JP2004271532A - Method for determining life of battery - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method for determining the life of a battery, which can reduce, even if a discharge voltage fall-out amount is locally fluctuated, the influence of the fluctuation. <P>SOLUTION: The relation between the integrated value of the discharge voltage fall out amount from the discharge start of the battery and the life value thereof is stored in the life value table of a criterion determination part 44. The measurement data of the discharge voltage fall out amount from the start of discharge to a determination time is integrated by an integrating part 42. In a determination part 43, the life value of the battery is determined from the resulting integrated value by the integrating part 42 in reference to the life value table stored in the determination part 44. <P>COPYRIGHT: (C)2004,JPO&NCIPI

Description

  The present invention relates to a method for determining the life of a battery connected to or mounted on an electronic device or the like.

  2. Description of the Related Art Conventionally, only a few electronic devices having a battery have a function of checking the life of the battery. In the meantime, from the temperature and the load capacity, the judgment voltage drop amount and the judgment time as the judgment reference are judged as normal / abnormal by fuzzy inference. For example, when the judgment voltage drop amount is smaller than the reference, the normal judgment is made. I have.

  In the above-described battery life determination method, when a phenomenon that there is a load fluctuation occurs, the amount of drop locally largely changes near the determination point, and erroneous determination of the battery life occurs. Conventionally, the individual characteristics of the battery were treated as being the same, so that there was only one type of table for determining the life value from the integrated value. For this reason, it was not possible to accurately determine the life of a battery having a large variation.

  Further, in the above-described conventional method, the converted value based on the life value table is used as the determination result. For this reason, an infeasible determination result (for example, a new product but the life is already short) may be displayed as it is to the user, which is inconvenient.

  Further, since the life determination is performed ignoring the temperature change, when the temperature change occurs, the internal temperature and the surface temperature of the battery are different, and there is a problem that the accuracy of the life determination is deteriorated.

  In addition, in the conventional life determination method, since the battery life is measured every fixed period, for example, every month after the use of the battery, for example, in the case of a lead battery, for example, about two years (25 ° C.) after the start of use, There is a problem that the battery progresses rapidly, so that once a month of use elapses, the progress of the battery deterioration cannot be accurately grasped by the judgment once a month.

  Also, as shown in (a) of FIG. 26, the discharge of the battery is stopped so that the terminal voltage becomes a constant discharge end voltage after the start of the discharge. The reason why the discharge is terminated at a constant terminal voltage is that when the battery is discharged to a voltage equal to or lower than the discharge end voltage, the life of the battery is significantly shortened. However, in the case of a cell that causes a phenomenon of “cell drop”, such as a lead battery, if one of the cells connected in series (2 V cell in the case of a lead battery) rapidly deteriorates, FIG. As shown, there is a problem that the terminal voltage quickly reaches a certain discharge end voltage, and the ability of another cell without cell dropout cannot be sufficiently exhibited.

The present invention has been made in view of the above problems,
(1) To provide a battery life determination method capable of reducing the influence of a change in voltage drop amount even when a phenomenon such as a load change occurs and a voltage drop amount locally changes.

  (2) To provide a life determination method capable of accurately determining the life even for a battery whose characteristics are likely to vary.

  (3) Despite the fact that the product is new, the remaining life is short, or even though the product has been used for a considerable amount of time, a judgment result that cannot actually occur is output as if the service life was the same as that of a new product. Provided is a method for determining the life of a battery that can eliminate the possibility of the battery life.

  (4) A battery life determining method capable of performing a highly accurate life determination without being affected by a temperature change before and after the life determination.

  (5) To provide a battery life judging method capable of judging an appropriate life span from the start of use to the end of life.

  According to a first aspect of the present invention, there is provided a battery life determining method, comprising the steps of: storing a relationship between an integrated value of a discharge voltage drop amount from the start of battery discharge and a life value in storage means; Integrating the discharge voltage drop over time with the passage of time from the start of discharge, and the integrated value of the discharge voltage drop and the lifetime value stored in the storage means. And a step of performing a life determination by applying to the relationship described above.

