JP2015227840A - Battery charge state estimation method and battery charge state estimation device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To accurately estimate a battery charge state.SOLUTION: An error estimation unit 13 acquires an integration method charging state estimation error SOC_I, an OCV method charging state estimation error SOC_V, and an error average thereof SOC_VI. A multiplexer 15 selects one from an integration method estimation charging state SOC_I, an OCV method estimation charging state SOC_V, or an estimation charging state average value SOC_VI on the basis of the integration method charging state estimation error SOC_I, the OCV method charging state estimation error SOC_V, and the error average thereof SOC_VIacquired by the error estimation unit 13, and outputs the selected estimation charging state as a charging state of a battery BAT.

Description

  The present invention relates to a battery charge state estimation method and a battery charge state estimation device for estimating a state of charge (SOC) of a battery used in a hybrid electric vehicle (HEV) or the like.

  In recent years, hybrid vehicles and electric vehicles using a drive motor driven by electric power as one of the power sources have been widely developed. As a power source for such a drive motor, a battery composed of a large-capacity secondary battery is used. The battery mounted on the hybrid vehicle is charged when the drive motor regenerates power during traveling or by power generation accompanying the rotation of the engine. On the other hand, a battery mounted on an electric vehicle or a plug-in hybrid vehicle is charged from an external power source such as a household commercial power source.

  In the electric vehicle as described above, accurately grasping the SOC indicating the state of charge of the battery is very important for preventing the battery from being overcharged and overdischarged and for accurately communicating the remaining capacity of the battery during traveling to the passenger. Is important to. However, it is difficult to directly detect the state inside the battery.

  Conventionally, various methods are used for estimating the SOC of a battery. For example, the charge / discharge current of the battery is integrated over time to calculate the integrated charge amount and integrated discharge amount, and these integrated charge amount and integrated discharge amount are added to or subtracted from the initial state or the charge amount immediately before the start of charge / discharge. Input the battery input current value and terminal voltage value into the battery charge calculation method (hereinafter referred to as the current integration method) and the battery equivalent circuit model, and sequentially estimate the open circuit voltage of the model. Thus, there is a method for estimating the SOC of the battery from the open circuit voltage (hereinafter referred to as the OCV method).

  However, each of the above methods has advantages and disadvantages.

  In the latter OCV method, the open circuit voltage is calculated from the voltage / current / temperature / battery model. Therefore, the error in the estimated SOC value increases in the current region in the charged or discharged state due to the model error. There is no accumulation of offset current error, which is the sum of errors from the start of discharge.

  On the other hand, since the former current integration method does not require a battery model, the error does not increase even in a large current region, but there is a problem that the offset current error is accumulated with time and the SOC estimation error is deteriorated.

  The latter OCV method does not require constant measurement of charge current and discharge current as in the former current integration method, but an equivalent circuit model of the battery is used to calculate the OCV. There is a problem that an error occurs. This error is particularly noticeable when a large current (for example, 100 A to 200 A) flows.

  On the other hand, although the former current integration method is more accurate than the OCV method in short-term SOC estimation, it requires constant measurement of charging current and discharging current, and further, a current sensor for measuring an input current value. Since there is an error such as a gain error, there is a problem that the error due to the current sensor is integrated with the lapse of time from the start of charge / discharge of the battery, and the SOC estimation accuracy deteriorates.

  In Patent Document 1, for the purpose of calculating such an error of the current sensor, the offset of the current sensor is based on the battery capacity difference between the battery capacity obtained from the integrated current value and the battery capacity obtained from the open circuit voltage. A method for calculating an error and a gain error is disclosed.

JP-A-2005-261130

  By the way, in an electric vehicle equipped with a driving high voltage battery for driving the vehicle, it is necessary to electrically insulate the high voltage power supply system from the vehicle body in order to ensure safety. Sensors are commonly used. However, in the Hall element type current sensor, an offset current error occurs. Therefore, in the SOC estimation using the above-described integration method, the offset current error is accumulated with the lapse of integration time, and the SOC estimation accuracy may be deteriorated. is there.

  Further, the offset current error is estimated to be about 2A, considering that the error is 1% F.S. and the maximum current is about 200A. It can be seen that the SOC estimation result by the long-time current integration method cannot be used in a hybrid vehicle using a battery capacity of about 3 [Ah].

