JP6708956B2 - Battery system - Google Patents

Description

The present invention relates to a technique for controlling charge/discharge of a nickel hydrogen battery.

In order to accurately charge and discharge the secondary battery, it is desirable to accurately calculate the state of charge (hereinafter also referred to as “SOC”) of the secondary battery. As a SOC calculation method, a voltage method that uses the voltage of a secondary battery is known. In consideration of the fact that the larger the SOC, the higher the open circuit voltage of the secondary battery (Open Circuit Voltage, hereinafter also referred to as “OCV”), this voltage system is a curve showing the value of OCV with respect to SOC (hereinafter also referred to as “OCV curve”). ) Is obtained in advance by experiments and the like, and the SOC corresponding to the detected value of the voltage of the secondary battery is calculated by referring to this OCV curve.

It is known that a phenomenon called a memory effect occurs in a secondary battery as charging and discharging are repeated. When the memory effect occurs, the OCV curve deviates from that when it is new (when it is not used), so that the accuracy of SOC calculation by the voltage method deteriorates. In view of this point, Japanese Unexamined Patent Application Publication No. 2007-333447 (Patent Document 1) discloses a technique for correcting an OCV curve in consideration of a voltage shift due to a memory effect, and calculating an SOC with reference to the corrected OCV curve. Is disclosed.

JP, 2007-333447, A

Conventionally, nickel-hydrogen batteries have been widely used as secondary batteries for hybrid vehicles. The inventors of the present application have newly found through experiments and the like that a phenomenon different from the memory effect described above occurs in a nickel hydrogen battery. Another phenomenon is that when the time the SOC stays in a region lower than a predetermined value (hereinafter also referred to as “low SOC region”) exceeds a predetermined time, the slope of the OCV curve (OCV of the SOC when the SOC changes by a unit amount) This is a phenomenon in which the amount of change) becomes flatter (smaller) than that of a new product. In this specification, this phenomenon is also referred to as "flattening of OCV curve" for convenience of description.

If the OCV curve is flattened, there is a concern that the accuracy of SOC calculation by the voltage method may be reduced. In other words, even if the OCV curve is flattened, if the SOC is calculated using the OCV curve at the time of new product, the calculation accuracy of the SOC will be reduced. Further, even if the OCV curve at the time of a new product is corrected in consideration of the flattening, it is difficult to uniquely determine the SOC value with respect to the OCV in the corrected OCV curve due to the flattening. There is a concern that the accuracy will decrease.

The present invention has been made to solve the above-described problems, and an object thereof is to suppress a decrease in SOC calculation accuracy even when an OCV curve is flattened in a nickel-hydrogen battery. That is.

The battery system according to the present invention includes a nickel hydrogen battery and a control unit. The control unit uses the first charged amount calculated using the integrated current value of the nickel hydrogen battery and the second charged amount calculated using the voltage of the nickel hydrogen battery to determine the third charged amount of the nickel hydrogen battery. To calculate. The control unit weights a first contribution degree indicating a degree of reflecting the first charged amount on the third charged amount and a second contribution degree indicating a degree on which the second charged amount is reflected on the third charged amount. By doing so, the third charged amount, which is a weighted average of the first charged amount and the second charged amount, is calculated. The controller controls the slope of the OCV curve indicating the value of the open-circuit voltage (OCV) with respect to the amount of stored electricity of the nickel-hydrogen battery, when the time period during which the third stored amount of electricity is lower than the predetermined value exceeds the predetermined time. It is determined that “flattening of the OCV curve”, which is flatter than that of a new product, has occurred , and the first contribution rate is made larger than that when the staying time does not exceed the predetermined time.

According to the above configuration, the control unit controls the first contribution of the first stored amount (temporary SOC calculated using the current integrated value) and the second stored amount (provisional SOC calculated using voltage). The third charge amount (final SOC) of the nickel-hydrogen battery is calculated by weighting the second contribution.

As described above, in the nickel-hydrogen battery, when the SOC stays in the low SOC region for a predetermined time or longer, the SOC (second charge amount) is calculated using the voltage due to the flattening of the OCV curve. There is a concern that the accuracy will decrease. In view of this point, the control unit retains the third stored amount of electricity when the residence time in the first region (low SOC region) exceeds a predetermined time (when the OCV curve is flattened). The first contribution degree of the first storage amount (provisional SOC calculated using the integrated current value) is larger than that in the case where the charging time does not exceed the predetermined time (when the OCV curve is not flattened). To do. This increases the degree to which the first stored amount that is not affected by the flattening of the OCV curve is reflected in the third stored amount, and the second stored amount that is affected by the flattening of the OCV curve is changed to the third stored amount. To a lesser degree. As a result, even when the OCV curve is flattened in the nickel-hydrogen battery, it is possible to prevent the calculation accuracy of the SOC from decreasing.

