JP4215171B2 - Battery capacity detection method - Google Patents

Description

This invention relates to the detection how the battery capacity in a plurality of times of charge and discharge secondary battery such as a lithium ion battery.

  Conventionally, as a method for estimating the remaining capacity of a storage battery, the charge rate is measured from the voltage change when the current value flowing through the storage battery is equal to or less than a predetermined value, and the full charge capacity is estimated from the charge rate and the current integrated value. It has been shown that the remaining capacity can be estimated from the estimated full charge capacity and the charging rate (see Japanese Patent No. 3126591). However, since this estimation method corresponds to the full charge capacity measured when the battery current is very small, it is not suitable for the remaining amount estimation, and there are inconveniences such as a large error especially when the current is passed. large.

  On the other hand, as a method aiming at improving the estimation accuracy of the remaining usage time in the secondary battery, there is a “method for detecting the remaining capacity of the lead storage battery” disclosed in Japanese Patent Laid-Open No. 2-170372. This is because there are multiple data on the discharge characteristics when the temperature and discharge current change, and by referring to the timely data, the accuracy of the remaining dischargeable time is maintained even if the temperature and current change. To do.

Japanese Patent No. 3126591

  However, if the battery configuration is changed and used over many types like a battery pack, the amount of data that must be prepared will be enormous, and even a battery pack that has only a small difference will need to be prepared. If the specifications are different, it will be necessary to replace all data, leading to increased costs.

The present invention has been made in view of such inconveniences, and an object thereof is to provide a battery capacity detection method capable of accurately estimating the amount of charge .

  In the above description and the following description, “charging rate” refers to the total capacity that is input to the secondary battery until the secondary battery is fully charged from the state where the secondary battery is almost completely discharged and is fully charged. "Indicates the percentage of the capacity remaining in the secondary battery in"% ".

  The “open circuit voltage (hereinafter referred to as“ OCV ”)” is a voltage between terminals measured by opening both electrodes in a stable state when the secondary battery is left standing for a long time. Furthermore, “open circuit voltage characteristics (hereinafter referred to as“ OCV characteristics ”)” refers to the relationship between the charge rate and OCV measured in advance for a secondary battery whose capacity is to be displayed, such as a table, graph or database. Data that is stored in a form that is easy to process data.

  Furthermore, a terminal voltage in a state where a charge / discharge load is applied to the secondary battery and a charge / discharge current flows is simply referred to as “battery voltage”. In addition, the relationship between the charging / discharging time and the battery voltage during charging / discharging is called “battery voltage characteristics”, and the curve indicating the relationship between the discharging time and the battery voltage when discharging at a predetermined constant current during discharging is shown as “discharge characteristics” That's it.

  In the battery capacity detection method according to the present invention, when it is determined that the OCV measurement process is started during the period in which the load current as shown in FIG. The envelope of the voltage change during the non-period is obtained, and the OCV is estimated from the convergence value of the envelope.

Furthermore, sustained per charge rate, with a database for pre-correcting such relationship in Figure 22 with the reduction rate of the battery temperature and the full charge capacity, the state in which supply is sealed stop of the load current from the secondary battery 12 When it is determined that, for example, it has been determined that the OCV has continued beyond the time required for stabilization, the charging rate is estimated from the measured voltage value and the OCV characteristics, while the charging rate, the battery temperature, From the change of the charging rate, the correction process of the full charge capacity can be performed using the correction database.

  As described above, according to the present invention, in a load state in which load current is intermittently supplied, an envelope of voltage change during a period in which no current flows is obtained, and OCV is estimated from the convergence value of the envelope. Thus, even when used in a device such as a mobile phone that is driven and stopped even during standby, a charge amount estimation operation can be performed relatively accurately.

  In addition, the relationship between the battery temperature and the reduction rate of the full charge capacity with respect to each charging rate during the battery leaving period is created in advance as a database, and by correcting the full charging capacity in addition to the charging rate during the leaving period, in particular, Capacitance correction when left at high temperature can be performed accurately.

Hereinafter, the battery capacity detection apparatus 10 to which the battery capacity detection method according to the present invention is applied is a lithium ion battery unit cell having an OCV of about 4.2 V at full charge and a full charge capacity of about 1600 mAh in a state without deterioration. Further, a battery pack as an assembled battery in which a plurality of unit cells are combined will be described in more detail as an example, but it is needless to say that other types of secondary batteries can be implemented in substantially the same manner.

  In the present invention, the OCV characteristic illustrated in FIG. 2A is measured in advance for the secondary battery 12 whose battery capacity is to be detected. Then, using the fact that this OCV characteristic is maintained substantially constant regardless of the deterioration of the secondary battery 12 or the operating temperature, the open circuit voltage Va or Vb of the secondary battery 12 is set as shown in FIG. The basic configuration is to estimate the charging rate a or b at the time of measurement by measuring as necessary.

  Regarding the temperature characteristics of the OCV, the knowledge that the change in temperature characteristics itself is small and that the temperature characteristics are substantially constant due to the deterioration of the secondary battery was obtained from the measurement of characteristics of the batteries in various deteriorated states. Therefore, even if there is a temperature change, the charging rate can be estimated from the OCV characteristic taking into account the temperature characteristic.

  Further, for example, when the charging rate immediately before the start of charging is a [%] and the charging rate immediately after the end of charging is b [%], the charging rate immediately before the start of charging is subtracted from the charging rate b immediately after the end of charging. The “differential charging rate” is caused by charging the “charging amount” obtained by integrating the charging current supplied to the secondary battery during charging.

  Therefore, by dividing this charge amount by the differential charge rate and multiplying it by 100, the total capacity at the time of full charge (hereinafter referred to as “full charge capacity”) is calculated without continuing the charge to the full charge state. The

  Therefore, it is necessary to perform so-called full charge learning, such as measurement of charge capacity by measuring the charge amount from the end of discharge to the end of full charge or the amount of discharge from the full charge state to the end of discharge, which has been conventionally performed. Disappears. Although the full charge capacity can be calculated from the change in the accumulated capacity during discharge and the charge rate before and after the discharge, it is preferable to estimate the full charge capacity in the charge because the current can be accumulated due to the transient.

  Further, if the full charge capacity is obtained, the deterioration of the secondary battery 12 can be determined by multiplying the full charge capacity by the charging rate at the time of measurement to determine the remaining capacity, that is, how much capacity remains in the secondary battery 12. Calculated without considering the state.

  Therefore, in order to obtain the above-described OCV characteristics prior to the above-described detection operation of the charging rate, full charge capacity and remaining capacity, charging and discharging are performed by the following method illustrated in FIG. Let it be done.

  That is, by confirming that the terminal voltage of the battery has dropped below the set value, the steady state where the ambient temperature is room temperature (25 ° C.) from time t1 when the remaining capacity of the battery is assumed to be zero or close to it. Then, charging is performed while continuously checking changes in the charge amount and battery voltage by continuously integrating the charge current amount every minute time.

  Immediately after the start of charging, for example, constant current charging is performed with a constant current source of 1600 mA, for example, and charging is continued for, for example, 3 hours under the condition of switching to 4.2 V constant voltage charging at time t2 when the battery voltage reaches a set value. The total charge amount supplied to the secondary battery 12 from the start to the end of charging is stored as the initial value of the full charge capacity.

  For OCV characteristics, for example, charge / discharge is stopped every time 10% charge or discharge progresses, and the OCV at that time is measured, and the charge rate and OCV at each measurement point are stored as a pair sequentially. Thus, a graph as shown in FIG. 2A is obtained.

  By the way, it is known that the voltage actually appearing between the terminals of this type of secondary battery 12 is higher than the above-mentioned OCV during charging and lower than the OCV during discharging. Further, the value of the voltage is not constant over time, but decreases or increases with a predetermined time constant.

  Such a phenomenon is generally explained by an equivalent circuit for the secondary battery 12 illustrated in FIG. That is, among the elements constituting the equivalent circuit, the solution resistance R interposed in series has an early response characteristic on the order of milliseconds, and the charge transfer resistance r connected in parallel with the electric double layer capacitance C is on the order of seconds. Therefore, it converges if it waits about 10 seconds at most.

  On the other hand, the polarization voltage Ep has a response characteristic in the order of time, and it takes several hours to several days until the value is stabilized. Therefore, in order to measure an accurate OCV, at least several hours, preferably 8 hours. Some degree of downtime is required.

  However, in actual usage conditions of secondary batteries, energization and rest are frequently repeated during other discharges, except that a relatively long rest period is often provided immediately after the end of charging. It is generally difficult to even measure OCV with a rest period of several hours, not to mention 8 hours.

  Incidentally, as shown in FIG. 5, when the horizontal axis represents the reciprocal of time t or the reciprocal of the square root of time t after the energization of the secondary battery 12 is stopped, the OCV at each measurement is plotted on the vertical axis. The locus is a substantially straight line or a curve that can be approximated to a straight line. Therefore, in this embodiment, two points of OCV are measured at a predetermined time interval, and the two points are connected by a straight line as illustrated in FIG. 5 so as to estimate the OCV at the time of convergence. It is configured.