  On the other hand, in the OCV method, the SOC estimation error increases in the current region, particularly in the region where a large current flows, due to the battery model error.

  As described above, in the OCV method, the current integration method is performed in the large current charge / discharge section where the error is large, and the OCV method estimation result is obtained in the section where the integration time increases or the charge / discharge current decreases and the OCV method accuracy can be expected. The logic to choose is qualitatively said. However, since it is difficult to accurately estimate the offset error, it is impossible to correctly estimate the two estimation method errors.

  The present invention has been made in view of the above point, and cancels the offset error component by utilizing the fact that the signs of the current integration method error and the OCV method error are contrary to the offset error. An object of the present invention is to accurately estimate the state of charge of a battery without the need to accurately estimate an offset current value.

  The offset error component is canceled by utilizing the fact that the signs of the current integration method error and the OCV method error are opposite to the offset error. According to this method, it is not necessary to accurately estimate the offset current value.

  Specifically, the first invention is directed to a battery charge state estimation method for estimating the charge state of the battery based on the charge / discharge current and terminal voltage of the battery, and has taken the following solution.

  In other words, according to the first aspect of the present invention, integration method charging is performed in which the charge / discharge current of the battery detected by the current sensor is integrated over time, and the integration method estimation charge state, which is the state of charge of the battery, is estimated based on the time integration value. Obtaining an open circuit voltage of the battery based on a state estimation step, a terminal voltage of the battery detected by the voltage sensor, a charge / discharge current detected by the current sensor, and an equivalent circuit model of the battery; An OCV method charge state estimation step for estimating an OCV method estimated charge state that is a charge state of the battery from a voltage, and a time integration from the start of charge / discharge of the battery of an error of the integration method estimated charge state caused by the current sensor Charge and discharge start of the battery due to the error of the integration method that is the value and the error of the charged state of the OCV method estimated due to the equivalent circuit model An error acquisition step for obtaining an OCV method error, which is a time change, and the integration method estimated charge state or the OCV method estimated charge state are selected based on the integration method error and the OCV method error obtained in the error acquisition step. And a charging state selection step of setting the selected estimated charging state to the charging state of the battery.

  According to the first invention, the state of charge of the battery is estimated by the current integration method and the OCV method, and the time change from the start of battery charge / discharge in both methods is obtained. When obtaining the time variation of the error in the current integration method, the time variation of the error caused by the current sensor, which is a dominant factor, is obtained. On the other hand, when obtaining the time variation of the error in the OCV method, the time variation of the error caused by the equivalent circuit model of the battery, which is the main factor, is obtained. Therefore, it is possible to accurately obtain the time change of the error in each estimation method. Then, the estimated charging state by the smaller estimation method among the obtained time variations of the error is selected and set as the charging state of the battery. As described above, the time variation of the error by each estimation method is obtained with high accuracy, and the estimated charge state by the estimation method with a small error is selected as the battery charge state, so the battery charge state is estimated with higher accuracy than before. can do.

  According to a second invention, in the first invention, the method further comprises an average value obtaining step for obtaining an estimated charge state average value that is an average value of the integration method estimated charge state and the OCV method estimated charge state, and in the error obtaining step, An error average which is a time change of the average value of the integration method error and the OCV method error is further obtained, and the integration method estimated charging state is determined based on the integration method error, the OCV method error and the error average in the charging state selection step. OCV method estimated charge state or estimated charge state average value is selected.

  Here, if attention is paid to the sign of the integration method error and the OCV method error being opposite (positive or negative), the average of these errors may be the smallest than the integration method error or the OCV method error.

  Therefore, according to the second invention, in addition to the current integration method and the OCV method, an estimated charge state average value that is an average value of the charge states estimated by both methods is obtained, and further, the integration method error and the OCV method error are obtained. In addition to the above, an error average which is an average of these is obtained. Then, an estimated charge state is selected by the method of the smallest error among these three error time variations. Therefore, the state of charge of the battery can be estimated with higher accuracy.

  According to a third invention, in the first or second invention, the charge state selection step selects the battery based on an elapsed time from the start of charge / discharge of the battery.

