JP6834397B2 - Battery monitoring system - Google Patents

Description

The present invention relates to a battery monitoring system that monitors the capacity of a plurality of battery blocks.

Hybrid vehicles and electric vehicles that use a rotating electric machine as a drive source are equipped with a battery pack (battery module) that is a power source. The battery pack is composed of, for example, a plurality of battery blocks connected in series. The battery block is an assembly of batteries in which k (for example, k = 1 to 12) battery cells (cells) are connected in series.

In obtaining the SOC [%] which is the remaining capacity of the battery block, for example, in Patent Document 1, the SOC (SOC_I) based on the current integration is corrected by the SOC (SOC_V) based on the open circuit voltage. For example, the value obtained by multiplying the difference dSOC between SOC_V and SOC_I by the correction coefficient R is added to SOC_I (SOC = SOC_I + dSOC × R).

It is known that the accuracy of SOC_V based on the open circuit voltage, that is, the degree of agreement with the true value (true SOC) changes depending on the value. For example, the accuracy of SOC_V is relatively high in the region where the SOC is α (for example, 30%) or less and the region where the SOC is β (for example, 80%) or more, and the accuracy of SOC_V is relatively low in the region where α <SOC_V <β. It is known.

Therefore, in Patent Document 1, the correction coefficient R is changed according to the value of SOC_V. For example, in the case of SOC_V ≦ α and β ≦ SOC_V, the accuracy is high, so the correction coefficient is set relatively high (closer to 1.0). On the other hand, when α <SOC_V <β, the accuracy is low, so the correction coefficient is set relatively low (closer to 0). By doing so, for example, as shown after the time t1 in FIG. 7, the value of SOC can be matched (matched) with SOC_V in the region where the accuracy of SOC_V is high.

Japanese Unexamined Patent Publication No. 2005-114646

By the way, when SOC correction is performed by SOC_V for each battery block in the battery pack, the SOC variation (ΔSOC) of each battery block may increase due to the difference in the correction coefficient R accompanying the variation of the SOC_V of each battery block. There is.

Generally, the SOC and SOC_V of a battery block vary depending on individual differences such as internal resistance and temperature history. FIG. 8 illustrates the maximum SOC_MaxBl and the minimum SOC_MinBl of the SOCs of the plurality of battery blocks, and the open circuit voltage-based SOC_V_MaxBl and SOC_V_MinBl of both.

Here, when only the former of SOC_V_MaxBl and SOC_V_MinBl enters the region of β or more, only the correction coefficient R for SOC_V_MaxBl is raised, and as a result, the variation ΔSOC of SOC_MaxBl and SOC_MinBl increases as seen after time t1. ..

As a result, as shown in the lower part of FIG. 8, the difference E between the ΔSOC estimated value obtained from the difference between SOC_MaxBl and SOC_MinBl and the ΔSOC true value shown by the one-point difference line in the lower part of FIG. 8 is the difference E before the correction coefficient R is raised. It may expand from E1 to E2 after raising the correction coefficient R, and the accuracy of so-called variation estimation may decrease.

Therefore, an object of the present invention is to provide a battery monitoring system capable of suppressing an increase in SOC variation due to a difference in correction coefficient R applied to each SOC_V.

The battery monitoring system according to the present invention includes a voltage sensor, a current sensor, a current-based SOC calculation unit, a voltage-based SOC calculation unit, a correction coefficient setting unit, an SOC estimation unit, and a correction coefficient changing unit. The voltage sensor measures the voltage of a plurality of battery blocks. The current sensor measures the current of a plurality of battery blocks. The current-based SOC calculation unit calculates the current-based SOC based on the integrated value of the current for each battery block. The voltage-based SOC calculation unit calculates a voltage-based SOC based on the voltage for each battery block. The correction coefficient setting unit sets the first correction coefficient when the voltage-based SOC is included in the first region, and the voltage-based SOC is set in the second region where the accuracy of SOC estimation is higher than that of the first region. When included, a second correction factor higher than the first correction factor is set. The SOC estimation unit obtains an estimated SOC based on the first or second correction coefficient, the current-based SOC, and the voltage-based SOC for each battery block. When each voltage-based SOC is distributed over the first region and the second region, the correction coefficient changing unit sets all the correction coefficients set for each battery block as the second correction coefficient. Reset.

According to the present invention, when each voltage-based SOC is distributed over the first region and the second region, the correction coefficient is aligned with the second correction coefficient for all battery blocks. As a result, the expansion of SOC variation is suppressed.