  The time after the energization to the secondary battery is relatively stable at room temperature, so two points after 30 minutes and 2 hours can be measured. .

  However, since the estimation accuracy can be improved when a longer time elapses, the first estimation is performed by measuring the voltage at two points, for example, 30 minutes and 2 hours, after the completion of charging / discharging, and further for 4 hours and 8 hours. When the hibernation state continues as described above, the second or third estimation is performed using the measured voltage at that time, and the estimated value is updated.

  By the way, if the charging rate is high at room temperature and higher than normal temperature, the voltage drop due to self-discharge is larger than that under other conditions. To do. Therefore, in that case, it is preferable to end the estimation of the OCV convergence value at the point of time of 2 hours and use the value for the capacity determination.

  On the other hand, under low temperature use conditions, the amount of self-discharge during the pause period is small, but the inflection of the voltage locus increases from that at room temperature, and it takes a long time for the OCV to stabilize. It is preferable to estimate the convergence value at two points of 8 hours.

  The time interval and the number of times of measurement for the OCV measurement and the convergence value estimation described above are examples, and it is needless to say that the time interval and the number of measurements can be changed as appropriate. For example, when the pause period is longer than 24 hours and continues for a long time, it is estimated that the OCV has already converged, so the charging rate is immediately obtained from the measured voltage value without performing the above-described estimation process. On the other hand, it is also possible to correct the decrease in the full charge capacity during the leaving period using a database. Details will be described later.

  On the other hand, not only during charging / discharging, but also when charging or discharging is resumed before 2 hours, which is the above-mentioned minimum pause period in which OCV can be estimated after charging / discharging is stopped. The error included in the obtained OCV is large even if it cannot be estimated by the above-described method.

Wherein, continue to continuously measure the amount of charge by integrating the current flowing through the secondary battery 12 during charging and discharging, charging and discharging before the start of the charging rate, charging rate converted from the measured amount of charge By integrating the amount of change, the charge rate and remaining capacity at the current time can be calculated by calculation.

  By using the OCV characteristics as described above, the charging rate and the remaining capacity at the time of measurement or estimation can be detected relatively accurately regardless of whether the secondary battery 12 is deteriorated or used. The remaining capacity detected in step 1 is based on the assumption that all the amount of coulomb input to the secondary battery 12 can be consumed.

  That is, it corresponds to the remaining capacity in the case of a minute current, and when the load is connected, all the remaining capacity detected by this method cannot be discharged. Therefore, the remaining capacity when the load is connected is calculated separately. There is a need.

  However, it has been detected as an unloaded full charge capacity under ideal conditions, and this capacity can be evaluated as one of the full charge capacity of the battery. In this application, based on this point of view, the distinction between the full charge capacity when there is no load and the remaining capacity when there is a load is clearly handled.

  This means that the full charge capacity in the case of no load detected as described above is simply treated as the full charge capacity of the battery, and the remaining capacity estimation database is unified based on the charge rate based on this full charge capacity. By constructing and calculating the discharge characteristics when there is a load based on this database and estimating the remaining capacity from this discharge characteristic, it is possible to construct a remaining capacity detection method that can accurately estimate the remaining capacity. ing.

Therefore, the battery capacity detection method will be described below, starting with an explanation of how the discharge characteristics are determined.

  First, in an actual use situation, as can be seen from the equivalent circuit shown in FIG. 4 and the characteristics shown in FIG. 6, from the ideal battery E whose voltage value changes according to the OCV characteristics, the secondary battery 12 A voltage value obtained by multiplying the load resistance I by the solution resistance R and the charge transfer resistance r, which are internal resistances, and the polarization voltage Ep are subtracted and output between the terminals.

  However, in an actual usage situation, as can be seen from the equivalent circuit shown in FIG. 4 and the characteristics shown in FIG. 6, from the ideal battery E whose voltage value changes according to the OCV characteristics, the secondary battery 12 A voltage value obtained by multiplying the load resistance I by the solution resistance R and the charge transfer resistance r, which are internal resistances, and the polarization voltage Ep are subtracted and output between the terminals.

  Furthermore, the element values of the solution resistance R, the charge transfer resistance r, and the polarization voltage Ep of the secondary battery 12 increase as the deterioration progresses due to repeated charge / discharge, etc. As a result, the value of the battery voltage taken out from the secondary battery 12 Tend to decrease.

  On the other hand, as illustrated in FIG. 6A, the discharge characteristics of the secondary battery 12 decrease while showing a substantially straight line during a period when the charging rate is high. However, when the voltage drops below a certain voltage value Vs, the voltage drop suddenly decreases. Generally has a tendency to become intense.

  However, it is inappropriate to use the electric device 18 in such a fluctuation range of the battery voltage, and when the discharge further proceeds, there is a high possibility that the battery deteriorates rapidly due to an overdischarge state.

  Therefore, usually, control is often performed in which the end voltage Vt is set in the middle of the fluctuation region and discharge is forcibly stopped at the end voltage Vt. Therefore, the remaining time (hereinafter referred to as “remaining time”) that can be actually used when the secondary battery 12 is connected to the load from when the discharge is started until the battery voltage reaches the end voltage Vt is accurately known. This is important when the secondary battery 12 is used in an electric device, and it is preferable that the remaining time can be accurately grasped at any time including immediately after the electric device is turned on.

Hereinafter, how to obtain the remaining time will be described as a reference example. This remaining time can be determined by measuring discharge characteristics under such conditions in advance when the use conditions such as load current and ambient temperature are constant, and detecting the battery voltage at the time of detecting the remaining time.

  However, the battery voltage at the time of actual discharge, particularly due to the presence of the polarization voltage Ep described above, gradually decreases from the OCV value when energization is started at time t11 as illustrated in FIG. This coincides with the theoretical discharge characteristics at time t12, and it is difficult to obtain the remaining time simply by detecting the terminal voltage of the secondary battery 12.

  Furthermore, the discharge characteristics of the secondary battery 12 are not only when the load current is changed as shown in FIG. 7A, but also when the secondary battery 12 is deteriorated as shown in FIG. As shown in c), since the ambient temperature of the secondary battery 12 changes greatly even if the ambient temperature increases or decreases, it is almost impossible to prepare discharge characteristics for all use conditions.

Wherein the element value of each element shown in the equivalent circuit of the rechargeable battery 12 is measured separately as in Figure 9 by using the method described below, is stored in a database of its value, the next The remaining time can be estimated by deriving the discharge characteristic by calculation using the stored value at the time of discharge.

Here , the solution resistance R and the charge transfer resistance r shown in FIG. 4 are measured using the transient response method shown in FIG. 8 by utilizing the fact that the transient response is relatively fast but the response speed of the both is different. is doing.

  That is, when a unit current consisting of a constant current having a step shape as shown in FIG. 8B is applied to the secondary battery 12 by a signal output from the measurement control means shown in FIG. ), A voltage drop VR due to the solution resistance R having a fast response speed is first detected at the time of rising of the current.

  After that, a voltage drop Vr due to the charge transfer resistance r occurs with a predetermined time constant τ (≈C · r). For example, when a time of about 10 seconds elapses from the start of energization, the voltage value thereafter becomes substantially linear. . Therefore, by extending the straight line until the charge start timing, the voltage drop VR due to the solution resistance R and the voltage drop Vr due to the charge transfer resistance r are separated and detected.

  The measurement of the solution resistance R and the charge transfer resistance r, which are internal resistances of the secondary battery 12, is performed simultaneously with the above-described measurement of the OCV characteristics, and is stored as a database with a charge rate of every 10%, for example. As a result, the graphs as shown in FIGS. 9A and 9B are obtained, and are set as initial values. Furthermore, it is also measured at an appropriate time such as a resting time at the time of discharge, and the value is normalized by converting it to room temperature, and stored together with the charging rate at the time of measurement.

  On the other hand, it is difficult to directly determine the value of the polarization voltage Ep. However, as shown in FIG. 6 (a), for example, at the time of discharging, the voltage value obtained by subtracting the battery voltage from the OCV matches the internal resistance of the secondary battery 12 and the voltage due to the polarization voltage Ep. Since the voltage drop can be directly detected by the above method, by subtracting the value, a graph as shown in FIG. 9C showing the relationship of the polarization voltage Ep at each charging rate can be calculated.

  However, since the polarization voltage Ep has a very long time constant of polarization, the current condition is stable after the start of charging shown in FIG. 3, and especially the time when the polarization characteristics are settled, for example, about 3 to 5 minutes have elapsed. Then, it is preferable to obtain from the change of the battery voltage after that.

  In addition, since the polarization voltage Ep depends on the state of ion movement due to diffusion, in addition to the method for determining the timing for evaluation over time, there is also a method for evaluating when a change in charging rate of about 3% occurs, for example. is there.

  However, since it depends on the diffusion rate of the battery, etc., it is desirable to change the measurement timing more accurately based on the deterioration, current, temperature, change in charge rate, or amount of change in time. In this case, the timing can be obtained by measuring the time during which the polarization is almost constant from the behavior of the voltage change after the start of discharge under various conditions such as the discharge current, and by measuring the change in the charging rate, and measuring the discharge condition versus By collecting timing data and obtaining the measurement timing from this database, the polarization voltage Ep can be evaluated more accurately and in a short time.

