JP5851514B2 - Battery control device, secondary battery system - Google Patents

Description

  The present invention relates to a technique for estimating the state of charge of a secondary battery.

  In a power supply device using a power storage means such as a battery, a distributed power storage device, and an electric vehicle, a battery control circuit that manages the state of the power storage means is mounted. The state of the power storage means managed by the battery control circuit includes a state of charge (SOC) indicating how much the power storage means is charged, or how much charge can be discharged, or a battery. A representative example is a deterioration state (State of Health: SOH) used for determining the lifetime of the battery and realizing control according to the deterioration.

  In the following Patent Document 1, a technique for calculating the remaining battery capacity is described. In the technique described in this document, when it is determined that the battery has not recovered to the original electromotive force, the capacity value stored in the battery capacity storage means is used as the initial value of the battery capacity. Calculate the remaining capacity. When it is determined that the original electromotive force of the battery has been recovered, an initial value of the battery capacity is obtained from the terminal voltage of the battery, and the remaining capacity of the battery being discharged is calculated based on the initial value.

Japanese Patent No. 2658159

  In the technique described in Patent Document 1, the remaining capacity of the power storage means can be accurately calculated using the terminal voltage of the power storage means. However, the calculation of the remaining capacity using the terminal voltage is complicated during charging / discharging, so the remaining capacity of the power storage means is obtained by integrating the current flowing into and out of the power storage means during charging / discharging. . Here, if the state of charge is estimated by integrating the current, the estimation error gradually increases as the distance from the estimation result of the remaining capacity estimated using the terminal voltage increases.

  The present invention has been made to solve the above-described problems, and an object of the present invention is to improve the estimation accuracy when estimating the state of charge by integrating the current input to and output from the secondary battery.

  The battery control device according to the present invention estimates the charge state by integrating the current flowing in the secondary battery with the charge state estimated in the past as an initial value, and in parallel with this, the charge estimated using the terminal voltage The state of charge is separately estimated by integrating the current flowing through the secondary battery with the state as an initial value. The battery control device estimates the state of charge after estimating the state of charge using the terminal voltage by weighting and adding these estimation results.

  According to the battery control device of the present invention, it is possible to accurately estimate the state of charge by performing a plurality of estimations of the state of charge in parallel and weighting and adding them.

It is a figure which shows the structure of the battery system 100 which concerns on Embodiment 1, and its periphery. It is a figure which shows the circuit structure of the cell control part 121. FIG. It is a figure which shows the example of the SOC table 181 which the 1st memory | storage part 180 has stored. It is a figure which shows the outline | summary of the SOC error which arises in Formula 1. It is the figure which showed the change of the terminal voltage of the cell 111 according to time as an example of the operation | movement pattern of rest / discharge / pause. It is a figure explaining the detail of the SOC error resulting from QmaxErr. It is a block diagram in the SOC calculation which the assembled battery control part 150 performs. It is a figure which shows the outline | summary of the process which the block diagram of FIG. 7 implements. It is a figure which shows the procedure which estimates SOC after the assembled battery control part 150 restarts. It is a block diagram of the SOC calculation which the assembled battery control part 150 in Embodiment 2 performs. It is a figure which shows transition of the SOC change in Embodiment 2. FIG.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. In the following embodiments, a case where the present invention is applied to a battery system constituting a power source of a plug-in hybrid vehicle (PHEV) will be described as an example.

  In the following embodiments, a case where a lithium ion battery is employed will be described as an example. However, a nickel metal hydride battery, a lead battery, an electric double layer capacitor, a hybrid capacitor, and the like can also be used. In the following embodiments, the assembled batteries are configured by connecting the cells in series. However, the assembled batteries may be configured by connecting the cells connected in parallel, or by connecting the cells connected in series. A battery pack may be configured by connecting batteries in parallel.

<Embodiment 1>
FIG. 1 is a diagram showing a configuration of a battery system 100 according to Embodiment 1 of the present invention and its surroundings. Battery system 100 is connected to inverter 400 via relays 300 and 310, and connected to charger 420 via relays 320 and 330. The battery system 100 includes an assembled battery 110, a single battery management unit 120, a current detection unit 130, a voltage detection unit 140, an assembled battery control unit 150, a first storage unit 180, and a second storage unit 190.

  The assembled battery 110 includes a plurality of single cells 111. The unit cell management unit 120 monitors the state of the unit cell 111. The current detection unit 130 detects a current flowing through the battery system 100. The voltage detection unit 140 detects the total voltage of the assembled battery 110. The assembled battery control unit 150 detects the state of the assembled battery 110 and also manages the state.

  The assembled battery control unit 150 includes the battery voltage and temperature of the single cell 111 transmitted by the single cell management unit 120, the current value flowing through the battery system 100 transmitted by the current detection unit 130, and the Receives the total voltage value. The assembled battery control unit 150 detects the state of the assembled battery 110 based on the received information. The result of the state detection by the assembled battery control unit 150 is transmitted to the single cell management unit 120 and the vehicle control unit 200.