It is a figure which illustrates the structure of the battery monitoring system which concerns on this embodiment. It is a figure which illustrates the functional block of a control part. It is a figure which illustrates the correction coefficient map. It is a figure explaining the mode of the distribution of SOC_V_k. 6 is a time chart showing a change in SOC when the correction coefficient according to the present embodiment is reset. It is a figure explaining the correction coefficient resetting flow. It is a figure explaining the conventional SOC estimation process. It is a figure explaining the conventional SOC estimation process, and is the figure which shows the example which the accuracy of ΔSOC decreases with variation of a correction coefficient.

FIG. 1 illustrates the configuration of the battery monitoring system according to the present embodiment. In addition, in order to simplify the illustration, the illustration of the configuration having low relevance to the battery monitoring system according to the present embodiment is omitted as appropriate in FIG. The arrow lines represent signal lines.

The battery monitoring system shown in FIG. 1 is mounted on a vehicle powered by a rotating electric machine, such as a hybrid vehicle, a plug-in hybrid vehicle, and an electric vehicle. In this vehicle, electric power is supplied from the battery pack 10 (main battery) to a load such as a rotary electric machine as a drive source.

The battery pack 10 includes a plurality of battery blocks 12_1 ... 12_n connected in series. The battery blocks 12_1 ... 12_n are an assembly in which i (for example, i = 1 to 12) battery cells 14 (cells) are connected in series. The battery cell 14 is composed of, for example, a lithium ion battery or a nickel hydrogen battery. When monitoring the battery status, SOC is managed in units of 12 battery blocks.

The battery monitoring system according to the present embodiment includes a voltage monitoring multiplexer 16 and a battery block voltage sensor 18 as voltage measurement system devices. Further, as the equipment of the temperature measurement system, the thermistors 20_1 ... 20_n, a temperature monitoring multiplexer 22, and a thermistor voltage sensor 24 are provided. The functions of the voltage monitoring multiplexer 16 and the temperature monitoring multiplexer 22 may be integrated into one multiplexer by using a multi-channel multiplexer having a voltage monitoring channel and a temperature measurement channel.

Further, the battery monitoring system according to the present embodiment includes a current sensor 28 that measures the current flowing through the high-voltage bus 26 connected to the battery pack 10. Further, it includes a battery block voltage sensor 18, a thermistor voltage sensor 24, and a control unit 30 that acquires various measured values from the current sensor 28.

The voltage Vb_k of each battery block 12_k (k = 1 to n) is acquired by appropriately switching the channel selection of the voltage monitoring multiplexer 16. Wiring drawn from the positive electrode of each battery block 12_1 to 12_n and the negative electrode of the terminal battery block 12_n are connected to each input side channel of the voltage monitoring multiplexer 16. The battery block voltage sensor 18 is connected to two of the output-side channels. Further, a signal line (switching signal line for channel selection) from the control unit 30 is connected to one of the output side channels.

The control unit 30 transmits a switching signal for selecting two adjacent channels to the voltage monitoring multiplexer 16. For example, channels CH1_Vb and CH2_Vb are selected. At this time, the battery block voltage sensor 18 acquires the voltage Vb_1 of the battery block 12_1 via the channels CH1_Vb and CH2_Vb. In the same manner below, the battery block voltage sensor 18 measures the voltage Vb_k of the battery block 12_k (k = 1 to n) by sequentially switching the channels of the voltage monitoring multiplexer 16.

The temperature Tb_k of each battery block 12_k (k = 1 to n) is acquired by appropriately switching the channel selection of the temperature monitoring multiplexer 22. Wiring connected to one end of each thermistor 20_1 to 20_n is connected to each input side channel of the temperature monitoring multiplexer 22. The other end of each thermistor 20_1 to 20_n is grounded. Further, one end of the thermistor voltage sensor 24 is connected to one of the output-side channels of the temperature monitoring multiplexer 22. The other end of the thermistor voltage sensor 24 is grounded. Further, a signal line (switching signal line for channel selection) from the control unit 30 is connected to one of the output side channels of the temperature monitoring multiplexer 22.

The control unit 30 transmits a switching signal for selecting one channel to the temperature monitoring multiplexer 22. For example, channel CH1_Vt is selected. At this time, the thermistor voltage sensor 24 acquires the voltage Vt_1 between the channel CH1_Vt and the ground. This voltage Vt_1 is a voltage divider of the reference resistor 34 and the thermistor 20_1 when the voltage V0 by the reference power supply 32 is applied. Each thermistor 20_k (k = 1 to n) is attached so as to be in contact with the casing or the like of each battery block 12_k (k = 1 to n), and responds to a temperature change of each battery block 12_k (k = 1 to n). The resistance value of each thermistor 20_k (k = 1 to n) changes. The voltage Vt_k (k = 1 to n) changes accordingly. The thermistor voltage sensor 24 sequentially acquires the voltage Vt_k by switching the channel of the temperature monitoring multiplexer 22. The acquired voltage Vt_k is sent to the control unit 30, and the temperature Tb_k of each battery block 12_k is obtained according to the voltage-temperature map (not shown) provided in the control unit 30.