  Alternatively, since there is a portion showing a substantially similar change characteristic of the polarization voltage Ep due to a constant charge rate change regardless of the current, it is also possible to obtain it by a voltage change when a constant charge rate change occurs. .

  Here, the OCV during charging is calculated by adding the estimated value of OCV corresponding to the charging rate converted from the charging amount obtained by integrating the current value to the value measured at the end of the last discharging before charging. it can.

  The values of the solution resistance R and the charge transfer resistance r are basically determined by the method described above. However, the DC resistance is measured under the condition that the charging is turned off at the timing when the OCV is calculated and the charging current is zero, or the transient response due to switching of the constant resistance load during charging is used. Can be measured.

  When the charging rate at the start of charging is small, the OCV also changes rapidly and the error is large as charging progresses and the charging rate increases. Therefore, until a predetermined charging rate is exceeded, the polarization voltage Ep near the start of charging can be estimated in more detail by charging for a certain time with a smaller current than normal and using the change in the battery voltage during that time.

  Here, when charging is performed by switching the charging method from constant current charging to constant voltage charging at time t2 in FIG. 3, the change in the battery voltage is directly measured during the constant current charging period to obtain the polarization voltage. This is used to estimate Ep. However, when entering the constant voltage charging region, a change in the charging current is measured, and the change in the current is converted into a voltage change and evaluated to be used for estimating the polarization voltage Ep.

  As described above, the values of the solution resistance R, the charge transfer resistance r, and the polarization voltage Ep constituting the equivalent circuit of the secondary battery 12 can be obtained in correspondence with the change in the charging rate as shown in FIG. Therefore, these values are used as initial values of element values indicating the characteristics of the secondary battery 12 and stored in a database.

  As mentioned above, although the method of measuring simply using a battery alone or using charge / discharge has been described, it goes without saying that there are other methods for continuously obtaining characteristics. As an example, the load current can be continuously changed in a pulse shape and obtained from the response.

  In this case, if the load is intermittently switched, for example, at intervals of 10 seconds, and a measurement load current in which a load current of a rectangular wave pulse with a cycle of 20 seconds is superimposed on a constant discharge load current is passed, as shown in FIG. A rectangular response waveform can be continuously obtained. Since the difference between the two envelopes passing through the upper and lower outermost portions of this waveform is the sum of the voltage drop of the solution resistance R and the charge transfer resistance r, if the solution resistance R is obtained separately, the charge is continuously charged every 10 seconds. The magnitude of the movement resistance r is obtained.

  Since the solution resistance R has almost no dependency on the charging rate, a value measured separately may be used, or a short rectangular pulse current of microsecond order may be passed through the load current and obtained from the response. Of course, it goes without saying that it may be obtained from the voltage and current changes at the instant when the rectangular wave pulses at intervals of 10 seconds change, but here an example is given assuming that measurement is performed with a dedicated measurement pulse.

  Further, since the difference between the two envelopes is a voltage drop due to the charge transfer resistance r and the solution resistance R with respect to the rectangular wave pulse current, it is assumed that the rectangular wave pulse current has increased by the average load current being measured. If the difference is increased, it corresponds to the voltage drop of the charge transfer resistance r and the solution resistance R due to the average current. Therefore, since the portion between the characteristic curve obtained by subtracting the found voltage drop from the OCV characteristic and the discharge characteristic curve can be evaluated as the voltage drop due to polarization, a database can be easily generated.

  On the other hand, regarding the temperature dependence of each element value in the secondary battery 12, as shown in FIG. 10, when the horizontal axis is a reciprocal of absolute temperature and the vertical axis is a logarithm of resistance value, for example, a predetermined line such as a substantially straight line is obtained. The shape can be approximated. Therefore, the temperature characteristics of each element value described above are measured in advance and stored.

  When the secondary battery 12 is actually used, the temperature correction of the element value is performed from the ambient temperature of the secondary battery 12 at the start of discharge, and the current correction is performed from the load current value to the drop voltage value caused by the element value. As a result, the discharge characteristics of the secondary battery 12 under the use conditions are calculated.

  On the other hand, since the charging rate is always accumulated and grasped, the discharge current immediately after the start of discharge or the discharge power described in detail later does not change from the charging rate at that time and the calculated discharge characteristics. The remaining time is estimated on the condition that it lasts.

  Furthermore, in the discharge characteristics, the area between the discharge characteristic curve and the output voltage zero, which is divided between the current charge rate and the end charge rate, that is, the area under the discharge characteristic curve (FIG. 6 ( The shaded portion in b) corresponds to the power output from the battery. Therefore, the remaining power of the battery can be estimated from this characteristic.

Hereinafter, how to obtain the remaining power will be described as a reference example. For this purpose, first, the area of this lower part is obtained as the area under the discharge characteristics. Next, the discharge charge amount from the present to the end is obtained by multiplying the full charge capacity by the difference between the current charge rate and the end charge rate. Thereafter, the remaining power is estimated by multiplying the area under the discharge characteristics by the discharge charge amount. The remaining power obtained in this way faithfully reproduces the discharge characteristics, and is a highly accurate remaining power estimation method.

  In addition, although it is necessary to correct the discharge characteristics based on the temperature characteristics, the temperature actually increases due to the internal resistance loss of the battery during discharge. Since this is almost determined by the battery pack and ambient conditions, the temperature characteristic change when discharging at a constant current is stored in the database together with the charging rate, or it is estimated from the temperature resistance and heat capacity and calculated together with the charging rate. Further, by further improving the estimation of the temperature dependence of the discharge characteristics at each charging rate, it is possible to estimate the remaining time and the remaining power with high accuracy.

  Next, in the following, as a method for evaluating the polarization voltage Ep, which is the voltage drop due to the polarization described above, the temperature-dependent element also depends entirely on the database by evaluating as the resistance value instead of the evaluation with the voltage. An example in which the calculation is possible without calculation is shown. Since the polarization voltage Ep is generated by diffusion, the voltage has current dependency. Therefore, the diffusion resistance Zw is assumed as the resistance that causes the voltage drop due to polarization with respect to the current dependency, and the remaining amount is estimated using this.

  Here, the diffused resistor Zw is defined and used as acting as a resistance component even when a current flows through the Wurburg impedance Zw of the internal equivalent circuit of the battery. This resistance value collectively indicates components that cause a voltage drop other than the solution resistance R and the charge transfer resistance r.

  Since the voltage drop due to diffusion has current dependency, the diffusion resistance Zw cannot often be expressed as a constant resistance component. In such a case, a database of diffusion resistance Zw taking into account the current dependency is constructed in advance, and then the polarization voltage Ep can be measured as described above at the appropriate time. By calculating and correcting, the diffusion resistance Zw to which current dependency is added is used for the remaining amount estimation. In addition, it is a matter of course that the diffusion resistance Zw is not used and the polarization voltage Ep is used as a database, and the value can be used at the time of calculation.

  Next, FIG. 11 shows an example of the circuit configuration of the capacity estimation circuit including the equivalent circuit model for obtaining the discharge characteristics in consideration of temperature dependency using the diffusion resistance Zw and the periphery thereof. In the following, according to this model example, a method for estimating discharge characteristics when discharging with constant power and a method for obtaining an estimated value of remaining power obtained therefrom will be shown. As a capacity meter when this model is used for remaining amount estimation An example of the remaining amount estimation process will be described.

  Even in this example, the equivalent circuit of the battery is the same as that shown in FIG. 4, but in order to show the temperature dependence, the “small lithium-ion secondary battery” proposed by Kameyama et al. From the heat balance equation that takes into account the entropy in "Heat generation analysis of the heat", a thermal change calculation formula suitable for the discharge characteristic simulation of the microcomputer is obtained and used for temperature estimation.

Here, the battery heat generation model considering the entropy described above is Qp for heat generation due to overvoltage, Qs for heat generation due to entropy change, Qb for heat generation around the battery, dVo / dT for temperature change of OCV, Tc for battery temperature, ambient temperature Qp, where Tr is the battery capacity, Cc is the heat capacity of the battery, I is the discharge current, Rη is the overvoltage resistance from the VI characteristics of constant current discharge, F is the Faraday constant, A is the surface area of the battery, and h is the heat transfer coefficient. = I ² · Rη, Qs = −Tc · ΔS · (I / F) = − Tc (dVo / dT) · I, Qb = −A · h · (Tc−Tr), and the heat balance equation is Cc It is pointed out that (dTc / dt) = Qp + Qs + Qb.

  From this heat balance equation, if the battery temperature rises by ΔTc in the minute time Δt, it can be expressed as Cc · ΔTc / Δt = Qp + Qs + Qb, and from this, ΔTc = Δt · (Qp + Qs + Qb) / C is obtained.

  Therefore, by obtaining Qp, Qs, and Qb from Tc, I, dVo / dT, and Rη of each item, ΔTc is obtained, and the estimated temperature is Σ (ΔTc) + starting temperature. The temperature change value is obtained, and the temperature is estimated by integration with the start temperature.