  The assembled battery 110 is configured by electrically connecting a plurality of unit cells 111 capable of storing and releasing electrical energy (charging and discharging DC power) in series. The unit cells 111 constituting the assembled battery 110 are grouped into a predetermined number of units when performing state management / control. The grouped unit cells 111 are electrically connected in series to form unit cell groups 112a and 112b. The number of the single cells 111 constituting the single cell group 112 may be the same in all the single cell groups 112, or the number of the single cells 111 may be different for each single cell group 112.

  The unit cell management unit 120 monitors the state of the unit cells 111 constituting the assembled battery 110. The unit cell management unit 120 includes a unit cell control unit 121 provided for each unit cell group 112. In FIG. 1, cell control units 121 a and 121 b are provided corresponding to the cell groups 112 a and 112 b. The unit cell control unit 121 monitors and controls the state of the unit cells 111 constituting the unit cell group 112.

  In the first embodiment, in order to simplify the description, four unit cells 111 are electrically connected in series to form unit cell groups 112a and 112b, and the unit cell groups 112a and 112b are further electrically connected in series. An assembled battery 110 including a total of eight unit cells 111 was connected.

  The assembled battery control unit 150 and the single cell management unit 120 transmit and receive signals via an insulating element 170 typified by a photocoupler and a signal communication unit 160.

  A communication unit between the assembled battery control unit 150 and the unit cell control units 121a and 121b constituting the unit cell management unit 120 will be described. The cell control units 121a and 121b are connected in series according to the descending order of potentials of the cell groups 112a and 112b monitored by each. A signal transmitted from the assembled battery control unit 150 to the unit cell management unit 120 is input to the unit cell control unit 121 a via the insulating element 170 and the signal communication unit 160. The output of the unit cell control unit 121a is input to the unit cell control unit 121b via the signal communication unit 160, and the output of the lowest unit cell control unit 121b is supplied to the assembled battery control unit via the insulating element 170 and the signal communication unit 160. 150. In the first embodiment, the insulating element 170 is not interposed between the unit cell control unit 121a and the unit cell control unit 121b, but signals can be transmitted and received through the insulating element 170.

  The first storage unit 180 includes an internal resistance characteristic of the assembled battery 110, the single battery 111, and the single battery group 112, a fully charged capacity, a polarization voltage, a deterioration characteristic, individual difference information, an SOC and an open circuit voltage (OCV). (Voltage) correspondence information and the like are stored. Furthermore, characteristic information such as the single cell management unit 120, the single cell control unit 121, and the assembled battery control unit 150 can be stored in advance. Even when the operations of the battery system 100, the assembled battery control unit 150, and the like are stopped, various information stored in the first storage unit 180 is retained.

  The second storage unit 190 stores various parameters including processing results after the assembled battery control unit 150 performs processing. Similar to the first storage unit 180, the stored parameters are retained even when the battery system 100, the assembled battery control unit 150, etc. stop operating. The various parameters stored in the second storage unit 190 are read out when the assembled battery control unit 150 performs processing next time, and are used as input parameters, as well as storing abnormal information, usage history, and the like of the battery system 100. If it is, the battery system 100 is read and used during maintenance.

  The assembled battery control unit 150 includes information received from the unit cell management unit 120, the current detection unit 130, the voltage detection unit 140, the vehicle control unit 200, information stored in the first storage unit 180 such as an SOC table 181 described later, and the first information. 2 Using the information stored in the storage unit 190, the SOC, SOH, chargeable / dischargeable current and power, abnormal state, calculation for controlling the charge / discharge amount, and the like of one or more single cells 111 are executed. . And based on a calculation result, information is output to the cell management part 120 and the vehicle control part 200. FIG. As described above, the processing result of the assembled battery control unit 150 is stored in the second storage unit 190 as necessary, and the previous processing stored in the second storage unit 190 when the assembled battery control unit 150 executes the calculation next time. The result can be used as input. The “state estimation unit” in the present invention corresponds to the assembled battery control unit 150.

  The vehicle control unit 200 controls the inverter 400 connected to the battery system 100 via the relays 300 and 310 using information transmitted from the assembled battery control unit 150. Moreover, the battery charger 420 connected to the battery system 100 via the relays 320 and 330 is controlled. During traveling of the vehicle, the battery system 100 is connected to the inverter 400 and drives the motor generator 410 using the energy stored in the assembled battery 110. At the time of charging, the battery system 100 is connected to the charger 420 and is charged by supplying power from a household power supply or a desk lamp.

  The charger 420 is used when charging the assembled battery 110 using an external power source typified by a home or a desk lamp. In the first embodiment, the charger 420 is configured to control a charging voltage, a charging current, and the like based on a command from the vehicle control unit 200, but the control may be performed based on a command from the assembled battery control unit 150. . The charger 420 may be installed inside the vehicle according to the configuration of the vehicle, the performance of the charger 420, the purpose of use, the installation conditions of the external power source, and the like, or may be installed outside the vehicle.

  When the vehicle system on which the battery system 100 is mounted starts and runs, the battery system 100 is connected to the inverter 400 under the control of the vehicle control unit 200, and the motor uses the energy stored in the assembled battery 110. Generator 410 is driven, and assembled battery 110 is charged by the power generated by motor generator 410 during regeneration. When a vehicle including the battery system 100 is connected to an external power source represented by a household or desk lamp, the battery system 100 and the charger 420 are connected based on information transmitted by the vehicle control unit 200, and the set The battery 110 is charged until a predetermined condition is met. The energy stored in the assembled battery 110 by charging is used when the vehicle is driven next time, or is used to operate electrical components inside and outside the vehicle. Further, if necessary, it may be discharged to an external power source represented by a household power source.

