JP2018169161A - Deterioration diagnosis apparatus, deterioration diagnosis method, and deterioration diagnosis system for battery - Google Patents

Abstract

To provide an apparatus, a method and a system for diagnosing the degree of deterioration from an operation history of an actual product even without obtaining a deterioration constant with respect to battery voltage and temperature beforehand in an experiment or the like, while in the conventional battery deterioration diagnosis it is necessary to prepare in advance a deterioration characteristic map showing the distribution of deterioration constants with respect to battery voltage and temperature, which is time-consuming and costly.SOLUTION: A battery deterioration diagnosis apparatus comprises: a communication unit that collects battery status; an operation history storage unit for storing the battery status; a characteristic measurement unit that measures changing characteristics from the battery status due to deterioration of the battery; a characteristic fluctuation model creation unit for creating a model of the relationship between the characteristics and the battery status; and a battery deterioration diagnostic unit for diagnosing deterioration of the battery on the basis of an estimated value relating to the characteristics calculated based on the model.SELECTED DRAWING: Figure 1

Description

  The present invention relates to battery deterioration diagnosis.

  As background art, there exists Unexamined-Japanese-Patent No. 2014-190763 (patent document 1). This Patent Document 1 calculates current battery characteristics including at least an internal resistance value with respect to a secondary battery, usage condition data including temperature time distribution and current value time distribution in an estimation period, battery voltage, Storing a deterioration characteristic map indicating a distribution of deterioration constants with respect to temperature, and integrating the battery voltage-temperature plane residence time in the deterioration characteristic map based on the current battery characteristic data and use condition data. It is described that the amount of deterioration is estimated.

JP 2014-190763 A

  Patent Document 1 describes a technique for estimating the deterioration amount of a secondary battery. However, it is necessary to prepare in advance a deterioration characteristic map showing the distribution of deterioration constants with respect to battery voltage and temperature. For this purpose, it is necessary to create a battery for experiments and perform a float test that maintains a constant potential, or a storage test that is stored after being adjusted to a constant potential, which takes time and cost. It was. In addition, since the test can be performed only under limited conditions, there is a case where sufficient estimation accuracy cannot be obtained unlike the condition in which the product is actually used.

  Therefore, the present invention provides a technique for diagnosing the degree of deterioration from the actual operation history of the product without obtaining the deterioration constant with respect to the battery voltage and temperature in advance through experiments or the like. Furthermore, a technique for estimating the degree of future deterioration and deriving a use condition for approaching an expected life when the deterioration progresses more than expected.

  In order to solve the above problems, for example, the configuration described in the claims is adopted. The present application includes a plurality of means for solving the above problems. To give an example, a battery deterioration diagnosis apparatus, which is a communication unit that collects battery states, and an operation history storage that stores battery states. A characteristic measuring unit for measuring characteristics that change due to deterioration of the battery from the state of the battery, a characteristic variation model creating unit for creating a model of the relationship between the characteristic and the state of the battery, and a characteristic calculated based on the model And a deteriorated battery diagnosis unit that diagnoses battery deterioration based on the estimated value.

  Deterioration can be accurately diagnosed from the actual operation history and characteristics of the product without obtaining a deterioration constant with respect to the battery voltage and temperature in advance through experiments or the like. Furthermore, the use condition for approaching the expected life can be obtained from a plurality of use conditions.

1 is a block diagram of a battery deterioration diagnosis system in Embodiment 1. FIG. It is a structure figure of the system master table in Example 1. It is a structure figure of the operation history table in Example 1. 6 is a structural diagram of a characteristic variation model table in Embodiment 1. FIG. It is a structure figure of the diagnostic result table in Example 1. 4 is a flowchart of battery deterioration diagnosis in the first embodiment. 6 is a flowchart of battery deterioration diagnosis in Example 2. 10 is a flowchart of battery deterioration diagnosis in Example 3. It is an example of the model abnormality display screen in Example 4. FIG. It is an example of the abnormally deteriorated battery display screen in Example 4. It is an example of the early deterioration battery display screen in Example 4. FIG. It is another example of the early deterioration battery display screen in Example 4. FIG. 10 is a block diagram of a battery deterioration diagnosis system according to a fifth embodiment. FIG. 10 is a structural diagram of an operation history table in the fifth embodiment. FIG. 10 is a structural diagram of a characteristic variation model table in Embodiment 5. 10 is a flowchart of battery deterioration diagnosis in Example 5. 10 is a flowchart of battery deterioration diagnosis in Example 6. 10 is an example of a characteristic estimation screen in Example 7. FIG. It is another example of the characteristic estimation screen in Example 7. It is another example of the characteristic estimation screen in Example 7.

  Hereinafter, a battery deterioration diagnosis apparatus, method, and system will be described with reference to the drawings.

  FIG. 1 shows a block diagram of a battery deterioration diagnosis system in the present embodiment. In FIG. 1, the power storage system 101 includes one or more batteries 102 and a controller 103 connected to the batteries 102. The controller 103 controls the power storage system 101 and the battery 102 and steadily collects measured values of the battery state such as voltage, current, and temperature as operation history. The collected operation history is regularly transmitted through the communication network 105 by the communication unit (storage system side) 104 connected to the controller. Note that a plurality of power storage systems 101 may exist. Further, in the power storage system 101, the power storage system may be structured so as to be easily managed, for example, by bundling a plurality of batteries 102 into a battery module. The power storage system 101 may be a power storage system that incorporates the communication unit (power storage system side) 104 and is integrated with the communication unit.

  The transmitted operation history is transmitted through the communication network 105 to the deterioration diagnosis device functioning as a monitoring device. The deterioration diagnosis apparatus includes a communication unit (storage unit side) 106, a storage unit 107, a processing unit 112, and a display unit 119 shown in FIG. As shown in FIG. 1, the transmitted operation history is regularly stored in the storage unit 107, and the storage unit 107 is a system master table 108 that stores basic data to be diagnosed, and an operation that stores an operation history. An operation history table 109 that is a history storage unit, a characteristic variation model table 110 that stores a characteristic variation model that indicates the relationship between the battery state and characteristics, and a difference between a measured value of the characteristic and an estimated value obtained from the characteristic variation model is stored. A diagnostic result table 111 to be stored. The storage unit 107 is connected to the processing unit 112 to exchange data.