  Further, the element value of the battery changes depending on the temperature and the like, and the temperature and the like change further according to the change of the element value. In this case, the current value is obtained according to the following conditions. The discharge characteristics that satisfy this condition are the required discharge characteristics.

  First, the solution resistance R, the charge transfer resistance r, and the diffusion resistance Zw, which are the element values of the battery, are measured using the temperature T, the charging rate Q, and the current I as parameters, and are preliminarily stored in a database. It can be derived if T, charging rate Q, and current I are known.

  If the load power consumption W is constant, the relationship of W = I × (open circuit voltage− (R + r + Zw) × I) is established. Therefore, when the OCV, the element value, and the power consumption W at the start of discharge are known, The current value I is calculated by calculation.

  Further, the change time Δt required for the charge rate Q to change by a minute amount ΔQ is obtained by “ΔQ × full charge amount / average current”.

  By the way, it is not impossible to calculate the discharge characteristics directly by calculation by simultaneously solving the above relational expressions. For example, a method of expanding to a linear simultaneous polynomial and the like can be considered.

  However, in general, the capacity meter for estimating the remaining amount is intended to reduce the cost, so that the capacity is estimated by using a one-chip microcomputer having a small calculation power. Therefore, conditions such as a small amount of calculation at one time and low memory consumption are considered necessary.

Therefore, the difference between the starting point and the next point values of calculated unloading calculates the value of the next point as small, while achieving improved accuracy by re-calculating the next point further using the calculation result, the database An example of a calculation method that makes it possible to estimate a curve indicating discharge characteristics while suppressing the use to a minimum is used.

  That is, the discharge characteristic and the temperature characteristic are obtained by repeatedly obtaining the first point as the start point and the second point when the charging rate slightly changes from the point by using the relational expression described above. is there. The point of calculation here is to first simulate the discharge characteristics by modifying other parts based on the determined values.

  Furthermore, by separately providing a database on the above-mentioned unit cells and a database as a battery pack such as a case and a battery configuration similar to thermal resistance, the remaining amount estimation by estimating discharge characteristics in consideration of deterioration, temperature, etc. This is possible using a smaller database.

  That is, when a capacity meter corresponding to various battery packs is configured by the database, it is possible to incorporate a unit cell database and a database related to the battery pack corresponding to the target battery pack.

  Here, when comparing the size of the database, the size of the database increases because the database related to the unit cells usually requires details such as current dependency. On the other hand, the database related to the battery pack is usually a connection configuration of unit cells of the battery pack, for example, “2S3P” (a configuration in which two parallel connections of three unit cells are connected in series, S is a series, P is The database does not increase because it includes only the values of thermal resistance, heat capacity, and other values determined by the type of (used in parallel), the battery pack case, and the like.

  Therefore, the microcomputer itself is changed by putting a database related to the battery pack in a memory such as an EEPROM from which the microcomputer data is read, and making it possible to update only the data related to the battery pack every time the configuration of the battery pack is changed. Thus, capacity meters corresponding to various battery packs can be configured.

  Of course, as described above, the capacity type corresponding to various battery packs can be configured by storing the value for correcting the cell database in an EEPROM or the like and responding to the change of the cell.

Na us, but the battery temperature has a determined in consideration of entropy, based on the discharge rate versus temperature characteristic, it is needless to say that may incorporate a temperature rise characteristic in discharge characteristic calculation.

  Further, when the discharge power is used as a parameter and the maximum temperature at the end of discharge is normalized with the charging rate as the horizontal axis, the temperature characteristic becomes a substantially constant pattern. Therefore, the discharge characteristics can be estimated without increasing the size of the database so much.

  In addition, the discharge characteristics are estimated on the assumption that the battery pack is used at a constant power. However, the original power consumption value is originally the electric device 18 that is the parent device in which the battery pack is mounted. The capacity meter calculates from the battery database based on the notification, but may differ from the actual power consumption. Therefore, when the actual power usage and the notification power deviate more than a certain level, it is desirable to estimate the remaining power based on the actual power usage. In that case, it is also desirable to notify the master device to that effect.

  As described above, calculation examples, calculation methods, and method examples such as cost reduction have been shown. FIG. 12 schematically shows examples of curves of discharge characteristics and temperature characteristics, and conditions for determining the characteristics. The calculation points of the discharge characteristic calculation processing method described below will be described more specifically.

  As a calculation point, after obtaining the first point as described above, the second point at which the charging rate has changed is obtained under the condition that the condition does not vary greatly.

  If the first point is the discharge start point, the ambient temperature Tr is measured as the temperature, and this temperature is used as the battery temperature Tc below.

  The sum of the voltage drop V1 due to the internal resistance and current of the battery pack and the voltage drop V2 due to the load of the electric device 18 provided as the parent device is OCV.

  Next, at the second point when the charging rate ΔQ is changed, if the current at the first point flows, the time Δt until the second point is determined, and therefore the temperature increase ΔTc is calculated and the second point is calculated. The battery temperature Tc at 2 points is calculated. Further, since the battery temperature Tc and the current I are known, the internal resistance and the like are obtained, and the voltage current at the second point is obtained.

  However, since the calculation result of the second point is not reflected on the second point as it is, the calculation result of the second point is further entered, and the discharge characteristic and temperature characteristic values of the second point are calculated again. By doing so, the accuracy of simulation is improved.

  Next, when the second point is obtained as described above, this is used as the new first point, and the state of the new second point is calculated in the same manner as the second point is obtained from the first point described above. To do. By continuing this process, it is possible to obtain a curve of discharge characteristics and temperature characteristics. Finally, the entire discharge characteristics are obtained by continuing until the discharge voltage becomes equal to or lower than the predetermined discharge end voltage Vt.

  Although the description has been made based on the charge transfer resistance r inside the battery as the internal resistance, the battery pack has an internal wiring resistance and the like, and it is necessary to make a voltage drop corresponding to the resistance including that. Furthermore, it is necessary to obtain the resistance component based on the battery pack configuration database.

  For example, in the case of 2S1P, the internal resistance of the battery is twice the resistance of one unit cell, and the OCV characteristic is twice the value. Therefore, it is necessary to change the timely calculation according to the battery pack configuration. Needless to say. Furthermore, in the case of 2S3P, the resistance is equivalent to 2 × 1/3 = 2/3 cells, and the OCV characteristic is equivalent to two cells.

  Therefore, the circuit resistance rc is measured as a database of battery packs, and is added to the internal resistance of the battery obtained from the connection configuration of the unit cells as the resistance, and the OCV characteristic is the battery connected in series. That's fine. Since this conversion formula itself can be easily processed based on the combination data of the unit cells, any battery pack can be supported if it is incorporated in the software of the microcomputer in advance.

  An example of a configuration method in such a case is shown in FIG. In the figure, a cell battery database 6 and a battery pack database 4 are provided, and the battery pack database and the like are stored in a rewritable memory such as an EEPROM, so that it is possible to flexibly cope with a change in the specifications of the battery pack.

  Next, FIG. 13 to FIG. 15 are flowcharts showing processing methods for obtaining the discharge characteristics, which will be described more specifically with reference to FIG. 13 is a flowchart showing the overall flow during discharge, FIG. 14 is a flowchart for obtaining the first step A in FIG. 13, and FIG. 15 is a flowchart for obtaining the second step B in FIG.

  When discharging is started in FIG. 13, first, the charging rate Q1 at the time of starting discharging and the used power W notified from the electric device 18 are detected, and the charging rate Q1 is further determined from the value of the charging rate Q1 and the data in the database. After the corresponding OVC is derived, the current value calculation process at the first point is entered.

  The calculation process of the first point shown in FIG. 14 shows a calculation method when it is unclear what element value or current value will be at the start of discharge. First, the OCV is obtained from the charging rate, The following conditions are set to determine the current.

  That is, when the discharge is started, the provisional first current determined only from the values of the used power W and the OCV is calculated by “W / OCV”. Further, assuming that the battery temperature Tc at the start of discharge is substantially equal to the ambient temperature Tr, the first element value is derived from the first current and the value of Q1 using an estimation database.

Here, when the current I is determined, the element value is derived from the database. When the element value is determined, the current is calculated from the relationship W = I × (OCV−I × element value). The current I1 at the first point is estimated by repeating the derivation and the calculation of the current using the derived element value a plurality of times (here, twice) to correct the current.

  As described above, when the current I1 at the first point which is the discharge start time is estimated, the process of estimating the current value I2 at the second point where the charging rate is decreased by the constant charging rate ΔQ is entered in FIG. Executed.

  Here, since the current and the like are still unknown at the second point, first, the arrival time and temperature are obtained on the assumption that the current I1 at the first point continues. Since the arrival time Δt required for the charge rate Q to change by a minute amount ΔQ is “ΔQ × full charge amount / I1”, the temperature change amount ΔTc that increases as the charge rate changes by ΔQ is Δt · (Qp + Qs + Qb). / C, thereby obtaining the battery temperature Tc at the second point.

  Further, when the battery temperature Tc is found, an element value is derived from the database, and the current is recalculated using the derived element value, whereby a more accurate second point current I2 'is calculated.

