JP2014163921A - Charging/discharging test system and control device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To effectively process a specific measurement value.SOLUTION: A charging/discharging test system 10 includes: an AC power source 14 for supplying a power for a charging/discharging test; a charging/discharging test unit 18 for charging or discharging a battery 12 to be tested, and for measuring physical quantity related to the battery 12 to be tested; and a control device 50 for controlling the charging/discharging test unit 18. The control device 50 includes: an acquisition part for acquiring the measurement value of the physical quantity such as currents from a measurement part 30 via an information processing part 20; an arithmetic part for performing an arithmetic operation for filtering on the basis of the change rate of the measurement value; a determination part for determining the validity of measurement data; and a threshold adjustment part for adjusting a threshold to be used for determination in accordance with the fluctuation tendency of the measurement data.

Description

  The present invention relates to data processing techniques, and more particularly to techniques for filtering singular measurements.

Due to the recent increase in awareness of environmental problems, rechargeable batteries, that is, secondary batteries, are widely used in various fields. Ensuring the reliability of a secondary battery is an important issue, and generally, a charge / discharge test is performed in the research and development and manufacture of the secondary battery.
In the charge / discharge test, charging / discharging of the secondary battery, which is a sample to be tested, is repeated, and physical quantities such as current, voltage, and temperature of the secondary battery accompanying charging / discharging are continuously measured at predetermined time intervals. Then, using the obtained measured values, control of input value determination during the charge / discharge test, switching of the test step, stop at the time of abnormality detection, and the like are performed, and the charge characteristics and discharge characteristics of the secondary battery are evaluated.

In various scenes, not limited to charge / discharge tests, measured values are continuously monitored, and control and evaluation are performed according to the measured values.
For example, in Patent Document 1, during charging of a charging vehicle, a change rate of current is calculated in addition to a moving average of current or a short-time current value, and it is determined that charging is completed by taking into account the history of change rate or abnormal And a charging control system that stops charging is disclosed.

JP 2012-5172 A

  For example, in a charge / discharge test of a secondary battery, measured values such as voltage, current, and temperature of the secondary battery are continuously acquired, and the progress of the charge / discharge test is controlled based on the acquired measured values.

  The charge / discharge test includes a plurality of steps, and the steps may be switched according to a trigger condition set in advance with respect to measured values of current and voltage. In general, when abnormal measured values exceeding a preset threshold value are obtained during a charge / discharge test, protection measures such as stopping the test are taken to protect the secondary battery being tested and to ensure the safety of the test. (This function is hereinafter also referred to as “protection function”).

  However, due to noise from measurement systems, AD converters, other circuits and peripheral devices, and shifts in access timing from cooperating units, physical quantities such as actual current and voltage are in the normal range. Therefore, an abnormal value exceeding the normal range may be erroneously detected as a measured value.

Due to such an erroneously detected abnormal measurement value, an unexpected operation may be caused by the protection function and feedback control function of the charge / discharge device.
For example, due to false detection of an abnormal value exceeding the threshold specified by the protection function, the test equipment is automatically stopped even though there is no actual problem, or the erroneously detected measurement value satisfies the step change trigger condition, causing an unplanned step. Or the charge / discharge test may not be completed.
For example, a large-scale charge / discharge test is performed for about 100 batteries under test in a single test while changing the conditions over the same conditions over several hundred hours. In some cases, the test of some batteries may be stopped. In such a case, the reliability of the entire test may be questioned, but it is difficult to perform a large-scale test again in terms of cost.

  The present invention has been made in view of these problems, and an object thereof is to effectively process singular measurement values.

  In order to solve the above-described problems, a charge / discharge test system according to an aspect of the present invention is a charge / discharge test system for a secondary battery, and includes a measurement unit that measures a physical quantity related to a battery under test, and a measurement value obtained by the measurement unit. And a control device for determining effectiveness. The control device uses the acquisition unit that acquires the measurement value from the measurement unit and the measurement value, and the measurement value at the measurement time t (n−k) before the nth measurement time t (n) (where n is an integer). The rate of change d1≡ | A (n) −A (n−k) | / Δt (where n is the natural number) and the measured value A (n) at the nth measurement time point t (n) Δt≡t (n) −t (n−k)), a first reference change rate d0 that is a change rate of a measurement value at a measurement time before the nth measurement time t (n), and an nth measurement time t. A calculation unit that calculates a first change rate ratio r, which is a ratio between the change rate d1 and the first reference change rate d0 in (n), and compares the first change rate ratio r with a threshold value, and the nth measurement time point and a determination unit that determines the effectiveness of the measurement value A (n) at t (n).

  Another aspect of the present invention is a control device. This device uses a measurement unit to acquire a measurement value and a measurement value A (n) at a measurement time t (n−k) prior to the nth measurement time t (n) (where n is an integer). −k) (where k is a natural number) the rate of change d1≡ | A (n) −A (n−k) | / Δt (where Δt≡t) at the nth measurement time point t (n) (N) −t (n−k)), the first reference change rate d0 that is the change rate of the measured value at the measurement time before the nth measurement time t (n), and the nth measurement time t (n). A calculation unit that calculates a first change rate ratio r, which is a ratio of the change rate d1 and the first reference change rate d0, and the first change rate ratio r is compared with a threshold, and the nth measurement time point t (n And a determination unit that determines the effectiveness of the measurement value A (n) in ().

  It should be noted that any combination of the above-described constituent elements and a representation obtained by converting the expression of the present invention between a method, an apparatus, a system, a program, and the like are also effective as an aspect of the present invention.

  According to the present invention, singular measurement values can be processed effectively.

It is a schematic diagram which shows the example of a change of the electric current value in a charging / discharging test system. It is a figure which shows the structure of the charging / discharging test system concerning embodiment. It is a figure which shows the structure of the control apparatus and user interface of the charging / discharging test system of FIG. It is a figure which shows the filtering process when it determines with the measured value of the last round being effective. It is a flowchart which shows the filtering process when it determines with the measured value of the last round being invalid. It is a flowchart which shows the example of mounting of the filtering process by the charging / discharging test system concerning embodiment. It is a flowchart which shows the change rate ratio calculation process of FIG. It is a flowchart which shows the determination / update process of FIG. It is a figure which shows the example of an electric current measured value for demonstrating a determination process. It is a figure which shows the example of an electric current measured value for demonstrating a determination process. It is a figure which shows the example of a change of an electric current value. It is a flowchart which shows the threshold value adjustment process of FIG.