  According to the third aspect, since the estimated charge state is selected based on the elapsed time from the start of charge / discharge of the battery, the charge state of the battery can be estimated easily and with relatively high accuracy.

  According to a fourth invention, in any one of the first to third inventions, in the charge state selection step, the integration method estimated charge state is selected from a start of charge / discharge of the battery to a lapse of a predetermined time. It is characterized by.

  According to the fourth aspect of the invention, the integration method error is the time integration of the sensor error of the current sensor. That is, the integration method error is the smallest at the start of charging / discharging (current integration) of the battery. Therefore, from the start of charge / discharge (current integration) of the battery to the elapse of a predetermined time, the state of charge of the battery can be estimated with higher accuracy by selecting the estimated charge state by the current integration method with the smallest error.

  The fifth aspect of the present invention is directed to a battery state-of-charge estimation device that estimates the state of charge of a battery based on the terminal voltage and charge / discharge current of the battery, and has taken the following solution.

  That is, the fifth aspect of the present invention integrates the current sensor for detecting the charging / discharging current of the battery, the voltage sensor for detecting the terminal voltage of the battery, and the charging / discharging current detected by the current sensor. An integration method charging state estimation means for estimating an integration method estimated charging state which is a charging state of the battery based on an integration value, a terminal voltage detected by the voltage sensor, a charging / discharging current detected by the current sensor, and the above An open circuit voltage is obtained based on the equivalent circuit model of the battery, and an OCV method charge state estimating means for estimating an OCV method estimated charge state that is a charge state of the battery from the open circuit voltage; Accumulation method estimation charge state error is an integration method error, which is the time integration value from the start of charging / discharging of the battery, and the equivalent circuit model. Based on the error obtaining means for obtaining the OCV method error, which is a time change from the start of charging / discharging of the battery, and the integration method error and the OCV method error obtained by the error obtaining means. Charging state selecting means for selecting the estimated charging state or the OCV method estimated charging state and setting the selected estimated charging state to the charging state of the battery.

  According to the fifth aspect, the same operational effects as the first aspect can be achieved.

  As described above, according to the present invention, the state of charge of the battery can be accurately estimated.

It is a block diagram which shows the structure of the battery charge condition estimation apparatus which concerns on embodiment of this invention. (A) is a block diagram of the control of the current integration method charging state estimation unit, (b) is a block diagram of the control of the OCV method charging state estimation unit, (c) is a diagram showing an equivalent circuit model of the battery. is there. (A) is a control block diagram for calculating an integration method charging state estimation error, (b) is a control block diagram for calculating an OCV method charging state estimation error, and (c) is a battery It is a figure which shows an equivalent circuit model. It is a figure which shows the control flow of a charge condition estimation apparatus. It is a figure which shows the relationship between an integration method charging state estimation error and OCV method charging state estimation error, (a) is a figure which shows the case where OCV method charging state estimation error is comparatively large, (b) is OCV method It is a figure which shows the case where a charging condition estimation error is comparatively small. It is a figure which shows an example of the time transition of the charge condition estimation error in each method of a battery. It is a block diagram which shows the structure of the charge condition estimation apparatus which concerns on the modification of embodiment of this invention.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. In addition, the following description of preferable embodiment is only an illustration essentially, and is not intending restrict | limiting this invention, its application thing, or its use.

  FIG. 1 is a block diagram illustrating a configuration of a battery charge state estimation device 1 (hereinafter simply referred to as a charge state estimation device 1) according to the present embodiment. The charging state estimation device 1 is a device that estimates the charging state of a battery, particularly a battery BAT mounted on a HEV. The battery BAT is composed of a secondary battery such as a lead storage battery and is electrically insulated from the vehicle body.

The charging state estimation device 1 includes a current sensor 3 and a voltage sensor 5 that detect a charging / discharging current I and a terminal voltage V _BAT of the battery BAT, respectively, and a current integration method charging state that estimates a charging state of the battery BAT by a current integration method. An estimation unit 7 (accumulation method charging state estimation unit), an OCV method charging state estimation unit 9 (OCV method charging state estimation unit) that estimates the charging state of the battery BAT by the OCV method, and a current integration method charging state estimation unit 7 The estimated state of charge of the battery BAT (hereinafter referred to as the integration method estimated state of charge SOC_I) and the state of charge of the battery BAT estimated by the OCV method state of charge estimation unit 9 (hereinafter referred to as the OCV method estimated state of charge SOC_V). An average value calculation unit 11 that calculates an estimated charge state average value SOC_VI that is an average value, and estimates an error of each of the above estimation methods The error estimation unit 13 (error acquisition means) and the integration method estimated charge state SOC_I, OCV method estimated charge state SOC_V, or estimated charge state average value SOC_VI are selected based on the error estimated by the error estimate unit 13 And a multiplexer 15 for charging the battery BAT.