Further, the control unit 30 acquires the current Ib from the current sensor 28. Since all the battery blocks 12_k are connected in series in the battery pack 10 according to the present embodiment, the current Ib measured by the current sensor 28 is the current value of each battery block 12_k.

The control unit 30 is composed of, for example, a computer, and includes a storage unit such as a CPU and a memory which are arithmetic circuits. The storage unit stores a program that executes SOC estimation, a program that executes a correction coefficient resetting flow described later, and the like. By executing the program, the functional block shown in FIG. 2 is constructed in the control unit 30.

The control unit 30 includes a current-based SOC calculation unit 30A, a voltage-based SOC calculation unit 30B, a correction coefficient setting unit 30C, a correction coefficient changing unit 30D, a voltage correction term calculation unit 30E, and an SOC estimation unit 30F.

The current-based SOC calculation unit 30A calculates the SOC estimated value SOC_I (current-based SOC) of each battery block 12_1 to 12_n based on the integrated value ΣIb dt of the battery current Ib detected by the current sensor 28. SOC_I is calculated by, for example, SOC_I = SOC_I + A × ΣIb dt (A is an arbitrary constant). As described above, the battery blocks 12_1 to 12_n are connected in series, and the current Ib measured by the current sensor 28 is used as their own current. The constant A can be appropriately changed based on the temperature, internal resistance, etc. of each battery block 12_1 to 12_n.

The voltage-based SOC calculation unit 30B calculates the SOC estimated value SOC_V (voltage-based SOC) of each battery block 12_1 to 12_n based on the voltages Vb_1 to Vb_n detected by the battery block voltage sensor 18. When the voltage drop ΔV obtained by multiplying the voltage Vb_1 to Vb_n (closed circuit voltage CCV) detected by the battery block voltage sensor 18 by the internal resistance of each battery block 12_1 to 12_n and the current Ib detected by the current sensor 28 is added, each battery is added. The open circuit voltage OCV of blocks 12_1 to 12_n is obtained (OCV = CCV + ΔV). It is known that there is a correlation between the open circuit voltage OCV and the SOC, and the voltage-based SOC calculation unit 30B obtains the SOC_V corresponding to the obtained OCV from the OCV-SOC map stored in the storage unit of the control unit 30. get.

In calculating the SOC_V shown above, the voltage-based SOC calculation unit 30B acquires the voltages Vb_1 to Vb_n from the battery block voltage sensor 18. Further, in calculating the voltage drop ΔV, the current Ib is acquired from the current sensor 28. Further, in calculating the internal resistance, the voltages Vt_1 to Vt_n are acquired from the thermistor voltage sensor 24, and the temperatures Tb_1 to Tb_n of each battery block 12_1 to 12_n are obtained.

The correction coefficient setting unit 30C obtains the correction coefficient R based on the SOC_V calculated by the voltage-based SOC calculation unit 30B. As described above, the accuracy of SOC_V, that is, the degree of agreement with the true value of SOC, changes depending on the value of SOC_V. That is, the accuracy of SOC_V is relatively high in the high SOC region (for example, 80% or more) and the low SOC region (for example, 30% or less), and in the region between them (for example, 30% <SOC_V <80%), the accuracy of SOC_V is high. The accuracy is relatively low.

Based on the above characteristics, the correction coefficient setting unit 30C stores the correction coefficient map illustrated in FIG. The correction coefficient map has SOC_V on the horizontal axis and correction coefficient on the vertical axis. In this map, if SOC_V is included in a relatively inaccurate region (first region) that exceeds α% (eg 30%) and is less than β% (eg 80% or more), it is relative. A low correction coefficient R1 (first correction coefficient) is set in. On the other hand, when SOC_V is included in a region with relatively high accuracy (second region) of α% (for example, 30%) or less and β% (for example, 80%) or more, a relatively high correction coefficient R2 (for example) The second correction coefficient) is set.

For example, the first correction coefficient R1 is a predetermined constant. Further, the second correction coefficient R2 is set so that when it is α% or less, the value becomes higher as it approaches 0%, and when it is β% or more, it becomes higher as it approaches 100%. For example, the second correction coefficient R2 is represented by a function R2 (SOC_V) having SOC_V as a variable.

