery is carried out, and then the gradual voltage change when charging / discharging is stopped, that is, the real part of the internal impedance even from a fragmentary waveform for one period. And the imaginary part can be calculated. When charging / discharging is stopped, there is no change in the charge amount, and there is no heat generation in the secondary battery, so that the real part and imaginary part of the internal impedance in the low frequency range can be calculated with high accuracy. Thus, even in the operating system, the details of the internal state of the secondary battery can be accurately estimated.

The electrical block diagram of 1st Embodiment. The functional block diagram of a battery internal state estimation apparatus. The conceptual diagram showing the characteristic (correspondence relationship between OCV and SOC) of a storage battery. The figure which shows the equivalent circuit model of a storage battery. The conceptual diagram of the vector locus diagram of the internal impedance of a storage battery. The conceptual diagram showing transition of the electric current and voltage during operation | movement and a stop of a vehicle. The flowchart which shows the calculation process of diffused resistance. The figure showing the measurement result of a diffused resistance, and a model estimation result. The figure which shows the calculation precision of SOC. The electrical block diagram of a modification.

(First embodiment)
FIG. 1 shows an electric circuit diagram including the assembled battery B and the control device 50 as a battery internal state estimating device in the present embodiment. The assembled battery B is an assembled battery in which a plurality of storage batteries 10 (battery cells) made of lithium ion storage batteries are connected in series or in parallel, and the control device 50 includes the storage battery 10 (battery cells) constituting the assembled battery B. Each internal state is estimated. Assume that the assembled battery B and the control device 50 are mounted on a vehicle, for example. The vehicle is assumed to be a hybrid vehicle including an internal combustion engine and a motor generator.

  The storage battery 10 is connected to a motor generator 22 via an inverter 21 that performs AC-DC conversion. When the motor generator 22 functions as a generator, the storage battery 10 is charged with electric power supplied from the motor generator 22. Further, when the motor generator 22 functions as a motor as a power source, electric power is supplied to the motor generator 22 by discharging from the storage battery 10. The vehicle has an EV (Electric Vehicle) travel mode in which combustion in the internal combustion engine is stopped, output of driving force is stopped, and the motor generator 22 functions as a motor. Further, an HV (Hybrid Vehicle) traveling mode is provided in which combustion in the internal combustion engine is performed to output driving force, and the motor generator 22 functions as a motor or a generator.

  The storage battery 10 is connected to the electrical load 20 and supplies power to the electrical load 20. Moreover, it is possible to charge the storage battery 10 from a commercial power source or the like. Further, relay switches 33 and 34 are installed between the storage battery 10 and the electric load 20 and the inverter 21 and the commercial power source, and the storage battery 10 and the main path are connected and disconnected according to a command from the control device 50. A current sensor 30 for detecting a charge / discharge current flowing in the storage battery 10 is provided on a path connecting the storage battery 10 to the electric load 20 and the inverter 21. Between the terminals of the storage battery 10, a voltage sensor 31 for detecting a voltage between the terminals is provided. The storage battery 10 is provided with a temperature sensor 32. The current sensor 30 (current detection means), the voltage sensor 31 (voltage detection means), and the temperature sensor 32 output detection signals corresponding to the charge / discharge current, the voltage between terminals, and the battery temperature, respectively. Is input.

  The control device 50 acquires the detection value I of the current sensor 30, the detection value V of the voltage sensor 31, and the detection value T of the temperature sensor 32 based on the input detection signal. The control device 50 calculates the SOC, the deterioration degree, and the predicted power of the storage battery 10 based on the acquired detection values I, V, and T.

  Subsequently, the calculation method of the present embodiment performed by the control device 50 will be described below. FIG. 2 is a functional block diagram showing functions of the control device 50 of the present embodiment. The control device 50 includes an internal impedance processing unit 51, an RC component calculation unit 62, an SOC processing unit 52, a deterioration degree processing unit 53, and a predicted power processing unit 54.

  First, the internal impedance processing unit 51 will be described. The control apparatus 50 memorize | stores the detected value V of the voltage between the terminals of the storage battery 10 detected by the current sensor 30 and the voltage sensor 31, and the detected value I of the charging / discharging current in time series. Then, the detected value V (t) of the inter-terminal voltage and the detected value I (t) of the charge / discharge current stored in time series are wavelet transformed by the wavelet transform unit 60, respectively.

  The principle of the technology used in the present invention will be described. The continuous wavelet transform of the waveform f (t) is defined by equation (1).

（Ｗψｆ）（ａ，ｂ）はウェーブレット変換係数であり、周波数特性を持つ。ウェーブレット変換係数は、対応する周波数では強度のピークが現れ、対応する周波数から外れると、積分による効果で正負が打ち消され、強度が小さくなる特性がある。式（１）の右辺に用いられているマザーウェーブレットψａ，ｂ（ｔ）は式（２）で定義され、上線は複素共役を表す。ここで、スケールファクタａは拡大縮小量を表すパラメータであり、周波数の逆数に比例する。シフトパラメータｂはｔ軸上のシフト量を表すパラメータであり、時間に比例する。

(Wψf) (a, b) is a wavelet transform coefficient and has frequency characteristics. The wavelet transform coefficient has a characteristic that an intensity peak appears at the corresponding frequency, and when it deviates from the corresponding frequency, the sign is canceled by the effect of integration, and the intensity becomes small. The mother wavelet ψa, b (t) used on the right side of equation (1) is defined by equation (2), and the upper line represents the complex conjugate. Here, the scale factor a is a parameter that represents the amount of enlargement / reduction, and is proportional to the reciprocal of the frequency. The shift parameter b is a parameter representing the shift amount on the t axis and is proportional to time.

なお、マザーウェーブレットとして様々な関数を用いることができ、一般的なものとして、ガボール関数、メキシカンハット関数、モルレー関数が挙げられる。

Various functions can be used as the mother wavelet, and general examples include a Gabor function, a Mexican hat function, and a Morray function.

  Various mother wavelets were applied to the voltage and current waveforms when the secondary battery was mounted on a vehicle. As a result, comparing the case where the Morray function is applied and the case where the Gaussian function is applied, the obtained spectrum is narrower and the internal impedance of the secondary battery is calculated with higher accuracy when the Morray function is applied. it can.

As a discrete wavelet transform, there is a method obtained by decomposition using a Haar scaling function. In this method, the relationship of a = 1/2 ^ j and b = k / 2 ^ j is established for the expression (2).
Equation (3) shows the wavelet transform coefficient WψV of the inter-terminal voltage V and the wavelet transform coefficient WψI of the charge / discharge current I.