  FIG. 2 is a diagram illustrating a circuit configuration of the unit cell control unit 121. The cell control unit 121 includes a voltage detection circuit 122, a control circuit 123, a signal input / output circuit 124, and a temperature detection unit 125. The voltage detection circuit 122 measures the voltage between the terminals of each unit cell 111. The control circuit 123 receives measurement results from the voltage detection circuit 122 and the temperature detection unit 125, and transmits the measurement results to the assembled battery control unit 150 via the signal input / output circuit 124. In addition, it is determined that the circuit configuration that is generally implemented in the unit cell control unit 121 and that equalizes the voltage and SOC variation between the unit cells 111 generated due to self-discharge and variation in consumption current is known. The description is omitted.

  The temperature detection unit 125 included in the unit cell control unit 121 in FIG. 2 has a function of measuring the temperature of the unit cell group 112. The temperature detection unit 125 measures one temperature as the entire cell group 112 and treats the temperature as a temperature representative value of the cell 111 constituting the cell group 112. The temperature measured by the temperature detection unit 125 is used for various calculations for detecting the state of the cell 111, the cell group 112, or the assembled battery 110. Since FIG. 2 is based on this assumption, the single battery control unit 121 is provided with one temperature detection unit 125. A temperature detection unit 125 may be provided for each single cell 111 to measure the temperature for each single cell 111, and various calculations may be performed based on the temperature for each single cell 111. In this case, the number of temperature detection units 125 Therefore, the configuration of the cell control unit 121 and its surroundings becomes complicated.

  In FIG. 2, the temperature detector 125 is simply shown. In practice, a temperature sensor is installed on the temperature measurement target, and the installed temperature sensor outputs temperature information as a voltage, and the measurement result is transmitted to the signal input / output circuit 124 via the control circuit 123. Outputs the measurement result outside the unit cell control unit 121. A function for realizing this series of flows is implemented as a temperature detection unit 125 in the single cell control unit 121, and the voltage detection circuit 122 can be used for measuring temperature information (voltage).

  The configuration of the battery system 100 has been described above. Below, the calculation method in which the assembled battery control part 150 estimates SOC is demonstrated.

  FIG. 3 is a diagram illustrating an example of the SOC table 181 stored in the first storage unit 180. The SOC table 181 is a data table describing a correspondence relationship between the OCV of the single battery 111 and the SOC of the single battery 111. The data format may be arbitrary, but here, for convenience of explanation, an example of data is shown in a graph format. Although the data table is used in the first embodiment, the correspondence relationship between the OCV and the SOC can be expressed by using mathematical formulas. It is characteristic information indicating the correspondence between OCV and SOC, and any means can be used as long as it can convert from OCV to SOC or from SOC to OCV.

  The OCV is a voltage when the unit cell 111 is not loaded. Before the relays 300, 310, 320, 330 are closed, or when the relays 300, 310, 320, 330 are closed but charging / discharging of the assembled battery 110 is not started, etc. It can be determined that the voltage between the terminals is OCV. Furthermore, when the assembled battery 110 is charged or discharged, but the current value is weak, it can be regarded as OCV.

As one of the methods for calculating the SOC, a voltage OCV at the time of no load of the unit cell 111 provided in the battery system 100 is measured and converted to SOC by referring to the data table of FIG. 3 (SOCv). The SOC change from the SOCv is obtained by integrating (∫I (t) dt) the current flowing in and out and dividing by the full charge capacity (Qmax) of the cell 111 (ΔSOC (t)), and the initial SOC (SOCv) And a method of obtaining SOC based on SOC change (ΔSOC (t)) is known (SOCi).

FIG. 4 is a diagram showing an outline of the SOC error generated in Equation 1. The SOCvErr (SOCv error) in FIG. 4A is generated by converting to SOC including the voltage measurement error VErr that occurs when measuring the voltage of the unit cell 111 as shown in Equation 2. However, since SOCv is calculated only under conditions that can be regarded as no load or no load (ΔSOC (t) is calculated while fixing the value of SOCv during charging / discharging), after obtaining SOCv including SOCvErr once After the voltage of the unit cell 111 can no longer be regarded as OCV, SOCvErr becomes a constant value with respect to time.

ΔSOCErr (ΔSOC error) in FIG. 4B is an error included in ΔSOC (t) calculated when charge / discharge is performed after SOCv is determined. As shown in Equation 3, ΔSOCErr is caused by two error factors: a current measurement error IErr that occurs during current measurement and a calculation error QmaxErr of the battery full charge capacity Qmax.

  IErr is the sum of offset error and gain error, and is a value corresponding to the current measurement accuracy of the current detector 130. In the first embodiment, it is assumed that the current detection unit 130 having an accuracy suitable for satisfying Expression 1 is used, and calibration is performed as necessary to reduce the error IErr. Therefore, in the following description, the influence of IErr is ignored.