  The processing unit 112 includes a characteristic measurement unit 113 that measures characteristics related to deterioration that changes due to deterioration based on the operation history, and a characteristic variation model generation unit that calculates a relationship between the state of the battery and the characteristics based on the operation history and creates a model. 114, an abnormally deteriorated battery diagnosis unit 115 that diagnoses a battery that deteriorates abnormally from the operation history and the model, and an early deterioration battery diagnosis unit 116 that diagnoses a battery that deteriorates early from the operation history and the model. The abnormally deteriorated battery diagnosis unit 115 and the early deteriorated battery diagnosis unit 116 are collectively used as a deteriorated battery diagnosis unit, and a deteriorated battery diagnosis unit that diagnoses battery deterioration based on an estimated value related to characteristics calculated based on a model. Good. In addition, the processing unit 112 includes an alert issuing unit 117 that issues an abnormal sign as an alert by a desired method such as e-mail, telephone, or fax, and a result display unit 118 that generates a screen for displaying the contents of the alert. Then, the display unit 119 connected to the processing unit 112 displays the screen generated by the result display unit 118.

  The deterioration diagnosis device functions as remote monitoring, and may be, for example, a stand-alone type including a personal computer and a monitor, or a client server type including a server, a personal computer, and a monitor. Moreover, it is good also as a cloud type which preserve | saves the transmitted operation history on a cloud and performs deterioration diagnosis using those data.

  FIG. 2 shows the structure of the system master table 108 in this embodiment. In FIG. 2, the power storage system ID column stores an ID uniquely assigned to the power storage system to be diagnosed, the power storage system type column stores the type of the power storage system, and the battery type column stores the type of battery constituting the power storage system. In addition, the table may have information such as information on customers who own the power storage system, installation location, usage, and the like.

  FIG. 3 shows the structure of the operation history table 109 in this embodiment. The operation history table 109 is a table that records the state of the battery 102 constituting the power storage system, which is the state of the power storage system 101, in time series. The time column is the time when the state is measured, the power storage system ID column is an ID for uniquely identifying the power storage system to be diagnosed, the battery ID column is the ID for uniquely identifying the battery to be diagnosed, and current / voltage -SOC (State of Charge), temperature, humidity column stores the value of the battery state measured with a sensor. The internal resistance / full charge capacity column stores respective calculated values obtained from the actually measured values of the battery state. The internal resistance represents a phenomenon in which a voltage drop occurs when a load is connected to an open circuit voltage as a resistance. The full charge capacity is an amount of electricity that can be taken out until the battery is fully charged and fully discharged.

  FIG. 4 shows the structure of the characteristic variation model table 110 in this embodiment. In the characteristic variation model table 110, based on the data of the operation history table 109, the relationship between the state of the battery and the characteristic is digitized as a model, and the result is stored. Note that the shape and configuration of the power storage system varies depending on the type, and the material and shape of the battery differ depending on the type. Therefore, the relationship between the state and characteristics may vary depending on the type of power storage system and battery. Therefore, in this embodiment, the characteristic variation model is calculated for each combination of the power storage system type and the battery type.

  In FIG. 4, the power storage system type column stores the type of the power storage system 101, and the battery type column stores the type of the battery 102. The characteristic column stores the characteristic of the object of the model. In this embodiment, the internal resistance and the full charge capacity are treated as characteristics.

  The internal resistance and the full charge capacity vary depending on the battery state at the time of measurement and the influence of deterioration until measurement. The fluctuations due to each influence are hereinafter referred to as measurement fluctuations and deterioration fluctuations. In the present embodiment, it is assumed that the initial (product shipping) characteristic, the measurement-time fluctuation, and the deterioration fluctuation are linear with respect to the characteristic, and the characteristic fluctuation model is expressed by the following formulas 1, 2, and 3. To do.

Characteristics (t) = Initial characteristics + Measurement fluctuation (t) + Degradation fluctuation (t) (Equation 1)
Fluctuation during measurement (t) = a11 x f11 (x11 (t)) + a12 x f12 (x12 (t)) (Equation 2)
Deterioration fluctuation (t) = a21 x f21 (x21 (t)) + a22 x f22 (x22 (t)) (Equation 3)
Here, t is a certain time, and x11 (t), x12 (t), x21 (t), and x22 (t) each represent a battery state at time t. f11, f12, f21, and f22 represent functions suitable for expressing the relationship between the battery state and characteristics. For example, a power function (√), a logarithmic function (log), an exponential function (exp), and the like, which have good fitness are selected. a11, a12, a21, and a22 are coefficients indicating the degree of influence on the characteristics.

  In FIG. 4, in the characteristic variation model table 110, the battery state to be assigned to the term of the model formula is selected from the column name of the operation history table 109 and stored in the variable column. In the variable column, a calculated value using a function can be input based on the value of the operation history table. For example, in the data 401 on the fourth row in FIG. 4, “sum (abs (current))” is stored in the variable column. sum is a function that outputs a cumulative sum, and abs is a function that outputs an absolute value. In this case, it represents the amount of current that has been accumulated and charged up to time t in the operation history table 109. In addition, the variable string “time-init (time)” of the data 402 in the fifth row represents an elapsed time since the start of recording.

  In addition, when the initial characteristic, the fluctuation at the time of measurement, and the deterioration fluctuation are not linear with respect to the characteristic and there is an interaction, a term of interaction may be added. The term of interaction is expressed by putting the product of variables in a variable string, such as f3 (x1 × x2).

  The constraint condition column stores a constraint condition based on the physical relationship between the characteristics and the battery state in order to guarantee the validity of the model. For example, in the data 403 in the second row in FIG. 4, the constraint condition is described as “<0”, and in the model expression of the characteristic “internal resistance”, the coefficient “only” is a negative value in the variable “temperature” term. Represents the constraint condition of taking This is because when the temperature rises, the chemical reaction in the battery is accelerated, so the internal resistance becomes smaller due to the influence. Normally, the coefficient of temperature at the time of measurement can only take a negative value. It shows that.

  In the measurement variation / degradation variation column, “measurement” is input if the variation is measurement, and “deterioration” is input if the variation is degradation. Coefficients calculated based on the data of the operation history table 109 are stored in the coefficient column.