First, an outline will be described.
The charge / discharge test system according to the embodiment of the present invention monitors physical quantity measurement values such as current, voltage, temperature, and the like related to a battery to be tested. And when the value which shows the tendency different from the tendency of the change of the measured value until then is obtained, it detects as a singular value. Furthermore, it is determined whether the singular value indicates a change in the actual physical quantity of the battery or is a false detection value due to noise or the like. If it is a false detection, the data is excluded from the sampling target (also referred to as “invalid data”). remove. As a result, it is possible to prevent the test from being automatically stopped or inadvertently transferred to the step due to the erroneously detected measurement value.

FIG. 1 is a schematic diagram illustrating an example of a change in current value in a charge / discharge test.
In general, the content of the charge / discharge test is determined by parameters such as test time, cycle, voltage, current, power, and battery capacity. The charge / discharge test system charges or discharges the test target battery according to the planned execution schedule, and the battery current and the like are continuously measured in a short measurement unit time of microseconds or nanoseconds. In FIG. 1, the measurement unit time Δt is indicated by one scale on the time axis.
In FIG. 1, the actual current value change is indicated by an actual current value curve 1, and the measured current value is indicated by a black circle.

In the present specification, a measured value that shows a variation different from the variation tendency so far is also referred to as a “singular value”. For example, the measurement value 2 and the measurement value 3 in FIG. 1 are singular values.
The charge / discharge test system according to the embodiment automatically detects a singular value based on a state of fluctuation of a measured value. The state of fluctuation of the measurement value can be expressed as a numerical value by using, for example, the time change rate between the two measurement values. The time change rate corresponds to the slope of a straight line (not shown) connecting the measurement points in the graph of FIG. For example, in FIG. 1, the rate of change at time t2 can be expressed as ΔI2 / Δt.

The charge / discharge test system according to the embodiment compares the current change rate with the previous change rate, and determines whether or not the current measurement value is a singular value.
For example, in FIG. 1, since the change rate ΔI3 / Δt at time t3 is considerably larger than the change rate ΔI2 / Δt at the previous time t2, the measured value 2 at time t3 is determined to be a singular value.
Similarly, since the rate of change ΔI9 / Δt at time t9 is considerably larger than the rate of change ΔI8 / Δt at the previous time t8, the measured value 3 at time t9 is determined to be a singular value.

Even if there is a fluctuation as a whole, such as the actual current value curve 1 from the time t1 to the time t4 in FIG. 1 by the determination using the ratio between the current change rate and the previous change rate. Thus, a singular value indicating a variation different from the variation tendency can be detected.
The detected singular value is excluded from the data used for the protection function, feedback control function, or test result analysis of the charge / discharge test system. Hereinafter, the measurement value not used for the protection function, the feedback control function, or the test result analysis of the charge / discharge test system is also referred to as “non-target measurement value” or “non-target data”. A measured value that is not a singular value is also referred to as an “effective” measured value.

In this specification, when a singular value is detected in a short time, for example, 1 to 2 measurement unit times, for example, as measured value 2 in FIG. 1, and the subsequent measured value shows the same fluctuation tendency as the original, the singular value is Called a single singular value.
As described above, a value different from the actual current value may be erroneously detected as a measurement value due to noise from the system or peripheral devices, a shift in access timing, or the like.
A single singular value is considered to be such a misdetected measurement.
Since the charge / discharge test system according to the present embodiment detects singular values and handles them as non-target data, it can prevent inadvertent operations of the protection function and feedback control function of the charge / discharge device due to erroneous detection data.

In calculating the rate of change, the charge / discharge test system according to the embodiment calculates the rate of change from a measured value determined to be valid among the previous measured values.
For example, as the rate of change at time t4 in FIG. 1, the rate of change ΔI4 / 2Δt with respect to the measured value at time t2, which is an effective measured value, is calculated instead of the rate of change with respect to measured value 2 at time t3, which is a non-target measured value.

Further, the change rate of the measured value that is validated among the previous change rates is used as the previous change rate (also referred to as “reference change rate”) compared with the current change rate.
For example, when determining the validity of the measured value at time t4, the rate of change at time t4 is not compared with the rate of change at time t3, but the rate of change at time t4 is compared with the rate of change at time t2. Since the rate of change ΔI4 / 2Δt at time t4 is similar to the rate of change ΔI2 / Δt at the previous time t2, the measured value at time t4 is determined to be valid.
As a result, the normal measured value after the single singular value can be determined as valid data while detecting the single singular value as non-target data.

In general, since the charge / discharge test measures the change in the state of the battery by changing the power supply and load, the rate of time change such as current and voltage usually fluctuates even during the test.
For example, the rate of change, that is, the slope in the graph is different between time t8 in FIG. 1 and after time t8. As described above, a state in which the rate of change greatly fluctuates as a whole from the previous value is also referred to as a “transition state” in the present specification. In general, a measured value indicating a transition state is considered to reflect a change in actual current, voltage, etc., not a false detection value.

  For example, like the measured value 3 in FIG. 1, there is a singular value indicating a variation different from the variation tendency so far even in the transition state. In the charge / discharge test system according to the embodiment, for example, the measurement value 3 at time t9 is determined to be a singular value, but the measurement value 4 at the next measurement time t10 is singular value even if it differs from the tendency of fluctuations so far. Instead, it is judged as a transition state and treated as an effective measurement value.

However, according to the above singular value determination method, the rate of change of the measured value 4 at the measurement time t10 is expressed as the rate of change ΔI10 / 2Δt of the measured value 4 with respect to the measured value at the time t8, which is an effective measured value among the previous measured values. Calculated.
When the rate of change ΔI8 / Δt at time t8, which was validated among the previous rates of change, is compared with the rate of change ΔI10 / 2Δt of the measured value 4, the rate of change, that is, the slope has changed greatly. The value is also determined as singular data.