  In this battery life determination method, the discharge voltage drop amount until a predetermined time has elapsed since the start of discharge is integrated, and the life is determined from the integrated value. Hateful.

  According to a second aspect of the present invention, in the method of the first aspect, in order to correct an individual difference in a discharge characteristic of the battery, a process of performing the initial discharge of the battery a plurality of times is provided. Calculating a correction coefficient from the discharge characteristics obtained from the initial discharge and the standard discharge characteristics, and using this correction coefficient, the integrated value of the discharge voltage drop amount stored in the storage means and the life. Correcting the relationship between the values, and the life is determined by applying the integrated value of the measured discharge voltage drop to the relationship between the corrected integrated value of the discharge voltage drop and the life value. .

  In this battery life determination method, in order to correct individual differences in the discharge characteristics of the battery, initial discharge is performed a plurality of times, a correction coefficient is calculated, and the integrated value of the discharge voltage drop amount and the life value are calculated using the correction coefficient. Is corrected, and the life value is obtained by applying the actually measured discharge voltage drop amount to the corrected one, so that the life can be determined without concern for the variation in the characteristics of the battery.

  According to a third aspect of the present invention, in the method of the first aspect, a temperature change is detected when the life is determined, and when the temperature change is equal to or more than a predetermined value, the life is determined after a predetermined time elapses. I'm trying to do it.

  In this battery life determination method, if there is a temperature change equal to or more than a predetermined value at the time of life determination, the life determination is performed after a predetermined time has elapsed, so that the surface and the inside of the battery have almost the same temperature, Life determination can be performed, and accurate determination not affected by temperature change can be performed.

  According to a fourth aspect of the present invention, in the method of the first aspect, the life is determined at a relatively long first cycle until a predetermined period elapses after the use of the battery is started. Performing, and, after the predetermined period elapses, performing a life determination for each second cycle shorter than the first cycle.

  In this method of determining the life of a battery, the life of the battery is determined for each first cycle of a long period for a predetermined period after the start of use of the battery. Since the life is determined at most, the life can be determined without delay even at the end of the life.

  According to a fifth aspect of the present invention, there is provided a battery life judging method according to the first aspect, further comprising: detecting a battery cell drop from the discharge voltage drop amount; Correcting the life of the battery, and performing a life determination on the basis of the corrected measured discharge voltage drop amount when the battery has dropped.

  According to the invention of this application, the discharge voltage drop amount from the start of battery discharge to the elapse of a predetermined time is integrated, and the life is determined from the integrated value of the discharge voltage drop amount. Less susceptible to fluctuations.

  According to the present invention, since the correction coefficient is obtained by a plurality of initial discharges and the variation in individual characteristics is corrected, the life can be determined without worrying about the variation in the characteristics of the battery.

  According to the present invention, if there is a temperature change equal to or more than a predetermined value at the time of determining the life, the life is determined after a predetermined time. It is possible to make accurate determinations without being affected by temperature changes.

  According to the present invention, the frequency of the life determination is changed according to the elapsed time from the start of use of the battery. Therefore, even at a point near the end of the life, appropriate life determination can be performed without delay.

  According to the present invention, when there is a cell drop, the life is determined by correcting the measured discharge voltage drop amount by detecting the drop, so that the life can be determined without worrying about the characteristic change due to the cell drop.

  Hereinafter, the present invention will be described in more detail with reference to embodiments.

  FIG. 1 is a block diagram showing a configuration of an uninterruptible power supply according to the present invention. This uninterruptible power supply includes a battery unit 10 and a control unit 20.