  QmaxErr is a calculation error of the full charge capacity Qmax of the single battery 111. The arithmetic expression will be described later. The full charge capacity of the unit cell 111 decreases with deterioration. If Equation 1 is applied without grasping this small full charge capacity, a large SOC error occurs. Therefore, Qmax is calculated as follows.

  FIG. 5 is a diagram showing a change in the terminal voltage of the unit cell 111 according to time, taking an operation pattern of pause / discharge / pause as an example. Since the unit cell 111 has an internal resistance, an IR drop occurs at the moment when discharge is started. During discharge, the voltage gradually decreases as the SOC decreases, and immediately after the discharge ends, the voltage gradually increases due to the polarization voltage. After the influence of the polarization voltage disappears, the voltage becomes constant, and the OCV after being lowered in the discharge can be obtained. Hereinafter, a method of obtaining the full charge capacity (maximum chargeable charge amount) Qmax of the unit cell 111 will be described with reference to FIG.

The OCV before discharge is OCV1, and the OCV after discharge is OCV2. By using the data table of FIG. 3, two OCVs can be converted into SOC1 and SOC2, respectively. If the remaining capacities corresponding to these are Q1 and Q2, and the integrated current value between them is dQ (= ∫I (t) dt), the current full charge capacity Qmax can be obtained as shown in equations 4-7. That is, as shown in Equation 7, the full charge capacity Qmax can be calculated using the current integrated value dQ during discharge and the SOC difference before discharge and after discharge (SOC1-SOC2).

  Here, discharge has been described as an example, but Qmax can be similarly obtained even during charging. Furthermore, even if charge / discharge is mixed, Qmax can be calculated in the same manner when an SOC difference occurs before and after charge / discharge.

  However, in the calculation example of Qmax here, since it is necessary to measure the OCV when charging / discharging is not performed, the value cannot be updated in real time during operation of the battery system 100. When the operation of the battery system 100 stops, the assembled battery control unit 150 detects that the charging / discharging of the assembled battery 110 is completed, and preferably after detecting that the influence of the polarization voltage is reduced, the unit cell 111 It is necessary to obtain the OCV.

  The assembled battery control unit 150 obtains Qmax using the last acquired OCV (OCV2 in the case of FIG. 5 as an example), and stores it in the second storage unit 190. After the activation of the assembled battery control unit 150 next time, the SOC calculation can be executed using the Qmax value stored in the second storage unit 190 as the Qmax of Equation 1.

  In the above description, the method of obtaining the OCV2 when the battery system 100 stops is described. However, when the assembled battery control unit 150 is activated next time, the OCV of the unit cell 111 is acquired, and this is used as the previous OCV final value. Assuming that Qmax is obtained as OCV2 and applied to the SOC calculation of Equation 1. In this case, the second storage unit 190 stores the OCV before the previous charge / discharge and the dQ at the previous charge / discharge of the assembled battery 110. Whichever method is used, the calculation accuracy of the SOC of Equation 1 can be secured using the current full charge capacity Qmax.

Although the method for obtaining Qmax has been described above, since Qmax is equivalent to the deterioration state of the unit cell 111, the deterioration state may be stored in the second storage unit 190 instead of Qmax. For example, as shown in Equation 8, a capacity maintenance ratio (1 when the battery cell 111 is new and smaller than 1) is calculated using the initial capacity Qmax0, and this is calculated as a deterioration degree parameter SOH (State of Health). It is also possible to define and store the SOH calculation result in the second storage unit 190 and update Qmax.

  In this case, after the next activation of the assembled battery control unit 150, Qmax is re-established using Equation 9 using the SOH stored in the second storage unit 190 and the initial capacity Qmax0 stored in the first storage unit 180 in advance. Calculate and apply the latest Qmax obtained to Equation 1. In the first embodiment, an example in which SOH is calculated and stored in the second storage unit 190 will be described.

  As described above, Qmax cannot be updated in real time. More specifically, after Qmax is updated after the assembled battery control unit 150 is activated, there is no opportunity to update Qmax until the assembled battery control unit 150 is stopped and then activated. Therefore, if an error occurs in a direction in which the Qmax calculation result is too large compared to the true value, the excessively large Qmax is continuously used until the assembled battery control unit 150 is stopped and restarted. On the other hand, even if the Qmax calculation result is too small, the Qmax that is too small will continue to be used until the assembled battery control unit 150 is restarted.

  FIG. 4B shows the influence of ΔSOCErr caused by this Qmax error (QmaxErr) on the SOC error. When Qmax is too large (+ QmaxErr), ΔSOC (t) is obtained by dividing ∫I (t) dt by Qmax that is too large, so that an SOC having a gentler slope than that of the true SOC can be obtained. On the contrary, when Qmax is too small (−QmaxErr), ΔSOC (t) is obtained by dividing ∫I (t) dt by Qmax which is too small, so that the slope is steeper compared to the true SOC. A correct SOC calculation result can be obtained.

  FIG. 6 is a diagram for explaining the details of the SOC error caused by the above-described QmaxErr. FIG. 6A shows an example of change in SOC when discharged from a high SOC and charged, and FIG. 6B shows an example of change in SOC when charged and discharged from low SOC. In order to simplify the description, the influence of SOCvErr described with reference to FIG.