インピーダンス算出部６１において、式（４）で示すように、周波数毎で、電圧ウェーブレット変換係数ＷψＶを、電流ウェーブレット変換係数ＷψＩで除算することで、周波数毎の内部インピーダンスＺを算出することができる。

The impedance calculation unit 61 can calculate the internal impedance Z for each frequency by dividing the voltage wavelet transform coefficient WψV by the current wavelet transform coefficient WψI for each frequency, as shown in Expression (4).

内部インピーダンスＺの実部Ｒｅ（Ｚ）と虚部Ｉｍ（Ｚ）は、式（５）となる。

The real part Re (Z) and the imaginary part Im (Z) of the internal impedance Z are expressed by Equation (5).

各シフトパラメータｂにおけるＲｅ（Ｚ（ａ，ｂ）），Ｉｍ（Ｚ（ａ，ｂ））の平均値として、Ｒｅ（Ｚ（ａ）），Ｉｍ（Ｚ（ａ））をそれぞれ算出することができる。また、スケールファクタａと周波数ｆとは一対一の関係にあるため、Ｒｅ（Ｚ（ａ））及びＩｍ（Ｚ（ａ））をＲｅ（Ｚ（ｆ））及びＩｍ（Ｚ（ｆ））に変換することができる。内部インピーダンスＺの実部Ｒｅ（Ｚ）と虚部Ｉｍ（Ｚ）を算出する際、具体的に３周期程度（周波数ｆの逆数の３倍程度）に相当する電流及び電圧の検出値Ｉ（ｔ），Ｖ（ｔ）をサンプリングするのが望ましい。

Re (Z (a)) and Im (Z (a)) can be calculated as average values of Re (Z (a, b)) and Im (Z (a, b)) for each shift parameter b. it can. Since scale factor a and frequency f have a one-to-one relationship, Re (Z (a)) and Im (Z (a)) are changed to Re (Z (f)) and Im (Z (f)). Can be converted. When calculating the real part Re (Z) and the imaginary part Im (Z) of the internal impedance Z, the current and voltage detection values I (t corresponding to about three cycles (about three times the reciprocal of the frequency f) are specifically obtained. ), V (t) is preferably sampled.

  As described above, the real part Re (Z (f)) and the imaginary part Im (Z (f)) of the internal impedance Z for each frequency f of the storage battery 10 can be obtained by the wavelet transform by the wavelet transform unit 60.

  Next, the voltage correction unit 56 will be described. In FIG. 3, the conceptual diagram showing the relationship between charge amount SOC and the open end voltage OCV as a characteristic which the storage battery 10 has is shown. Attention is paid to the change in the open-circuit voltage OCV when the storage battery 10 is charged and discharged. As a characteristic of the storage battery 10, there is a one-to-one relationship between the charge amount SOC and the open-circuit voltage OCV. Generally, a linear region is used in consideration of controllability and is used for calculation of the charge amount SOC. . When the storage battery 10 is charged / discharged, the charge amount SOC changes, and the open end voltage OCV also changes with the change.

  If the internal impedance Z is calculated without considering the change in the open circuit voltage OCV of the storage battery 10 due to the change in the SOC, the behavior in which the change in the open circuit voltage OCV corresponds to a part of the capacity component of the internal impedance Z. It becomes. That is, a capacity component different from the internal impedance of the storage battery is erroneously added, and the accuracy of the calculated internal impedance Z is reduced. Therefore, when calculating the internal impedance Z, it is necessary to take measures such as aligning the reference of the terminal voltage except for the change of the open-end voltage OCV included in the detected waveform of the terminal voltage.

  Therefore, in the present embodiment, the OCV in the current SOC is calculated by the OCV calculation unit 57, and the change amount ΔOCV of the open-end voltage OCV due to the change in the SOC is calculated by the ΔOCV calculation unit 58 (ΔOCV = OCVini−OCV). OCV is a current value, OCVini is a reference value, and OCVini is an OCV when the storage unit 55 starts storing Va. Then, a value Va corrected by subtracting ΔOCV from the detected value V of the terminal voltage by the subtractor 59 is stored in the storage unit 55 (Va = V−ΔOCV). Then, the corrected voltage detection value Va (t) stored in time series by the storage unit 55 is wavelet transformed, and the real part Re (Z) and imaginary part Im () of the internal impedance Z are used by using the wavelet transformation coefficient WψVa. Z) was calculated.

  Next, the RC component calculation unit 62 will be described. The RC component calculation unit 62 calculates the equivalent of the internal impedance Z based on the real part Re (Z (f)) and the imaginary part Im (Z (f)) of the internal impedance Z for each frequency f calculated by the impedance calculation unit 61. The resistance component and the capacitance component in the circuit model are calculated.

  An equivalent circuit of the storage battery 10 is shown in FIG. The equivalent circuit of the storage battery 10 has a configuration in which an open-circuit voltage OCV11 and a model 12 of the internal impedance Z are connected in series. The model 12 of the internal impedance Z includes a direct current resistance 13 representing the current resistance of the electrode in the solution and the electrodes, reaction resistances 14 and 15 representing the electrode interface reaction in the positive electrode and the negative electrode, and diffusion representing the ion diffusion in the active material and in the solution. The resistor 16 is connected in series.

  The diffused resistor 16 has a configuration in which a plurality of parallel connections of resistance components and capacitance components are connected in series. An equivalent circuit in which a resistance component and a capacitance component are connected in parallel is called a Foster-type equivalent circuit. In addition, there is a method in which the diffused resistor 16 is expressed by a Cowell equivalent circuit. It should be noted that the Cowell equivalent circuit representation and the Foster equivalent circuit representation are equivalent because they can be converted although the values of the resistance component and the capacitance component are different.

  The DC resistor 13 is configured as a resistance component Rs. The reaction resistor 14 is configured by connecting a resistance component R1 and a capacitance component C1 in parallel. The reaction resistor 15 is configured by connecting a resistance component R2 and a capacitance component C2 in parallel.

  The diffused resistor 16 is configured by connecting a plurality of resistance components Rwi and a plurality of capacitance components Cwi in parallel. However, i is the number of the parallel connection bodies of the resistance component and the capacitance component constituting the diffused resistor 16 and takes a range of i = 0, 1,.

  The DC resistors 13, reaction resistors 14, 15 and diffusion resistors 16 have different time constants. The time constant of the DC resistor 13 is 0.001 second, and can be handled as approximately zero. The time constant of the reaction resistances 14 and 15 is about 0.1 second to about 1 second. The time constant of the diffused resistor 16 is on the order of several tens of seconds to several thousand seconds.