  FIG. 5 shows the structure of the diagnosis result table 111 in this embodiment. The diagnosis result table 111 stores the result of diagnosing the battery using the characteristic variation model. In FIG. 5, the time column is the battery characteristic measurement time, the storage system ID, the battery ID column is the diagnosis target ID, and the internal resistance (residual) column is the difference between the value estimated from the characteristic variation model and the measured value (residual). Store the difference. The estimation in the characteristic variation model is performed by substituting the data in the corresponding columns of the current, voltage, SOC (State of Charge), temperature, and humidity columns in the operation history table 109 into the battery states of Equations 1, 2, and 3. This is performed by substituting the function of the function sequence of the characteristic variation model table 110 for the function and substituting the coefficient of the coefficient sequence of the characteristic variation model table 110 for the coefficient. Since the estimated value is a value obtained by estimating an average battery characteristic, the difference between the estimated value and the measured value is an index indicating how far the target battery is separated from the normal battery.

  Further, the internal resistance (deterioration variation) column stores the value obtained by estimating the variation at the time of measurement using the characteristic variation model according to the following formula 4, and removing the variation at the time of measurement from the measured value.

Internal resistance (deterioration fluctuation) = Internal resistance measurement
-A1 x (f1 (x1 (t)-(average value of x1 over all periods))
−a2 × (f2 (x2 (t) − (average value of x2 over all periods)) (Expression 4)
Since the internal resistance (deterioration variation) column is a value obtained by extracting the initial characteristics and the degradation variation, the tendency of degradation is confirmed by confirming the trend in time series. Similarly, the full charge capacity (residual) column and the full charge capacity (deterioration variation) column calculate and store the full charge capacity residual and degradation variation.

  FIG. 6 shows a flowchart of battery deterioration diagnosis in this embodiment. In FIG. 6, first, the characteristic measurement unit 113 obtains the measured value of the characteristic affected by the deterioration based on the data read from the operation history table 109, and writes it in the operation history table (601). Among the characteristics, the internal resistance can be measured by dividing the voltage fluctuation amount by the current fluctuation amount according to the following equation 5 based on Ohm's law.

Internal resistance = (Voltage (t)-Voltage (t-Δt) / (Current (t)-Current (t-Δt))) (Equation 5)
Here, t is a certain time, and Δt is a battery state measurement interval. Voltage (t) indicates the battery voltage at time t, and current (t) indicates the current flowing through the battery at time t.

  The full charge capacity can be measured by extracting two points when the battery is fully charged and when it is completely discharged, and integrating the current value between the two points.

  Next, the characteristic variation model creation unit 114 creates a characteristic variation model based on the data read from the operation history table 109, and writes it in the characteristic variation model table 110 (602). In this embodiment, the model coefficient is obtained by multiple regression analysis. In the multiple regression analysis, the coefficients of the model formulas of Formula 1, Formula 2, Formula 3 are calculated for each term stored in the characteristic column and variable column of the characteristic variation model table 110 for each combination of power storage system type and battery type. The data of the operation history table 109 is calculated so that the residual sum of squares is minimized. The initial (product shipment) characteristic uses a value calculated as a constant term of an initial value of multiple regression analysis. In this embodiment, the variable name “initial value” is stored in the characteristic variation model table 110.

  Next, the validity of the model is confirmed (603). Here, it is confirmed whether the coefficient stored in the characteristic variation model table 110 satisfies the constraint condition stored in the constraint condition column. If the constraint condition is not satisfied, the process proceeds to model abnormality report / display (604). Here, an abnormality is reported through the alert reporting unit 117, and a model abnormality display screen is generated by the result display unit 118 and displayed on the display screen 119. Model abnormalities may be due to abnormal data in the operation history table 109, incorrect variables or constraint conditions set in the characteristic variation model table 110, etc. The data of each table is corrected (605). The model validity check (603) is necessary for the purpose of preventing erroneous determination, but is not essential and may be omitted.

  When the constraint condition is satisfied in the model validity check (603), the measured value and the model estimated value are compared for the characteristics of all the batteries (606). Here, the abnormally deteriorated battery diagnosis unit 115 calculates the residual of each characteristic based on the data of the operation history table 109 and the characteristic variation model table 110. In this embodiment, the data of the internal resistance (residual) column and the full charge capacity (residual) column of the diagnosis result table 111 are calculated and stored.

  Next, it is investigated whether there is a battery deviating from the model (607). Here, the latest residual for each battery in the diagnosis result table 111 is confirmed for each combination of the power storage system type and the battery type, or for each power storage system ID, and a check is made to see if there is a residual deviating from variation. For example, if there is a battery that deviates from the upper limit value or the lower limit value with + 3σ (a value that is three times the standard deviation) as the upper limit value and −3σ as the lower limit value, it is determined that there is a battery that deviates from the model. If there is a battery that deviates from the model, an alert is issued and a screen is displayed with the deviated battery as an abnormally deteriorated battery (608). An abnormality is reported through the alert reporting unit 117, and an abnormally deteriorated battery display screen is generated by the result display unit 118 and displayed on the display screen 119.

  The abnormally deteriorated battery may be replaced with a normal battery as a countermeasure. In this case, since the data such as the battery ID in the operation history table 109 is changed, the data in the operation history table 109 is corrected as necessary (609).

  If there is no battery deviating from the model, it is determined that there is no abnormally deteriorated battery, and the diagnosis ends. Since the diagnosis needs to be performed regularly, the flowchart of this diagnosis is repeatedly executed from the beginning (601) at an appropriate interval (for example, every day).

  As described above, the present embodiment is a battery deterioration diagnosis device, and includes a communication unit that collects battery states, an operation history storage unit that stores battery states, and changes depending on battery deterioration from battery states. A characteristic measurement unit that measures characteristics to be measured, a characteristic variation model creation unit that creates a model of the relationship between the characteristics and the state of the battery, and a deterioration that diagnoses the deterioration of the battery based on an estimated value for the characteristic calculated based on the model It consists of a battery diagnostic unit.