  FIG. 6A shows a state in which ΔSOCErr expands as discharging from a high SOC, and deviates from the true SOC as the SOC decreases. As described above, Qmax cannot be updated in real time, and can be updated after the battery pack control unit 150 is restarted. Since Qmax and QmaxErr included in the discharge are also fixed during the discharge, the deviation from the true SOC is increasing with the discharge, and the SOC accuracy is deteriorated in the low SOC where it is desired to ensure the accuracy in order to use up the stored amount of the battery. To do. However, when the battery is subsequently charged, if the Qmax and the QmaxErr included therein remain fixed, the SOC change occurs with the same slope. Therefore, the SOC error improves as the SOC before the discharge approaches.

  FIG. 6B is an example in which charging is started from a low SOC, which is the reverse of FIG. Since Qmax and QmaxErr included therein are fixed, the divergence from the true SOC is increasing with the charging, and the SOC accuracy is deteriorated in a high SOC where it is desired to ensure the accuracy in order to stop the charging of the battery with high accuracy. To do. However, when discharging after that, if Qmax and QmaxErr included therein remain fixed, the SOC change occurs with the same slope, so the SOC error improves as the SOC approaches before charging. The present invention utilizes these SOC error features.

  FIG. 7 is a block diagram of the SOC calculation performed by the assembled battery control unit 150. The assembled battery control unit 150 includes an SOCv calculation unit 151, a ΔSOC calculation unit 152, a first SOCi calculation unit 153, a second SOCi calculation unit 154, a weight coefficient W calculation unit 155, and an SOCc calculation unit 156.

  The SOCv calculation unit 151 converts the voltage acquired when charging / discharging is not performed into SOCv using the data table of FIG. The SOCv corresponds to the initial value of the SOC. The ΔSOC calculation unit 152 obtains Qmax using Equation 9 and obtains the SOC change (ΔSOC (t)) by current integration.

  The first SOCi computing unit 153 calculates SOC (SOCi1 (t) in FIG. 8 described later) using Equation 1 using SOCv and ΔSOC (t).

  The second SOCi calculation unit 154 reads a final SOC calculation result (SOC1Z in FIG. 8 described later) after charging and discharging the assembled battery 110 from the second storage unit 190, and uses this as an initial value to calculate the SOC (described later in FIG. 8 SOCi2 (t)) is calculated. That is, the second SOCi calculation unit 154 calculates the SOC using Equation 1 using an initial value different from that of the first SOCi calculation unit 153 as a starting point. Details will be described again with reference to FIGS.

  The weighting factor W calculation unit 155 calculates the weighting factor W using the SOCv, ΔSOC (t), and the SOCvZ read from the second storage unit 190. The SOCc calculation unit 156 adds SOCi1 (t) and SOCi2 (t) using the weighting coefficient W (t), and estimates SOCc (t) that is a charged state after resuming charging and discharging.

  FIG. 8 is a diagram showing an outline of processing performed by the block diagram of FIG. FIG. 8A shows the SOC before the assembled battery control unit 150 is started and charging / discharging is started. FIG.8 (b) shows the change of SOC until it stops after implementing charging / discharging. FIG.8 (c) shows SOC when the assembled battery control part 150 is restarted after stopping charging / discharging. FIG. 8D shows a change in SOC after resuming charge / discharge.

  In the above description, it has been described that the Qmax can be updated after the assembled battery control unit 150 is restarted. However, the condition that is considered inappropriate for the calculation of the Qmax (for example, the influence of the polarization voltage is not relaxed when obtaining the OCV2). Etc.), the SOC may be calculated many times using the same Qmax. In the present embodiment, an operation example regarding a period in which Qmax is not updated will be described.

  In FIG. 8A, the SOCv calculation unit 151 converts the OCV of the unit cell 111 into the SOCv using the data table of FIG. 3 under the condition that is regarded as no load or no load. In order to simplify the description, the influence of the voltage detection error is omitted, and the SOCv obtained at this time is assumed to be equal to the true SOC. When charging / discharging is started, ΔSOC calculation unit 152 calculates an SOC change (ΔSOC (t)) from SOCv, and calculates the current SOC (SOCi1 (t)) using Equation 1. Since Qmax in Equation 1 includes an error, the true value and the calculated value of the SOC gradually deviate as charging / discharging continues.

  When the assembled battery control unit 150 is stopped at the time of FIG. 8B, the SOCv (initial value in FIG. 8A) before charging / discharging the assembled battery 110 is stored in the second storage unit 190 as SOCvZ. The final value of SOCi1 (t) after charging / discharging is stored in the second storage unit 190 as SOCi1Z.

  At the time of FIG. 8C, when the assembled battery control unit 150 is restarted, the SOCv calculation unit 151 obtains the SOCv from the OCV of the unit cell 111 using the data table of FIG. Further, the second SOCi calculation unit 154 reads the SOC calculation result (SOCi1Z) up to the previous time from the second storage unit 190.