  When the current I1 at the first point and the current I2 ′ at the second point are obtained, it is assumed that the average current “(I1 + I2 ′) / 2” flows up to the second point. Recalculate the arrival time and temperature.

  Further, the second point current I2 is obtained by using the temperature calculated from the recalculation, and the second point state is calculated and the second point operation is terminated. That is, in the first calculation, the current of the second point is calculated on the assumption that the current of the first point lasts up to the second point, and in the second calculation, the current of the second point calculated as the first point is calculated. The current value at the second point is corrected assuming that the average current is sustained.

  Thereafter, the current I2 at the second point is replaced with the current I1 at the first point, and the current at the next point is calculated. This is continued until the output voltage becomes equal to or lower than the final voltage Vt, and the overall discharge characteristics are obtained.

  Here, only the current has been described in the above description. However, since the voltage is obtained immediately when the current is obtained, the explanation is omitted.

  Although an example of the calculation method has been described above, it is needless to say that the accuracy can be improved by returning the current finally obtained in the second step to the average current and performing the calculation.

  Moreover, it is of course possible to change the OCV and resistance according to the battery connection state using a database related to the battery pack, and to obtain the resistance including the wiring resistance of the circuit. It is also possible to improve the discharge characteristic accuracy by calculating including the behavior.

  That is, in order to improve the remaining amount estimation accuracy, the efficiency of the power source of the electric device 18 provided as the master unit is also an important factor. For example, when the power supply configuration of the master unit is a DD converter, the conversion efficiency of the DD converter differs depending on the input voltage. Therefore, when calculating the remaining power, the remaining power is calculated by taking into account the conversion efficiency of the DD converter. The remaining time estimation accuracy is improved. That is, the power used including the conversion efficiency is obtained as the power used for calculating the discharge characteristics, and this is used for calculating the discharge characteristics.

  Usually, since it depends on the input voltage of the DD converter, a voltage vs. conversion efficiency database is provided, and when estimating the remaining power, the conversion efficiency is added to the target power consumption, and the target power consumption is changed in magnitude. Estimate the characteristic curve.

  In this case, it is desirable that the conversion efficiency of the power supply is passed from the master unit to the capacity estimation process as a database. As the database itself, the data of a pair of the voltage and the conversion efficiency in that case is several points within the operating voltage range. It is desirable to provide. As the database, for example, in the case of a battery pack and a parent device, the database may be passed by communication before the start of battery use. In such a case, it goes without saying that it can be easily realized if the communication specifications are determined, but it is necessary to prepare these conversion efficiency data before estimating the remaining amount. Not even. Therefore, when used as a pair with the battery pack, it may be written in advance in an EEPROM of a capacity meter in the battery pack. Alternatively, the parent device may write its own database in the EEPROM of the battery pack.

  Although it seems that it is possible to improve the accuracy of the simulation by repeating the calculation, it is necessary to determine the calculation amount, the calculation time, the calculation power, and the accuracy of the calculation result. It has been confirmed that the calculation results from the actual battery data have sufficient accuracy even in the present method shown in FIGS.

  In addition, the battery pack database 4 needs to include a battery configuration, an internal resistance, and the like, but can be configured to cope with changes in the unit cell itself. Since the discharge characteristic curve of the unit cell itself depends substantially on the electrode material, the OCV and the like are substantially the same even if the size of the battery is changed, and the charge transfer resistance r and the like due to the change in the electrode area and the like. It is the extent to which the value changes, and the behavior is similar. Therefore, in the unit cell database 6, data for correcting each element value can be recorded in an EEPROM or the like so as to cope with changes in the unit cell.

  Therefore, by having such a database, it is possible to construct a remaining amount estimation method and apparatus that can flexibly cope with changes in the configuration of the battery pack and the unit cell itself. Needless to say, this data may be included in the database 4 of the battery pack having the configuration shown in FIG.

  In the discharge characteristic estimation, the battery temperature Tc is estimated from the rise from the ambient temperature Tr to improve the accuracy of the discharge characteristic estimation. In this case, the ambient temperature Tr is used after the discharge is stopped for a long time. The battery temperature Tc before the start of discharge, which is estimated to be the temperature, is obtained as the ambient temperature Tr, and the subsequent temperature rise is estimated.

  However, the ambient temperature Tr is rarely used under a certain condition, and when the ambient temperature Tr changes, the battery temperature estimated using the above method is between the battery temperature actually measured and the actually measured battery temperature. An error occurs. Therefore, on the contrary, the change in the ambient temperature Tr can be estimated from the difference between the measured value and the estimated value of the current battery temperature Tc. When the change in the ambient temperature Tr is estimated, the remaining amount can be estimated corresponding to the change in the ambient temperature by using the changed ambient temperature to estimate the discharge characteristics.

  Specifically, the discharge characteristic calculation is started under the estimation start condition, and the battery temperature Tc and the charge rate at the two points of the start time and the current time are measured, while the current charge rate estimated at the start time of the discharge is measured. The battery temperature Tc is estimated by calculation. If the difference between the measured battery temperature and the estimated battery temperature exceeds the set value and is large, it is estimated that the ambient temperature Tr has changed, and the ambient temperature Tr is corrected to the following estimated value.

  Here, an example of an equation for calculating the estimated value of the ambient temperature is shown. This is the difference between the current estimated battery temperature (current battery temperature estimate before correction) obtained from the start of discharge and the current measured battery temperature (current battery temperature) (battery temperature error before correction). It is assumed that the ambient temperature before correction (ambient temperature before correction), which was originally measured and saved, corresponds to the difference between the current actual ambient temperature (estimated ambient temperature) and the ambient temperature before correction. The electric temperature error before correction is added to the temperature, and first, the ambient temperature assumed to have been corrected (ambient temperature after correction) is obtained. Next, the current battery temperature is estimated again (corrected current battery temperature estimated value) based on the corrected ambient temperature, and the final ambient temperature is calculated from these values. The final ambient temperature is: (Ambient temperature after correction-Ambient temperature before correction)-(Current battery temperature-Current battery temperature estimate before correction) / (Current battery temperature estimate after correction-Correction) (Previously estimated current battery temperature) + ambient temperature before correction.

  In the database construction, the step response as shown in FIG. 8 and the frequency response as shown in FIG. 19 are described to measure the value of the resistance component inside the battery. However, the present invention is not limited to this. Needless to say.

  For example, when the voltage detection accuracy of the A / D converter is low, the A / D conversion value is measured many times, and the average value thereof is taken to perform high-accuracy voltage measurement to improve the detection accuracy of the element value. Is possible.

  In this case, since the response of the charge transfer resistance r is fast, a short pulse, for example, a square wave on the order of μ seconds is output in the circuit that performs the step response in FIG. 8, and the on-state and off-state voltages of the response waveform are output. It is possible to improve the resolution of measurement by sampling many times to improve the detection accuracy of element values.

  Furthermore, since the voltage change is slow in the charge transfer resistor r, the measurement resolution can be improved by increasing the voltage measurement before energization and the number of voltage measurements during energization. Needless to say, this method is effective in a microcomputer having low A / D conversion accuracy.

  Furthermore, when an actual AV device is used, the step response may be changed, and the element value can be obtained from the response in this case. Further, when the current changes in this way, it is also possible to obtain the internal resistance from the change in current and voltage and the passage of time in that case.

  In this case, a resistance component with respect to a response of 10 seconds is obtained from a voltage / current change before and after a certain time, for example, 10 seconds, and this is obtained as a charge transfer resistance r, and further, resistance components with respect to many time responses are obtained. Therefore, the resistance component when the response time is short is evaluated as the solution resistance R. This corresponds to the frequency analysis of the resistance component, and it goes without saying that it becomes an effective means when there is a margin in the calculation power and the like.

  Although the battery remaining amount estimation method has been described above, the entire estimated value in the remaining amount estimation can be corrected. Therefore, a detailed description thereof will be given below.

  As the discharge of the secondary battery 12 progresses, the estimated remaining power amount and the actually measured power amount (power integrated value) change. Focusing on two points during discharge, the estimation at the two points A change amount of the remaining power amount and a change amount of the actually measured remaining power amount are obtained. In this case, if the amount of change is equal, the estimated value change and the measured value change between them are in good agreement regardless of the middle of the discharge. This is considered to be one proof.

  If the two are different, the estimated value and the measured value are different. If the two estimated values are different, the estimated value and the measured value are different, and the accuracy of the estimated value is low. Therefore, if a relational expression is defined as the coefficient K, the difference between the previous measured value = K × the difference between the previous estimated values, K = 1 when the estimated value is correct. However, in practice, K = difference between previous measured values / difference between previous estimated values≈1.

  Accordingly, if there is no change in the remaining amount estimation algorithm and the estimation is repeated in the same manner, it is expected that “difference in current measured value = K × difference in current estimated value”. That is, it is expected that the difference between the current measured value and the difference between the current estimated value that would be obtained by the current discharge do not match.

  Therefore, “difference of current measured value = difference of corrected estimated value = K × difference of current estimated value = (difference of previous measured value / difference of previous estimated value) × difference of current estimated value ", It is possible to correct the difference between the estimated values.