  The deteriorated battery diagnostic unit compares the measured value of the characteristic output from the characteristic measuring unit with the estimated value of the characteristic based on the model, and determines whether the difference between the measured value and the estimated value exceeds a predetermined reference value. An abnormally deteriorated battery diagnosis unit for diagnosing battery abnormality is provided.

  Also, a method for diagnosing battery degradation, collecting battery status, measuring characteristics that change from battery status due to battery degradation, creating a model of the relationship between characteristics and battery status, and based on the model The battery is configured to be diagnosed for deterioration based on the estimated value related to the calculated characteristic.

  Also, the estimated value for the characteristic is an estimated value of the characteristic by the model, and the measured value of the characteristic is compared with the estimated value of the characteristic, and the battery is determined by whether or not the difference between the measured value and the estimated value exceeds a predetermined reference value. It is configured so as to diagnose abnormalities.

  The battery deterioration diagnosis system includes a power storage system, a communication network, and a deterioration diagnosis device. The power storage system is connected to one or a plurality of batteries and the measured value of the battery state as an operation history. A controller that collects, a communication unit that transmits the collected operation history to the communication network, the deterioration diagnosis device includes a communication unit that collects the operation history via the communication network, and an operation history storage unit that stores the operation history A characteristic measurement unit that measures characteristics that change due to battery deterioration from the operation history, a characteristic variation model creation unit that creates a model of the relationship between the characteristics and the operation history, and an estimated value related to the characteristics calculated based on the model And a deteriorated battery diagnosis unit that diagnoses battery deterioration.

  This makes it possible to accurately diagnose deterioration from the actual operation history and characteristics of an actual product without obtaining a deterioration constant with respect to the battery voltage and temperature in advance through experiments, etc., and a battery that exhibits an abnormal deterioration tendency unlike a normal battery. Can be detected at a stage before it leads to an abnormality of the power storage system, and countermeasures can be taken in advance to prevent an abnormality of the power storage system.

  FIG. 7 shows a flowchart of battery deterioration diagnosis in this embodiment. In FIG. 7, processing steps 601 to 605 are the same processing steps as in FIG. 6, and the same reference numerals as those in FIG.

  In FIG. 7, when the constraint condition is satisfied in the model validity confirmation (603), in the processing step 610, the fluctuation at the time of measurement is removed for each characteristic, the initial characteristic + deterioration fluctuation is extracted, and the diagnosis result table 111. Is stored in a column for the deterioration variation of the characteristic of. Next, for each combination of the power storage system type and the battery type, or for each power storage system ID, the data for deterioration change based on the latest data for each battery in the diagnosis result table 111 is confirmed to determine whether there is an early deterioration battery ( 611). For example, if there is a battery that deviates from the upper limit value / lower limit value with the average value of deterioration fluctuation + 3σ (three times the standard deviation) as the upper limit value and the average value −3σ as the lower limit value, It is determined that there is a deteriorated battery. In addition, a warning line is determined in advance for the deterioration variation of characteristics, and when the warning line is exceeded, it is determined that there is an early deteriorated battery. For example, if the internal resistance is 5 times the initial characteristic, the warning line is used. If the capacity is full, half of the initial characteristics is used as a warning line. Here, the early deterioration battery diagnosis unit 116 calculates the deterioration fluctuation amount of each characteristic based on the data of the operation history table 109 and the characteristic fluctuation model table 110.

  If there is an early deteriorated battery, an alert is issued and a screen is displayed (612). An abnormality is reported through the alert reporting unit 117, and an early deterioration battery display screen is generated by the result display unit 118 and displayed on the display screen 119. The early deteriorated battery may be replaced with a normal battery as a countermeasure. In this case, since the data such as the battery ID in the operation history table 109 is changed, the data in the operation history table 109 is corrected as necessary (609).

  If there is no prematurely deteriorated battery, it is determined that there is no sign of abnormality in the power storage system, and the diagnosis ends. Since the diagnosis needs to be performed regularly, the flowchart of this diagnosis is repeatedly executed from the beginning (601) at an appropriate interval (for example, every day).

  As described above, the present embodiment is a battery deterioration diagnosis device, and includes a communication unit that collects battery states, an operation history storage unit that stores battery states, and changes depending on battery deterioration from battery states. A characteristic measurement unit that measures the characteristics to be measured, a characteristic variation model creation unit that creates a model of the relationship between the characteristics and the state of the battery, and a measured value that is output from the characteristic measurement unit varies depending on the battery state at the time of measurement. It is divided into deterioration fluctuations due to the influence of deterioration until measurement, and whether or not the deterioration fluctuation amount excluding the estimated value of fluctuation at the time of measurement from the measurement value of the characteristic output from the characteristic measurement unit exceeds a predetermined reference value It is composed of an early deterioration battery diagnosis unit that diagnoses a sign of battery abnormality.

  In addition, it is a battery deterioration diagnosis method that collects the battery state and measures the characteristics that change from the battery state due to the battery deterioration. The measured value of the characteristic depends on the battery state at the time of measurement and the measurement A model of the relationship between the characteristics and the state of the battery was created due to the effects of the deterioration of the battery, and whether the deterioration fluctuation amount excluding the estimated value of the fluctuation at the time of measurement from the measured value of the characteristics exceeded the predetermined reference value It is configured to diagnose a sign of battery abnormality depending on whether or not.

  As a result, according to the present embodiment, it is possible to diagnose whether there is an early deteriorated battery without being affected by variations during measurement. In addition, it is possible to diagnose deterioration with accuracy from the actual operation history and characteristics of an actual product without obtaining a deterioration constant with respect to battery voltage and temperature in advance by experiments, etc., and the deterioration progresses quickly and cannot be used at an earlier stage than expected. The battery can be detected at an early stage before it leads to an abnormality in the power storage system, and measures can be taken in advance to prevent an abnormality in the power storage system.

  FIG. 8 shows a flowchart of battery deterioration diagnosis in this embodiment.

  The present embodiment is a processing flow in which FIGS. 6 and 7 are combined, and the same processing steps as those in FIGS. 6 and 7 are denoted by the same reference numerals. Therefore, although a detailed description is omitted, an outline thereof will be described below.

  In FIG. 8, a characteristic variation model is created and the validity of the model is confirmed (601-605), an abnormal battery is detected by comparing the measured value and the estimated value by the model (606-609), and the variation at the time of measurement is removed. Then, the early deterioration battery is detected based on the initial characteristics + deterioration fluctuation (610-612).