  As shown in FIG. 8D, when charging / discharging is restarted, the ΔSOC calculation unit 152 obtains the SOC change (ΔSOC (t)) from the SOCv, and using this, the first SOCi calculation unit 153 obtains the equation (1). To calculate the current SOC (SOCi1 (t)). At the same time, the second SOCi calculation unit 154 calculates SOCi2 (t) using the SOCi1Z read from the second storage unit 190 and ΔSOC (t) obtained by the ΔSOC calculation unit 152. As described with reference to FIG. 6, when SOCi2 (t) is obtained without changing Qmax used in the calculation and approaches SOCvZ which is the starting SOC, the difference between SOCi2 (t) and the true value gradually decreases. In this embodiment, since it is assumed that the SOC is calculated many times using the same Qmax, the first SOCi calculation unit 153 also uses the same Qmax, so that SOCi1 (t) and SOCi2 (t) are SOCs. The degree of change is the same.

  FIG. 9 is a diagram illustrating a procedure for estimating the SOC after the assembled battery control unit 150 is restarted. Hereinafter, the concept of SOC estimation will be described with reference to FIG.

  When the assembled battery control unit 150 is restarted, as shown in FIG. 8C, the SOCv obtained from the terminal voltage of the unit cell 111 and the SOCi1Z calculated by integrating the current are obtained. The SOCv at this time point indicates an accurate state of charge, but there is a Qmax error. Therefore, when current integration is performed using the SOCv as an initial value, the estimation error increases as shown in SOCi1 (t) shown in FIG.

  Therefore, in the present invention, SOC is estimated with high accuracy by using SOCi1 (t) and SOCi2 (t) in a complementary manner. The first SOCi calculation unit 153 performs calculation of Formula 1 using SOCv as an initial value, and the second SOCi calculation unit 154 performs calculation of Formula 1 using SOCi1Z as an initial value. ΔSOC (t) is a common parameter because the value of Qmax is fixed. Since ΔSOC (t) is the amount of change in SOC from the initial SOC value, if ΔSOC (t) is common, SOCi1 (t) and SOCi2 (t) are as shown in FIG. 8D and FIG. It moves in parallel.

  Since it is difficult to think that the deterioration of the cell 111 rapidly proceeds in a short time, the assembled battery control unit 150 does not need to update Qmax frequently. Therefore, the assembled battery control unit 150 may delay the update cycle of Qmax, and may execute the SOC calculation many times using the same Qmax. In addition, as described above, when the condition is inappropriate for the calculation of Qmax, it may be difficult to update Qmax. In view of this, the case where the same value of Qmax is continuously used is taken up, and in this embodiment, the transition of the SOC is described with reference to FIGS. Subsequently, a specific calculation formula of the SOC in the present invention will be described.

  A specific arithmetic expression of the SOC in the present invention will be described using Expressions 10 and 11. The weighting factor W calculation unit 155 uses the SOCv and ΔSOC (t) and the SOCvZ stored in the second storage unit 190 (the SOC at the previous activation of the assembled battery control unit 150) based on Equation 10 to Calculate W (t).

The value of the weighting factor W (t) changes depending on where the current SOC is located between the previous initial SOC (SOCvZ) and the initial SOC (SOCv) before the current charging / discharging. When the current SOC is close to SOCv, the weighting factor W (t) tends to be close to 0, and when close to SOCvZ, the weighting factor W (t) tends to be close to 1. By using the weighting coefficient W (t), it is possible to perform a weighted calculation according to the current SOC.

Expression 11 represents an arithmetic expression for calculating a final SOC (denoted as SOCc (t)) using SOCi1 (t), SOCi2 (t), and a weight coefficient W (t). When the current SOC is close to SOCv, the weight coefficient W (t) is close to 0. Therefore, the output of the first SOCi calculation unit 153 that performs the SOC calculation based on the SOCv acquired when the assembled battery control unit 150 is activated is strongly adopted. . When the current SOC is close to SOCvZ, the weight coefficient W (t) approaches 1, so that the SOC calculation is performed based on SOCi1Z that is the final value of SOCi1 (t) stored when the assembled battery control unit 150 was stopped last time. The output of the second SOCi calculation unit 154 to be performed is strongly adopted.

  As a result, as shown in FIG. 9, the final calculation result SOCc (t) of the SOC that overlaps the true value of the SOC can be obtained. Actually, the effect of SOCvErr is added to this to obtain the final SOC error, but this is omitted for the sake of simplicity.

  Note that W (t) = 0 corresponds to the time when the calculation result of Formula 11 and SOCi1 (t) match, and W (t) = 1 corresponds to the time when the calculation result of Formula 11 matches SOCi2 (t). To do. Therefore, the weight coefficient W (t) needs to satisfy 0 ≦ W (t) ≦ 1. The assembled battery control unit 150 rounds down the excess when W (t) exceeds this range. That is, when W (t) becomes less than 0, the correction is made to W (t) = 0, and when W (t) exceeds 1, the correction is made to W (t) = 1.

<Embodiment 1: Summary>
As described above, the battery system 100 according to the first embodiment determines the initial SOC based on the voltage after starting the assembled battery control unit 150, obtains the SOC by integrating the current SOC from the starting point, and determines the previous assembled battery control. The SOCc (t) is estimated by combining with the SOC obtained by restarting the current integration starting from the SOC final value acquired by operating the unit 150. As a result, the SOC can be accurately estimated even under a constraint condition where the OCV cannot be acquired at an arbitrary time point and when Qmax includes an error.