  Therefore, now, as a difference, from full charge to the end of discharge, “corrected current estimated value = (difference of previous measured value / difference of previous estimated value) × current estimated value” can be expected, The estimation accuracy will be improved.

  Actually, the first point is the discharge start point, the estimated value and the power integrated value in that case are recorded, the second point is the discharge end point, and the estimated value and the power integrated value in that case are recorded, By calculating the coefficient K from the four data thus obtained and correcting the next remaining amount estimated value, a more accurate remaining amount estimated value can be obtained as a learning effect.

  In addition, this can expect a good result when discharge conditions are the same. Therefore, when the discharge condition is different from the previous time, it is needless to say that a remaining amount detection method with high remaining amount estimation accuracy can be obtained in various cases by taking measures such as bringing the coefficient closer to 1.

  Hereinafter, based on the example shown in FIG. 16, a battery in which a battery capacity detection device 10, a secondary battery 12, and a battery protection unit 14 configured by using the above-described battery capacity detection method are integrally housed in a case. The configuration of the pack 16 will be described. Since the characteristics of the secondary battery 12 and the detection method in the battery capacity detection device 10 are substantially the same as those described above, detailed description thereof will be omitted below.

  The battery pack 16 is attached to various electric devices 18 such as AV devices, personal computer devices, and mobile phones, and sends detection data from the battery pack 16 to the communication circuit 20 on the electric device 18 side capable of bidirectional communication. Shows the display of the battery capacity using the display in the electric device 18, but the battery pack 16 is provided with an integral display means on the case of the battery pack 16, and the battery pack 16 alone detects the battery capacity. In addition to the above, a display operation is also possible. Conversely, a part of the control operation performed by the battery capacity detection device 10 can be performed on the electric device 18 side.

  The secondary battery 12 accommodated in the battery pack 16 is configured such that the capacity and output voltage required for the battery pack are achieved by connecting a plurality of unit cells in series or in parallel.

  Further, in addition to the database 6 of element values and OCV characteristics in the unit cell, a battery pack database 4 that can discriminate the difference in the pack state such as the configuration of the unit cell or the temperature parameter is individually provided.

  Here, the battery protection unit 14 includes a switching unit 24 including two FETs 22 and 22 and a protection circuit 26, and the switching unit 24 is interposed in series in an energization circuit from the secondary battery 12 to the electrical device 18. It is the thing of the structure substantially the same as the past used.

  That is, the protection circuit 26 includes a comparator composed of, for example, an operational amplifier or a comparator and a reference voltage, and can be operated only by hardware without using computer software, thereby suppressing the occurrence of malfunction as much as possible and improving safety. It is increasing.

Then, the battery voltage of the secondary battery 12 and the magnitude of the load current flowing through the switching unit 24 are constantly checked, and the magnitude of the load current has exceeded the set value or the battery voltage has exceeded the set value. When it has descended is detected Te, as a basic structure that prevented from forcibly stop and a secondary battery 12 and power supply to the load by sending a signal to the switching unit 24 is broken, but further In addition, it is possible to send a signal corresponding to the control state to the battery capacity detection device 10 to notify that an abnormal state has occurred.

  The battery capacity detection apparatus 10 is mainly composed of an external circuit group 28 that performs analog value signal processing and a control unit 30 that performs digital value signal processing using a one-chip microcomputer as the center of the control. Various measurement values output from the external circuit group 28 are sent to the control unit 30, and the measurement values are calculated while referring to the database 2 by a program built in the ROM provided in the control unit 30. Thus, the entire operation is controlled by software.

  Here, the external circuit group 28 includes a temperature detection circuit 32 that outputs a signal corresponding to the battery temperature Tc of the secondary battery 12, a voltage detection circuit 34 that outputs a signal corresponding to the terminal voltage of the secondary battery 12, The current detection circuit 36 outputs a signal corresponding to the current flowing through the secondary battery 12, and the load circuit 38 supplies a load current for detecting the circuit constant of the secondary battery 12 to the secondary battery 12.

  The temperature detection circuit 32 includes a temperature detection unit 40 such as a thermistor close to the secondary battery 12, converts an analog value output from the temperature detection unit 40 into a digital value, and converts the analog value into a digital value. Send to.

  The voltage detection circuit 34 is connected to both ends of the secondary battery 12 and takes out the voltage output between the terminals, converts it to a digital value, and sends it to the control unit 30 for measuring the battery voltage and the OCV. used.

  The current detection circuit 36 converts the voltage value generated at both ends of the resistor 44 connected in series with the secondary battery 12 into a digital value and sends the digital value to the control unit 30, and the current value flowing through the secondary battery 12 Used to measure.

  The load circuit 38 is for measuring the solution resistance R and the charge transfer resistance r separately by the transient response method shown in FIG. 8, and the energization from the secondary battery 12 to the electric device 18 is stopped. During the idle period, the switching transistor 42 is turned on by a signal sent from the control unit 30, and the resistors 46 and 44 connected in series with the transistor 42 are energized. The transient response state of the secondary battery 12 illustrated in FIG. 8B or FIG. 8C due to this energization is simultaneously measured by the voltage detection circuit 34 and the current detection circuit 36, whereby the internal resistance of the secondary battery 12 is determined. Enable measurement.

  Hereinafter, the operation procedure of the battery pack 16 will be described in more detail with reference to the flowcharts shown in FIGS. 17 and 18.

  First, on a predetermined storage means such as an EEPROM, as the unit cell database 6, in addition to the OCV characteristics of the unit cells illustrated in FIG. 2, the discharge characteristics illustrated in FIG. 6 of the secondary battery in the standard state and FIG. Each element value illustrated in (1) is measured in advance and stored in a database as an initial value during normal control.

Here, the OCV characteristic data stored as a database, and includes at least two kinds and for the remaining amount calculation for calculating the charging rate. That is, the OCV characteristic actually measured for the unit cell generally has a shoulder (inflection point) as shown by a broken line in FIG. 21A, not a smooth curve as shown in FIG. Therefore, when estimating the charging rate using the OCV characteristics as they are, by preparing a database that faithfully reproduces the first OCV characteristics in which the measured values are converted into a database, the error in estimating the charging rate is minimized. Suppress.

  On the other hand, when estimating the discharge characteristics, unevenness such as a shoulder portion in the actual measurement value shown in FIG. 21A is smoothed to obtain a curve as shown by a solid line in FIG. 2 OCV characteristics are prepared. Based on this second OCV characteristic, the element values such as R, r, and Zw are made into a database. As a result, the change of the element values is suppressed, and the data amount when making the database can be reduced. It is said.

  Further, the battery pack database 4 stores data capable of determining the difference in pack state such as the configuration of the cells in series or in parallel or the temperature parameter. Various data as an assembled battery are derived or calculated from the estimation database 2 including a database of unit cells and battery packs.

  Therefore, in step ST1 of FIG. 17, after performing predetermined initial setting based on the above-described initial value, the battery capacity detection process from step ST2 is entered.

  In step ST2 of FIG. 17, it is determined whether or not the secondary battery 12 is at rest. If it is not at rest, it is further determined whether or not the secondary battery 12 is left at step ST6 and whether or not it is being discharged at step ST3. In step ST5, it is further determined whether or not the battery protection unit 14 is operating, and the processing operations shown in FIG. 18 are performed based on the determination results.

  Here, if the determination in step ST2 is paused, the process moves to step ST21, and it is determined whether the set time of 30 minutes, 2 hours, 4 hours, or 8 hours has elapsed since the start of the sleep state. Then, the OCV detection process shown in FIG. However, if it is determined in step ST6 in FIG. 17 that 24 hours or more have elapsed, it is determined that the device is left unattended. For example, every time it is determined in step 61 that 24 hours have elapsed, the process proceeds to step 62 and remains. A quantity correction step is performed.

  In the OCV detection step shown in FIG. 18A, the terminal voltage of the secondary battery 12 is measured in step ST22. Further, in step ST23, it is determined whether or not the measured terminal voltage is after the second point. If it is after the second point, the process proceeds to step ST24 and the OCV estimation operation shown in FIG. 5 is performed. If not, save the measured terminal voltage and return.

  The remaining amount correction process during standing performed in step ST62 in FIG. 17 is performed each time a predetermined long period of time such as 24 hours elapses. In this stage, the OCV value is also stable. Therefore, the OCV is actually measured, not based on the above-described estimation, and the current charging rate is determined based on the actually measured value and the first OCV characteristic, and the OCV is satisfied according to the detected OCV change. Correct the charge capacity.

  If the battery is left for a long time, as shown in FIG. 22A, for example, if the charging rate is 50% or less, a decrease in the full charge capacity is almost recognized regardless of the battery temperature Tc. However, when the battery temperature Tc exceeds 30 ° C., as shown in FIG. 22 (b) or (c), the decrease in the full charge capacity tends to increase as the charging rate during standing increases.

  Therefore, a correction amount correction database indicating a change in full charge capacity with respect to a change in charge rate with the passage of time and temperature is created and stored in advance for each charge rate, for example, every 10%, according to the relationship shown in FIG. Then, as described above, the OCV is measured to determine the charging rate, and at the same time, the average standing temperature is detected, the amount of decrease in the full charge capacity is estimated using the correction database, and necessary correction is performed.