  This makes it possible to diagnose the deterioration of the battery status during measurement in accordance with the usage conditions of the product, determine whether there is an abnormally deteriorated battery, and diagnose whether there is an early deteriorated battery without being affected by fluctuations during measurement. It becomes possible. Therefore, there is an effect that the reliability of the deterioration diagnosis of the used battery is further improved.

  In this embodiment, a display example on the display unit in the battery deterioration diagnosis system will be described.

  FIG. 9 shows a model abnormality display screen. In FIG. 9, the model type display unit 701 displays the power storage system type, battery type, and characteristics of the model to be displayed. Then, the coefficient stored in the characteristic variation model table 110 is displayed on the graph display unit 702. The data of the variable column and measurement variation / degradation variation column are displayed as labels, and the coefficient size is displayed as a bar graph. At this time, instead of the coefficient, the standard partial regression coefficient when the multiple regression analysis is performed may be displayed. Then, the coefficients that deviate from the constraint condition are highlighted, and the labels and the bar graph plot are surrounded by a bold line. In the example of FIG. 9, since the voltage does not satisfy the constraint condition, it is surrounded by a thick line.

  FIG. 10 shows an example of the display screen of the abnormally deteriorated battery. In FIG. 10, the type display unit 801 displays the storage system type, battery type, storage system ID, and characteristics to be displayed. This display screen can be displayed for each power storage system type, each battery type, and each power storage system ID. In this example, an abnormally deteriorated battery related to internal resistance is displayed for each power storage system ID.

  A characteristic and residual graph display unit 802 displays characteristics and residuals in time series. As for the characteristics, the data in the characteristic column of the operation history table 109 is displayed for the battery of the type display unit 801. In the residual graph, the residual data of the corresponding characteristic in the diagnosis result table 111 for the battery of the type display unit 801 is displayed.

  The latest residual distribution display section 803 displays the data at the time point indicated as “latest” in the residual graph of the characteristic and residual graph display section 802 as a histogram. Here, the battery ID is displayed for abnormally deteriorated batteries exceeding the upper limit value / lower limit value. In this example, it can be seen that the battery with battery ID 3 is abnormal.

  FIG. 11 shows another example of the early deterioration battery display screen. The model type display unit 901 displays the storage system type, battery type, storage system ID, and characteristics to be displayed. This display screen can be displayed for each power storage system type, each battery type, and each power storage system ID. In this example, an example of displaying an early deteriorated battery related to internal resistance for each power storage system ID is shown.

  A characteristic and deterioration change graph display unit 902 displays characteristics and deterioration fluctuations in time series. As for the characteristics, the data in the characteristic column of the operation history table 109 is displayed for the battery of the type display unit 901. In the graph of the deterioration variation, the deterioration variation of the corresponding characteristic in the diagnosis result table 111 for the battery of the type display unit 901, that is, the data of the initial characteristic + degradation variation is displayed.

  In the distribution display unit 903 for the latest deterioration variation, data at a time point indicated as “latest” in the characteristic and deterioration variation graph display unit 902 is displayed as a histogram. Here, the battery ID is displayed for the early deteriorated battery exceeding the upper limit value / lower limit value. In this example, it can be seen that the battery with a battery ID of 5 is an early deteriorated battery.

  FIG. 12 shows an example different from FIG. 12, unlike the characteristic and deterioration variation graph display unit 902 in FIG. 11, the warning line is shown in the characteristic and deterioration variation graph display unit 1002 in FIG. As a result, it is possible to know whether the deterioration fluctuation amount increases and exceeds the warning line. Unlike the latest degradation change distribution display section 903 in FIG. 11, the latest degradation change distribution display section 1003 also shows a warning line, and there is a battery that exceeds the warning line, and its battery ID is the same. It turns out that it is 6,7.

  FIG. 13 shows a block diagram of the battery deterioration diagnosis system in the present embodiment. In FIG. 13, components 101 to 119 are the same as those in FIG. 1, and are given the same reference numerals as those in FIG. In FIG. 13, the operation history estimation unit 122 estimates a future operation history based on the past operation history stored in the operation history table 109, and adds the estimation result to the operation history table 109. The characteristic estimation unit 123 estimates future battery characteristics based on the data of the operation history table 109 and the characteristic variation model table 110, and stores the estimation result in the diagnosis result table 111.

  FIG. 14 shows the structure of the operation history table 109 in this embodiment. In FIG. 14, case No. is added to the operation history table 109 of the first embodiment shown in FIG. Since the structure is such that columns are added, only the added columns will be described. Case No. The column stores information for distinguishing whether the stored data is a measured value or a value estimated by the operation history estimation unit 122. In this example, the actually measured value stores 0, and the estimated value stores other than 0.

  FIG. 15 shows the structure of the diagnosis result table 111 in this embodiment. In FIG. 15, case No. is stored in the diagnosis result table 111 of the first embodiment shown in FIG. Since the structure is such that columns are added, only the added columns will be described. Case No. The column stores information for distinguishing whether the stored data is a value diagnosed based on an actually measured operation history or a value diagnosed based on an estimated operation history. In this example, 0 is stored when diagnosis is performed using the actually measured operation history, and non-zero is stored when diagnosis is performed using the estimated operation history.

  FIG. 16 shows a flowchart for performing a deterioration diagnosis by estimating a future deterioration of characteristics in this embodiment. In FIG. 16, processing steps 601 to 605 are the same processing steps as those in FIG. 6, and are denoted by the same reference numerals as those in FIG. 6, and description thereof is omitted. In FIG. 16, when it is confirmed in the processing step 603 that the model satisfies the constraints, a future operation history is estimated and added to the operation history table 109 (613). Here, based on the actually measured operation history, the future operation history is estimated according to the rule. In the operation history table 109, the actually measured operation history is stored in the case number. Since the column is registered as 0, the power storage system ID column matches the power storage system ID to be diagnosed. Get data whose column is 0. A future value is estimated according to the rule for each battery ID, and case No. Register the column as 1. In this example, in consideration of the influence of the season and the like, in order to acquire the state of the past battery for a certain period, the operation history of the last one year is repeated until the time when a predetermined expected life (for example, 10 years) elapses. Estimate by things. Next, the future battery characteristics (deterioration variation) are estimated (614). Future battery characteristics (deterioration variation) can be estimated by Equation 1, Equation 3, and Equation 6 below.