  Further, according to the battery system 100 according to the first embodiment, the assembled battery control unit 150 transmits the SOCc (t) to the vehicle control unit 200, so that the vehicle control unit 200 is based on the highly accurate SOCc (t). Therefore, it is possible to appropriately determine whether to stop charging or discharging the assembled battery 110.

<Embodiment 2>
In the first embodiment, the case where the same Qmax is used in the previous process in the assembled battery control unit 150 and the current process in the assembled battery control unit 150 has been described. This is an effect that can be seen when Qmax cannot be updated or when the same Qmax value is used less frequently and continues to be used. In the second embodiment of the present invention, an operation example will be described in which a SOH calculation is performed to acquire a new SOH and a new Qmax is acquired accordingly. Others are the same as those described in the first embodiment, and in the second embodiment, description will be made focusing on the changes in the assembled battery control unit 150.

  The second SOCi calculation unit 154 performs the SOC calculation based on the SOC calculation final value (SOCi1Z) stored during the previous operation of the assembled battery control unit 150, and uses the Qmax (or SOH) used in the processing of the previous assembled battery control unit 150 ) Is used as it is. This is to ensure the effect of reducing the estimation error when approaching the SOC that is the starting point described in FIG.

  On the other hand, when the first SOCi calculation unit 153 obtains a new Qmax value (or SOH), the first SOCi calculation unit 153 can perform the SOC calculation using the new value. This is because it is not affected by the previous operation result of the assembled battery control unit 150.

  FIG. 10 is a control block diagram for the assembled battery control unit 150 to calculate the SOC in the second embodiment. The second storage unit 190 outputs two SOHs (SOH1, SOH2). SOH1 is the latest SOH, and SOH2 is the SOH used in the previous processing of the assembled battery control unit 150. The ΔSOC calculation unit 152 obtains ΔSOC1 (t) and ΔSOC2 (t) by using SOH1 and SOH2, respectively. ΔSOC1 (t) is used in the first SOCi calculation unit 153 and also in the weight coefficient W calculation unit 155. ΔSOC2 (t) is used in the second SOCi calculation unit 154.

  FIG. 11 is a diagram showing the transition of the SOC change in the second embodiment. In the first embodiment, since the first SOCi calculation unit 153 and the second SOCi calculation unit 154 use the same ΔSOC (t) (by using the same Qmax and SOH), SOCi1 (t) and SOCi2 (t) have shifted in parallel. However, the present embodiment 2 shows a different SOC change.

  The processing before reaching FIG. 11 is the same as (a) to (c) of FIG. SOCi2 (t) obtained based on SOCi1Z uses the same Qmax as in FIG. 8D and FIG. 9, and the inclination is gentle with respect to the true SOC change.

  On the other hand, SOCi1 (t) is calculated using ΔSOC1 (t) obtained using Qmax determined based on the latest SOH (SOH1). In the example shown in FIG. 11, an example is shown in which the latest SOH, SOH1, gives a Qmax smaller than the true Qmax, and as a result, obtains SOCi1 (t) having a steeper slope than the true SOC change. did.

  The behavior of changes in SOCi1 (t) and SOCi2 (t) is as shown in FIG. The calculation result of SOCc (t) that combines these based on Equation 10 and Equation 11 is also shown. Thus, even when SOH (or Qmax) is not constant and the value is updated, an SOC closer to the true value is obtained than SOCi1 (t) or SOCi2 (t) alone.

<Embodiment 3>
In Embodiment 1 or 2, a plurality of SOCs are combined based on a weight W (t) having a value between 0 and 1, and an SOC that is closer to the true value than a single unit is obtained. In this embodiment, a new weight W ′ (t) having only a value of 1 or 0 is used by applying the following determination process to the weight W (t) obtained by Expression 10.
・ When W (t) ≧ 0.5 W ′ (t) = 1
When W (t) <0.5 W ′ (t) = 0

  When W ′ (t) is applied instead of the weight W (t) used in Equation 11, when it is determined that the error is smaller in SOCi1 (t) (W (t) <0.5), the SOCi1 is immediately When (t) is adopted and it is determined that SOCi2 (t) has a smaller error (W (t) ≧ 0.5), SOCi2 (t) can be immediately adopted as well. The SOC selection result is output as SOCc (t) to the vehicle control unit 200 and the like, and charging / discharging control of the assembled battery 110 such as charging limitation or discharging limitation is performed based on the SOC having higher accuracy at all times.

  However, since the weight W ′ (t) has two values of 1 and 0, for example, when the weight W (t) obtained by Equation 10 varies around 0.5, the value of W ′ (t) is 1 to 0. Therefore, SOCi1 (t) and SOCi2 (t) are frequently switched from 0 to 1, and as a result, SOCc (t) may become unstable. When the vehicle control unit 200 receives such SOCc (t) and performs charge restriction or discharge restriction on the assembled battery 110, the charge / discharge control of the assembled battery 110 becomes unstable.