  The current sensor 3 is a Hall element type because the battery BAT is electrically insulated from the vehicle body as described above. This current sensor 3 is connected in parallel with a current integration method charging state estimation unit 7, an OCV method charging state estimation unit 9 and an error estimation unit 13, and transmits a signal indicating the detected charging / discharging current I to these units. The voltage sensor 5 is connected to an OCV method charging state estimation unit 9, and transmits a signal indicating the detected terminal voltage to the OCV method charging state estimation unit 9.

  The current integration method charging state estimation unit 7 calculates the integration method estimation charging state SOC_I based on the signal transmitted from the current sensor 3. FIG. 2A is a block diagram of the control of the current integration method charging state estimation unit 7. Based on the charge / discharge current value I detected by the current sensor 3 and the known battery capacity Cap of the battery BAT, the current integration method charging state estimation unit 7 calculates the time integration value ΔSOC_I (ΔSOC_I = (Idt / Cap) is obtained. On the other hand, the current integration method charging state estimation unit 7 acquires the charging state initial value SOC0 of the battery BAT from the OCV method charging state estimation unit 9. Then, current integration method charging state estimation unit 7 adds time integration value ΔSOC_I to charging state initial value SOC0 to obtain integration method estimated charging state SOC_I of battery BAT, and outputs the result to multiplexer 15.

The OCV method charging state estimation unit 9 calculates the OCV method estimated charging state SOC_V based on the signals transmitted from the current sensor 3 and the voltage sensor 5. FIG. 2B is a block diagram of the control of the OCV method charging state estimation unit 9. In the OCV method charging state estimation unit 9, an equivalent circuit model of the battery BAT shown in FIG. 2C is set in advance, and the resistance value R of the battery BAT set by the equivalent circuit model, the current sensor 3, and Based on the charge / discharge current value I and the terminal voltage value V _BAT detected by the voltage sensor 5, the open circuit voltage OCV (OCV = V _BAT −IR) of the battery BAT is calculated. Here, the OCV method charging state estimation unit 9 stores an OCV-SOC map indicating the correspondence between the open circuit voltage OCV and the charging state SOC of the battery BAT. The OCV method charge state estimation unit 9 refers to this map, acquires the OCV method estimated charge state SOC_V of the battery BAT corresponding to the calculated open circuit voltage OCV, and outputs it to the multiplexer 15.

  The average value calculation unit 11 is connected to the current integration method charging state estimation unit 7 and the OCV method charging state estimation unit 9, and acquires the integration method estimation charging state SOC_I and the OCV method estimation charging state SOC_V. Then, the average value calculation unit 11 calculates an estimated charge state average value SOC_VI (SOC_VI = (SOC_V + SOC_I) / 2), which is an average value of the estimated charge states SOC_I and SOC_V, and outputs the calculated value to the multiplexer 15.

The error estimation unit 13 includes an integration method charging state estimation error SOC_I _err , which is a time integration value from the start of charging / discharging of the battery BAT, due to an error in the integration method estimated charging state SOC_I caused by the offset error of the current sensor 3, and FIG. The OCV method charge state estimation error SOC_V _err , which is a time change from the start of charge / discharge of the battery BAT, due to the error of the OCV method estimated charge state SOC_V caused by the equivalent circuit model of the battery BAT shown in (c), and both of these charges An error average SOC_VI _err that is an average value of the state estimation errors is obtained.

FIG. 3A shows a block diagram of control for calculating the integration method charging state estimation error SOC_I _err . The error estimation unit 13 obtains an integration method charging state estimation error SOC_I _err (SOC_I _err = ∫I _err dt / Cap) from the time integration value of the current error I _err caused by the offset error of the current sensor 3.