  Therefore, in the charge / discharge test system according to the embodiment, in the validity determination of the next measured value of the singular value, even if it is not effective based on the past effective rate of change, the rate of change immediately before is used as a reference. Judgment is made. If the criterion is satisfied, the transition state is determined and treated as valid data.

For example, in the case of the example in FIG. 1, the slope of a line segment (not shown) connecting the measurement value at time t8 and the measurement value at time t9 and the line segment connecting the measurement value at time t8 and the measurement value at time t10 (not shown). As can be seen from the fact that the slope of () is comparable, the rate of change ΔI10 / 2Δt at the measurement time t10 and the rate of change ΔI9 / Δt immediately before are the same. Therefore, the measured value at time t10 is determined to be valid.
As a result, even when the rate of change greatly fluctuates, if the rate of change after the change shows a certain tendency, it is determined as a transition state and processed as valid data.

Details will be described below.
FIG. 2 shows a configuration of the charge / discharge test system 10 according to the embodiment of the present invention.

The charge / discharge test system 10 charges or discharges the AC power supply 14 that supplies power for the charge / discharge test, the battery under test 12 to be tested in the charge / discharge test, and measures the physical quantity related to the battery under test 12. The discharge test unit 18, the control device 50 that controls the charge / discharge test unit 18, the user interface 63 that acquires instructions from the user and presents the test results to the user, and the discharge energy is recovered when the battery under test 12 is discharged. Then, a regenerative inverter 16 that returns to the AC power source 14 is included.
Although only one charge / discharge test unit 18 is shown in FIG. 1, the charge / discharge test system 10 includes a plurality of charge / discharge test units 18 and can be configured to perform a charge / discharge test on a plurality of channels. Good. In this case, each charge / discharge test unit 18 tests one battery under test 12, and the control device 50 comprehensively controls the plurality of charge / discharge test units 18. Each battery under test 12 may include a plurality of cells.

The charge / discharge test unit 18 includes a converter unit 40 that charges or discharges the battery under test 12 in accordance with a test cycle specified in advance, a measurement unit 30 that measures a physical quantity related to the battery under test 12 every measurement unit time, and a preset value. An information processing unit 20 is provided that sends an instruction to the converter unit 40 in accordance with the charged / discharge test plan and generates log data for processing the measured raw data and using it for evaluation and analysis. The information processing unit 20 includes a CPU 22 and a memory 24.
The log data includes measurement values acquired at every sampling time. The log data is sent to the user interface 63 and presented to the user.

The converter unit 40 includes a converter control unit 44 that controls the converter unit 40 in accordance with an instruction from the information processing unit 20, and a bidirectional converter 42 that charges or discharges the battery under test 12 by adjusting a load or output to the battery under test 12. Is provided.
The charge / discharge test unit 18 may include a plurality of converter units 40 and a plurality of measurement units 30.

The measurement unit 30 includes a current measurement unit 32 that measures the current of the battery under test 12, a voltage measurement unit 34 that measures the voltage of the battery under test 12, and a temperature measurement unit 36 that measures the temperature of the battery under test 12.
The measurement unit 30 measures physical quantities such as voltage, current, and temperature for each measurement unit time. Each measurement unit time and each measurement step for each measurement unit time is also referred to as “round”. The measurement unit time may be the minimum response time, and may be determined based on the CPU frequency, for example.
The measurement value measured by the measurement unit 30 is recorded every sampling time. Sampling time is defined by the specification of the device and user designation. In general, the sampling interval is an integral multiple of the minimum response time. Hereinafter, an example in which the measurement unit time is equal to the sampling time will be described.

FIG. 3 shows the configuration of the user interface 63 and the control device 50 of FIG.
The user interface 63 includes an input device 62 that acquires instructions and settings from the user of the charge / discharge test system 10, an output device that includes a display for presenting real-time test status and statistical values to the user, and the like. 64.
The control device 50 includes an overall control unit 52 that comprehensively controls one or more charge / discharge test systems 10, a filter processing unit 68 that determines the validity of measured values and detects singular values, an implementation plan for charge / discharge tests, A storage unit 66 for storing user instructions, measurement data, and the like is provided. The filter processing unit 68 includes an acquisition unit 54 that acquires a measurement value such as a current from the measurement unit 30 via the information processing unit 20, a calculation unit 56 that performs an operation for filtering based on the measurement value, and the validity of the measurement data And a threshold adjustment unit 60 that adjusts a threshold used in the determination based on the variation tendency of the measurement data.

  Each block shown in FIG. 2 and FIG. 3 can be realized in hardware by an element such as a CPU or a mechanical device, and is realized in software by a program or the like. Draw functional blocks. Therefore, those skilled in the art will understand that these functional blocks can be realized in various forms by a combination of hardware and software.

The operation according to the above configuration is as follows.
4 and 5 are flowcharts showing filtering processing by the charge / discharge test system 10 according to the embodiment.
Here, “measured value” refers to a measured voltage value, a measured current value, a measured temperature value, or a measured value of other physical quantities.

FIG. 4 shows the filtering process in the next round when it is determined that the measurement value of the previous round is valid.
In this specification, n is an integer, and m, j, k, and s are natural numbers.
Let A (n) be the measurement value of the nth round at the current round, that is, at the measurement time t (n).

First, the acquisition unit 54 acquires the current round measurement value A (n) from the measurement unit 30 via the information processing unit 20 (S10), and passes it to the calculation unit 56. The calculation unit 56 uses the measurement value A (n) acquired by the acquisition unit 54 and the data up to the previous round stored in the storage unit 66 to change the current round of the measurement value A (n). The rate d1 and the ratio between the current rate of change and the previous rate of change are calculated (S12).
Specifically, using the following (Equation 1), the measurement value at the most recent measurement time point t (n−k) at which the measurement value is determined to be valid among the measurement values at the measurement time point before the nth measurement time point t (n). A time change rate d1 of the measurement value A (n) at the nth measurement time point t (n) with respect to the measurement value A (n−k) is calculated.
d1≡ | A (n) −A (nk) | / Δt (Formula 1)
However, Δt≡t (n) −t (n−k).
In the present embodiment, the time change rate and the reference change rate of the measurement value are defined as absolute values, but the present invention is not limited to this, and may be defined as a signed variable. In this case, the inequality at the time of the change rate ratio determination described later may be corrected as appropriate.