Variation during measurement (t) = a11 x f11 (average value over the entire period of x11 (t))
+ A12 × f12 (average value over the entire period of x12 (t)) (Expression 6)
Here, x11 (t), x12 (t) and f11, f12, a11, and a12 in Expression 6 are the same variables as in Expression 1. From the operation history table 109, the storage system ID column matches the storage system ID to be diagnosed, and the case No. Data in which the column is other than 0 is acquired, and the obtained data is substituted into Equation 1, Equation 3, and Equation 6 to estimate future battery characteristics (deterioration variation), and the estimated value is stored in the diagnosis result table 111. Store. Case No. The case No. acquired from the operation history table 109 is shown in the column. Stores the same value as the column. Next, it is determined whether the battery characteristics exceed the use limit by the expected life (615). Here, it is determined whether or not the battery characteristics (deterioration fluctuation) exceed the use limit that is a predetermined reference value by the expected life (for example, 10 years) that is a predetermined use period. If it exceeds, an alert is issued and a screen is displayed (616). An abnormality is reported through the alert reporting unit 117, and a characteristic estimation screen is generated by the result display unit 118 and displayed on the display unit 119. A battery that has not reached its expected life may be replaced with a normal battery as a countermeasure. In this case, since the data such as the battery ID in the operation history table 109 is changed, the data in the operation history table 109 is corrected as necessary (609).

  As a result, according to the present embodiment, it is possible to accurately estimate future characteristics from the actual operation history and characteristics of the product without diagnosing the deterioration constant with respect to the battery voltage and temperature in advance through experiments or the like, and diagnose deterioration. It is possible to detect batteries that are rapidly deteriorated and become unusable at an earlier stage than expected, before they lead to abnormalities in the power storage system, and it is possible to prevent power storage system abnormalities in advance by taking countermeasures. Become.

  FIG. 17 shows a flowchart of battery deterioration diagnosis in this embodiment. In FIG. 17, processing steps 601 to 615 are the same processing steps as those in FIG. 16, and the same reference numerals as those in FIG. In FIG. 17, when the characteristic exceeds the use limit by the expected life in step 615, a revision plan of the future operation history is created in order to achieve the expected life (617). First, among the columns of the operation history table 109, a column that affects deterioration is specified. However, columns related to time are excluded because they cannot be controlled. The column corresponding to the variable stored in the variable column is specified for the row in which the measurement variation / degradation variation column of the characteristic variation model table 110 is stored as “degradation”. In the example of FIG. 4, the “current” and “time” included in the variable sequence “sum (abs (current))” and “time-init (time)” in which the measurement variation / degradation variation sequence is “degradation”. ”Is excluded, and the variable“ time ”relating to time is excluded, so that“ current ”is specified. The revision plan is created by the following two methods.

  The first method is a method of accepting a correction proposal via a screen. The result display unit 118 displays a variable that affects the deterioration and accepts a change amount of the variable. Next, as in step 613, the operation history estimation unit 122 performs estimation by repeating the operation history of the most recent year until the time when a predetermined expected life elapses. A correction plan is created by multiplying or subtracting the received variable change amount by the value obtained from the operation history table 109. Then, the created correction plan is stored in the operation history table 109. At that time, case no. The stored case number is stored in the column. Stores the value obtained by incrementing the maximum value of the column.

  Next, the second method is a method of creating a correction plan with reference to the operation history of another power storage system stored in the operation history table 109. First, all the power storage system IDs having the same value as the power storage system to be diagnosed are acquired from the system master table 108. Next, a case No. indicating that the power storage system ID column from the operation history table 109 matches the acquired power storage system ID and is a measured value. Data for which the column matches 0 is acquired for the most recent year from the time of diagnosis. Next, for a column that affects deterioration, an average value is calculated for a plurality of different batteries of all the battery IDs of the storage system ID to be diagnosed from the acquired data, and the representative value is set to 1 as a representative value. The average value for each combination of ID and battery ID is the amount of change. Then, among the values acquired from the operation history table 109, a plurality of modification proposal candidates are created by multiplying or gradually subtracting each change amount for the value of the column that affects the deterioration. The created revision proposal is stored in the operation history table 109. At that time, case no. The column stores a unique number for each change amount.

  Next, the future characteristics (deterioration variation) are estimated for all the prepared revision proposals (618). As in step 614, the data in which the power storage system ID string matches the power storage system ID to be diagnosed is obtained from the operation history table 109, and the obtained data is estimated by substituting the data into Equations 1, 3, and 6. The estimated battery characteristics (deterioration variation) are stored in the diagnosis result table 111. The estimation is case no. This is performed for each column value, and the diagnosis result table 111 shows the case number. Stores the estimated result together with the value of.

  Next, it is determined whether there is a correction plan that does not exceed the use limit by the expected life (619). The storage system ID column from the diagnosis result table 111 acquires all data that matches the storage system ID to be diagnosed, and does not exceed a predetermined battery characteristic usage limit by a predetermined expected life (for example, 10 years). Determine if a case exists. If there is a case where the usage limit is not exceeded, the result is displayed (620). In that case, all the data in which the power storage system ID column stored in the diagnosis result table 111 matches the diagnosis target and the amount of change when the operation history correction plan is created are displayed. If there is no case where the usage limit is not exceeded, an alert is issued / displayed that the expected battery has not been reached (621). An abnormality is reported through the alert reporting unit 117, and a characteristic estimation screen is generated by the result display unit 118. The storage unit ID string stored in the diagnosis result table 111 is displayed on the display unit 119 as in step 621. Together with all the data and the amount of change when the revision plan of the operation history is created.

  As a result, according to the present embodiment, it is possible to accurately estimate future characteristics from the actual operation history and characteristics of the product without diagnosing the deterioration constant with respect to the battery voltage and temperature in advance through experiments or the like, and diagnose deterioration. Can be detected at the stage of signs before deterioration leads to abnormalities in the power storage system, and the degree of deterioration is more advanced than expected. From a plurality of use conditions, use conditions that approach the expected life can be obtained.