It is a figure which shows roughly the structure of the vehicle in which a battery system is mounted. 6 is a flowchart showing an example of a processing procedure performed when the ECU calculates a control SOC. It is a figure which shows the voltage curve after the endurance test 1. It is a figure which shows the voltage curve after the endurance test 2. It is a figure which shows the voltage curve after the endurance test 3. It is a figure which shows the OCV curve of a nickel hydrogen battery typically. It is a figure which shows the voltage curve when charging/discharging is repeated twice after the endurance test 2. It is a figure which shows the voltage curve when charging/discharging is repeated twice after the endurance test 3. 4 is a flowchart showing an example of a processing procedure when the ECU 100 cancels flattening of an OCV curve and adjusts a weighting coefficient K. It is a figure which shows an example of a change of the weighting factor K.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. It should be noted that the same or corresponding parts in the drawings are designated by the same reference numerals and the description thereof will not be repeated.

<Battery system configuration>
FIG. 1 is a diagram schematically showing a configuration of a vehicle 1 in which a battery system 2 according to the present embodiment is mounted. Note that the vehicle 1 shown in FIG. 1 is a so-called hybrid vehicle, but the battery system 2 according to the present embodiment is not limited to being applied to a hybrid vehicle, and is applicable to all vehicles equipped with a nickel hydrogen battery, and other than vehicles. It is also applicable to.

The vehicle 1 includes a battery system 2, a power control unit (PCU) 30, motor generators (hereinafter referred to as “MG”) 41 and 42, an engine 50, a power split mechanism 60, and a drive. A shaft 70 and a drive wheel 80 are provided. The battery system 2 includes a nickel hydrogen (NiMH) battery 10, a monitoring unit 20, and an electronic control unit (ECU: Electronic Control Unit) 100.

The engine 50 outputs kinetic energy by the combustion energy of fuel. MGs 41 and 42 can function as both a generator and an electric motor.

MG 41 mainly operates as a generator that generates electric power by using a part of the output of engine 50 transmitted through power split device 60. The electric power generated by the MG 41 is supplied to the MG 42 or the nickel hydrogen battery 10 via the PCU 30.

MG 42 is driven by at least one of the power from nickel hydrogen battery 10 and the power generated by MG 41. The drive force of MG 42 is transmitted to drive shaft 70. During regenerative braking of vehicle 1, MG 42 operates as a generator by being driven by the rotational force of the wheels. The nickel-hydrogen battery 10 is charged via the PCU 30 with the electric power generated by the MGs 41 and 42.

Nickel hydrogen battery 10 stores electric power for driving MG 41, 42. The nickel hydrogen battery 10 includes a plurality of nickel hydrogen battery cells connected in series. The monitoring unit 20 includes a voltage sensor 21, a current sensor 22, and a temperature sensor 23. The voltage sensor 21 is configured to be able to detect the terminal voltage of the nickel-hydrogen battery 10 (hereinafter also referred to as “battery voltage VB”). The current sensor 22 is configured to be able to detect the charge/discharge current of the nickel hydrogen battery 10 (hereinafter also referred to as “battery current IB”). The temperature sensor 23 is configured to be able to detect the temperature of the nickel hydrogen battery 10 (hereinafter also referred to as “battery temperature TB”). Each sensor outputs a signal indicating the detection result to ECU 100.

PCU 30 is configured to perform bidirectional power conversion between nickel hydrogen battery 10 and MGs 41, 42 according to a switching command from ECU 100. The PCU 30 is configured to be able to control the states of the MGs 41 and 42 separately. For example, PCU 30 can put MG 41 in a power running state while putting MG 41 in a regenerative (power generating) state.

ECU 100 is configured to include a CPU (Central Processing Unit), a memory, and an input/output interface (none of which is shown). The ECU 100 controls charging/discharging of the nickel hydrogen battery 10 by controlling the engine 50 and the PCU 30 based on the signals from the respective sensors and the information stored in the memory.