  The battery unit 10 has a built-in chargeable / dischargeable battery (storage battery). The control unit 20 executes an AC plug 21 for connecting to a commercial AC power supply, a power circuit 22, an output outlet 23, a control circuit 24 for controlling the power circuit 22, and a control process for the power circuit 22. Microcomputer 25, a BMCU microcomputer 26 for monitoring the connection state of the battery unit 10, determining the life of the battery, etc., an alarm and LED display unit 27, and a power supply unit for supplying power in the control unit 10. 28. The battery unit 10 and the control unit 20 are connected to each other by a cable 11 and a connector 12. The number of lines of the cable 11 and the number of terminals of the connector 12 are provided at least for charging and discharging and for monitoring. The BMCU microcomputer 26 has an A / D converter for monitoring the voltage of the battery of the battery unit 10 and a battery for backup. In addition, the BMCU microcomputer 26 has various functions for determining the life of the battery. A temperature sensor 14 is attached to the battery unit 10, a load sensor 32 such as a current transformer is attached to a load 30 to which an AC plug 31 is connected to the output outlet 23, and the temperature sensor 11 and the load sensor 32 are attached. Are taken into the BMCU microcomputer 26.

  In this uninterruptible power supply, an AC plug 31 of an electronic device is usually connected to the output outlet 23 as the load 30. Then, during a non-power failure, the AC voltage input from the AC plug 21 is rectified and converted into a DC voltage, and then converted again into an AC voltage by an inverter. Supplying voltage. Further, the battery of the battery unit 10 can be charged via the power circuit 22 and the power supply unit 28. When an AC voltage is not supplied from the AC plug 21 due to a power failure, the power circuit 22 is switched by the control circuit 24 under the control of the UMCU microcomputer 25, and the voltage from the battery of the battery unit 10 is output to the power circuit 22 for output. The power is supplied to each electronic device via the outlet 23. At this time, the BMCU microcomputer 26 continuously monitors the battery connection state, the voltage, the usage time, and the like of the battery unit 10 regardless of the presence or absence of the AC power (power ON / OFF).

  The uninterruptible power supply of this embodiment has a function of discharging the battery of the battery unit 10 and calculating an integrated value of a discharge voltage drop amount from the start of discharging, and a table for determining a life from the integrated value at a battery life determination point. To reduce the effects of local fluctuations. This feature is achieved by the integration unit 42 shown in FIGS. 2 and 3, the measured data 41 of the discharge voltage drop is integrated by the integration unit 42, and the lifetime stored in the determination criterion determination unit 44 by the determination unit 43. With reference to the value table (life value storage means), the life of the battery is determined from the integrated value of the discharge voltage drop amount.

  The uninterruptible power supply according to this embodiment has a function of performing initial discharge three times and correcting a life value table from a characteristic discharge curve thereof. A table can be provided, and accurate life determination can be performed for each individual. This feature is achieved by the life value table calculator 45 shown in FIGS. The life value table calculator 45 counts the number of times of the initial discharge 71 three times by the counter 72. During this time, the life value table of the basic life value table 74 is corrected from the characteristic discharge curve, and the individual life value table is corrected. Is used to determine the life of the battery.

  The uninterruptible power supply of this embodiment has a function of measuring the elapsed time and a function of measuring the charging history, and can judge the life by comprehensively fuzzy inferring the life value, the elapsed time, and the charging history. As a result, it is possible to eliminate the possibility that a new determination result that the life is the same as that of a new product that has been used for a considerable number of years has been obtained. This feature is achieved by inferring the total fuzzy inference unit 52 shown in FIG. 2 using the life value from the determination unit 43, the input charging history 52, and the elapsed time 53.

  Further, the uninterruptible power supply according to this embodiment has a temperature determination function, and when the battery life determination is started immediately after a certain temperature change occurs, the life determination is performed immediately after the battery life determination is waited. Has functions. As a result, the life can be determined while keeping the state of the battery constant, and the life of the battery can be accurately determined. This feature is achieved by the standby unit 50 shown in FIGS. The temperature stored in the temperature memory 81 30 minutes ago and the temperature 82 measured this time are compared by the comparing unit 83, and if there is a change in temperature by a certain amount or more, the life judgment is started after a while.