  In this embodiment, a display example on the display unit in the battery deterioration diagnosis system will be described. FIG. 18 shows an example of the characteristic estimation screen. In FIG. 18, the type display unit 1101 displays the power storage system type, battery type, characteristics, and power storage system ID to be diagnosed. This display screen can be displayed for each power storage system type, each battery type, and each power storage system ID. In the present example, an example is shown in which a battery that has not reached its expected life regarding internal resistance is displayed for each power storage system ID. A characteristic and deterioration fluctuation graph 1102 displays characteristics and deterioration fluctuations in time series. Regarding the characteristics, case No. which is a measured value among the data in the characteristic column of the operation history table 109 for the battery of the type display unit 1101. Display data with column 0. In the graph of the deterioration variation, the deterioration variation of the corresponding characteristic in the diagnosis result table 111 for the battery of the type display unit 1101 is displayed. Here, the case number of the diagnosis result table 111 is changed. Data with a column of 0 is plotted with a solid line to indicate that it has been calculated based on the measured operating history. Case No. Data other than 0 in the column is plotted with a broken line to indicate that it is calculated based on the estimated operation history. The use limit is indicated by a broken horizontal line in the deterioration change graph, and when the deterioration change of characteristics exceeds the use limit by the expected life, the battery ID is displayed as a battery that has not reached the expected life. In this example, it can be seen that the battery ID is 3.

  FIG. 19 shows an example in which the change amount used for estimating the operation history is received on the screen from the characteristic estimation screen. In FIG. 19, the type display unit 1103 displays the power storage system type, battery type, characteristics, power storage system ID, and battery ID to be diagnosed. This display screen can be displayed for each power storage system type, each battery type, each power storage system ID, and each battery ID. In this example, an estimation example of a battery ID of a specific power storage system ID is shown. The operating history change amount receiving unit 1104 identifies and displays variables that affect deterioration. Further, the change amount can be input and accepted by a slide bar or a text box. When the input of the change amount is received, the operation history is estimated based on the change amount, and the characteristic (deterioration variation) based on the estimated operation history is estimated and displayed on the characteristic and deterioration variation graph 1105. Regarding the characteristics, case No. indicating the measured value of the data in the characteristic column of the operation history table 109 for the battery of the type display unit 1103. Display data with column 0. In the deterioration variation graph, the deterioration variation amount of the corresponding characteristic in the diagnosis result table 111 for the battery of the type display unit 1103 is displayed. Here, the case number of the diagnosis result table 111 is changed. Data with a column of 0 is plotted with a solid line to indicate that it has been calculated based on the measured operating history. Case No. Data other than 0 in the column is plotted with a broken line to indicate that it is calculated based on the estimated operation history. In addition, the variables changed during estimation and the amount of change are also shown. In this example, when the operation history is estimated with the current change amount set to 1.0, the usage limit is reached by the expected life, but when the operation history is estimated with the change amount set to 0.8, the expected value is expected. It can be seen that the service life can be continued without reaching the service life limit.

  FIG. 20 shows an example in which a plurality of proposals are created and presented from the characteristic estimation screen without accepting the change amount at the time of estimating the operation history from the screen. This display screen can be displayed for each power storage system type, each battery type, each power storage system ID, and each battery ID. In this example, an estimation example of a battery ID of a specific power storage system ID is shown. In FIG. 20, a characteristic and deterioration change graph 1106 displays a plurality of operation history estimations and estimation results of characteristics (deterioration fluctuations) based on the estimated operation history. Regarding the characteristics, case No. indicating the measured value of the data in the characteristic column of the operation history table 109 for the battery of the type display unit 1103. Display data with column 0. In the deterioration variation graph, the deterioration variation amount of the corresponding characteristic in the diagnosis result table 111 for the battery of the type display unit 1103 is displayed. Here, the case number of the diagnosis result table 111 is changed. Data with a column of 0 is plotted with a solid line to indicate that it has been calculated based on the measured operating history. Case No. Data other than 0 in the column is plotted with a broken line to indicate that it is calculated based on the estimated operation history. In addition, the variables changed during estimation and the amount of change are also shown. In this example, when the operation history is estimated with the current change amount set to 1.2 or 1.0, the use limit is reached by the expected life, but the change amount is set to 0.8 or 0.6. When the estimation is performed, it is understood that the use limit can not be reached until the expected life and the use can be continued. From the above description, the battery abnormality predictive diagnosis system according to the present invention allows the battery to be deteriorated at an abnormally fast rate without obtaining a deterioration constant with respect to the battery voltage and temperature by a prior experiment or the like. It is possible to detect the battery that will be used as a sign before the storage system malfunctions due to deterioration, and when the degree of deterioration is more advanced than expected, it is possible to derive conditions that allow it to be used until the expected life It becomes.

  The present invention is not limited to the above-described embodiments, and includes various modifications. The above-described embodiments have been described in detail for easy understanding of the present invention, and are not necessarily limited to those having all the configurations described. Further, a part of the configuration of one embodiment can be replaced with the configuration of another embodiment, and the configuration of another embodiment can be added to the configuration of one embodiment. Further, it is possible to add, delete, and replace other configurations for a part of the configuration of each embodiment.

DESCRIPTION OF SYMBOLS 101 ... Power storage system, 102 ... Battery, 107 ... Memory | storage part, 108 ... System master table, 109 ... Operation history table, 110 ... Characteristic variation model table, 111 ... Diagnosis result table, 112 ... Processing part, 119 ... Display part

Claims (22)