On the other hand, FIG. 3B shows a block diagram of control for calculating the OCV method charge state estimation error SOC_V _err . In the error estimation unit 13, an equivalent circuit model of the battery BAT shown in FIG. The resistance in the equivalent circuit model is represented by the sum of the resistance true value R _ref and the resistance error R _err , while the charge / discharge current is the current error I due to the current true value I _ref and the offset error of the current sensor 3. It is expressed as the sum of _err .

In this equivalent circuit model, the error estimation unit 13 calculates the open circuit voltage error OCV _err (OCV _err = {V _BAT − (I _ref + I _err ) (R + R _err )} − (V _BAT −I _ref · R) = − I _err · R _ref + (I _ref + I _err ) R _err ) is calculated. Since R _err can take a positive or negative value, the open circuit voltage error OCV _err is −I _err · R _ref − (I _ref + I _err ) | R _err | ≦ open circuit error OCV _err ≦ −I _err · R _ref + (I _ref + I _err ) | R _err | Here, | R _err | is an absolute value of R _err .

Further, the error estimating unit 13 obtains the OCV method estimated charging state SOC_V from the OCV method charging state estimating unit 9, and refers to the OCV-SOC map to determine the slope of OCV in the OCV method estimated charging state SOC_V (dOCV / dSOC). ) | Obtain the _SOC . Then, the error estimation unit 13 calculates the OCV method charging state estimation error SOC_V _err (SOC_V _err = (dOCV / dSOC) OCV _err ) based on the open circuit voltage error OCV _err and the slope (dOCV / dSOC) | _SOC. Is calculated.

Further, the error estimation unit 13 calculates an error average SOC_VI _err (SOC_VI _err = (SOC_I _err + SOC_V _err ) / 2) of the integration method charging state estimation error SOC_I _err and the OCV method charging state estimation error SOC_V _err .

Then, the error estimating unit 13 outputs the calculated integration method charging state estimation error SOC_I _err , OCV method charging state estimation error SOC_V _err and error average SOC_VI _err to the multiplexer 15.

The multiplexer 15 receives one of the inputs from the current integration method charging state estimation unit 7, the OCV method charging state estimation unit 9, and the average value calculation unit 11 based on the selection control input from the error estimation unit 13. Output as an estimated charge state of the battery BAT. Specifically, the multiplexer 15 determines the magnitude relationship among the integration method charging state estimation error SOC_I _err , the OCV method charging state estimation error SOC_V _err and the error average SOC_VI _err input from the error estimation unit 13, and sets the smallest error. The charging state estimated by the corresponding method is selected, and the selected estimated charging state is output as the charging state of the battery BAT.

  Next, the operation of the charging state estimation device 1 will be described with reference to the flowchart shown in FIG.

When the charging / discharging of the battery BAT starts, the charging state estimation device 1 acquires the estimated charging state previous value SOC ₍₋₁₎ estimated by the OCV method charging state estimation unit 9 at the end of the previous charging / discharging of the battery BAT, Referring to the OCV-SOC map of the OCV method charging state estimation unit 9, the slope of the OCV with respect to the SOC at the estimated charging state previous value SOC _(-1) (dOCV / dSOC) | _SOC is calculated (step S1).

Next, the charging state estimation device 1 calculates an open circuit voltage error OCV _err based on the charging / discharging current I and the terminal voltage V _BAT detected by the current sensor 3 and the voltage sensor 5, respectively, and an upper limit OCV _err U (OCV _err U = −I _err · R _ref + (I _ref + I _err ) | R _err |) and lower limit OCV _err L (OCV _err L = −I _err · R _ref − (I _ref + I _err ) | R _err |) Is calculated (step S3). At this time, it is assumed that the current error I _err resulting from the offset error of the current sensor 3 is I _err > 0.

Subsequently, the state-of-charge estimation device 1 determines the OCV slope (dOCV / dSOC) | _SOC with respect to the SOC ₍₋₁₎ calculated in step S1 and the open circuit voltage error OCV _err calculated in step S3. Then, an OCV method charge state estimation error SOC_V _err is calculated (step S5).

On the other hand, the charging state estimation device 1 determines whether or not the upper limit value OCV _err U of the open circuit voltage error OCV _err calculated in step S3 is smaller than 0 (step S7).