  In the example of FIG. 4, since it is assumed that the measurement current value A (n−1) of the previous round, that is, the (n−1) -th round is determined to be valid, d1≡ | A (n) −A (n− 1) | / Δt. Here, Δt≡t (n) −t (n−1). In this case, Δt coincides with the measurement unit time.

The calculation unit 56 calculates the ratio between the first reference change rate d0 calculated in advance using the values up to the previous round and the change rate d1 of the current round.
The first reference change rate d0 is a measured value A (nk) at a measured time t (nk) at which the measured value is determined to be valid among measured values before the nth measured time t (n). ), The rate of change with respect to the measured value A (n−j) at the previous measurement time t (n−j), and is expressed by the following (Equation 2).
d0≡ | A (n−k) −A (n−j) | / Δt (Formula 2)
However, here, Δt≡t (n−k) −t (n−j).

The example of FIG. 4 assumes that the measured current value A (n−1) of the previous round, that is, the (n−1) -th round is effective, and therefore d0≡ | A (n−1) −A (n− j) | / Δt.
However, Δt≡t (n−1) −t (n−j).

The computing unit 56 calculates a first change rate ratio r shown in the following (Equation 3) for comparison between the first reference change rate d0 and the current round change rate d1 (S12).
r = max {d1, d0} / min {d1, d0} (Formula 3)
The first change rate ratio r is the ratio of the current change rate to the previous change rate that is not a singular value. In the present specification, the first rate of change ratio r and the second rate of change ratio r ′, which will be described later, are always defined to be 1 or more. However, the present invention is not limited to this. For example, r = d1 / d0 is defined. May be. In this case, various inequalities used for determination to be described later may be appropriately adjusted.

The determination unit 58 compares the first change rate ratio r with the threshold Th (S14), and when the first change rate ratio r is equal to or less than the threshold Th (Y in S14), the change in the change rate is within a certain range. The measured value A (n) is determined as valid data and can be used for analysis and test control condition determination (S16). On the other hand, when the first change rate ratio r exceeds the threshold Th (N in S14), the change rate has changed beyond the allowable range, and the measured value A (n) is determined to be a singular value, that is, invalid (S18). .
Then, the calculation unit 56 sets the reference value for the next round according to each of the valid and invalid cases (S20), and the threshold adjustment unit 60 compares the first change rate ratio r with the threshold Th, and the result Accordingly, the threshold value Th is adjusted for the next round (S22). Specifically, for example, when the first rate-of-change ratio r is smaller than the threshold value Th, the threshold value Th is lowered by a predetermined width in a range that is equal to or greater than the lower limit value Tha of the threshold value, and the first rate-of-change ratio r is larger than the threshold value Th. In this case, the threshold value Th may be increased by a predetermined width within a range where the threshold value Th is equal to or less than the upper limit value Thb. Thus, the measurement value processing for the current round is completed.

FIG. 5 shows a filtering process in the n-th round when the determination unit 58 determines that the measurement value A (n−1) at the previous round, that is, the (n−1) -th measurement time t (n−1) is invalid.
Let A (n) be the measurement value of the nth round at the current round, that is, at the measurement time t (n).

  First, the acquisition unit 54 acquires the measurement value A (n) of the current round from the measurement unit 30 (S30), and passes it to the calculation unit 56. The calculation unit 56 uses the measurement value A (n) acquired by the acquisition unit 54 and the data up to the previous round stored in the storage unit 66, and the change rate of the measurement value A (n) in the current round. d1 is calculated (S32).

Specifically, the most recent measurement time point t (n−k) at which the measurement value is determined to be valid among the measurement values at the measurement time point before the nth measurement time point t (n) using the above-described equation (1). The time change rate d1 of the measurement value A (n) of the current round with respect to the measurement value A (n−k) of is calculated. However, Δt≡t (n) −t (n−k).
For example, the measurement current value A (n-1) in the previous round, that is, the (n-1) th round is determined to be invalid, and the measurement current value A (n-2) in the previous round, that is, the (n-2) th round is determined to be valid. If
d1≡ | A (n) −A (n−2) | / (t (n) −t (n−2)).

The calculation unit 56 calculates the ratio between the first reference change rate d0 calculated in advance using the values up to the previous round and the change rate d1 of the current round.
The first reference change rate d0 is a measured value A (nk) at a measured time t (nk) at which the measured value is determined to be valid among measured values before the nth measured time t (n). ) With respect to the previous measurement value A (n−j). That is, the first reference change rate d0 is expressed by the above-described (Expression 2). Here, Δt≡t (n−k) −t (n−j).
The computing unit 56 further calculates the first change rate ratio r expressed by the above-described (Equation 3) as the ratio between the first reference change rate d0 and the current round change rate d1 (S34).

The calculation unit 56 further calculates a second change rate ratio r ′, which is a ratio between the second reference change rate d ′ calculated in advance and the change rate d1 of the current round (S36).
The second reference change rate d ′ is a change rate of the measurement value A (n−1) at the n−1 measurement time point t (n−1), which is the previous round, and is represented by the following (formula 4).
d′ ≡ | A (n−1) −A (ns) | / Δt (Formula 4)
Here, A (ns) is a measured value at a measurement time t (ns) before t (n-1), and Δt≡t (n-1) -t (ns). .
A (ns) may be an effective latest measurement value before the measurement time t (n-1), and may be a measurement time t (n-) immediately before the measurement time t (n-1). It may be the measured value in 2).

The calculation unit 56 calculates a second change rate ratio r ′ shown in the following (Equation 5) in order to compare the change rate d1 of the current round with the second reference change rate d ′.
r′≡max {d1, d ′} / min {d1, d ′} (Formula 5)
The second change rate ratio r ′ is a ratio of the change rate of the current round to the change rate of the previous round determined to be a singular value.