The voltage correction term calculation unit 30E calculates a voltage correction term for correcting SOC_I. Specifically, the difference dSOC_1 to dSOC_n between SOC_V_1 to SOC_V_n and SOC_I_1 to SOC_I_n for each battery block 12_1 to 12_n is obtained. Further, the difference dSOC_1 to dSOC_n is multiplied by the correction coefficients R_1 to R_n to obtain a voltage correction term.

The correction coefficients R_1 to R_n are transmitted from the correction coefficient setting unit 30C to the voltage correction term calculation unit 30E in consideration of resetting the correction coefficients R_1 to R_n by the correction coefficient changing unit 30D described later. For example, instead of sequentially transmitting the obtained correction coefficient R_k to the voltage correction term calculation unit 30E, the correction coefficient setting unit 30C obtains all the correction coefficients R_1 to R_n, and then the set of the correction coefficients R_1 to R_n is voltage-corrected. It is transmitted to the coefficient calculation unit 30E.

The SOC estimation unit 30F obtains an SOC estimated value from SOC_I and the voltage correction term. Specifically, the SOC estimated value SOC_k is obtained by SOC_k = SOC_I_k + dSOC_k × R_k. The lower limit of the discharge amount and the upper limit of the charge amount are determined based on the obtained SOC estimated value.

The correction coefficient changing unit 30D resets the correction coefficients R_1 to R_n set by the correction coefficient setting unit 30C according to the distribution of SOC_V_1 to SOC_V_n. FIG. 4 shows an example of setting the correction coefficients R_1 to R_n according to the distribution of SOC_V_1 to SOC_V_n. The x mark in FIG. 4 represents one value (plot) of SOC_V_k, and in FIG. 4, it is SOC_V_k (k = 1 to 5).

As described above, the correction coefficients R_1 to R_n vary greatly depending on whether SOC_V is in the first region (α <SOC_V <β) or in the second region (SOC_V ≦ α, β ≦ SOC_V). Therefore, as shown in Cases 2 and 4 of FIG. 4, if the SOC_V_1 to SOC_V_n of the battery blocks 12_1 to 12_n are distributed over the first region and the second region, the difference between the correction coefficients R_1 to R_n is large. As a result, the variation ΔSOC of the SOC estimated value may be expanded.

Therefore, the correction coefficient changing unit 30D is assigned to the SOC_V of β% or more when only a part of SOV_V is β% or more, that is, when SOC_V_k is distributed over β% as in case 2 of FIG. A resetting command for setting (resetting) the second correction coefficient R2 to be performed for all other SOC_Vs is transmitted to the correction coefficient setting unit 30C.

Further, the correction coefficient changing unit 30D is assigned to the SOC_V of α% or less when only a part of SOV_V is α% or less, that is, when SOC_V_k is distributed over α% as in case 4 of FIG. A reset command for setting (resetting) the second correction coefficient R2 to be performed for all other SOC_Vs is transmitted to the correction coefficient setting unit 30C.

When there are a plurality of SOC_Vs having β% or more or α% or less, and thus a plurality of second correction coefficients R2 exist, an arbitrary value is selected from them. For example, the maximum or minimum of the plurality of second correction coefficients R2 is selected, and all the other correction coefficients are updated (reset) to the selected second correction coefficient R2.

As described above, in the present embodiment, when SOC_V_1 to SOC_V_n are distributed across the first region and the second region, the correction coefficient R_1 to R_n given corresponding to SOC_V_1 to SOC_V_n is set to the second correction coefficient R2. Align to. By doing so, as shown after the time t1 in FIG. 5, the variation ΔSOC of the SOC estimated value is suppressed from expanding.

FIG. 6 illustrates a correction coefficient resetting flow by the correction coefficient changing unit 30D. The correction coefficient changing unit 30D sequentially acquires the SOC_V_k (k = 1 to n) obtained by the voltage-based SOC calculation unit 30B, and SOC_V_MinBl, which takes the minimum value among these, is the boundary between the first region and the second region. It is determined whether or not the value (lower boundary value) is α% or less (S10).

When SOC_V_MinBl is equal to or less than the lower boundary value α%, the distribution of SOC_V_k is either Case 4 or Case 5 in FIG. When SOC_V_MinBl ≦ α% in step S10, the correction coefficient changing unit 30D determines whether or not SOC_V_MaxBl, which takes the maximum value among SOC_V_k (k = 1 to n), is equal to or less than the lower boundary value α% ( S12).