  Furthermore, the full charge capacity can be corrected more accurately by measuring the value of the solution resistance R and the charge transfer resistance r simultaneously with the measurement of the charge rate, but only when the change in the charge rate exceeds a certain value. By performing the measurement, the power consumption of the battery is preferably suppressed.

  Next, when it is determined in step ST3 in FIG. 17 that discharging is in progress, the process proceeds to the discharging process in FIG. In such a discharge treatment process, after the discharge current value is measured in step ST31 and the ambient temperature of the secondary battery 12 is measured in step ST32, the element value shown in FIG. 9 is corrected using the measured value. Thus, the discharge characteristics illustrated in FIG. 7 are calculated in step ST33, and the remaining time is estimated in step ST34 using the calculated discharge characteristics.

  In addition, the current charging rate is calculated by integrating the charging rate calculated from the previously measured charging rate from the measured discharge current in step ST35, and then the process proceeds to the next step ST36.

  In step ST36, it is determined whether or not the battery voltage is lower than a preset minimum voltage. If it is determined that the battery voltage is lower, the switching unit 24 is forcibly sent by sending a signal to the battery protection unit 14 in step ST37. Turn off and end the discharge.

  In addition, the above estimation of the remaining capacity assumes a case of discharging at a constant power. However, in the case of application to a device where there is a large opportunity for surge current to flow, a mechanism that does not cause the device to shut down even if surge current flows is necessary.

  Therefore, when a surge current is detected or when a surge current is predicted in advance, the discharge characteristics at a constant power considering only the internal resistance and polarization as shown by the thick line in FIG. As shown by the alternate long and short dash line in FIG. 5, the discharge characteristic is calculated by taking into account the voltage drop calculated by multiplying the internal resistance and circuit resistance of the battery by the surge current, and until the point when the discharge characteristic due to the surge decreases beyond the shutdown voltage Vt. By estimating the remaining time, remaining amount estimation with a margin is performed. The effect of surge is greater when the battery is deteriorated, and is an effective means.

  Note that this is not limited to cases where the voltage actually drops significantly, such as surge currents, and especially for equipment that needs to avoid battery shortage during operation as much as possible. Thus, a part of the remaining amount can be secured in advance as insurance, and the remaining remaining amount can be notified by the above remaining amount estimation.

Furthermore, to set the final stop voltage Vt, was constructed so as to notify the time to below its voltage Vt as a remaining time is not limited to, using the estimating function of the current and the battery temperature during discharge as described above, battery It is also possible to preset the maximum current to flow through and the maximum temperature that can be increased, and to set the end point of the device when the current estimated by the calculation and the battery temperature exceed the set values.

  Further, in step ST38 in FIG. 18B, it is determined whether or not the discharge has ended. If the discharge has ended, the method and load circuit 38 shown in FIG. In step 39, the solution resistance R and the charge transfer resistance r are individually measured in step 39 and stored, and a learning process operation is performed in step 40.

  The learning processing operation stores the solution resistance R, charge transfer resistance r and diffusion resistance Zw measured or calculated as described above together with the charging rate, while correcting the current dependency, correcting the temperature coefficient, and correcting the charging rate. And comparison processing with similar numerical values stored in the past.

  If the numerical value compared here is within a certain range, the new numerical value is used instead of the past numerical value, but if it is out of the range, the value is only saved without being used. .

  When it is determined that the value measured at the next time is within a certain range from the previous time, the previous and current values are averaged and used, and the data after the previous time are stored.

  Conversely, when the next measured value is also out of the range, it is determined that the deterioration has rapidly progressed in a high temperature state, and the value is used for use and the previous data is discarded.

  If the drive current flows in a pulsating manner even though it is very small during standby, like a mobile phone, the discharge current value can be measured accurately or the OCV can be measured. Both are difficult. In this case, the voltage when the current is off is intermittently detected, the envelope indicated by the broken line in FIG. 24 is obtained, and the current is obtained by using the fact that the envelope gradually decreases and approaches the OCV. As a result, it is possible to estimate the OCV by estimating the final voltage when it is assumed to be off.

  Next, when it is determined in step ST4 of FIG. 17 that charging is in progress, the charging process shown in FIG. In this process, the charging current is measured in step ST41, and the process proceeds to step ST42 to calculate the charging rate.

  In step ST42, the current charging rate is calculated by integrating the charging rate calculated from the charging current measured this time with respect to the charging rate calculated last time. After the resistance Zw is measured, the measured value is stored together with the current charging rate.

  If charging is performed by a pulse power supply and the above voltage and current measurements are performed by sampling at a predetermined cycle, the gap between pulses at the time of charging may be sampled continuously, possibly making measurement impossible. is there. Therefore, when such a condition is determined, the measurement loss can be reduced by shifting the phase at the time of voltage-current measurement by a predetermined amount, for example, every second to obtain an average value. .

  Further, when the operation of the battery protection unit 14 is detected in step ST5 of FIG. 17, the charging rate calculated or measured as described above is corrected in accordance with the value corresponding to the detected content.

  FIG. 19A shows a circuit configuration in the case where the above-described charge transfer resistance r is measured by a frequency response method. By sending an AC signal from the measurement control means 48 to the transistor switch 52 via the operational amplifier 50, FIG. It is possible to apply current modulation to the load 54. Then, while changing the modulation frequency, the current and voltage are measured by the voltage detection means and the current detection means shown in FIG. 8A, and the relationship between the modulation amplitude and the phase is complex as shown in FIG. 19B. By plotting the impedance, the value of the solution resistance R and the charge transfer resistance r can be individually measured from the value of the intersection with the real part.

  Further, when the secondary battery 12 is deteriorated, the equivalent circuit may change from FIG. 4A to FIG. 20A. In such a case, when the element value is measured using the method of FIG. 19, the complex impedance plot result is as shown in FIG. 20B, and the component of r2 increased by the degradation is also present. As a result of separation and detection, the degree of deterioration can be determined more specifically.

  The measuring method of the internal resistance of the secondary battery 12 can be further measured by using the switching unit 24 provided in the battery protection unit 14 and stopping energization of the electric device 18 at an appropriate time such as during discharge. That is, the voltage response curve immediately after the stop of energization has a shape similar to that in FIG. 8C, and the solution resistance R and the charge transfer resistance r can be detected separately by the transient response characteristics.

  As described above, how to separate and detect the internal resistance has been described above, but it goes without saying that it is not always necessary to separately detect the internal resistance. To collectively evaluate the internal resistance as a DC resistance component is to redefine that the discharge characteristics are obtained by subtracting the voltage drop due to the internal resistance and the voltage drop due to polarization from the OCV characteristics. Therefore, since it is clear that the configuration described so far can be applied as it is, detailed description thereof is omitted.

  In order to simplify the explanation when measuring the internal resistance, each resistance component is estimated by observing the response as a constant current load, but it goes without saying that a constant resistance load may be used.

The batteries parameters measured, it is essential to be next used capacity estimation The voltage varies with load fluctuation, because the separation measure resistance component from the change of the current is easy for those skilled in the art Detailed description will be omitted.

  Furthermore, the load fluctuation can be observed not only by the constant current load and the resistance load but also by the load fluctuation of the connected device, and the battery parameter can be observed as a result. This is to obtain R = ΔVR / ΔIR by detecting the difference between voltage and current, ΔVR, ΔIR between two points having the time difference τR, where τR is the time difference for measuring the solution resistance R. In this case, when τR is too short, several τR are defined, and a convergence value (for example, a value estimated as the original time difference τR) in the plot of R and time τR in each case is detected. The solution resistance R can be obtained. Similarly, a time difference for measuring the charge transfer resistance r is set as τr, and is obtained from voltage and current fluctuations therebetween. Furthermore, when a steep current change cannot be expected, it goes without saying that the solution resistance R and the charge transfer resistance r may be collectively measured as a DC resistance component to estimate the discharge characteristics.

  In addition, the full charge capacity has been explained by obtaining the charge amount obtained by integrating the charge current supplied to the secondary battery during charging and dividing by the difference between the charge rates before and after the charge. As described above, the discharge amount obtained by integrating the discharge current during discharge can also be obtained by dividing by the difference in charge rate before and after the discharge.

  In this case, the full charge capacity detected by the discharge and the full charge capacity detected by the charge are generally substantially the same, but differ when experiencing high-temperature standing deterioration or the like. Therefore, while it is left unattended, ambient temperature conditions can be measured at regular intervals with a clock circuit inside the microcomputer, etc., so that it is possible to detect high temperature neglect, etc.In addition, the full charge capacity detected by discharging and the full charge capacity detected by charging When it is detected that the difference is large, it can be confirmed that the deterioration has progressed all at once due to being left at a high temperature or the like, so that it is possible to reliably notify and display the connected devices.

  Also, the remaining time is estimated by estimating the discharge characteristics at the start of discharge, and then the remaining time is changed while detecting the change in the charging rate by current integration, but the actual discharge voltage becomes close to the discharge end voltage. The change may be different from the estimated change in the discharge voltage.