When the charging / discharging current I of the battery BAT is relatively large, when the charging rate of the battery BAT is relatively small, or when both conditions are satisfied, the upper limit value OCV _err U of the open circuit voltage error OCV _err is There is a high possibility of 0 or more (NO in step S7). In this case, the device for estimating charged state 1 does not select the error mean SOC_VI _err, the absolute value of the integration method for estimating charged state error SOC_I _err | SOC_I _err | is OCV method for estimating charged state error SOC_V _err lower limit _{OCV err} L It is determined whether or not the absolute value is smaller than | OCV _err L | (step S9).

When absolute value | SOC_I _err | of integration method charging state estimation error SOC_I _err is smaller than absolute value | OCV _err L | of lower limit OCV _err L of OCV method charging state estimation error SOC_V _err (YES in step S9) As shown in FIG. 5A, since the integration method charging state estimation error SOC_I _err indicated by the broken line is the smallest, the charging state estimation device 1 selects the integration method estimated charging state SOC_I as the charging state of the battery BAT ( In step S11), the integration method estimated charging state SOC_I is output as the charging state of the battery BAT, and the process returns to step S1.

On the other hand, when absolute value | SOC_I _err | of integration method charging state estimation error SOC_I _err is larger than absolute value | OCV _err L | of lower limit value OCV _err L of OCV method charging state estimation error SOC_V _err (NO in step S9) Since the OCV method charging state estimation error SOC_V is the smallest, the charging state estimation device 1 selects the OCV method estimated charging state SOC_V as the charging state of the battery BAT (step S13), and sets the OCV method estimated charging state SOC_V to Output as the state of charge of the battery BAT. Then, the charging state estimation device 1 sets the integration method charging state estimation error SOC_I _err to 0, resets the error due to the current integration method (step S15), and returns the process to step S1.

In step S7, when the charging / discharging current I of the battery BAT is relatively small, when the charging rate of the battery BAT is relatively large, or when both of these conditions are satisfied, the open circuit voltage error OCV _err The upper limit OCV _err U is likely to be smaller than 0 (YES in step S7). In this case, as shown in FIG. 5B, the charging state estimation device 1 has an OCV method charging state estimation error SOC_V _err indicated by a one-dot chain line that is relatively small and the error average SOC_VI _err is the smallest. State average value SOC_VI is selected (step S17), and this estimated charge state average value SOC_VI is output as the state of charge of battery BAT (step S19). And the charge condition estimation apparatus 1 returns a process to step S1.

FIG. 6 is a diagram illustrating an example of a time transition of a charging state estimation error in each method of the battery BAT. It is. In FIG. 6, the vertical axis indicates the magnitude of the error, and the horizontal axis indicates the elapsed time t from the start of charging / discharging of the battery BAT. In this example, the upper limit SOC _err U of the OCV method charging state estimation error SOC_V _err is greater than zero. In FIG. 6, the integration method charging state estimation error SOC_I _err is indicated by a broken line, the OCV method charging state estimation error SOC_V _err is indicated by a broken line, and the error average SOC_VI _err is indicated by a broken line.

When the battery BAT starts charging / discharging, the charging state estimation device 1 selects the integration method estimated charging state SOC_I and sets the charging state of the battery BAT because the integration method charging state estimation error SOC_I _err is the smallest.

When the error due to the offset error of the current sensor is integrated over time, the integration method charging state estimation error SOC_I _err gradually increases, and at the elapsed time t1, the integration method charging state estimation error SOC_I _err becomes larger than the error average SOC_VI _err. The error average SOC_VI _err is the smallest. Therefore, the charging state estimation device 1 selects the estimated charging state average value SOC_VI and sets the charging state of the battery BAT.

When the elapsed time t2 is reached, the error average SOC_VI _err becomes larger than the integration method charging state estimation error SOC_I _err , and the integration method charging state estimation error SOC_I _err becomes the smallest. Therefore, the charging state estimation device 1 selects the integration method estimated charging state SOC_I and sets the charging state of the battery BAT.

When the elapsed time t3 is reached, the integration method charging state estimation error SOC_I _err again becomes larger than the error average SOC_VI _err , and the error average SOC_VI _err becomes the smallest. Therefore, the charging state estimation device 1 selects the estimated charging state average value SOC_VI and sets the charging state of the battery BAT.