  Transition of each resistance component and each capacitance component when categorized by DC resistance 13, reaction resistances 14 and 15 and diffusion resistance 16 in ascending order of time constant in a deterioration durability test on a consumer battery (lithium ion secondary battery) Will be described. The values of each resistance component and each capacitance component both increase due to deterioration compared to the values before the test. Moreover, the increase rate (deterioration degree) which each differs is shown. As described above, since the degree of deterioration of each resistance component and each capacity component is different, it is necessary to individually calculate each resistance component and each capacity component in order to appropriately estimate the degree of deterioration. In particular, the resistance component R2 and the capacitance component C2 of the reaction resistor 15 have a result that increases remarkably compared to the other components.

  Here, the frequency range of the current / voltage of the storage battery 10 in the vehicle will be described. During operation of the vehicle (on-vehicle power supply system), the storage battery 10 is charged and discharged aperiodically. On the high frequency side of the change in the current / voltage of the storage battery 10 during the operation of the vehicle, the carrier frequency of the inverter 21 is several kHz, and the frequency at which the DC resistance 13 of the general storage battery and the reaction resistances 14 and 15 of the electrode interface can be obtained. Is in range. On the other hand, on the low frequency side, the low frequency region in the vicinity of the reaction resistor 15 can be acquired for the frequency region of the diffused resistor 16. However, the ultra-low frequency range which is a lower frequency range and is located at the end point of the diffused resistor 16 is not included during vehicle travel, and the internal impedance of the ultra-low frequency range may not be acquired.

  The diffused resistor 16 has a combined impedance in which a parallel circuit of a plurality of resistance components and capacitance components overlaps. Therefore, it is impossible to separate each pair of resistance component and capacitance component in parallel circuits, and it is necessary to match them to the combined impedance. In consideration of the above, the following embodiment was adopted.

  The RC component calculation unit 62 calculates a resistance component and a capacitance component using an RC component calculation method 1 and an RC component calculation method 2 described below. In the RC component calculation method 1, the DC resistance 13 and the reaction resistances 14 and 15 at the electrode interface are calculated. In the RC component calculation method 2, the diffusion resistance 16 is calculated.

  First, the RC component calculation method 1 will be described. A method using a vector locus diagram is applied. The vector locus diagram has a more complicated equivalent circuit model than the Bode diagram.

  FIG. 5 shows a conceptual diagram of a vector locus diagram representing the frequency characteristics of the internal impedance Z. The real part Re (Z) of the internal impedance Z increases as the frequency f of the change in the inter-terminal voltage of the storage battery 10 decreases. For convenience, the imaginary part Im (Z) has a negative value on the positive side. In FIG. 5, it divides | segments into the area | regions D0-D2 according to the frequency f for description. The relationship between the frequency f and the time constant τ is 2πf = 1 / τ = 1 / (R · C). A region where the frequency f is f >> 1 / (2π · τ1) is a region D0, a region where the frequency f is near fa1 = 1 / (2π · τ1) is a region D1, and the frequency f is fa2 = 1 / (2π · τ2). ) Let D2 be a nearby region.

  In the region D0 where the frequency f is f >> fa1, the DC resistor 13 contributes as the internal impedance Z. That is, the real part of the internal impedance Z is Rs. When the frequency f falls below the lowest point P1 (Re (Z) = Rs, Im (Z) ≈0) at the boundary between the region D0 and the region D1 where the frequency f is fb1 (fb1> fa1), the internal impedance The real part and the imaginary part of Z increase.

  In the vector locus diagram in the region D1 where the frequency f is near fa1, an arc is formed. In the region D1, the reaction resistance 14 contributes as the internal impedance Z. When the frequency f is f = fa1, the vertex Q1 of the arc-shaped vector locus in the region D1 is obtained. At the vertex Q1, Re (Z) = Rs + R1 / 2 and Im (Z) = R1 / 2. Further, when the frequency f becomes fb2 (fa1> fb2> fa2), it becomes the lowest point P2 of the boundary between the region D1 and the region D2. At the lowest point P2, Re (Z) = Rs + R1, Im (Z) ≈0.

  In the vector locus diagram in the region D2 where the frequency f is in the vicinity of fa2, an arc is formed as in the region D2. In the region D2, the reaction resistance 15 contributes as the internal impedance Z. When the frequency f becomes f = fa2, it becomes the vertex Q2 of the arc-shaped vector locus in the region D2. At the vertex Q2, Re (Z) = Rs + R1 + R2 / 2, Im (Z) = R2 / 2. Further, when the frequency f becomes fb3 (fa2> fb3> fa3), the lowest point P3 is obtained. At the lowest point P3, Re (Z) = Rs + R1 + R2, Im (Z) ≈0.

  Each resistance component Rs, R1, R2 can be calculated based on the respective real parts of the lowest points P1 to P3 searched in this way. Specifically, the real part of the lowest point P1 corresponds to the resistance component Rs. Further, a value obtained by subtracting the real part of the lowest point P1 from the real part of the lowest point P2 corresponds to the resistance component R1. Similarly, a value obtained by subtracting the real part of the lowest point P2 from the real part of the lowest point P3 corresponds to the resistance component R2.

  Further, the capacity components C1 and C2 can be calculated based on the searched vertices Q1 and Q2. Specifically, since the frequency fa1 of the vertex Q1 is equal to 1 / (2π · C1 · R1), the capacitance component C1 can be calculated based on the frequency fa1 and the resistance component R1 (C1 = 1 / (2π · R1). -Fa1)). Similarly, the capacitance component C2 can be calculated based on the frequency fa2 and the resistance component R2 of the vertex Q2 (C2 = 1 / (2π · R2 · fa2)). In the calculation of the capacitance components C1 and C2, the resistance components R1 and R2 calculated based on the real parts of the lowest points P1 to P3 are used. Note that a predetermined value may be used as the resistance components R1 and R2.

  Here, in the vector locus diagram on the lower frequency side than the region D2, the diffused resistor 16 has a combined impedance in which a parallel circuit of a plurality of resistance components and capacitance components overlaps. The diffused resistance is often calculated using the RC component calculation method 2 shown below because the lowest point and the apex of the arc cannot be distinguished.