  In addition, the uninterruptible power supply of this embodiment determines the battery life every month for two years after the start of use of the battery, and performs the life determination every week after two years. In other words, a function is provided for shortening the life determination cycle with the passage of time. As a result, it is possible to accurately grasp the progress of deterioration of the battery whose life is approaching. This feature is achieved by the timer section 48 shown in FIGS. For two years (24 months) from the start of use of the battery by the timer selection unit 93 of the timer unit 48, the one-month timer 91 activates the determination unit 43 every month to determine the life. When two years have elapsed after the start of use of the battery, the timer selection unit 93 selects the one-week timer 92, and the timer 92 activates the determination unit 43 every week to determine the life. Here, the period of one month and one week may be appropriately changed depending on the type of the battery, or may be divided into several times and the cycle may be sequentially shortened.

  Next, a battery life determining process in the uninterruptible power supply according to the embodiment will be described. In the apparatus of this embodiment, the point in the life determination process is that the correction coefficient is calculated, the integrated value is corrected with the correction coefficient for the actually measured discharge voltage drop amount, and the life is calculated from the integrated value of the discharge voltage drop amount. That is, a value is obtained. This correction coefficient is determined by three initial discharges. Until the correction coefficient is determined, the battery life is determined using a discharge voltage drop curve interpolated from the standard discharge voltage drop curve. After the correction coefficient is determined, the battery life is determined using the corrected discharge voltage drop amount.

The correction coefficient is calculated by fuzzy interpolation of the table of integrated values of load and temperature conditions (the integrated value of discharge voltage drop amount) from the integrated value table of 16 points of temperature × load (see the example of FIG. 12). 17, see the membership function in FIG. 18). Compare the integrated value obtained by this fuzzy interpolation with the integrated value of the actual measurement value,
Correction coefficient = [integrated value at actual measurement (digit S) / [integrated value at 100% fuzzy interpolation (digit S: digit / second, indicating unit of {quantity (numerical value) x time}]]] Ask for.

  In updating the correction coefficient, the correction coefficient is stored at the time of the first initial discharge, and the average is obtained by adding the first correction coefficient and the second correction coefficient at the time of the second initial discharge. . In the case of the third initial discharge, the final correction coefficient is defined by the final correction coefficient = {(second correction coefficient × 2) + third correction coefficient} / 3.

  Next, the overall operation of life determination of the uninterruptible power supply according to the embodiment will be described with reference to flowcharts shown in FIGS.

  Upon entering the life determination routine, a first discharge start command is issued [step ST (hereinafter abbreviated as ST) 1]. To activate the battery, the normal state (power supply from the AC power supply) is switched to the backup state using the battery, and the battery of the battery unit 10 is discharged. Next, at the point when an arbitrary few seconds (basic time 10 seconds) have elapsed after the start of the discharge, the load current is measured eight times continuously. The load current is converted to load power, an average of eight times is obtained, and a first discharge time is calculated (ST2).

  Subsequently, it is measured whether the measured temperature is within a predetermined range (ST3). If the measured temperature is not within the predetermined range, the discharge is terminated, a judgment impossible is transmitted (ST4), and the process returns. If the temperature is within the predetermined range, it is further determined whether the load is within the predetermined range (ST5), and if not, the discharge is terminated, a determination impossible is transmitted (ST6), and the routine returns. If both the temperature and the load are within the predetermined ranges, it is further determined whether the load fluctuation is smaller than 10% (ST7). Here, the moving average of the load power is calculated eight times, and when a load fluctuation of 10% or more occurs in the measured average load power, the discharge is terminated and the determination impossible is transmitted (ST8).

  Next, it is checked whether the discharge time obtained in ST2 has elapsed (ST9), and the loop of ST7 and ST8 is performed every second until the discharge time has elapsed. When the discharge time has elapsed, a first discharge stop command is issued (ST10), the battery returns from the backup state to the normal state, and charging of the battery is started. Then, it waits for two minutes to elapse from the recharged state (ST11). Here, the first discharge and the charge for 2 minutes are performed by performing a discharge for activating the state of the battery and then entering a main measurement after a pause to make a correct determination. That's why.