When it is necessary to charge the nickel-hydrogen battery 10, the ECU 100 causes the MG 41 to generate power by using a part of the power of the engine 50, and charges the nickel-hydrogen battery 10 with the electric power generated by the MG 41. Control (MG41, MG42). Hereinafter, control of charging nickel-hydrogen battery 10 with electric power generated by MG 41 using a part of the power of engine 50 is also referred to as “P charge”.

The ECU 100 calculates the charge amount (SOC) of the nickel-hydrogen battery 10 in order to accurately control the charge/discharge of the nickel-hydrogen battery 10. Generally, SOC is represented by the ratio of the remaining capacity to the full charge capacity.

Generally, as the SOC calculation method, a method using the voltage of the secondary battery (hereinafter referred to as “voltage method”), a method using the integrated current value of the secondary battery (hereinafter referred to as “current integration method”), a voltage method There is known a method (hereinafter, also referred to as a “weighting method”) in which both the SOC according to 1) and the SOC according to the current integration method are weighted and used. In the voltage method, an OCV curve showing an OCV value with respect to SOC is obtained in advance by experiments or the like and stored in a memory, and the SOC corresponding to the detected voltage value of the secondary battery is calculated with reference to the OCV curve. .. In the current integration method, the amount of change in SOC from the start of integration is calculated by integrating the charge/discharge current of the secondary battery, and the calculated amount of change in SOC is added to the initial value of SOC (the value at the start of integration). The SOC is calculated by The weighting method is to calculate both the SOC by the voltage method and the SOC by the current integration method and weight the contributions of both to calculate the final SOC.

ECU 100 according to the present embodiment calculates SOC by a weighting method. Specifically, the ECU 100 calculates the first provisional SOCa by the current integration method, calculates the second provisional SOCb by the voltage method, and weights the first provisional SOCa and the second provisional SOCb to obtain the control SOC. calculate. This control SOC is used for charge/discharge control of the nickel hydrogen battery 10.

FIG. 2 is a flowchart showing an example of a processing procedure performed when the ECU 100 calculates the control SOC. This flowchart is repeatedly executed at a predetermined cycle.

In step (hereinafter, step is abbreviated as “S”) 10, ECU 100 calculates first provisional SOCa by the current integration method. Specifically, the ECU 100 obtains the amount of change in SOC from the start of integration by integrating the battery current IB, and adds the obtained amount of change in SOC to the initial value of SOC (the value at the start of integration). The first provisional SOCa is calculated. Since the current integration method itself is known as described above, further detailed description will not be repeated here.

In S11, the ECU 100 calculates the second provisional SOCb by the voltage method. Specifically, the ECU 100 stores in advance a new OCV curve (refer to the alternate long and short dash line in FIG. 6 described later) obtained in an experiment or the like in a memory, and refers to the OCV curve to determine the battery voltage VB (voltage The SOC corresponding to the value detected by the sensor 21) is calculated. Since the voltage system itself is publicly known as described above, further detailed description will not be repeated here.

After that, the ECU 100 calculates the control SOC by weighting the contributions of both the first provisional SOCa and the second provisional SOCb (S12, S13). Specifically, the ECU 100 first reads the weighting coefficient K stored in the memory in S12. The weighting factor K indicates the degree to which the first provisional SOCa is reflected in the control SOC (the contribution of the first provisional SOCa to the control SOC). The weighting factor K is set to a value of 0 or more and 1 or less. The degree to which the second provisional SOCb is reflected in the control SOC (the degree of contribution of the second provisional SOCb to the control SOC) is represented by "1-K". That is, in the present embodiment, the sum of the contribution of the first provisional SOCa and the contribution of the second provisional SOCb is set to "1". The weighting factor K is normally set to a predetermined initial value K _ini .

Then, the ECU 100 calculates the control SOC by substituting the first provisional SOCa, the second provisional SOCb, and the weighting factor K into the following calculation formula (1) in S13.

Control SOC=K·SOCa+(1-K)·SOCb (1)
Note that the calculation formula (1) is merely an example, and the control SOC may be calculated by another calculation formula or another method if a weighting method is used.

The ECU 100 controls the charging/discharging of the nickel-hydrogen battery 10 so that the control SOC calculated as described above falls within a predetermined region (a normal SOC region or a high SOC region described later). For example, when the SOC falls below the lower limit value of the predetermined region, ECU 100 increases the SOC by performing the above-mentioned P charge to charge nickel hydrogen battery 10.