  In such a case, it is preferable to match the estimated remaining time estimated by integration with the estimated remaining time based on the actual change in the discharge voltage. Therefore, the change in the battery voltage is monitored, and when it differs from the estimated discharge characteristic, it is used to estimate the remaining time so that the value indicated by the estimated discharge characteristic approaches the actually measured value. It goes without saying that the accuracy of the remaining time estimated value may be improved by combining the estimated value and the actually measured value by changing the charging rate.

  Furthermore, when the battery voltage that was being measured changes so as to rapidly approach the end voltage, it is close to the end voltage, so an emergency communication that prompts the connected device to end is sent because there is no remaining capacity. Needless to say.

  Although it has been described above that a temperature transition database may be created by actual measurement and the temperature dependence may be taken into account, the temperature dependence may be calculated by calculation using the equivalent circuit of FIG. This makes it possible to reduce the size of the database. The calculation method is as described above.

  Further, although the battery remaining amount estimation method has been described above, it is needless to say that this estimation means does not have to be on the battery pack side. In this case, the estimation process itself can be removed from the battery pack side by, for example, incorporating the remaining amount estimation process into the control means on the device side where the battery pack is mounted. In that case, in order to reduce costs, improve the speed of the estimation process, etc., it is desirable to place most of the databases necessary for the estimation process on the device side together with the estimation processing means.

  More specifically, the unit cell database 6 constituting the battery pack is placed on the electric device 18 side, and the battery pack database 4 and the database for correcting the unit cell database are not on the battery pack 16 side. When the battery pack 16 and the electric device 18 that uses the battery pack 16 are connected, the database 2 necessary for estimation is obtained by capturing the necessary database on the electric device 18 side by means of communication between the battery pack 16 and the electric device 18. Can be provided on the device side. Needless to say, according to such a configuration, it is possible to cope with a change in the unit cell itself or a change in the configuration of the battery pack.

  Furthermore, it was stated that the internal resistance of the unit cell was measured and the database was corrected. However, including this measuring means, the battery voltage and current detection means were left on the battery pack side, and the current integrated value and internal resistance were It goes without saying that the remaining amount estimation method described above can be realized as it is by measuring the measured value in the battery pack and transmitting it to the device side.

Also, even when housed in integral part or all of the battery capacity detection device 10 in the battery pack 16, showing an example of supplying a driving power from the secondary battery 12 in the battery pack 16. However, as illustrated in FIG. 25, a dedicated power source 60 for driving an electronic circuit composed of a small secondary battery or a capacitor is provided in the battery pack 16 at the same time, and by driving the circuit of the battery capacity detecting device 10 with the dedicated power source 60, The influence of current consumption on the secondary battery whose remaining amount is to be detected is prevented, and the OCV and the remaining battery amount can be accurately measured.

  When the voltage value of the dedicated power source 60 is constantly detected by the voltage monitoring unit 61 and it is determined by the voltage determining unit 62 that the value has fallen below the set value, the switch control unit 63 switches the switch 64 to switch the second value. The dedicated power source 60 is charged from the secondary battery 12. When the charging is completed, the charge amount detection means 65 detects the charge amount in the meantime, and the battery remaining amount calculation means 9 subtracts the detected charge amount from the current remaining amount in the secondary battery 12 to obtain the remaining amount. Correction is performed. In this case, if the charge amount for the dedicated power source 60 is constant, the constant amount is stored in advance, and the constant amount may be subtracted without detecting the charge amount each time the dedicated power source 60 is fully charged. Is possible.

  Furthermore, the calibration in the A / D converter provided in the battery pack 16 can be performed by a command from the electric device 18 side to which the battery pack 16 is attached. In this case, a subroutine for calibration is incorporated in advance in the software of the battery capacity detection device 10 provided in the battery pack 16, and the subroutine is executed by an instruction from the electric device 18 side.

  As a specific processing procedure, an instruction to perform A / D conversion is sent from the electric device 18 side to the battery pack 16 side, and the electric device 18 reads the converted A / D conversion value, and whether the value is optimal. Determine whether or not. If it is not the optimum value, the calibration value is sent to the battery pack 16 side, the A / D conversion value is read again, and it is repeated until it is determined as the optimum value.

  Costs can be reduced by automating the calibration operation described above at the time of initial assembly. Further, the calibrated optimum value is preferably written in an electrically erasable non-volatile memory, so that it can cope with a change with time after assembly and can cope with a reset start by a system reset.

It is a block diagram which shows roughly an example which implemented this invention in the battery pack. It is a graph which shows an example of the open circuit voltage characteristic of a secondary battery, Comprising: (a) shows the relationship between a charging rate and an open circuit voltage, (b) shows the determination process of the charging rate using an open circuit voltage characteristic. It is a graph which shows the change state of the battery voltage and charging current at the time of charge with respect to a secondary battery. It is explanatory drawing which shows the equivalent circuit of a secondary battery, and the condition of a frequency response. It is explanatory drawing which shows the estimation process of an open circuit voltage. It is a graph which shows the relationship between an open circuit voltage characteristic and a battery voltage characteristic. It is a graph which shows an example of the discharge characteristic of a secondary battery, (a) is a change of the graph by the difference in load current, (b) is a change of the graph with deterioration of a battery, (c) is ambient temperature. Each of the graph changes with the change is shown. It is explanatory drawing which shows the process of detecting the internal resistance of a secondary battery, (a) is a circuit structure for measuring the internal resistance of a battery, (b) is a change of the electric current supplied to load, (c ) Indicates a change in terminal voltage. It is a graph which shows the relationship between each element value which comprises the equivalent circuit of a secondary battery, and a charging rate. It is a graph which shows an example of the temperature dependence in the element value of a secondary battery. It is an equivalent circuit of a battery pack considering the temperature dependence of the battery. FIG. 12 is a discharge characteristic diagram illustrating a method for obtaining the discharge characteristic based on the equivalent circuit of FIG. 11. 13 is a flowchart for explaining main processing for obtaining the discharge characteristic diagram of FIG. 12. It is a flowchart explaining the 1st process which calculates | requires the discharge characteristic figure of FIG. It is a flowchart explaining the 2nd process which calculates | requires the discharge characteristic figure of FIG. It is a block diagram which shows an example which implemented this invention in the battery pack. It is a whole flowchart which shows the capacity | capacitance detection procedure performed in a control part. It is a flowchart which shows the control procedure in each operation mode. It is explanatory drawing which shows the other process of detecting the internal resistance of a secondary battery, Comprising: (a) shows the circuit structure for measuring the internal resistance of a battery, (b) shows the plot result of complex impedance, respectively. It is explanatory drawing when a secondary battery deteriorates, Comprising: (a) shows the equivalent circuit in that case, (b) shows the plot result of complex impedance, respectively. The graph which illustrated the 1st OCV characteristic and the 2nd OCV characteristic is shown. It is a graph which illustrates the relationship between the battery temperature and the correction amount of the full charge capacity over time with the charging rate as a parameter. It is a graph which shows the discharge characteristic when a surge current flows. It is explanatory drawing which shows the estimation procedure of OCV in case an electric current flows into a battery intermittently also in standby. It is an electric circuit diagram which shows the other Example of a battery capacity detection apparatus.

Explanation of symbols

2 Database for estimation 4 Battery pack database 6 Cell battery database 7 Discharge characteristic estimation means 8 Charge / discharge amount detection means 9 Battery remaining amount calculation means 10 Battery capacity detection device 12 Secondary battery 14 Battery protection unit 16 Battery pack 18 Electrical equipment 20 Communication Circuit 22 FET
24 switching unit 26 protection circuit 28 external circuit group 30 control unit 32 temperature detection circuit 34 voltage detection circuit 36 current detection circuit 38 load circuit 40 temperature detection means 42 switching transistor 44 current detection resistor 46 load resistance 48 measurement control means 50 operational amplifier 52 Transistor switch

Claims (2)

While measuring the open circuit voltage characteristic indicating the relationship between the charging rate and the open circuit voltage in advance for the secondary battery whose battery capacity is to be detected,
A battery capacity detection method for measuring an open circuit voltage of a secondary battery, and estimating a charging rate at the time of measurement from the open circuit voltage obtained by the measurement and the database,
When it is determined that the open circuit voltage measurement process has been started during the period when the load current is intermittently supplied from the secondary battery, an envelope of the voltage change during the period when the current is not flowing is obtained. A battery capacity detection method for estimating an open circuit voltage from a convergence value of an envelope. While measuring the open circuit voltage characteristic indicating the relationship between the charging rate and the open circuit voltage in advance for the secondary battery whose battery capacity is to be detected,
A battery capacity detection method for measuring an open circuit voltage of a secondary battery, and estimating a charging rate at the time of measurement from the open circuit voltage obtained by the measurement and the database,
For each charging rate, the relationship between the battery temperature and the reduction rate of the full charge capacity is provided as a correction database in advance,
When the state in which supply is sealed stop of the load current from the secondary battery is judged that persisted,
While estimating the charging rate from the measured voltage value and open circuit voltage characteristics,
A battery capacity detection method, wherein a correction process of a full charge capacity is performed using a database for correction based on a change in charge rate, battery temperature, and charge rate .

Images