  Following the two-minute rest, a second discharge start command is issued (ST12), and the main discharge for life determination is started. That is, the state is switched from the normal state to the backup state. Also, at the point of time when an arbitrary number of seconds (basic time 10 seconds) have elapsed after the start of discharge, the load current is measured eight times continuously (ST13). Next, it is determined whether the load fluctuation is smaller than 10% (ST14). If the load fluctuation is 10% or more, the determination in ST14 is NO, the discharge is terminated, and a determination impossible is transmitted (ST15). If the load fluctuation is smaller than the predetermined value in ST14, the load fluctuation value is stored for use in 1 for updating the correction coefficient (ST16).

  Next, an integrated value (integrated value table) of the discharge voltage drop amount at the measured temperature and the load is obtained (ST17), and an integrated power value (warning determination timing) is obtained from the measured temperature and the load from the determined elapsed power rule table. (ST18). In this integrated value table, a load and temperature condition integrated value table at the time of measurement is obtained by fuzzy interpolation from an integrated value table of 16 points of temperature × load. In the calculation of the fuzzy interpolation, a rule is created for each life step (100%, 90%, 80%,...) Of the integrated value table, and the temperature and load membership functions (see FIGS. 17 and 18) are used. Perform fuzzy interpolation. This calculation is repeated for every life value.

  For example, taking interpolation at 180 W and 20 ° C. as an example, a 100% interpolation rule shown in FIG. 12 is created from the 16 integrated value tables shown in FIG. , And an interpolation result of 180 W and 20 ° C. shown in FIG.

The power integrated value (judgment timing) determines how many WS (watt seconds) the battery life should be determined at a certain temperature and load, and is usually about 16700 WS. Fuzzy interpolation is performed using the membership function of temperature and load and the power integration value (judgment timing) rule. Then, the power integrated value is divided by 10 to calculate a power integrated value of one section obtained by dividing the integrated area into ten sections (ST19). Hereinafter, cell drop detection is performed in the processing of ST19,..., ST44. The power integrated value of one section is divided by the power value to obtain the duration T of one section (ST20). Since the measurement time varies depending on the temperature and the load, this one section is easy to change, but is usually about 1670 WS, for example, about 7 seconds for a 250 W load. Further, to enter the operation of each section, clears the integrated value S _s of the measured discharge voltage drop amount (ST21), the section I: performing the calculation of (I = 0, ..., 9 sequentially incremented from 0) (ST22).

First, a section initial value D _ss [I] of the measured discharge voltage drop D _s is obtained (ST23). Next, an interpolated discharge voltage drop curve at the measured temperature and load is obtained (ST24). The interpolated discharge voltage drop amount curve is obtained as follows. For example, in the case of performing interpolation at 180 W and 20 ° C., first, from the table of the discharge voltage drop amount and the accumulated power integration shown in FIG. As shown in FIG. 15, an interpolation rule for each digit is created from one digit. From these, the interpolation result at a load of 180 W and a temperature of 20 ° C. shown in FIG. 16 can be obtained. Further, the section initial value D _hs [I] of the interpolated discharge voltage drop amount D _h is obtained (ST25). Then, it is determined whether or not the discharge is terminated (ST26). At first, the discharge is not terminated, but the time comes soon, and the determination of ST26 is YES. That is, in the case of the discharge terminated, the discharge is terminated (ST64). Complete degradation communication is performed, and alarm communication is performed when a warning is determined.

If the determination of whether or not to stop discharging in ST26 is NO, it is determined whether the power integrated value has elapsed 500 WS [Watt seconds] (ST27). During NO, the process returns to ST26 until the power integrated value exceeds 500 WS. , ST26 and ST27, and when YES is determined, D _s -D _ss [I] is integrated with the integrated value S _sk [I] of the section-measured discharge voltage drop (ST28), and the measured discharge voltage drop is integrating the D _s to the integrated value S _st (ST29). Here, if the increments 500WS crosses a section boundary, the discharge voltage drop amount D _s, measured in the increments were apportioned WS values belonging to each section integrates the prorated belonging to the self interval. D _s is calculated by the interval initial value D _hs of the previous period as a reference.