  The RC component calculation method 2 will be described. A method for calculating the resistance component Rwi and the capacitance component Cwi of the diffused resistor 16 will be described below. However, i is the number of the parallel connection bodies of the resistance component and the capacitance component constituting the diffused resistor 16 and takes a range of i = 0, 1,. For example, i = 4. The diffusion resistance has a large time constant, and in order to calculate the diffusion resistance with high accuracy, it is necessary to use the detection values I and V detected over a long period of time according to the time constant. When the detection values I and V detected over a long period of time are used, the open-circuit voltage OCV change due to charging / discharging during the detection period and the temperature increase of the storage battery greatly affect the accuracy. Therefore, it is necessary to measure the inter-terminal voltage V and the charge / discharge current I in a situation where the battery state change is small.

  The internal impedance processing unit 51 of the present embodiment acquires the vehicle operation end point when the vehicle shifts from the operation state to the stop state, and the detection values I and V when the vehicle is stopped, and the internal impedance based on the detection value. The real part Re (Z) and the imaginary part Im (Z) are calculated. Then, using the real part Re (Z) and the imaginary part Im (Z) of the internal impedance, the RC component calculation unit 62 calculates the resistance component Rwi and the capacitance component Cwi of the diffused resistor 16.

  Specifically, as shown in FIG. 6, the voltage V and the current I in the charge / discharge stop state are acquired slightly before the storage battery 10 is switched from the charge / discharge state to the charge / discharge stop state. Here, the charge / discharge state of the storage battery 10 corresponds to the operation / stop state of the vehicle. That is, in the vehicle operating state, the storage battery 10 is in a charge / discharge state, and in the vehicle stop state, the storage battery 10 is in a charge / discharge stop state. When the vehicle is stopped, the voltage V between the terminals of the storage battery 10 changes slowly, and the change in the voltage V corresponds to a change caused by the diffusion resistor 16. Therefore, the diffused resistor 16 can be accurately calculated by using the detection values I and V during the period in which the voltage V changes slowly.

  When calculating the internal impedance, the current and voltage detection values I and V just before the charge / discharge stop state is switched are acquired so that the denominator current I is not zero. As for the open-circuit voltage change due to charging / discharging at this time, the voltage V is corrected as described above. Further, since the diffusion resistor 16 has a time constant of several tens of seconds or more, which is larger than the time constant of the DC resistor 13 and the reaction resistors 14 and 15, the detection value sampling period is calculated by the DC resistor 13 and the reaction resistors 14 and 15. It may be larger than sometimes.

  Next, a method for calculating the resistance component and the capacitance component of the diffused resistor will be described. Since the diffused resistor 16 is a composite impedance, the lowest point handled in the RC component calculation method 1 is far from the imaginary part, and the resistance component and the capacitance component may not be separated accurately. Therefore, the resistance component Rwi and the capacitance component Cwi of the diffused resistor 16 are compared with the combined impedance of the equivalent circuit model with reference to the real part and the imaginary part of the internal impedance Z for each frequency f calculated by the impedance calculation unit 61. Is calculated.

  FIG. 7 shows a flowchart of the calculation process of the resistance component Rwi and the capacitance component Cwi of the diffusion resistor 16. Further, as the internal impedance Z (f), the detection value I of the charging current and the distance between the terminals when the vehicle finishes running (when the storage battery changes from the charge / discharge state to the charge / discharge stop state) and stops (charge / discharge stop). The voltage calculated from the detected voltage value V by wavelet transform is used.

  First, in step S11, initial values of the resistance component Rwi and the capacitance component Cwi stored in the control device 50 are set. As the initial value, the previous value is applied after the first time in order to increase the calculation speed. In step S <b> 12, the internal impedance Z (f) at each frequency f corresponding to the measurement value is acquired from the impedance calculation unit 61.

  Next, in step S13, the estimated internal impedance Za (f) at each frequency f is calculated based on the initial values of the resistance component Rwi and the capacitance component Cwi. Next, in step S14, a deviation between the estimated internal impedance Za (f) calculated based on the initial value and the internal impedance Z (f) calculated by the impedance calculation unit 61 is calculated.

  Next, in step S15, it is determined whether or not the deviation calculated in step S14 is a predetermined value or less. You may make it select the thing with the smallest error. When the deviation calculated in step S14 is larger than the predetermined value (S15: NO), in step S16, the resistance component Rwi and the capacitance component Cwi are changed so that the deviation decreases. Then, in step S13, the estimated internal impedance Za (f) is calculated at each frequency f. If the deviation calculated in step S14 is equal to or smaller than the predetermined value (S15: YES), the values of the resistance component Rwi and the capacitance component Cwi are stored as current values in step S17, and the process is terminated.

  According to the above processing, based on the real and imaginary parts of the internal impedance Z (f) calculated by the impedance calculation unit 61 and the provisional values of the resistance component Rwi and the capacity component Cwi that constitute the equivalent circuit model in the storage battery 10. It is possible to compare the real part and the imaginary part of the calculated estimated internal impedance Za (f), and calculate the resistance component Rwi and the capacitance component Cwi based on the comparison result.

  FIG. 8 shows the internal impedance Z (f) calculated by wavelet transform using the measured values for the diffused resistor 16 that is the synthetic impedance in the high-capacity secondary battery for vehicles (lithium ion secondary battery) and the resistance to be used. The estimated internal impedance Za (f) obtained by combining the component and the capacitance component is shown. As the estimated internal impedance Za (f), the estimated internal impedance Za (2RC model) calculated by regarding two sets of resistance components and capacitance components and the diffused resistor 16 as four sets of resistance components and capacitance components were calculated. The estimated internal impedance Za (4RC model) is shown.

  As compared with the estimated internal impedance Za (2RC model), the estimated internal impedance Za (4RC model) can match the internal impedance Z calculated by the impedance calculating unit 61 with high accuracy. In this way, the configuration of the model of the diffused resistor 16 that is the combined impedance can be selected, and a plurality of resistance components and a plurality of capacitance components can be calculated.

  Next, the SOC processing unit 52 will be described. In the configuration shown in FIG. 2, the OCV calculation unit 63 of the SOC processing unit 52 receives the values of the resistance components Rs, R1, R2, and Rwi and the values of the capacitance components C1, C2, and Cwi from the RC component calculation unit 62. . Further, the current detection value I of the charge / discharge current is input from the current sensor 30 and the current detection value V of the inter-terminal voltage is input from the voltage sensor 31 to the OCV calculation unit 63. The OCV calculation unit 63 forms the internal impedance Z based on the values of the resistance components Rs, R1, R2, and Rwi and the values of the capacitance components C1, C2, and Cwi, detects the current charge / discharge current detection value I, and the current The open-circuit voltage OCV of the storage battery 10 is calculated based on the detected value V of the inter-terminal voltage.