The determination unit 58 compares the first change rate ratio r with the threshold Th (S38), and determines that the measured value A (n) is valid when the first change rate ratio r is equal to or less than the threshold Th (Y in S38). (S40).
In this case, since the fluctuation of the change rate is within a certain range excluding the singular value of the previous round, it can be estimated that the singular value of the previous round is a single-shot singular value that has been erroneously detected. Therefore, the measurement value A (n) of the nth round is used as analysis and condition determination as valid data.

  When the first change rate ratio r exceeds the threshold Th (N in S38), the determination unit 58 further compares the second change rate ratio r ′ with the threshold Th, and the measured value A (n) at the nth measurement time point t (n). The effectiveness of n) is determined (S42). When the second change rate ratio r ′ is equal to or less than the threshold Th (Y in S42), the determination unit 58 determines that the measurement value A (n) is valid (S44). In this case, when the previous effective measurement value is used as a reference, the current rate of change greatly fluctuates. However, since the rate of change in the previous round and the rate of change this time show the same tendency, it can be estimated that the current state is in the transition state. Therefore, the measurement value A (n) of the nth round is used as analysis and condition determination as valid data. In this case, the determination unit 58 or the overall control unit 52 may effectively correct the attribute information of the measurement value A (n−1) of the previous round once recorded as a singular value and include it in the sampling target.

On the other hand, when the second change rate ratio r ′ exceeds the threshold Th (N in S42), the determination unit 58 determines that the measurement value A (n) is invalid (S49).
After the determinations in steps 40, 44, and 49, the calculation unit 56 sets parameters such as a reference value for the next round according to each case (S46), and the threshold adjustment unit 60 sets the first change rate ratio r to The threshold Th is compared, and the threshold Th is adjusted for the next round according to the result (S48). Specifically, for example, when the first rate-of-change ratio r is smaller than the threshold value Th, the threshold value Th is lowered by a predetermined width in a range that is equal to or greater than the lower limit value Tha of the threshold value, and the first rate-of-change ratio r is larger than the threshold value Th. In this case, the threshold value Th may be increased by a predetermined width within a range where the threshold value Th is equal to or less than the upper limit value Thb. Thus, the current round process is completed.

FIG. 6 is a flowchart illustrating an implementation example of the filtering process in the charge / discharge test by the charge / discharge test system 10 according to the embodiment. In this implementation example, the minimum measurement unit time is set to 1 for simplicity. As a result, the burden on the CPU and memory can be reduced as compared with the case of strictly calculating.
Hereinafter, although description will be made using a current value as an example of a measured value, the present mounting example can be similarly applied to a measured value other than current, such as voltage and temperature.

At the start of the charge / discharge test, the calculation unit 56 acquires initial setting information related to the filtering process from the user via the input device 62 or reads it from the storage unit 66 to set initial values of parameters used for the filtering process (S50). ).
For example, a lower limit value Tha, an upper limit value Thb, and a threshold adjustment width dTh for automatically adjusting the threshold value Th are set, and an initial value of the threshold value Th is set to the lower limit value Tha. Appropriate initial values of these parameters vary depending on the conditions of the charge / discharge test, the specifications of the charge / discharge test system 10, and the use environment, and therefore may be determined by experiments such as empirical rules or simulations.
The computing unit 56 also sets the initial value of the second reference change rate d ′ used when the measurement value of the previous round is invalid to ∞.

In each round, the acquisition unit 54 first acquires the current measurement value of the current round from the measurement unit 30 and substitutes it for the variable I1 (S52). The calculation unit 56 calculates the current measurement value I1 of the current round and the latest measurement value I0 (the “reference current measurement value”) that has been determined to be valid among the previous current measurement values stored in the storage unit 66. The change rate d1 of the measured value in the current round is calculated using (Formula 6) below (S54).
d1≡ | I1-I0 | / Δt (Formula 6)
Here, assuming that the unit of time Δt is the measurement unit time, if it is determined that the measurement value of the previous round is valid, (Equation 6) becomes d1≡ | I1-I0 |. Further, when the measurement value before m times is the latest measurement value determined to be valid, (Expression 6) can be expressed as d1≡ | I1-I0 | / m.

  When the change rate d1 of the current round is 0 (Y in S56), since the measurement value of the current round is equal to the measurement value determined to be valid most recently, the determination unit 58 determines that the current measurement value is valid, Sampling is performed as log data (S58), the measurement value processing for the current round is terminated, and the measurement value processing for the next round is started (S52).

  When the change rate d1 of the current round is not 0 (N in S56), the validity determination process for the measurement value of the current round is executed. That is, the calculation unit 56 calculates the first change rate ratio r and the first change rate ratio r ′ (S60), the determination unit 58 determines the validity (S62), and the calculation unit 56 performs the next round determination process. The reference value for updating is updated (S62), and the threshold value Th for the next round determination process is adjusted (S64). While the charge / discharge test is continued (Y in S66), the filtering process from step S52 to step S64 is repeated for each round, and when the charge / discharge test is completed (N in S66), the filtering process is also ended.

FIG. 7 is a flowchart showing the change rate ratio calculation process (S60) of FIG.
The computing unit 56 calculates the ratio between the first reference change rate d0 and the current round change rate d1, that is, the first change rate ratio r≡max {d1, d0} / min {d1, d0} (S70).
The first reference rate of change d0 is a rate of change of the measured value determined to be valid among the previous measured values with respect to the measured value of the previous round, and is set by the calculation unit 56 in the previous round, and is stored in the storage unit 66. Stored in

Subsequently, the calculation unit 56 confirms the value of the variable d ′ indicating the validity of the previous round (S72). If the previous round is valid, that is, d ′ = ∞ (Y in S72), the second change rate r ′ = ∞ is set (S74), and the change rate ratio calculation process is completed. In this case, no substantial determination based on r ′ is made in this round determination process.
On the other hand, when the previous round is invalid, that is, d ′ ≠ ∞ (N in S72), the calculation unit 56 calculates r′≡max {d1, d ′} / min {d1, d ′} (S76), and changes The ratio ratio calculation process ends. In this case, even if it is determined that the first change rate ratio r is invalid in the round determination process, the determination using the second change rate ratio r ′ is further executed.