<Flatization of OCV curve>
As described above, the ECU 100 calculates the control SOC by using both the first provisional SOCa of the current integration method and the second provisional SOCb of the voltage method. Therefore, in order to accurately calculate the control SOC, it is desirable to calculate the first provisional SOCa and the second provisional SOCb with high accuracy.

However, the inventors of the present application have newly found that a phenomenon that affects the calculation accuracy of the second provisional SOCb by the voltage method may occur in the nickel-hydrogen battery 10. Specifically, the inventors of the present invention have found that when the SOC continuously stays in a region where the SOC is lower than a predetermined value (low SOC region) for a predetermined time or longer, the slope of the OCV curve (OCV when the SOC changes by a unit amount) It was newly found by various experiments including the following endurance tests 1 to 3 that the amount of change) is flatter than that when new (unused).

<<Experimental content and results>>
The inventors each carry out the following durability tests 1 to 3 on the nickel-hydrogen battery 10, and carry out an experiment to measure the behavior of the battery voltage VB with respect to the SOC (hereinafter also referred to as “voltage curve”) after each durability test. I did.

(Durability Test 1) P1 is left for a predetermined period at 25° C. in a state of almost complete discharge (state of SOC is about several%).

(Durability Test 2) P1 is left for a predetermined period at 65° C. in a state of being almost completely discharged.
(Durability Test 3) The battery is left at 65° C. for a predetermined period of P2 (P2>P1) in a state of being almost completely discharged.

The above durability tests 1 to 3 are tests in which SOC is continuously retained in a low SOC region (region lower than a predetermined value) for a predetermined time T1 or more. The above durability tests 1 to 3 are common in that they are left in a state of being almost completely discharged. The endurance test 2 has the same standing period as the endurance test 1, but is different in that the temperature is high. The endurance test 3 has the same temperature as the endurance test 2, but is different in that the leaving period is long.

The above experimental results are shown in FIGS. FIG. 3 is a diagram showing a voltage curve after the durability test 1. FIG. 4 is a diagram showing a voltage curve after the durability test 2. FIG. 5 is a diagram showing a voltage curve after endurance test 3. In each of FIGS. 3 to 5, the horizontal axis represents SOC, and the vertical axis represents battery voltage VB (CCV: Closed Circuit Voltage) when SOC is changed. Further, in each of FIGS. 3 to 5, the solid line represents the voltage curve after the endurance test, and the broken line represents the voltage curve when the product is new. In each of the charging voltage curves shown in FIGS. 3 to 5 (and FIGS. 7 and 8 to be described later), there is a region where the battery voltage VB becomes flat at the maximum level. This is because the voltage at the maximum level is intentionally maintained for a predetermined time in order to bring the battery into a nearly fully charged state.

From all the experimental results, it can be confirmed that after the durability test (solid line), the voltage curve is flatter (the slope of the voltage curve is smaller) than that of the new product (broken line). Further, it can be confirmed that the flattening of the voltage curve depends on the temperature and the standing period. Specifically, comparing FIG. 3 and FIG. 4 in which only the temperature is different, it is confirmed that the voltage curve of FIG. 4 having a high temperature is flatter than the voltage curve of FIG. 3 having a low temperature. it can. Further, when comparing FIG. 4 and FIG. 5 which are different only in the leaving period, it is confirmed that the voltage curve of FIG. 5 having a long leaving period is flatter than the voltage curve of FIG. 4 having a short leaving period. it can. The flattening of the voltage curve described above is caused by leaving it in the low SOC region, but even when charging/discharging in the low SOC region, the flattening of the voltage curve occurs as in the case of leaving. Applicant has confirmed in other experiments.

The voltage curves during charging and discharging shown in FIGS. 3 to 5 are the results of measurement with a small current close to no load. That is, it is considered that the OCV curve is also flattened in the nickel hydrogen battery in which the voltage curve is flattened as described above.

FIG. 6 is a diagram schematically showing an OCV curve of the nickel hydrogen battery 10. In FIG. 6, the horizontal axis represents SOC and the vertical axis represents OCV. Further, in FIG. 6, the solid line represents the OCV curve after the durability test, and the alternate long and short dash line represents the OCV curve at the time of new product.

As shown in FIG. 6, the OCV curve (dashed line) at the time of new product has a certain degree of inclination in the entire SOC range. Therefore, if the OCV is determined, the SOC is also uniquely easy to determine.