In ST30, an interpolating drop amount curve at the measured temperature and load is obtained. Next, the interpolated discharge voltage drop amount D _h -the interpolated initial value D _hs [I] of the interpolated discharge voltage drop amount is integrated with the integrated value S _sk [I] of the interpolated discharge voltage drop amount (ST31), It is determined whether or not one section is completed (ST32). The processing of ST26,..., ST32 is repeated until the one section is completed, and the interpolated discharge voltage drop amount of one section is integrated.

Next, the interpolated life value table is multiplied by the integrated value S _hk [I] of the interpolated discharge voltage drop amount and divided by the table integrated value of 100% to create a life value table for the interval I (ST33). . Then, the section life value J [I] is obtained from the integrated value S _hk [I] of the discharge voltage drop amount obtained through the section interpolation and the life value table of the section [I] (ST34). This is stored in the array (ST35). The format of the array is as shown in FIG. In FIG. 22, I, I-1, and I-2 indicate adjacent sections, J indicates a section life value, correction F indicates a correction flag, and numerical value H indicates a correction numerical value. D _ss [K] −D _hs [K].

  In order to determine whether or not the cell has dropped, first, the section life value J [I] is smaller than 40%, and either the section life value J [I-1] or the section life value J [I-2] is satisfied. It is determined whether it is 60% or more (ST36). If the determination is YES, the flag F [I] is set to 1 (ST37), and the process proceeds to ST38. If the determination is NO, the process directly proceeds to ST38. In ST38, it is determined whether the section life value J [I-1] is smaller than 40% and either the section life value J [I] or the section life value J [I-2] is 60% or more. If the determination in ST38 is YES, the flag F [I-1] is set to 1 (ST39), and the process proceeds to ST40. If the determination is NO, the process directly proceeds to ST40.

In ST40, it is determined whether the section life value J [I-2] is smaller than 40% and the section life value J [I] or the section life value J [I-1] is 60% or more. If the determination in ST40 is YES, the flag F [I-2] is set to 1 (ST41), and the process proceeds to ST42. If the determination is NO, the process directly proceeds to ST42. In ST42 the flag F (I-2) is determined one, if the determination YES is cell drop, multiplied by the correction of -H [I-1] the discharge voltage drop D _s, measured after continuing together (ST43), from the integrated value S _s of the measured discharge voltage drop amounts, calculates the accumulated number of H [I-1] × 2 × 1 section (ST44). As described above, the presence or absence of a cell drop is determined, and if there is a cell drop, correction is automatically made in the next section.

Next, the process proceeds to ST45, where it is determined whether or not I = 4. If the determination is YES, that is, if the section I is 4, the alarm level is determined in the processing after ST46. It checks for severely degraded batteries in the middle stage. First, life value calculated from the integration amount S _s of the measured discharge voltage drop amount, and stores (updates) the life value (ST46). Then, fuzzy inference is performed (ST47). The input values are the lifetime value membership function, the elapsed month membership function, and the charging history membership function, and the labels of the membership functions in the conclusion part are normal, warning, and alarm. The membership functions of the life value, the elapsed month, the charging history, and the conclusion part are as shown in FIGS. 20, 21, 22, and 24. The fuzzy rules are shown in FIG. 23, in which charging history Low (0), MID (120), and HIG (240) are shown in the upper, middle, and lower rows, and the elapsed month NEW (12) is shown in the row direction of each row. ), MED (36), and OLD (60) are further arranged in the column direction of each row, and their life values are 100%, 70%, and 30%, and are shown in three dimensions. The life value is calculated by a weighted average. The life value here is distinguished from the fuzzy input life value. The calculation method firstly (1) multiplies the fitness (27) for each matrix by 100%, 50%, and 0% of the conclusion part. (2) All calculated values (27) are added, and this value is divided by the overall goodness of fit (a value obtained by adding 27 goodnesses of fit). (3) Take a weighted average of the calculated values. As a result of this calculation, if the life value is 100 to 70, "normal", if 70 to 30, "warning", and if 30 to 0, "alarm".