If SOC_V_MaxBl exceeds the lower boundary value α%, the distribution of SOC_V_k is Case 4. In this case, the correction coefficient changing unit 30D instructs the correction coefficient setting unit 30C to reset the correction coefficient (S14). For example, the correction coefficient changing unit 30D gives an instruction to reset the correction coefficients R_1 to R_n obtained corresponding to SOC_V_1 to SOC_V_n, and the reset correction coefficient is a correction coefficient corresponding to SOC_V_MinBl (second correction coefficient). ) Is transmitted to the correction coefficient setting unit 30C. After resetting by the correction coefficient setting unit 30C, the SOC estimated value is calculated (S16).

When SOC_V_MaxBl is equal to or less than the lower boundary value α%, the distribution of SOC_V_k is Case 5. In this case, since all SOC_V_k are contained in the second region, the correction coefficient changing unit 30D does not reset the correction coefficients R_1 to R_n. Therefore, the correction coefficients R_1 to R_n corresponding to the respective SOC_V_k are maintained (S22). After that, the SOC estimated value is calculated (S16).

Returning to step S10, when SOC_V_MinBl exceeds the lower boundary value α, the distribution of SOC_V_k becomes one of Case 1, Case 2, and Case 3. Next, the correction coefficient changing unit 30D determines whether or not SOC_V_MaxBl, which takes the maximum value of SOC_V_k (k = 1 to n), is the upper boundary value β% or more (S18).

When SOC_V_MaxBl is less than the upper boundary value β%, the distribution of SOC_V_k is Case 1. In this case, since all SOC_V_k are contained in the first region, the correction coefficient changing unit 30D does not reset the correction coefficients R_1 to R_n. Therefore, the correction coefficients R_1 to R_n corresponding to the respective SOC_V_k are maintained (S22). After that, the SOC estimated value is calculated (S16).

When SOC_V_MaxBl is equal to or greater than the upper boundary value β%, the distribution of SOC_V_k is Case 2 or Case 3. The correction coefficient changing unit 30D determines whether or not the minimum value SOC_V_MinBl of SOC_V is the upper boundary value β% or more (S20).

When SOC_V_MinBl is the upper boundary value β% or more, the distribution of SOC_V_k is Case 3. Since all the SOC_V_k are contained in the second region, the correction coefficient changing unit 30D does not reset the correction coefficients R_1 to R_n. Therefore, the correction coefficients R_1 to R_n corresponding to the respective SOC_V_k are maintained (S22).

When SOC_V_MinBl is less than the upper boundary value β%, the distribution of SOC_V_k is Case 2. In this case, the correction coefficient changing unit 30D instructs the correction coefficient setting unit 30C to reset the correction coefficient (S24). For example, the correction coefficient changing unit 30D gives an instruction to reset the correction coefficients R_1 to R_n obtained corresponding to SOC_V_1 to SOC_V_n, and the reset correction coefficient is a correction coefficient corresponding to SOC_V_MaxBl (second correction coefficient). ) Is transmitted to the correction coefficient setting unit 30C. After resetting, the SOC estimate is calculated (S16).

10 Battery pack, 12 Battery block, 14 Battery cell, 16 Voltage monitoring multiplexer, 18 Battery block voltage sensor, 20 Thermista, 22 Temperature monitoring multiplexer, 24 Thermista voltage sensor, 28 Current sensor, 30 Control unit, 30A Current base SOC Calculation unit, 30B voltage-based SOC calculation unit, 30C correction coefficient setting unit, 30D correction coefficient change unit, 30E voltage correction term calculation unit, 30F SOC estimation unit.

Claims (1)

A voltage sensor that measures the voltage of each battery block of a plurality of battery blocks connected in series, and
A current sensor that measures the current flowing through the bus connected to the battery pack in which the plurality of battery blocks are connected in series, and
For each of the battery blocks, a current-based SOC calculation unit that calculates a current-based SOC based on the integrated value of the current, and a current-based SOC calculation unit.
A voltage-based SOC calculation unit that calculates a voltage-based SOC based on the voltage for each of the battery blocks,
When the voltage-based SOC is included in the first region, the first correction coefficient is set, and when the voltage-based SOC is included in the second region where the accuracy of SOC estimation is higher than that of the first region. In addition, a correction coefficient setting unit that sets a second correction coefficient higher than the first correction coefficient,
Each battery block obtains an estimated SOC obtained by adding a value obtained by multiplying the difference between the voltage-based SOC and the current-based SOC by the first correction coefficient or the second correction coefficient to the current-based SOC. SOC estimation unit to be calculated for each
With
When each of the voltage-based SOCs is distributed over the first region and the second region, when the estimated SOC is obtained , all the correction coefficients set for each of the battery blocks are used. A battery monitoring system including a correction coefficient changing unit that resets to a second correction coefficient.