  Specifically, in the equivalent circuit model shown in FIG. 4, the open-circuit voltage OCV of the storage battery 10 is based on the current detected value V of the inter-terminal voltage and the current detected value I of the charge / discharge current. It can be calculated as in 6). Note that s is a differential operator. m is the number of RC parallel circuits of the reaction resistors 14 and 15, for example, m = 2. n is the number of RC parallel circuits of the diffused resistors 16, and m = 4, for example, considering the reproducibility of the combined impedance.

検出値ＶとＩが離散値である場合は、式（６）に対し双一次変換などで離散化を行う。離散化後の開放端電圧ＯＣＶを算出する方法は周知であるため、ここでは詳細な説明を省略する。

When the detected values V and I are discrete values, the discretization is performed on the equation (6) by bilinear transformation or the like. Since the method of calculating the open end voltage OCV after discretization is well known, detailed description thereof is omitted here.

  The OCV-SOC conversion unit 64 of the SOC processing unit 52 stores the OCV-SOC characteristics shown in FIG. 3 as a map, and converts the open end voltage OCV calculated by the OCV calculation unit 63 into SOC. By using the SOC calculated by the SOC processing unit 52 in this way, it is possible to perform suitable charge / discharge control for the storage battery 10.

  FIG. 9 shows, as an effect of the present invention, the time change of the SOC calculated using the waveform when the vehicle travels in the high-capacity secondary battery for vehicles (lithium ion secondary battery). However, S1: True value of SOC, S2: SOC calculated by the SOC processing unit 52 of the present embodiment, S3: Only the DC resistance 13 and the reaction resistances 14 and 15 of the present embodiment (an equivalent circuit not including the diffusion resistance 16) SOC calculated using the model), S4: SOC obtained by applying the least square method to the current I and the voltage V and using the intercept as the open-circuit voltage OCV (SOC by IV measurement method).

  From time T1 to T2, the vehicle travels in the EV travel mode, and after time T2, the vehicle travels in the HV travel mode. In the EV travel mode at times T1 to T2, the SOC (S3) calculated using the equivalent circuit model not including the diffused resistor 16 and the SOC (S4) calculated using the IV measurement method are errors during EV travel. Is big. This is because when a high current typified by the EV traveling mode continuously flows in one of the charge and discharge, the diffusion phenomenon becomes remarkable and causes an apparent voltage drop. That is, when calculating the SOC with the voltage V as an input, the diffusion resistance 16 must be considered in the EV travel mode. Therefore, the equivalent circuit model and the IV measurement method not including the diffused resistor 16 are not suitable for the EV traveling mode.

  On the other hand, the SOC (S2) calculated by the SOC processing unit 52 of the present embodiment accurately follows the true value (S1) of the SOC in the EV traveling mode. Further, when the EV travel mode is switched to the HV travel mode, the SOC (S4) by the IV measurement method is canceled by the diffusion resistance 16, and returns to the true value (S1) of the SOC. That is, it shows that the impedance of the diffusion resistor 16 can be calculated from the voltage change after switching from the EV travel mode to the HV travel mode. Therefore, the component of the diffusion resistance 16 can be obtained even when the vehicle is running.

  Next, the deterioration degree processing unit 53 will be described. Transition of each resistance component and each capacitance component when categorized by DC resistance 13, reaction resistances 14 and 15 and diffusion resistance 16 in ascending order of time constant in a deterioration durability test on a consumer battery (lithium ion secondary battery) Will be described. The values of each resistance component and each capacitance component both increase due to deterioration compared to the values before the test. Moreover, the increase rate (deterioration degree) which each differs is shown. As described above, since the degree of deterioration of each resistance component and each capacity component is different, it is necessary to individually calculate each resistance component and each capacity component in order to appropriately estimate the degree of deterioration.

  In the configuration shown in FIG. 2, the degradation level processing unit 53 includes a reference value calculation unit 65 and a degradation level calculation unit 66. The reference value calculation unit 65 calculates values of the resistance components Rsini, R1ini, R2ini, Rwiini and the values C1ini, C2ini, Cwiini of the respective resistance components in the non-degraded state as the reference values of the RC components. These take the temperature T of the storage battery 10 and the charge amount SOC as arguments.

  The degradation component calculation unit 66 receives the current resistance components Rs, R1, R2, and Rwi of the storage battery 10 and the capacity components C1, C2, and Cwi from the RC component calculation unit 62, and also receives the reference resistance from the reference value calculation unit 65. A component and a reference capacitance component are input. The deterioration degree calculation unit 66 calculates the deterioration degree K of the storage battery 10 as the ratio between the resistance component and the reference resistance component and the ratio between the capacity component and the reference capacity component.

  Next, the predicted power processing unit 54 will be described. The predicted voltage calculation unit 68 uses the current resistance component R and capacitance component C, the reference current Iu, the open-circuit voltage OCV, and the applied voltage ΔVc of the parallel circuit (internal impedance) of the resistance component and capacitance component shown in Equation (6). Thus, the predicted voltage V (t + t1) after 1 second is estimated from the equation (6). The reference current Iu corresponding to the predicted current is, for example, the maximum allowable current Imax. The applied voltage ΔVc of the parallel circuit is a voltage change that changes in accordance with the current charge amount of the capacitance component, and corresponds to the third and fourth terms on the right side of Equation (6). For example, the current value of the OCV calculation unit 63 of the SOC processing unit 52 can be used for calculation. Predicted power P (t + t1) after t1 seconds = predicted voltage V (t + t1) after t1 seconds × reference current Iu. A suitable control method is determined by comparing the predicted power P (t + t1) after t1 seconds with the discharge allowable power Wout or the charge allowable power Win of the storage battery 10.

  Hereinafter, effects in the present embodiment will be described.

  The real part Re (Z (f)) and imaginary part Im (Z (f)) of the internal impedance Z for each frequency of the storage battery 10 can be calculated using wavelet transform. Based on the real part Re (Z (f)) and imaginary part Im (Z (f)) of the calculated internal impedance Z for each frequency, a plurality of resistance components corresponding to a plurality of time constants and a plurality of capacitances are obtained. The component can be calculated.

  The wavelet transform can calculate the internal impedance Z for each frequency even if the object to be transformed is an aperiodic waveform. Thereby, the internal impedance Z can be calculated from the waveform of the storage battery mounted on the vehicle. In particular, the internal impedance Z can be calculated in a complicated charging / discharging waveform under a vehicle environment, a point where switching from charging / discharging state to charging / discharging stop and a subsequent slow voltage change waveform during charging / discharging stopping, and a rectangular waveform. . Of course, the internal impedance Z can also be calculated with a sine wave.