  Returning to the flowchart of FIG. 10, if the result of the fuzzy inference in ST47 is an alarm, the discharge ends and alarm communication is performed (ST48). However, if the inference result is normal or a warning, the process returns to ST22 without any problem and shifts to the calculation of the next section.

If it is determined in ST45 that I is not I = 4 but another section, the process proceeds to ST49, and it is determined whether I = 9. If the determination is YES, that is, if the section I is 9, the warning level is set in the processing after ST50. Make a decision. This is the battery life value at the final stage. First, the life value is calculated from the accumulated amount S _s of the measured discharge voltage drop amount, and stores (updates) the life value (ST50). Then, similarly to ST47, fuzzy inference is performed using the membership functions of the life value, charging history, and elapsed month as inputs. Each membership function and calculation method are the same as those described in ST47. As a result of the fuzzy inference, when the discharge is normal, the discharge is terminated, the normality determination communication is performed to the UMCU microcomputer 25 (ST52), a second discharge end command is issued (ST53), and the normality determination communication is performed (ST54). Next, it is determined whether or not the discharge is the third discharge (ST55). If the initial discharge is the third discharge, the process proceeds to ST56. In ST56, a correction coefficient is calculated. As described above, the calculation method is the integrated value for actual measurement (digit S) / the integrated value for 100% fuzzy interpolation (digit S). Next, it is determined whether the load fluctuation is smaller than 5%. If the determination is YES, the measurement counter is incremented (ST58), and it is determined whether the correction coefficient is updated (ST59). If the determination in ST59 is YES and the update is performed by the third initial discharge, the update is calculated by {(second correction coefficient × 2) + third correction coefficient} / 3 and stored (updated) as the final correction coefficient ( ST60). The last calculated life value is stored (ST61), and the routine returns.

  If the result of the fuzzy inference in ST51 is a warning, the discharge is terminated, a warning communication is performed (ST62), and the process returns. If the fuzzy inference result is an alarm, the discharge is terminated, an alarm is communicated (ST63), and the process returns.

  The above-described life determination is periodically executed as shown in a part of the main flowchart of FIG. Unless 24 months have passed since the start of use of the battery of the connected battery unit, the determination of whether or not 24 months has passed in ST71 is NO. In this case, 30 days have passed since the last life determination date. It is determined whether the day has passed (ST72). The routine returns until 30 days have elapsed, but when 30 days have elapsed, the 30-day timer is reset (ST73) and the life determination process is started. That is, as long as the month and day have not elapsed since the start of use of the battery, the life determining process is executed every 30 days.

  If 24 months have elapsed since the start of use of the battery, the determination in ST71 becomes YES, and it is determined whether seven days have passed since the last life determination date (ST74). The routine returns until seven days have elapsed, but when seven days have elapsed, the seven-day timer is reset (ST75) and the life determination process is started. In this case, years have passed since the start of use of the battery, and in the life determination every 30 days, even if the life is determined to be normal in the current life determination, the battery may deteriorate by the next determination after 30 days. Therefore, the life is determined every seven days so that the end of the life can be immediately determined. Although the uninterruptible power supply of the above embodiment has shown the case where the battery unit and the control unit are separate bodies, the present invention can be applied to a case where both are integrated or a case where the control unit is integrated with an electronic device. It goes without saying that it can be applied. In the uninterruptible power supply of this embodiment, when a battery drop is detected, as shown in FIG. 26 (c), the discharge end voltage is set to be lowered by V, and other cells without cell drop are set. That can be used effectively.

  Further, in the uninterruptible power supply of the above embodiment, the relationship between the integrated value of the discharge voltage drop amount and the life value is stored in a table. A value or the like may be calculated.