FIG. 8 is a flowchart showing the determination / update process (S62) of FIG.
The determination unit 58 compares the first change rate ratio r with the threshold Th (S80), and if it is determined that the first change rate ratio r exceeds the threshold Th (N in S80), the second change rate ratio r ′ Is compared with the threshold Th (S82). Thereby, the determination unit 58 determines the effectiveness of the measurement value A (n) at the nth measurement time point t (n).
When the current measurement value of the previous round is valid, the second change rate ratio r ′ is set to ∞ (S74), and the determination condition for the first change rate ratio r is not satisfied (N in S80). The second change rate ratio r ′ is always larger than the threshold value Th, and it is determined that the measured value of the current round is invalid (N in S82). That is, the determination based on the second change rate ratio r ′ (S82) has no substantial meaning when the previous round is valid, and has a substantial meaning when the previous round is a singular value.

9, 10 and 11 show examples of current measurement values for explaining the determination processing of FIG.
In FIG. 9, if the current round is n, the change rate d1 of the current round corresponding to the slope 82 and the first reference change rate d0 corresponding to the slope 80 in the previous (n−1) -th round are greatly different. It is determined that the first rate-of-change ratio r, which is the ratio of, exceeds the threshold Th (N in S80). In this case, since the measurement value in the previous (n−1) -th round is determined to be valid, the second change rate ratio r ′ is set to ∞ and is larger than the threshold Th (N in S82). The measurement value in n rounds is determined to be invalid.

  Similarly, if the current round is n in FIG. 10, the change rate d1 of the current round corresponding to the slope 92 in the figure and the first reference change rate d0 corresponding to the slope 90 in the previous (n−1) -th round are large. Since they are different, it is determined that the first change rate ratio r, which is the ratio, exceeds the threshold Th (N in S80). In this case, since the measurement value in the previous (n−1) -th round is determined to be valid, the second change rate ratio r ′ is set to ∞ and is larger than the threshold value Th (N in S82). The measurement value in round n is determined to be invalid.

  Returning to FIG. When the second change rate ratio r ′ exceeds the threshold Th (N in S82), the calculation unit 56 measures the reference current measurement value I0 and the first reference change rate d0 used for calculating the change rate in the next round in the current round. It is not updated with the value I1 or the change rate d1. In other words, since the measured value of the current round is a singular value, it is not used as a reference when determining the next round. In the next round, the value of the previous round determined to be valid is used as the reference current measured value. It is used as I0 or the first reference change rate d0.

Instead, the calculation unit 56 records the change rate d1 of the current round as the second reference change rate d ′ for the next round (S100), and comprehensively controls the Mark signal indicating that the measurement value of the current round is singular data. It transmits to the part 52 or the information processing part 20 (S102). The overall control unit 52 or the information processing unit 20 removes the current round measurement value from the log data or records that it is a specific measurement value as an attribute of the round measurement value.
The computing unit 56 increments the time m for calculating the current change rate in the next round (S104).

On the other hand, when the first change rate ratio r is equal to or less than the threshold Th, the determination unit 58 determines that the measurement value of the current round is valid (Y in S80).
For example, when the rate of change is about the same as before, as in the (n-1) th round in FIG. 9 and the (n-1) th round in FIG. 10, the first rate of change ratio r is equal to or less than the threshold Th, and the determination unit 58 It is determined that the measurement value of the (n-1) th round is valid (Y in S80).
In addition, as in the (n + 1) th round in FIG. 9, even when the measured value after the single singular value shows the same tendency as the change rate before the single singular value, the first change rate ratio r is equal to or less than the threshold Th. The determination unit 58 determines that the measured value is valid (Y in S80).

For example, when the current round in FIG. 9 is n + 1, the calculation unit 56 calculates the change rate d1 in the (n + 1) th round as the change rate from the measurement value in the (n−1) -th round, which is the latest effective measurement value. That is, a value corresponding to the slope 86 is calculated as the rate of change d1 in the (n + 1) -th round in FIG.
In addition, since the measurement value of the previous n-th round is determined to be a singular value, the calculation unit 56 does not update the first reference change rate d0 in the n-th round. Accordingly, the value of the first reference change rate d0 is set to the change rate of the (n-1) th round corresponding to the slope 80 in FIG.
Since the change rate d0 of the current round corresponding to the slope 86 and the first reference change rate d0 corresponding to the slope 80 are approximately the same, the first change rate ratio r, which is the ratio thereof, is equal to or less than the threshold value Th. 58 determines that the measurement value of the (n + 1) -th round is valid (Y in S80).

  When the determination unit 58 determines that the measurement value is valid using the first change rate ratio r (Y in S80), the calculation unit 56 sets a reference value for the filtering process of the next round. Specifically, the value of the change rate d1 of the current round determined to be valid is set as the first reference change rate d0 of the next round (S90), and the measurement value I1 of the current round is calculated when calculating the change rate of the next round. Is set as the reference measurement value I0 (S92), and the second reference change rate d 'for the next round is set to ∞ (S94). By setting the second reference change rate d ′ to ∞, in the next round after the effective round, it is determined whether or not the data is singular based on the first change rate ratio r.

Then, the determination unit 58 transmits a Pass signal indicating that the measurement value of the current round is valid data to the overall control unit 52 or the information processing unit 20. The overall control unit 52 or the information processing unit 20 samples the measurement value of the current round.
The computing unit 56 resets the time m for calculating the current change rate of the next round to 1 (S98).
Thus, by defining the rate of change and the reference rate of change based on the latest effective measurement value, it is possible to detect and filter a single singular value due to erroneous detection.
Further, by determining the effectiveness based on the ratio of the change rates, the change rates before and after the singular value can be appropriately compared.

  On the other hand, when the first change rate ratio r exceeds the threshold Th (N in S80) and the second change rate ratio r ′ is equal to or less than the threshold Th, the determination unit 58 determines that the measured value of the current round is in the transition state. (Y of S82).