A communication unit that collects battery status;
An operation history storage unit for storing the state of the battery;
A characteristic measuring unit for measuring characteristics that change due to deterioration of the battery from the state of the battery;
A characteristic variation model creating unit for creating a model of the relationship between the characteristics and the state of the battery;
A deterioration diagnosis device for a battery, comprising: a deterioration battery diagnosis unit that diagnoses battery deterioration based on an estimated value related to the characteristic calculated based on the model. The battery deterioration diagnosis device according to claim 1,
The battery deterioration diagnosis device characterized in that the state of the battery includes at least one of current, voltage, SOC (State of Charge), temperature, and humidity. The battery deterioration diagnosis device according to claim 1,
The battery deterioration diagnosis apparatus, wherein the characteristic includes an internal resistance or a full charge capacity. The battery deterioration diagnosis device according to claim 1,
The battery deterioration diagnosis apparatus, wherein the characteristic variation model creation unit creates the model by multiple regression analysis. The battery deterioration diagnosis device according to claim 1,
The deteriorated battery diagnosis unit compares the measured value of the characteristic output from the characteristic measuring unit with the estimated value of the characteristic based on the model, and whether the difference between the measured value and the estimated value exceeds a predetermined reference value A battery deterioration diagnosis device, which is an abnormal deterioration battery diagnosis unit that diagnoses an abnormality of a battery based on whether or not. The battery deterioration diagnosis device according to claim 1,
The deteriorated battery diagnosis unit divides the measurement value of the characteristic output from the characteristic measurement unit into a variation at the time of measurement due to a battery state at the time of measurement and a deterioration variation due to the influence of deterioration until the measurement, and is output from the characteristic measurement unit It is an early deterioration battery diagnosis unit for diagnosing a sign of battery abnormality based on whether or not a deterioration fluctuation amount excluding an estimated value of fluctuation at the time of measurement by the model from a measured value of characteristics exceeds a predetermined reference value. Battery deterioration diagnosis device. The battery deterioration diagnosis device according to claim 1,
The battery deterioration diagnosis apparatus characterized in that the characteristic variation model creation unit and the deteriorated battery diagnosis unit convert the temperature by a logarithmic function to create the model and diagnose the deteriorated battery. The battery deterioration diagnosis device according to claim 1,
The battery deterioration diagnosis apparatus characterized in that the characteristic variation model creation unit and the deteriorated battery diagnosis unit perform the model preparation and the deteriorated battery diagnosis by converting the battery usage time by a power function. The battery deterioration diagnosis device according to claim 1,
It has an alert issuing unit that issues an alert by any of screen display, e-mail, telephone, and fax, and when the abnormality occurs in the deteriorated battery diagnosis unit, it is reported through the alert issuing unit. Battery deterioration diagnosis device. Collect battery status,
Measure the characteristics that change due to the deterioration of the battery from the state of the battery,
Create a model of the relationship between the characteristics and the state of the battery,
A battery deterioration diagnosis method comprising diagnosing battery deterioration based on an estimated value relating to the characteristic calculated based on the model. The battery deterioration diagnosis method according to claim 10,
The estimated value for the property is an estimate of the property by the model;
A battery abnormality is diagnosed by comparing the measured value of the characteristic with the estimated value of the characteristic and determining whether or not a difference between the measured value and the estimated value exceeds a predetermined reference value. Degradation diagnosis method. The battery deterioration diagnosis method according to claim 10,
The measured value of the characteristic is a fluctuation at the time of measurement due to the battery state at the time of measurement and a deterioration fluctuation due to the influence of deterioration until the measurement,
The estimated value related to the characteristic is an estimated value of variation during measurement by the model,
A battery deterioration diagnosis method characterized by diagnosing a sign of battery abnormality based on whether or not a deterioration fluctuation amount excluding the estimated value at the time of measurement from the measurement value of the characteristic exceeds a predetermined reference value. A battery deterioration diagnosis system comprising a power storage system, a communication network, and a deterioration diagnosis device,
The power storage system includes:
One or more batteries;
A controller connected to the battery and collecting an actual measured value of the state of the battery as an operation history;
A communication unit that transmits the collected operation history to the communication network;
The deterioration diagnosis device is:
A communication unit for collecting the operation history via the communication network;
An operation history storage unit for storing the operation history;
A characteristic measuring unit for measuring characteristics that change due to deterioration of the battery from the operation history;
A characteristic variation model creating unit for creating a model of the relationship between the characteristics and the operation history;
A deterioration diagnosis system for a battery, comprising: a deterioration battery diagnosis unit that diagnoses deterioration of the battery based on an estimated value related to the characteristic calculated based on the model. The battery deterioration diagnosis device according to claim 1,
The battery deterioration diagnosis device, wherein the deterioration battery diagnosis unit diagnoses battery deterioration based on a future estimated value related to the characteristic calculated based on the model. The battery deterioration diagnosis device according to claim 14,
Based on the battery state stored in the operation history storage unit, an operation history estimation unit that estimates the state of the future battery;
The degradation battery diagnosis unit diagnoses degradation of a battery based on a state of the future battery and a future estimated value related to the characteristic calculated based on the model. The battery deterioration diagnosis device according to claim 14,
The deterioration battery diagnosis unit diagnoses battery deterioration based on whether or not a future estimated value related to the characteristic exceeds a predetermined reference value during a predetermined period of use. apparatus. The battery deterioration diagnosis device according to claim 15,
The operation history estimation unit obtains a past battery state stored in the operation history storage unit for a certain period, and estimates a future battery state as a repetition of the past battery state. Battery deterioration diagnosis device. The battery deterioration diagnosis device according to claim 15,
The operation history estimation unit acquires a past battery state stored in the operation history storage unit for a certain period, and multiplies or divides a future battery state by a certain value to the past battery state. A battery deterioration diagnosis device characterized in that the estimation is made as a repetition of a value. The battery deterioration diagnosis device according to claim 18,
The operation history estimation unit creates a plurality of candidates for the certain value, and estimates the characteristics of the future battery for each of the plurality of candidates. The battery deterioration diagnosis device according to claim 15,
The operation history estimation unit acquires a past battery state stored in the operation history storage unit for a plurality of different batteries for a certain period, and based on the past battery state acquired for the plurality of different batteries. A battery deterioration diagnosis device characterized by estimating a state of the future battery. The battery deterioration diagnosis method according to claim 10,
Based on the state of the battery, estimate the future state of the battery,
A battery deterioration diagnosis method comprising diagnosing battery deterioration based on a state of the future battery and a future estimated value related to the characteristic calculated based on the model. The battery deterioration diagnosis system according to claim 13,
Based on the battery state stored in the operation history storage unit, an operation history estimation unit that estimates the state of the future battery;
The battery deterioration diagnosis system, wherein the deterioration battery diagnosis unit diagnoses battery deterioration based on a state of the future battery and a future estimated value related to the characteristic calculated based on the model.

Images