  Therefore, when SOCc (t) is stabilized, W ′ (t) is switched between 1 and 0, and the adopted SOC is also frequently switched between SOCi1 (t) and SOCi2 (t). However, the SOCc (t) is applied to the SOCc (t) output as a result by applying a change amount limiting process for limiting the change to a certain SOC, and applying a change amount limiting process for preventing the SOC change above the threshold. t) is transmitted to the vehicle control unit 200 as the SOC final value. In this way, it is possible to provide a battery system capable of avoiding an SOC change in which charging / discharging control of the assembled battery 110 performed by the vehicle control unit 200 becomes unstable while immediately adopting an SOC with few errors. it can.

<Embodiment 4>
In the embodiments so far, the SOCi1 (t) based on the SOCv obtained by the SOCv calculation unit 151 and the SOCi2 (t) based on the SOCi1Z stored in the second storage unit 190 are stored in the SOCv and the second storage unit 190. The method of combining based on the weight W (t) determined by the SOCvZ and ΔSOC (t) from the ΔSOC calculation unit 152 has been described. However, there may be cases where the SOCv cannot be used, such as when there is an abnormality in the measured voltage information. Therefore, in the present embodiment, a method for obtaining a highly accurate SOCc (t) by applying a process of combining SOCs even when the SOCv cannot be used for some reason will be described.

  In the present embodiment, one more combination of SOCvZ and SOCi1Z is stored in the second storage unit 190. In this way, even if the SOCv cannot be used, SOCi2 (t) is calculated based on the two SOCi1Z stored in the second storage unit 190, and the weight is calculated based on the two SOCvZ and ΔSOC (t). By calculating W (t) and using the weights W (t) for the two SOCi2 (t), the combination calculation process can be realized. Even when there is an abnormality in the voltage information, the battery system 100 that obtains the SOCc (t) and outputs the highly accurate SOC to the outside can be provided.

  As mentioned above, the invention made by the present inventor has been specifically described based on the embodiment. However, the present invention is not limited to the embodiment, and various modifications can be made without departing from the scope of the invention. Needless to say.

  In addition, each of the above-described configurations, functions, processing units, etc. can be realized as hardware by designing all or a part thereof, for example, with an integrated circuit, or the processor executes a program for realizing each function. By doing so, it can also be realized as software. Information such as programs and tables for realizing each function can be stored in a storage device such as a memory or a hard disk, or a storage medium such as an IC card or a DVD.

  100: battery system 110: assembled battery 111: single battery 112: single battery group 120: single battery management unit 121: single battery control unit 122: voltage detection circuit 123: control circuit 124: signal input Output circuit, 125: temperature detector, 130: current detector, 140: voltage detector, 150: battery pack controller, 151: SOCv calculator, 152: ΔSOC calculator, 153: first SOCi calculator, 154: first 2SOCi calculation unit, 155: weight coefficient W calculation unit, 156: SOCc calculation unit, 160: signal communication means, 170: insulation element, 180: first storage unit, 181: SOC table, 190: second storage unit, 200: Vehicle control unit, 300 to 330: relay, 400: inverter, 410: motor generator, 420: charger.

Claims (9)

A device for controlling the operation of a secondary battery,
A voltage measuring unit for measuring a terminal voltage of the secondary battery;
A current measuring unit for measuring a current flowing through the secondary battery;
A state estimation unit for estimating a state of the secondary battery;
With
The state estimation unit
Based on the voltage information measured by the voltage measurement unit, the initial state of charge is estimated, and the state of charge under charge / discharge is estimated by integrating the initial state of charge and the current information measured by the current measurement unit. Has the function to
The battery controller according to claim 1, wherein the state estimator estimates two states of charge in parallel and estimates the state of charge by weighted addition of the two estimated states of charge.   The battery control device according to claim 1, wherein the charge state initial values of the two charge states estimated by the state estimation unit are estimated from different voltage information. A storage unit for storing results of processing performed by the state estimation unit in the past;
The at least one state of charge among the two states of charge estimated by the state estimation unit is estimated based on a processing result of the state estimation unit stored in the storage unit. Battery control device. The storage unit
At least a charge state initial value estimated based on voltage information measured by the voltage measurement unit,
The battery control apparatus according to claim 3, wherein a charge state final value at the end of charging / discharging of the secondary battery is stored. The state estimation unit
Two charge state initial values estimated based on voltage information measured by the voltage measurement unit;
The battery control according to any one of claims 1 to 4, wherein a weighting coefficient is calculated based on a change amount of a state of charge estimated by integrating current information measured by the current measuring unit. apparatus. The state estimation unit
Wherein if the calculated result of the weighting factors is smaller than 0 is set to 0 the weighting factor, when the calculation result is greater than 1 of the weighting coefficients according to claim 5, wherein the set of 1 to the weighting factor Battery control device.   The two charge states estimated in parallel are estimated based on full charge capacity or deterioration information of full charge capacity set at the time when the charge state initial value is estimated from voltage information, respectively. The battery control apparatus of any one of Claim 1 to 6. The state estimation unit
7. The battery control according to claim 5 , wherein charging or discharging of the secondary battery is limited based on a charging state of the secondary battery estimated by combining two charging state estimation results with the weighting coefficient. apparatus. The battery control device according to any one of claims 1 to 8,
The secondary battery whose operation is controlled by the battery control device;
A secondary battery system comprising:

Images