For example, when the current round is n + 1 in FIG. 10, the calculation unit 56 calculates the change rate d1 in the (n + 1) th round as the change rate from the measurement value in the (n−1) -th round, which is the latest effective measurement value. That is, the change rate d1 of the (n + 1) th round is a value corresponding to the slope 94 in FIG.
In the n-th round determined as the singular value, the calculation unit 56 does not update the first reference change rate d0. Therefore, the value of the first reference change rate d0 is set to the change rate of the (n-1) th round corresponding to the slope 90 of FIG.
Since the change rate d1 of the current round corresponding to the slope 94 and the first reference change rate d0 corresponding to the slope 90 are greatly different, the first change rate ratio r, which is the ratio thereof, exceeds the threshold Th (N in S80). ).

The determination unit 58 further compares the second change rate ratio r ′ with the threshold Th (S82).
When the current round is n + 1 in FIG. 10, the second reference change rate d ′ is the change rate in the immediately preceding n-th round and corresponds to the slope 92 in FIG.
Since the change rate d1 of the current round corresponding to the slope 94 and the second reference change rate d ′ corresponding to the slope 92 are approximately the same, the second change rate ratio r ′ that is the ratio thereof is equal to or less than the threshold Th. The determination unit 58 determines that the measurement value of the (n + 1) -th round is in the transition state and is valid (Y in S82).

When it is determined that the transition state exists and is valid (Y in S82), the calculation unit 56 defines the first reference change rate d0 used for the filtering process of the next round as the change rate from the measurement value immediately before in the current round.
Specifically, the computing unit 56 first redefines the variable d1 as d1≡ | I1-I0 | (S84). Here, the variable d1≡ | I1-I0 | corresponds to the displacement 98 in the vertical axis direction of FIG.
Then, the calculation unit 56 defines the first reference change rate d0 of the next round as d0≡ | d1-d ′ | (S86). Since the variable d1 corresponds to the displacement 98 and d ′ corresponds to the displacement 96 in the vertical axis direction of FIG. 10, the first reference change rate d0≡ | d1-d ′ | of the next round is the vertical axis of FIG. This corresponds to a displacement 100 in the direction. In this implementation example, the minimum measurement unit time is 1, and the displacement 100 corresponds to the rate of change of the measured value of the (n + 1) -th round from the measured value of the n-th round.

Thereby, the first reference change rate d0 used for the filtering process of the next round can be defined as the change rate from the measurement value immediately before the current round.
That is, instead of setting the rate of change calculated as a value across multiple rounds as the next first reference rate of change d0, the rate of change calculated in one measurement unit time in the current round as shown in steps S84 and S86. It can be defined as the first reference change rate d0 of the next round.

In the charge / discharge test system according to the embodiment, in the next round in which a singular value is detected, the rate of change ratio is determined based on the last effective round instead of the previous round, and a single singular value is detected. To do.
On the other hand, when the transition state is determined, the slope, that is, the current change rate is gradually increased as shown in FIG. 11, for example, by using the change rate calculated in the immediately preceding measurement unit time as the first reference change rate d0 of the next round. Even in the case of a change, the change in the change rate can be finely determined in round units.
FIG. 11 is a schematic diagram showing a change in current value, which is an example of a measured value. The actual current value change is shown by an actual current value curve 102, and the current value measured every measurement unit time is shown by a black circle. Yes.
The charge / discharge test system according to the embodiment can detect a single singular value and can prevent a situation in which a change in the change rate is detected as a singular value each time in a transition state in which the change rate gradually changes. .

When the set d0 is not 0 (N in S88), the calculation unit 56 uses the first reference change rate defined in step S86 as the next first reference change rate d0, but when d0 becomes 0. (Y in S88), the calculation unit 56 sets the change rate d1 of the current measurement value determined to be valid as the next first reference change rate d0 (S90).
Subsequently, the calculation unit 56 sets another reference value for the filtering process of the next round. Specifically, the measurement value I1 of the current round determined to be valid is set as the reference measurement value I0 for calculating the rate of change of the next round (S92). Further, the second reference change rate d ′ used in the next round is set to ∞ (S94). As a result, the validity of the measurement value is determined based on the first rate of change ratio r in the round following the round determined to be effective.

Then, the determination unit 58 transmits a Pass signal indicating that the measurement value of the current round is valid data to the overall control unit 52 or the information processing unit 20. The overall control unit 52 or the information processing unit 20 samples the measurement value of the current round.
The computing unit 56 also resets the time m for calculating the current change rate of the next round to 1 (S98).

FIG. 12 is a flowchart showing the threshold adjustment process (S64) of FIG. The threshold adjustment unit 60 compares the first change rate ratio r and the threshold Th and adjusts the threshold Th for the next round.
First, the threshold adjustment unit 60 determines whether or not the first change rate ratio r is greater than or equal to the current threshold Th and less than or equal to the upper limit Thb (S110). When the first change rate ratio r is greater than or equal to the current threshold Th and less than or equal to the upper limit value Thb (Y in S110), the threshold adjustment unit 60 increases the threshold Th and sets the value to the first change rate ratio r (S112). ).
When the first change rate ratio r is not less than the current threshold Th and not more than the upper limit Thb (N in S110) and the first change rate ratio r is not less than the threshold Th (N in S114), the threshold adjustment unit 60 Terminates the threshold adjustment process without changing the threshold Th.

  On the other hand, when the first change rate ratio r is not greater than or equal to the current threshold Th and not greater than the upper limit Thb (N in S110), and the first change rate ratio r is smaller than the threshold Th (Y in S114), a threshold adjustment unit 60 lowers the threshold Th by the threshold adjustment width dTh (S116). As a result, when the threshold Th becomes smaller than the first change rate ratio r (Y in S118), the threshold adjustment unit 60 sets the threshold Th to the value of the first change rate ratio r (S120). When the threshold value Th is smaller than the lower limit value Tha (Y in S122), the threshold value adjusting unit 60 sets the threshold value Th to the value of the lower limit value Tha (S124).

  In this way, the charge / discharge test system 10 is provided with a learning function, and the threshold value is dynamically adjusted automatically, so that when the change rate change is large, the change rate change is larger than when it is not so extreme. In the steady state where the rate of change does not change so much, the singular value can be detected with higher sensitivity.