Then, at the elapsed time t4, the integration method charging state estimation error SOC_I _err increases, and the error average SOC_VI _err becomes larger than the OCV method charging state estimation error SOC_V _err . At this timing, since the OCV method charging state estimation error SOC_V _err is the smallest, the charging state estimation device 1 selects the OCV method estimated charging state SOC_V and sets the state of charge of the battery BAT. At the same time, the charging state estimation device 1 sets the integration method charging state estimation error SOC_I _err to 0 and resets it.

  The charging state estimation device 1 repeats the above processing while the battery BAT is charging / discharging.

As described above, the charging state estimation device 1 constantly monitors the time transition of the integration method charging state estimation error SOC_I _err , the OCV method charging state estimation error SOC_V _err and the error average SOC_VI _err from the start of charging / discharging of the battery BAT. By selecting the state of charge estimated by the method corresponding to the smallest error, it is possible to accurately estimate the state of charge.

-Effects of the embodiment of the invention-
According to the above embodiment, the state of charge of the battery BAT is estimated by the current integration method and the OCV method, and the time change from the start of battery charge / discharge in both methods is obtained. When obtaining the time change of the error in the current integration method, the time change of the integration method charging state estimation error SOC_I _err caused by the current sensor, which is the dominant factor, is obtained. On the other hand, when obtaining the time variation of the error in the OCV method, the time variation of the error due to the equivalent circuit model of the battery BAT, which is the main factor, is obtained. Therefore, it is possible to accurately obtain the time change of the error in each estimation method. Then, the estimated charging state by the smaller estimation method among the obtained time variations of the error is selected, and the charging state of the battery BAT is set. In this way, the time variation of the error by each estimation method is obtained with high accuracy, and the estimated charge state by the estimation method with a small error is selected to obtain the charge state of the battery BAT. Can be estimated.

Here, when the charging / discharging current of the battery BAT is relatively small, depending on the lapse of time from the start of charging / discharging of the battery BAT, these are more than the integration method charging state estimation error SOC_I _err and the OCV method charging state estimation error SOC_V _err. The average error average SOC_VI _err may be the smallest.

Therefore, according to the above embodiment, in addition to the current integration method and the OCV method, an estimated charge state average value SOC_VI _err that is an average value of the charge states estimated by both methods is obtained, and further, an integration method charge state estimation error is obtained. An error average SOC_VI _err which is an average of the SOC_I _err and the OCV method charge state estimation error SOC_V _err is obtained. Then, an estimated charge state is selected by the method of the smallest error among these three error time variations. Therefore, the state of charge of the battery can be estimated with higher accuracy.

  Moreover, according to the said embodiment, since the estimated charge state is selected based on the elapsed time t from the start of charging / discharging of the battery BAT, the charge state of the battery BAT can be estimated easily and with relatively high accuracy. .

Furthermore, according to the above embodiment, the integration method charging state estimation error SOC_I _err is obtained by integrating the offset error of the current sensor 3 over time. That is, at the charge / discharge start stage of the battery BAT, the integration method charge state estimation error SOC_I _err is the smallest. Therefore, from the start of charging / discharging of the battery BAT to the elapse of a predetermined time, the charging state of the battery BAT can be estimated with higher accuracy by selecting the estimated charging state SOC_I by the current integration method with the smallest error.

(Modification of Embodiment of Invention)
FIG. 7 is a block diagram showing a configuration of the battery charge state estimation device 2 according to a modification of the embodiment. This charging state estimation device 2 differs from the charging state estimation device 1 according to the above embodiment in that the average value calculation unit 11 is not provided. Then, the charging state estimation device 2 estimates the integration method charging state estimation error SOC_I _err and the OCV method charging state estimation error SOC_V _err in the error estimation unit 13, and the multiplexer 15 estimates the estimation method corresponding to the smaller error. The charged state is set as the charged state of the battery BAT.

  Thus, while simplifying the configuration, it is possible to perform estimation with higher accuracy than estimation of the state of charge of the battery BAT using only the current integration method or the OCV method.