On the other hand, the OCV curve (solid line) after the endurance test changes from the OCV curve at the time of a new product, and has a long flat region (a region where the OCV is almost constant regardless of SOC). In this flat region, it is difficult to uniquely determine SOC even if OCV is determined.

In this way, when the time during which the SOC remains in the low SOC region exceeds the predetermined time T1, the actual OCV curve is flattened like the OCV curve (solid line) after the durability test. The flattening of the OCV curve depends not only on the standing time but also on the temperature. Nevertheless, if the second provisional SOCb is calculated using the OCV curve (dashed line) at the time of new product, the calculation accuracy of the second provisional SOCb decreases. Further, even if the OCV curve at the time of a new product is corrected in consideration of flattening, the corrected OCV curve is difficult to uniquely determine the SOC value with respect to the OCV due to the flattening. There is a concern that the calculation accuracy of SOCb may decrease. If the calculation accuracy of the control SOC decreases due to the decrease in the calculation accuracy of the second provisional SOCb, there is a concern that the charge/discharge of the nickel-hydrogen battery 10 cannot be controlled appropriately.

<Elimination of flattening of OCV curve>
The inventors conducted various studies to eliminate the flattening of the OCV curve. As a result, the inventors have found that the flattening of the OCV curve can be resolved by charging to a SOC that is somewhat high.

FIG. 7 is a diagram showing a voltage curve when charging and discharging are repeated twice after the durability test 2. FIG. 8 is a diagram showing a voltage curve when charging and discharging are repeated twice after the durability test 3. 7 and 8, the alternate long and short dash line represents the voltage curve during the first charging/discharging, and the alternate long and two short dashes line represents the voltage curve during the second charging/discharging.

From the examination results of FIGS. 7 and 8, it is confirmed that by repeating charging and discharging, the state where the voltage curve is flattened gradually recovers to the state of a new product having a slope. In particular, it can be confirmed that the inclination recovery after the first charging is remarkable. From the results of this examination, it is considered that the flattening of the OCV curve can be resolved by charging to a high SOC to some extent.

In view of this point, the ECU 100 according to the present embodiment sets the SOC to a predetermined value (low value) when the time during which the SOC stays in the low SOC region (hereinafter, also referred to as “low SOC stay time”) exceeds the predetermined time T1. The value is increased to a value larger than the upper limit of the SOC region), and charging and discharging are repeated in a relatively high SOC region for a predetermined period thereafter. The flattening of the OCV curve is gradually eliminated as the charge and discharge are repeated in the high SOC region. By completely eliminating the flattening of the OCV curve, the problem of deterioration in the calculation accuracy of the second provisional SOCb is also eliminated.

<Adjustment of Weighting Factor K (Contribution of First Provisional SOCa to Control SOC)>
As described above, when the low SOC residence time exceeds the predetermined time T1, the ECU 100 gradually cancels the flattening of the OCV curve by repeating charging and discharging in the high SOC region. As a result, when the flattening of the OCV curve is completely eliminated, the problem of deterioration of the calculation accuracy of the second provisional SOCb is also solved.

However, during the period until the flattening of the OCV curve is completely eliminated, the calculation accuracy of the second provisional SOCb still decreases, and the problem that the calculation accuracy of the control SOC decreases due to the influence remains. ..

Therefore, the ECU 100 according to the present embodiment, when the low SOC staying time exceeds the predetermined time T1, than when the low SOC staying time does not exceed the predetermined time T1, the weighting factor K (the first provisional value for the control SOC). SOCa contribution) is set to a large value. This increases the degree to which the first provisional SOCa that is not affected by the flattening of the OCV curve is reflected in the control SOC, and the second provisional SOCb that is affected by the flattening of the OCV curve is reflected in the control SOC. The degree of being reduced is reduced. As a result, even if the calculation accuracy of the second provisional SOCb temporarily decreases due to the flattening of the OCV curve, it is possible to make it difficult to decrease the calculation accuracy of the control SOC.

Note that, as described above, the flattening of the OCV curve is gradually eliminated by repeating charging and discharging in the high SOC region. In consideration of this point, the ECU 100 sets the weighting coefficient K to the maximum value Kmax at the beginning of the period in which the charging/discharging in the high SOC region is started, and thereafter, the weighting coefficient K is gradually initialized at a predetermined rate. Return to the value Kini. It is assumed that the elimination speed of the flattening of the OCV curve depends on the temperature. Therefore, the ECU 100 adjusts the “predetermined rate” for returning the weighting factor K to the initial value Kini, not taking it constant but taking the battery temperature TB into consideration.