The usage scenes of the charge / discharge test system 10 configured as described above are as follows.
The user uses the user interface 63 to create a charge / discharge test plan by designating control steps and parameters such as start / stop / pause of the charge / discharge operation, for example. When the user instructs the start of the charge / discharge test, the charge / discharge test system 10 performs the charge / discharge test according to the charge / discharge test plan, and measures the physical quantity of the battery under test 12 every measurement unit time. The information processing unit 20 creates log data from the measured raw data. The control device 50 performs a filtering process on each measurement value, and removes a singular value that is considered to be a false detection value from the log data. As a result, it is possible to prevent the test from being automatically stopped or inadvertently transferred to the step due to the erroneously detected measurement value.

  In the above, this invention was demonstrated based on embodiment. The present invention is not limited to the above-described embodiment, and an appropriate combination of the elements of the embodiment is also effective as an embodiment of the present invention. Various modifications such as design changes can be added to the embodiments based on the knowledge of those skilled in the art, and embodiments to which such modifications are added can be included in the scope of the present invention.

  For example, in the embodiment, the example of the charge / discharge test system 10 has been described. However, the scope of application of the present invention is not limited to this, and in various other measurement devices that acquire data such as observation values and measurement values at predetermined intervals. Can be applied in the same way, and noises mixed in data can be filtered.

  In the embodiment, the control device 50 mainly performs the filtering process. However, the filtering process may be executed by the information processing unit 20 of the charge / discharge test unit 18, or the control device 50 and the information processing unit 20. May be executed in cooperation. In this case, the information processing unit 20 may include some or all of the components of the control device 50 illustrated in FIG.

  As described above, the present invention can be used for measurement value processing.

  DESCRIPTION OF SYMBOLS 10 Charging / discharging test system, 30 measurement part, 50 control apparatus, 54 acquisition part, 56 calculating part, 58 determination part, 60 threshold value adjustment part.

Claims (7)

A charge / discharge test system for a secondary battery,
A measurement unit for measuring a physical quantity related to the battery under test;
A control device for determining the validity of a measurement value by the measurement unit;
With
The controller is
An acquisition unit for acquiring a measurement value from the measurement unit;
Using the measured value,
At the nth measurement time t (n) with respect to the measurement value A (nk) (where k is a natural number) at the measurement time t (nk) before the nth measurement time t (n) (where n is an integer). Change rate d1≡ | A (n) −A (nk) | / Δt (where Δt≡t (n) −t (n−k)) of the measured value A (n),
A first reference change rate d0 that is a change rate of a measurement value at a measurement time before the n-th measurement time t (n);
A first rate-of-change ratio r, which is a ratio of the rate of change d1 at the n-th measurement time t (n) and the first reference rate of change d0;
An arithmetic unit for calculating
A determination unit that compares the first rate of change ratio r with a threshold value and determines the effectiveness of the measurement value A (n) at the n-th measurement time point t (n);
A charge / discharge test system comprising:   The rate of change d1 at the nth measurement time point t (n) is the measurement value A at the measurement time point t (n−k) determined to be valid among the measurement values at the measurement time point before the nth measurement time point t (n). 2. The charge / discharge test system according to claim 1, wherein the charge / discharge test system is a rate of change of a measured value A (n) at an nth measurement time point t (n) with respect to (n−k). The first reference change rate d0 is the measurement time t (n−j) before the measurement time t (n−k) determined to be valid among the measurement values at the measurement time prior to the nth measurement time t (n). ) In the measurement value A (n−j) (j is an integer of 2 or more), the rate of change d0≡ | A ( n−k) −A (n−j) | / Δt (where Δt≡t (n−k) −t (n−j))
The charge / discharge test system according to claim 1, wherein: When the determination unit determines that the measurement value A (n−1) at the n−1 measurement time point t (n−1) is invalid,
The calculation unit further includes:
The (n-1) th measurement time point t (n) for the measurement value A (ns) (s is an integer of 2 or more) at the measurement time point t (ns) prior to the (n-1) th measurement time point t (n-1). -1) is the second reference change rate d′ ≡ | A (n−1) −A (ns) | / Δt (where Δt≡t (n− 1) -t (ns)),
A second rate of change ratio r ′, which is a ratio of the rate of change d1 at the n-th measurement time t (n) and the second reference rate of change d ′;
To calculate
The determination unit further compares the second change rate ratio r ′ with a threshold value to determine the effectiveness of the measurement value A (n) at the nth measurement time point t (n). The charge / discharge test system described in 1.   5. The charge / discharge test system according to claim 1, further comprising a threshold adjustment unit configured to change the threshold based on the first change rate ratio r. An acquisition unit for acquiring measurement values;
Using the measured value,
At the nth measurement time t (n) with respect to the measurement value A (nk) (where k is a natural number) at the measurement time t (nk) before the nth measurement time t (n) (where n is an integer). Change rate d1≡ | A (n) −A (nk) | / Δt (where Δt≡t (n) −t (n−k)) of the measured value A (n),
A first reference change rate d0 that is a change rate of a measurement value at a measurement time before the n-th measurement time t (n);
A first rate-of-change ratio r, which is a ratio of the rate of change d1 at the n-th measurement time t (n) and the first reference rate of change d0;
An arithmetic unit for calculating
A determination unit that compares the first rate of change ratio r with a threshold value and determines the effectiveness of the measurement value A (n) at the n-th measurement time point t (n);
A control device comprising: A charge / discharge test control program for a charge / discharge test control device,
An acquisition function to acquire measurement values;
Using the measured value,
At the nth measurement time t (n) with respect to the measurement value A (nk) (where k is a natural number) at the measurement time t (nk) before the nth measurement time t (n) (where n is an integer). Change rate d1≡ | A (n) −A (nk) | / Δt (where Δt≡t (n) −t (n−k)) of the measured value A (n),
A first reference change rate d0 that is a change rate of a measurement value at a measurement time before the n-th measurement time t (n);
A first rate-of-change ratio r, which is a ratio of the rate of change d1 at the n-th measurement time t (n) and the first reference rate of change d0;
A calculation function for calculating
A determination function for comparing the first change rate ratio r with a threshold value and determining the effectiveness of the measurement value A (n) at the nth measurement time point t (n);
A program for running

Images