  The internal impedance Z can be calculated from the resistance component and capacity component of the DC resistor 13, reaction resistors 14, 15 and diffused resistor 16, so that, for example, the SOC of the storage battery, the predicted power, and the degree of deterioration of the storage battery 10 can be accurately calculated. it can. Identification of individual deterioration modes occurring inside the storage battery can be performed.

  Further, by obtaining from the real part Re (Z (f)) and the imaginary part Im (Z (f)) of the internal impedance, each resistance component and each capacitance component can be calculated based on the vector locus diagram.

  By correcting the detected value V of the inter-terminal voltage with the change amount ΔOCV of the open-end voltage OCV in order to suppress the influence of the change in the open-end voltage OCV accompanying the charge / discharge of the storage battery 10 on the calculation of the internal impedance Z, The internal impedance Z can be calculated with high accuracy.

(Other embodiments)
In the above embodiment, the internal impedance Z is modeled as a series connection body of the DC resistance 13, the reaction resistances 14 and 15, and the diffusion resistance 16, but this may be changed. For example, if the characteristics of the storage battery and the surrounding environment are restricted and the reaction resistances 14 and 15 and the diffusion resistance 16 can be suppressed, any of them may be omitted.

  In the above embodiment, the diffusion resistance is calculated from the point of switching from the charge / discharge state to the charge / discharge stop and the waveform of the slow voltage change during the subsequent charge / discharge stop, but this may be changed. For example, the diffusion resistance can be calculated from the point of switching from charging / discharging with a large average current to charging / discharging with a small average current and the waveform of the slow voltage change during charging / discharging with a small average current, during charging / discharging. Good. Specifically, the diffusion resistance can be calculated when the voltage drop recovers at the point where the EV traveling is switched to the HV traveling and the subsequent HV traveling. Thereby, since the voltage change is remarkable during traveling, the diffusion resistance can be accurately calculated during traveling. Further, for example, the diffusion resistance may be calculated from the point at which the charge stop state is switched to the charge / discharge state and the waveform of the slow voltage change during the subsequent charge / discharge. Specifically, the secondary battery may be rapidly charged. In the above-described waveform including the above embodiment, at least one of direct current resistance, reaction resistance, and diffusion resistance may be calculated.

  As shown in FIG. 9, in a high capacity secondary battery for vehicles (lithium ion secondary battery), SOC (S2) calculated using an equivalent circuit model including a diffused resistor 16 and an equivalent circuit not including the diffused resistor 16 The difference from the SOC (S3) calculated using the model continues to increase from time T1 and becomes the largest at time T2. Then, after time T2, the difference between S2 and S3 disappears. This is considered to be because the polarization generated in the diffusion resistor 16 increases as a large current continues to be output from the storage battery 10 as the EV travels, and then the polarization is eliminated by switching to the HV travel. That is, the influence of the diffusion resistor 16 on the voltage V becomes significant between the time when the EV running is switched to the HV running (time T2) and the time after which the polarization generated in the diffusion resistor 16 is eliminated. Therefore, the internal impedance is calculated based on the current I and the voltage V detected during the time point (time T2) when the EV travel is switched to the HV travel and until the polarization generated in the diffusion resistor 16 is eliminated. As a result, the diffusion resistance 16 can be calculated with high accuracy.

  -In the said embodiment, although the vehicle in which the storage battery 10 is mounted showed the hybrid vehicle, you may change this. For example, the present invention may be applied to any vehicle that requires a power supply system including a storage battery. Similarly, you may apply not only to a vehicle but to any system which requires a power supply system including a storage battery.

  In the above embodiment, a lithium ion secondary battery is used as the storage battery 10, but this may be changed. For example, a nickel metal hydride storage battery may be used.

  In the above embodiment, the change amount ΔOCV of the open circuit voltage OCV due to the change in the charge amount SOC accompanying charge / discharge is calculated, and the detected value V of the terminal voltage is corrected using the change amount ΔOCV. This may be changed. As another correction method, a method for correcting the internal impedance Z will be described below. The relationship between the charge amount SOC and the open-circuit voltage OCV shown in FIG. 3 is converted in advance as an equivalent capacitance component Cocv of internal impedance and stored in the control device 50. The equivalent capacity component Cocv can be calculated, for example, as Cocv = Af · ΔSOC / ΔOCV (where Af is the full charge capacity of the storage battery 10). When calculating the internal impedance Z, Cocv may be subtracted (−1 / sCocv) as a correction term with respect to the internal impedance Z calculated in a state in which the open-circuit voltage variation ΔOCV is included (where s is a differential) operator). Thereby, the influence of the open end voltage change by the charge amount change which is a component different from the internal impedance Z of the storage battery 10 can be removed. In addition to this, when calculating the SOC using the equivalent circuit model, the equivalent capacitance component Cocv may be subtracted (−1 / sCocv) as a correction term from the equivalent circuit model expression of Expression (6). The relationship between the charge amount SOC and the open-circuit voltage OCV shown in FIG. 3 is almost unaffected by the deterioration of the storage battery and the temperature change of the storage battery, and can be handled as having no characteristic change. Therefore, the equivalent capacity component Cocv is converted into a constant in advance. And stored in the control device 50. The equivalent capacitance component Cocv may be stored in the control device 50 as a constant not depending on the SOC, or may be stored in the control device 50 as a map using the SOC as a parameter. It can be said that the method of calculating the equivalent capacity component Cocv from the map using the SOC as a parameter corrects the imaginary part of the internal impedance Z based on the charge amount SOC. The internal impedance correction method using these equivalent capacitance components Cocv requires a relatively small amount of storage and calculation by the control device 50.

  In the above embodiment, the RC component calculation unit 62 uses the calculation method of the resistance component and the capacitance component based on the vector locus diagram. However, this may be changed. Instead of the vector locus diagram, a calculation method of a resistance component and a capacitance component by a Bode diagram may be used. A calculation method using a Bode diagram will be described. Determine the frequency characteristics for the absolute values of the real and imaginary parts of the internal impedance. The shape of the frequency characteristic is stepped. The distance of each step corresponds to each resistance component of the DC resistance 13 and the reaction resistances 14 and 15. Further, each capacitance component can be calculated from the frequency at the break point of the step (break point frequency) and each resistance component obtained previously. As the RC component calculation method, both the method using the vector locus diagram and the method using the Bode diagram require a real part and an imaginary part of the internal impedance for each frequency.