(Other embodiments)
In the above embodiment, the integration method estimated charge state SOC_I, the OCV method estimated charge state SOC_V, or the estimated state average value based on the integration method charge state estimation error SOC_I _err , the OCV method charge state estimation error SOC_V _err, and the error average SOC_VI _err Although SOC_VI is selected, the present invention is not limited to this, and the SOC_VI may be selected based on the elapsed time t from the start of charging / discharging of the battery BAT. In this case, since the selection is made based only on the elapsed time, the control can be simplified.

  As described above, the battery charge state estimation method and the battery charge state estimation device according to the present invention can be applied to a use for estimating the charge state of the battery with higher accuracy.

1 Charge state estimation device (battery charge state estimation device)
3 Current sensor 5 Voltage sensor 7 Current integration method charge state estimation unit (integration method charge state estimation means)
9 OCV method charge state estimation unit (OCV method charge state estimation means)
11 Average value calculation unit 13 Error estimation unit (error acquisition means)
15 Multiplexer (charging state selection means)
BAT battery

Claims (5)

A battery charge state estimation method for estimating a charge state of the battery based on a charge / discharge current and a terminal voltage of the battery,
An integration method charging state estimation step of integrating the charging / discharging current of the battery detected by a current sensor over time, and estimating an integration method estimation charging state that is a charging state of the battery based on the time integration value;
An open circuit voltage of the battery is obtained based on a terminal voltage of the battery detected by the voltage sensor, a charge / discharge current detected by the current sensor, and an equivalent circuit model of the battery, and the battery's open circuit voltage is determined from the open circuit voltage. OCV method charge state estimation step of estimating the charge state of the OCV method estimated charge state;
The error of the integration method estimated charge state caused by the current sensor is an integration method error that is a time integration value from the start of charging / discharging of the battery, and the OCV method estimation charge state error caused by the equivalent circuit model. An error acquisition step for obtaining an OCV method error that is a time change from the start of charge / discharge of the battery;
A charging state in which the integration method estimated charging state or the OCV method estimated charging state is selected based on the integration method error and the OCV method error obtained in the error acquisition step, and the selected estimated charging state is the charging state of the battery. A battery charge state estimation method comprising: a selection step. The battery charge state estimation method according to claim 1,
An average value obtaining step of obtaining an estimated charge state average value that is an average value of the integration method estimated charge state and the OCV method estimated charge state;
In the error acquisition step, an error average that is a time change of the average value of the integration method error and the OCV method error is further obtained,
In the charge state selection step, the battery charge is characterized in that the accumulation method estimated charge state, OCV method estimated charge state, or estimated charge state average value is selected based on the accumulation method error, OCV method error, and error average. State estimation method. The battery charge state estimation method according to claim 1 or 2,
In the charging state selection step, the battery charging state estimation method is selected based on an elapsed time from the start of charging / discharging of the battery. The battery charge state estimation method according to any one of claims 1 to 3,
In the charge state selection step, the battery charge state estimation method is characterized in that the integration method estimated charge state is selected from the start of charging / discharging of the battery to a lapse of a predetermined time. A battery charge state estimation device for estimating a charge state of a battery based on a terminal voltage and a charge / discharge current of the battery,
A current sensor for detecting the charge / discharge current of the battery;
A voltage sensor for detecting the terminal voltage of the battery;
Integration method charging state estimation means for integrating the charging / discharging current detected by the current sensor over time, and estimating the integration method estimation charging state that is the charging state of the battery based on the time integration value;
An open circuit voltage is obtained based on the terminal voltage detected by the voltage sensor, the charge / discharge current detected by the current sensor, and the equivalent circuit model of the battery, and the OCV that is the state of charge of the battery is determined from the open circuit voltage. OCV method charging state estimating means for estimating the method estimating charging state;
The error of the integration method estimated charge state caused by the current sensor is an integration method error that is a time integration value from the start of charging / discharging of the battery, and the OCV method estimation charge state error caused by the equivalent circuit model. An error acquisition means for obtaining an OCV method error that is a time change from the start of charging and discharging of the battery;
A charging state in which the integration method estimated charging state or the OCV method estimated charging state is selected based on the integration method error and the OCV method error obtained by the error acquisition means, and the selected estimated charging state is set as the charging state of the battery. A battery charge state estimation device comprising: a selection unit;

Images