<Process flow of ECU>
FIG. 9 is a flowchart showing an example of a processing procedure when the ECU 100 cancels the flattening of the OCV curve and adjusts the weighting coefficient K.

In S20, ECU 100 determines whether the control mode of nickel-hydrogen battery 10 is the high SOC mode.

In the present embodiment, the control mode of nickel hydrogen battery 10 includes a normal SOC mode and a high SOC mode.

The normal SOC mode is a mode in which the charging/discharging of the nickel-hydrogen battery 10 is controlled so that the control SOC falls within the region from the control lower limit value SN1 to the control upper limit value SN2 (hereinafter also referred to as “normal SOC region”). For example, when the control lower limit value SN1 and the control upper limit value SN2 of the normal SOC region are 30% and 70%, respectively, the SOC control center value in the normal SOC mode is 50%. The above values are merely examples, and the present invention is not limited to these.

The high SOC mode is a mode for eliminating the flattening of the OCV curve. The high SOC mode is a mode in which charging/discharging of the nickel-hydrogen battery 10 is controlled so that the control SOC falls within a region from the control lower limit value SH1 to the control upper limit value SH2, which is relatively high (hereinafter, also referred to as “high SOC region”). is there. For example, when the control lower limit value SH1 and the control upper limit value SH2 in the high SOC region are 60% and 80%, respectively, the SOC control center value in the high SOC mode is 70%. The above values are merely examples, and the present invention is not limited to these. However, the control lower limit value SH1 in the high SOC region is at least larger than the control lower limit value SN1 in the normal SOC region, and is larger than the upper limit value (predetermined value) in the low SOC region where the flattening of the OCV curve is concerned. Must be set to.

When the control mode of nickel-hydrogen battery 10 is not the high SOC mode (NO in S20), that is, when the control mode of nickel-hydrogen battery 10 is the normal SOC mode, ECU 100 determines in S21 that the low SOC retention time (for control) The time during which the SOC continues to stay in the low SOC region) is calculated. The low SOC retention time is calculated from the history of control SOCs up to the present. The low SOC retention time is reset when the control SOC exceeds a predetermined value. The control SOC is calculated from the above calculation formula (1).

In S22, ECU 100 determines whether or not the low SOC residence time exceeds predetermined time T1. This determination is a process for determining whether or not the OCV curve is flattened. As described above, the flattening of the OCV curve depends not only on time but also on temperature. Therefore, the ECU 100 sets the predetermined time T1 to an appropriate value according to the history of the battery temperature TB during the low SOC retention time. For example, it is considered that the higher the battery temperature TB, the earlier the flattening of the OCV curve occurs, so the ECU 100 shortens the predetermined time T1. This makes it possible to accurately determine whether or not the OCV curve is flattened.

When the low SOC retention time exceeds the predetermined time T1 (YES in S22), it is considered that the OCV curve is flattened, and therefore the ECU 100 shifts the processing to S23 and sets the control mode of the nickel hydrogen battery 10 to the same. , The current normal SOC mode is changed to the high SOC mode. At this time, ECU 100 first performs forced charging (first charging) for forcibly charging nickel-hydrogen battery 10 by performing the above-mentioned P-charge until the control SOC increases to the control center value in the high SOC region. By this forced charging (first charging), the SOC increases to a value larger than the predetermined value, so that the flattening of the OCV curve can be eliminated early. After that, the ECU 100 controls the charging/discharging of the nickel hydrogen battery 10 so that the control SOC falls within the high SOC range. This suppresses the SOC from immediately returning to the low SOC region (that is, the OCV curve returns to the flattened state) after the forced charging.

After changing the control mode of the nickel hydrogen battery 10 to the high SOC mode, the ECU 100 sets the weighting coefficient K (contribution degree of the first provisional SOCa to the control SOC) stored in the memory to the initial value Kini in S24. Set to a value greater than. This increases the degree to which the first provisional SOCa that is not affected by the flattening of the OCV curve is reflected in the control SOC, and the second provisional SOCb that is affected by the flattening of the OCV curve is reflected in the control SOC. The degree of being reduced is reduced. As a result, even when the OCV curve is flattened, it is possible to prevent the calculation accuracy of the control SOC from being lowered.