  The method in the RC component calculation unit 62 may be changed for the DC resistance 13, the reaction resistances 14 and 15, and the diffusion resistance 16. For example, the application target of the RC component calculation method 1 and the RC component calculation method 2 may be changed. Specifically, the total resistance component and the total capacitance component may be calculated using the RC component calculation method 1. In addition, the total resistance component and the total capacitance component may be calculated using the RC component calculation method 2.

  -Although the internal battery 10 of the said embodiment applied the storage battery 10 in the state mounted in the vehicle, you may change this. For example, as shown in FIG. 10, the internal state estimation device may be applied by changing to a state in which the storage battery 10 is connected to the battery characteristic measurement device 23. Even when the storage battery 10 is connected to the battery characteristic measurement device 23, a current or voltage waveform simulating non-periodic charge / discharge of the storage battery 10 in the vehicle is charged / discharged from the battery characteristic measurement device to the storage battery. Good. In this case, the internal state estimation device may be applied to a control device included in the battery characteristic measurement device 23 instead of the control device 50. Thereby, the state in which the storage battery 10 is mounted on the vehicle can be simulated, and the transition of the internal impedance during traveling can be easily measured.

  -It is good also as a structure which abbreviate | omits at least any one of the SOC process part 52, the deterioration degree process part 53, and the estimated electric power process part 54. FIG. The voltage correction unit 56 may be omitted.

  DESCRIPTION OF SYMBOLS 10 ... Storage battery (secondary battery), 30 ... Current sensor, 31 ... Voltage sensor, 54 ... Memory | storage part (memory | storage means), 60 ... Wavelet transformation part (wavelet transformation means), 61 ... Impedance calculation part (internal impedance calculation means) .

Claims (20)