FIG. 1 is a block diagram showing a configuration of an uninterruptible power supply according to the present invention. It is a block diagram showing the whole functional composition of the uninterruptible power supply of the embodiment. It is a block diagram which shows the structure of the integrating | accumulating part of the same functional structure. It is a block diagram which shows the structure of the life value table calculation part of the same functional structure. It is a block diagram which shows the structure of the standby part of the same functional structure. It is a block diagram which shows the structure of the timer part of the same function structure. It is a flowchart for demonstrating the life determination processing operation | movement of the said uninterruptible power supply. It is a flowchart for explaining the life determination processing operation of the uninterruptible power supply of the embodiment together with the flowchart of FIG. 9 is a flowchart for explaining a life determining process operation of the uninterruptible power supply according to the embodiment, together with the flowcharts of FIGS. 7 and 8. 10 is a flowchart for explaining the life determination processing operation of the uninterruptible power supply according to the embodiment, together with the flowcharts of FIGS. 7, 8, and 9. It is a figure explaining the life value table of the battery for every load and temperature. It is a figure showing an interpolation rule at the time of 100% of a life value. It is a figure which shows the example of an interpolation result of 180W and 20 degreeC. It is a figure explaining a standard discharge voltage drop amount curve. It is a figure which shows the interpolation rule at the time of 1 digit of a standard discharge voltage drop amount and an integrated value. It is a figure which shows the example of an interpolation result of 180W and 20 degreeC. It is a figure which shows the membership function of the temperature made into the input of the fuzzy inference part of the uninterruptible power supply of the said embodiment. It is a figure showing the membership function of the load made into the input of the same fuzzy inference part. It is a figure showing the membership function of the life value used as the input of the same fuzzy inference part. It is a figure which shows the membership function of the elapsed month used as the input of the same fuzzy inference part. FIG. 3 is a diagram showing a membership function of a charging history as an input to the fuzzy inference unit. FIG. 4 is a diagram for explaining array storage used for cell drop detection. It is a figure showing the rule used by the above-mentioned fuzzy inference part. It is a figure showing the membership function of the conclusion part of the same fuzzy inference part. It is a flowchart which shows a part of main flow of the said uninterruptible power supply. FIG. 4 is a diagram illustrating a cell drop of a battery.

Explanation of reference numerals

41 Measurement data 42 Accumulator 43 Judgment unit 44 Judgment criterion determination unit

Claims (5)

Storing the relationship between the integrated value of the discharge voltage drop from the start of battery discharge and the life value in the storage means,
Discharging the battery and integrating the discharge voltage drop over time from the start of discharging; and
Applying the discharge voltage drop amount integrated value at the elapse of the predetermined time to the relationship between the discharge voltage drop amount integrated value and the life value stored in the storage means, and performing a life determination;
A method for determining the life of a battery, comprising: In order to correct individual differences in the discharge characteristics of the battery, a process of performing the initial discharge of the battery a plurality of times,
A process of calculating a correction coefficient from the discharge characteristics obtained from the plurality of initial discharges and the standard discharge characteristics,
Using the correction coefficient, correcting the relationship between the integrated value of the discharge voltage drop amount and the life value stored in the storage means,
Further comprising
2. The battery life judging method according to claim 1, wherein the life judgment is performed by applying the measured discharge voltage drop amount integrated value to the corrected relationship between the integrated value of the discharge voltage drop amount and the life value.   2. The battery life judging method according to claim 1, wherein a temperature change is detected when the life judgment is started, and when the temperature change is equal to or more than a predetermined value, the life judgment is performed after a predetermined time elapses. Performing a life determination for each relatively long first cycle until a predetermined period elapses after the use of the battery is started;
Performing the life determination for each second cycle shorter than the first cycle when the predetermined period elapses;
The method according to claim 1, further comprising: A process of detecting a battery cell drop from the discharge voltage drop amount,
Correcting the measured discharge voltage drop amount according to the cell drop detection,
2. The battery life judging method according to claim 1, wherein the life judgment is performed based on the corrected measured discharge voltage drop amount at the time of detection of the battery cell drop.

Images