On the other hand, when the control mode of nickel-hydrogen battery 10 has already been changed to the high SOC mode (YES in S20), ECU 100 determines in S25 the time during which the high SOC mode is continued (hereinafter referred to as “high SOC continuation time”). (Also referred to as “”) exceeds a predetermined time T2. This determination is for determining whether the flattening of the OCV curve has been resolved. As described above, it is assumed that the elimination speed of the flattening of the OCV curve depends on the temperature. Therefore, the ECU 100 sets the “predetermined time T2” according to the battery temperature TB.

Further, until it is determined that the flattening of the OCV curve has been resolved, the weighting coefficient K (contribution of the first provisional SOCa by the current integration method) is maintained at a value larger than the initial value Kini, but the first provisional Since SOCa includes a current integration error, if the time for maintaining the weighting coefficient K at a large value becomes too long, the influence of the current integration error becomes too large, which rather reduces the calculation accuracy of the control SOC. There is a concern that it will end up. Therefore, the ECU 100 sets the “predetermined time T2” in consideration of the influence of the current integration error.

If the high SOC continuation time does not exceed the predetermined time T2 (NO in S25), there is a concern that the OCV curve will immediately return to the flattened state if the normal SOC mode is restored at this time. The ECU 100 shifts the processing to S23 to maintain the control mode of the nickel-hydrogen battery 10 in the high SOC mode, and in S24, maintains the weighting coefficient K stored in the memory at a value larger than the initial value Kini. At this time, considering that the flattening of the OCV curve gradually disappears according to the high SOC duration, the ECU 100 gradually returns the weighting factor K to the initial value Kini at a predetermined rate according to the high SOC duration (( (See FIG. 10 below).

When the high SOC continuation time exceeds the predetermined time T2 (YES in S25), it is considered that the flattening of the OCV curve has been completely eliminated, and therefore the ECU 100 shifts the processing to S26 and controls the nickel hydrogen battery 10 in the control mode. Is returned from the high SOC mode to the normal SOC mode. After that, the ECU 100 returns the weighting coefficient K to the initial value Kini in S27.

FIG. 10 is a diagram showing an example of changes in the weighting factor K when the low SOC residence time exceeds the predetermined time T1 under a constant temperature. FIG. 10 shows an example in which the initial value (Kini) of the weighting factor K is "0.5".

Before time t1, the control mode of the nickel-hydrogen battery 10 is the normal SOC mode, and the weighting coefficient K is set to the initial value Kini.

When the low SOC residence time exceeds the predetermined time T1 at time t1, it is considered that the OCV curve is flattened, so the control mode of the nickel-hydrogen battery 10 is changed to the high SOC mode. As a result, charging and discharging are repeated in the high SOC region, so that the flattening of the OCV curve is gradually eliminated.

However, there is a concern that the flattening of the OCV curve is not completely eliminated during the period from the time t1 when the high SOC mode is changed to the time t2 when the predetermined time T2 elapses. Therefore, the weighting factor K (contribution of the first provisional SOCa) is set to the maximum value Kmax larger than the initial value Kini at the beginning when the high SOC mode is changed, and thereafter gradually toward the initial value Kini at a predetermined rate. Be lowered. Accordingly, “1-K”, which is the degree of contribution of the second provisional SOCb to the control SOC, is greater than the initial value (=1-Kini, 0.5 in FIG. 10) at the beginning when the high SOC mode was changed. Is set to a small minimum value "1-Kmax", and thereafter gradually increased toward an initial value (=1-Kini, 0.5 in FIG. 10) at a predetermined rate.

Then, at the time t2 when the high SOC duration time reaches the predetermined time T2, the control mode of the nickel hydrogen battery 10 is returned to the low SOC mode, and the weighting coefficient K is returned to the initial value Kini.

Note that FIG. 10 shows an example in which the rate at which the weighting coefficient K is returned to the initial value Kini is a constant rate (straight line), but when the temperature changes, the rate is made variable while considering the temperature. Good.