Storage means for storing the detection values by the voltage detection means (31) for detecting the voltage between the terminals of the secondary battery (10) and the current detection means (30) for detecting the charge / discharge current flowing in the secondary battery in time series ( 5 5 )
A voltage wavelet transform coefficient is calculated by performing wavelet transform on the detected value of the inter-terminal voltage stored in the time series, and wavelet transform is performed on the detected charge / discharge current value stored in the time series. Wavelet transform means (60) for calculating a current wavelet transform coefficient by performing,
An internal impedance calculating means (61) for calculating a real part and an imaginary part of the internal impedance of the secondary battery for each frequency according to a ratio of the voltage wavelet transform coefficient and the current wavelet transform coefficient;
Equipped with a,
The secondary battery is mounted as a power source of a driving motor in a vehicle,
The secondary battery is supplied with a current or voltage imitating non-periodic charge / discharge in the secondary battery in the operation state and stop state of the vehicle from a battery characteristic measurement device,
The wavelet transform unit performs wavelet transform on the detection values detected by the voltage detection unit and the current detection unit when the non-periodic charge / discharge is performed by the battery characteristic measurement device. A battery internal state estimating device. Storage means for storing the detection values by the voltage detection means (31) for detecting the voltage between the terminals of the secondary battery (10) and the current detection means (30) for detecting the charge / discharge current flowing in the secondary battery in time series ( 55) and
A voltage wavelet transform coefficient is calculated by performing wavelet transform on the detected value of the inter-terminal voltage stored in the time series, and wavelet transform is performed on the detected charge / discharge current value stored in the time series. Wavelet transform means (60) for calculating a current wavelet transform coefficient by performing,
An internal impedance calculating means (61) for calculating a real part and an imaginary part of the internal impedance of the secondary battery for each frequency according to a ratio of the voltage wavelet transform coefficient and the current wavelet transform coefficient;
With
The secondary battery is mounted as a power source of a driving motor in a vehicle,
The vehicle uses the traveling motor and the internal combustion engine as driving power, and motor traveling using only the traveling motor as driving force, and hybrid traveling using both the traveling motor and internal combustion engine as driving force. Done
The wavelet transform unit detects the detected value detected by the voltage detection unit and the current detection unit during the hybrid travel after the switching from the motor travel to the hybrid travel in the vehicle. The voltage wavelet transform coefficient and the current wavelet transform coefficient are calculated by performing wavelet transform on the
The internal impedance calculation unit calculates a real part and an imaginary part of the internal impedance of the secondary battery based on a ratio between the voltage wavelet transform coefficient and the current wavelet transform coefficient . Storage means for storing the detection values by the voltage detection means (31) for detecting the voltage between the terminals of the secondary battery (10) and the current detection means (30) for detecting the charge / discharge current flowing in the secondary battery in time series ( 55) and
A voltage wavelet transform coefficient is calculated by performing wavelet transform on the detected value of the inter-terminal voltage stored in the time series, and wavelet transform is performed on the detected charge / discharge current value stored in the time series. Wavelet transform means (60) for calculating a current wavelet transform coefficient by performing,
An internal impedance calculating means (61) for calculating a real part and an imaginary part of the internal impedance of the secondary battery for each frequency according to a ratio of the voltage wavelet transform coefficient and the current wavelet transform coefficient;
With
A battery internal state estimation device comprising: a voltage correction means (56) for correcting a detected value of the voltage between the terminals by the voltage detection means based on a change in charge amount due to charging / discharging of the secondary battery . The voltage correction means (56) which correct | amends the detected value of the voltage between terminals by the said voltage detection means based on the charge amount change by charging / discharging of the said secondary battery is provided, The Claim 1 or 2 characterized by the above-mentioned. Battery internal state estimation device. Storage means for storing the detection values by the voltage detection means (31) for detecting the voltage between the terminals of the secondary battery (10) and the current detection means (30) for detecting the charge / discharge current flowing in the secondary battery in time series ( 55) and
A voltage wavelet transform coefficient is calculated by performing wavelet transform on the detected value of the inter-terminal voltage stored in the time series, and wavelet transform is performed on the detected charge / discharge current value stored in the time series. Wavelet transform means (60) for calculating a current wavelet transform coefficient by performing,
An internal impedance calculating means (61) for calculating a real part and an imaginary part of the internal impedance of the secondary battery for each frequency according to a ratio of the voltage wavelet transform coefficient and the current wavelet transform coefficient;
With
The internal impedance calculation means, when calculating the internal impedance, the imaginary part of the internal impedance calculated by the ratio of the voltage wavelet transform coefficient and the current wavelet transform coefficient, the charge amount due to charging and discharging of the secondary battery A battery internal state estimation device comprising correction means for correcting based on a change. The internal impedance calculation means, when calculating the internal impedance, the imaginary part of the internal impedance calculated by the ratio of the voltage wavelet transform coefficient and the current wavelet transform coefficient, the charge amount due to charging and discharging of the secondary battery battery internal state estimating device according to claim 1 or 2, characterized in that it comprises a correcting means for correcting, based on the change. Storage means for storing the detection values by the voltage detection means (31) for detecting the voltage between the terminals of the secondary battery (10) and the current detection means (30) for detecting the charge / discharge current flowing in the secondary battery in time series ( 55) and
A voltage wavelet transform coefficient is calculated by performing wavelet transform on the detected value of the inter-terminal voltage stored in the time series, and wavelet transform is performed on the detected charge / discharge current value stored in the time series. Wavelet transform means (60) for calculating a current wavelet transform coefficient by performing,
An internal impedance calculating means (61) for calculating a real part and an imaginary part of the internal impedance of the secondary battery for each frequency according to a ratio of the voltage wavelet transform coefficient and the current wavelet transform coefficient;
With
The internal impedance calculation means, when calculating the internal impedance, the imaginary part of the internal impedance calculated by the ratio of the voltage wavelet transform coefficient and the current wavelet transform coefficient, the charge amount due to charging and discharging of the secondary battery A battery internal state estimation device comprising correction means for correcting based on an equivalent capacity component representing a change in open circuit voltage of the secondary battery accompanying a change. The internal impedance calculation means, when calculating the internal impedance, the imaginary part of the internal impedance calculated by the ratio of the voltage wavelet transform coefficient and the current wavelet transform coefficient, the charge amount due to charging and discharging of the secondary battery battery internal state estimating device according to claim 1 or 2, characterized in that it comprises a correcting means for correcting, based on the equivalent capacitive component representing a change in the open circuit voltage of the secondary battery due to the change. The secondary battery is mounted as a power source of a driving motor in a vehicle, and charging / discharging is performed on the secondary battery according to a driving state of the vehicle,
The wavelet transform unit performs the wavelet transform on the detection value detected by the voltage detection unit and the current detection unit when performing non-periodic charge / discharge on the secondary battery. Calculating a wavelet transform coefficient and the current wavelet transform coefficient;
9. The internal impedance calculation unit calculates a real part and an imaginary part of the internal impedance of the secondary battery based on a ratio between the voltage wavelet transform coefficient and the current wavelet transform coefficient . The battery internal state estimation apparatus according to any one of the preceding claims. The secondary battery is mounted as a power source of a driving motor in a vehicle,
The wavelet transform unit performs the wavelet transform on the detection values detected by the voltage detection unit and the current detection unit during the charge / discharge stop time and the charge / discharge stop of the secondary battery, thereby performing the voltage wavelet transform. Calculating the coefficient and the current wavelet transform coefficient;
10. The internal impedance calculation unit calculates a real part and an imaginary part of the internal impedance of the secondary battery based on a ratio between the voltage wavelet transform coefficient and the current wavelet transform coefficient . The battery internal state estimation apparatus according to any one of the preceding claims. Component calculation means (62) for calculating at least one of a plurality of resistance components and a plurality of capacitance components constituting the internal impedance based on the real part and the imaginary part of the internal impedance calculated by the internal impedance calculation means The battery internal state estimation device according to any one of claims 1 to 10 , characterized by comprising: The internal impedance comprises a direct current resistance representing a conduction resistance, a reaction resistance representing a reaction at an electrode interface, and a diffusion resistance representing ion diffusion,
The DC resistance, the reaction resistance, and the diffusion resistance have different time constants, respectively.
The component calculating means is based on a real part and an imaginary part of the internal impedance calculated by the internal impedance calculating means, and includes at least a resistance component and a capacitance component that respectively constitute the DC resistance, the reaction resistance, and the diffusion resistance. The battery internal state estimation device according to claim 11 , wherein either one is calculated. The component calculating unit is configured to calculate the internal impedance at a minimum point where the imaginary part of the internal impedance is minimum in the vector locus diagram obtained from the real part and imaginary part of the internal impedance calculated by the internal impedance calculating unit. The battery internal state estimation device according to claim 11 , wherein the resistance component is calculated based on a unit. The component calculation means is based on a frequency corresponding to a maximum point at which the imaginary part of the internal impedance is maximum in the vector locus diagram obtained from the real part and imaginary part of the internal impedance calculated by the internal impedance calculation means. The battery internal state estimation device according to claim 11 , wherein the capacity component is calculated. The component calculation means includes a real part and an imaginary part of the internal impedance calculated by the internal impedance calculation means, and a real part and an imaginary part of the estimated internal impedance calculated from predetermined temporary values of the resistance component and the capacitance component. comparing the part, based on the comparison result, either one of claims 11 to 14 and calculates at least one of the plurality of resistance components and a plurality of capacitive components constituting the internal impedance The battery internal state estimation apparatus as described in. The component calculation means calculates a composite value of the real part and imaginary part of the internal impedance calculated by the internal impedance calculation means, and based on the break frequency calculated based on the composite value, the internal impedance 16. The battery internal state estimation device according to claim 11 , wherein at least one of a plurality of resistance components and a plurality of capacity components constituting the battery is calculated. The component calculation means calculates the resistance component and the capacitance component,
An open-end voltage calculation means for calculating an open-end voltage of the secondary battery based on the detection value by the voltage detection means and the current detection means and the resistance component and the capacity component calculated by the component calculation means. The battery internal state estimation device according to any one of claims 11 to 16 , further comprising (63). Based on the open circuit voltage of the secondary battery calculated by the open circuit voltage calculating means, to claim 17, characterized in that it comprises a charge amount calculating means (52) for calculating the amount of charge of the secondary battery The battery internal state estimation apparatus as described. Claim 11, characterized in that said component calculation means calculated by the based on at least one of the resistance component and the capacitive component includes a deterioration degree calculating means for calculating a degree of deterioration of the secondary battery (53) The battery internal state estimation device according to any one of 1 to 18 . The component calculation means calculates the resistance component and the capacitance component,
The resistance component and the capacitance component calculated by the component calculation means, the current value of the voltage generated by the charge accumulation in the capacitance component, the current value of the open-circuit voltage of the secondary battery, and a predetermined future time A predicted voltage that calculates a predicted voltage that is a predicted value of the inter-terminal voltage of the secondary battery at a predetermined future time based on a predicted current that is a predicted value of a charge / discharge current flowing in the secondary battery Means (68);
Based on the predicted current and the predicted voltage, claim 11, characterized in that it comprises a predicted power calculating means (54) for calculating the predicted power is the predicted value of the charge-discharge electric power in the secondary battery in the predetermined time The battery internal state estimation device according to any one of 1 to 19 .

Images