As described above, the ECU 100 according to the present embodiment, when the low SOC staying time exceeds the predetermined time T1, than when the low SOC staying time does not exceed the predetermined time T1, the weighting factor K (for the control SOC). The contribution of the first provisional SOCa) is set to a large value. This increases the degree to which the first provisional SOCa that is not affected by the flattening of the OCV curve is reflected in the control SOC, and the second provisional SOCb that is affected by the flattening of the OCV curve is reflected in the control SOC. The degree of being reduced is reduced. As a result, even when the OCV curve is flattened, it is possible to prevent the calculation accuracy of the control SOC from being lowered.

Further, the ECU 100 changes the control mode of the nickel hydrogen battery 10 from the normal SOC mode to the high SOC mode when the low SOC residence time exceeds the predetermined time T1. As a result, the nickel-hydrogen battery 10 is charged until the control SOC increases to a value higher than the predetermined value, and the control SOC is maintained within the high SOC region for the predetermined time T2 thereafter. As a result, when the OCV curve is flattened in the nickel-hydrogen battery 10, the flattening can be eliminated early.

In the above-described embodiment, when the low SOC retention time exceeds the predetermined time T1, the process of changing the control mode of the nickel-hydrogen battery 10 from the normal SOC mode to the high SOC mode (the OCV curve is flattened). Both the processing for early elimination) and the processing for increasing the weighting coefficient K were performed. However, it is also possible to perform only the process of increasing the weighting coefficient K while keeping the control mode of the nickel-hydrogen battery 10 in the normal SOC mode. Even in such a case, it is possible to prevent the calculation accuracy of the control SOC from being lowered due to the flattening of the OCV curve.

Further, in the above-described embodiment, in S22 of FIG. 9, it is determined whether or not the OCV curve is flattened by using the time (low SOC retention time) itself as an index. However, the index used for determining whether the OCV curve is flattened is not limited to the time itself. For example, it is possible to newly provide a standardized evaluation value in consideration of time and temperature, and use this evaluation value as an index to determine whether or not the OCV curve is flattened.

As a new evaluation value, for example, a temperature evaluation value E defined by the following formula is calculated and accumulated for each predetermined period Δt, and when the accumulated temperature evaluation value E exceeds a predetermined amount, the OCV curve It may be determined that flattening has occurred.

E(t+Δt)=E(t)+K·Δt
In the above equation, “t” represents time (hour), “E(t+Δt)” represents the current value of the temperature evaluation value E, and “E(t)” represents the previous value of the temperature evaluation value E. “K” is a flattening point at each temperature.

Further, “K” may have a fluctuation range dependency of SOC. By introducing such an evaluation value E, it is possible to more effectively carry out the present control while the vehicle is traveling in which the temperature greatly changes.

Further, as described above, the elimination speed of the flattening of the OCV curve depends on the battery temperature TB and the SOC fluctuation range. In view of this point, in S25 of FIG. 9, an evaluation index similar to the above-described evaluation value E is newly provided, and it is determined whether or not to return from the high SOC mode to the normal SOC mode by using this evaluation index. May be. By doing so, it is possible to more efficiently determine whether or not to return to the normal SOC mode.

The embodiments disclosed this time are to be considered as illustrative in all points and not restrictive. The scope of the present invention is shown not by the above description but by the claims, and is intended to include meanings equivalent to the claims and all modifications within the scope.

1 vehicle, 2 battery system, 10 nickel hydrogen battery, 20 monitoring unit, 21 voltage sensor, 22 current sensor, 23 temperature sensor, 30 PCU, 41, 42 MG, 50 engine, 60 power split mechanism, 70 drive shaft, 80 drive Wheel, 100 ECU.

Claims (1)

Ni-MH battery,
Using the first stored electricity amount calculated using the integrated current value of the nickel-hydrogen battery and the second stored electricity amount calculated using the voltage of the nickel-hydrogen battery, the third stored electricity amount of the nickel-hydrogen battery is calculated. And a control unit for calculating,
The control unit has a first contribution degree indicating a degree of the first stored amount being reflected on the third stored amount, and a second contribution indicating a degree of the second stored amount being reflected on the third stored amount. The weighted degree, the third charged amount which is a weighted average of the first charged amount and the second charged amount is calculated,
The control unit is an OCV curve indicating the value of the open-circuit voltage (OCV) with respect to the charged amount of the nickel-hydrogen battery, when the time during which the third charged amount is staying in a region lower than the specified value exceeds a specified time. It is determined that the “flattening of the OCV curve”, in which the slope of is flatter than that of a new product, occurs, and the first contribution degree is made larger than when the staying time does not exceed the predetermined time. Yes, the battery system.

Images