WO2015198631A1 - 蓄電池システムの劣化制御装置及びその方法 - Google Patents
蓄電池システムの劣化制御装置及びその方法 Download PDFInfo
- Publication number
- WO2015198631A1 WO2015198631A1 PCT/JP2015/055154 JP2015055154W WO2015198631A1 WO 2015198631 A1 WO2015198631 A1 WO 2015198631A1 JP 2015055154 W JP2015055154 W JP 2015055154W WO 2015198631 A1 WO2015198631 A1 WO 2015198631A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- unit
- storage battery
- value
- deterioration
- learning
- Prior art date
Links
Images
Classifications
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01M—PROCESSES OR MEANS, e.g. BATTERIES, FOR THE DIRECT CONVERSION OF CHEMICAL ENERGY INTO ELECTRICAL ENERGY
- H01M10/00—Secondary cells; Manufacture thereof
- H01M10/42—Methods or arrangements for servicing or maintenance of secondary cells or secondary half-cells
- H01M10/44—Methods for charging or discharging
- H01M10/441—Methods for charging or discharging for several batteries or cells simultaneously or sequentially
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/367—Software therefor, e.g. for battery testing using modelling or look-up tables
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/382—Arrangements for monitoring battery or accumulator variables, e.g. SoC
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/392—Determining battery ageing or deterioration, e.g. state of health
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/396—Acquisition or processing of data for testing or for monitoring individual cells or groups of cells within a battery
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/0265—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric the criterion being a learning criterion
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01M—PROCESSES OR MEANS, e.g. BATTERIES, FOR THE DIRECT CONVERSION OF CHEMICAL ENERGY INTO ELECTRICAL ENERGY
- H01M10/00—Secondary cells; Manufacture thereof
- H01M10/42—Methods or arrangements for servicing or maintenance of secondary cells or secondary half-cells
- H01M10/4285—Testing apparatus
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01M—PROCESSES OR MEANS, e.g. BATTERIES, FOR THE DIRECT CONVERSION OF CHEMICAL ENERGY INTO ELECTRICAL ENERGY
- H01M10/00—Secondary cells; Manufacture thereof
- H01M10/42—Methods or arrangements for servicing or maintenance of secondary cells or secondary half-cells
- H01M10/44—Methods for charging or discharging
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01M—PROCESSES OR MEANS, e.g. BATTERIES, FOR THE DIRECT CONVERSION OF CHEMICAL ENERGY INTO ELECTRICAL ENERGY
- H01M10/00—Secondary cells; Manufacture thereof
- H01M10/42—Methods or arrangements for servicing or maintenance of secondary cells or secondary half-cells
- H01M10/48—Accumulators combined with arrangements for measuring, testing or indicating the condition of cells, e.g. the level or density of the electrolyte
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01M—PROCESSES OR MEANS, e.g. BATTERIES, FOR THE DIRECT CONVERSION OF CHEMICAL ENERGY INTO ELECTRICAL ENERGY
- H01M10/00—Secondary cells; Manufacture thereof
- H01M10/42—Methods or arrangements for servicing or maintenance of secondary cells or secondary half-cells
- H01M10/48—Accumulators combined with arrangements for measuring, testing or indicating the condition of cells, e.g. the level or density of the electrolyte
- H01M10/486—Accumulators combined with arrangements for measuring, testing or indicating the condition of cells, e.g. the level or density of the electrolyte for measuring temperature
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J7/00—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries
- H02J7/0047—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries with monitoring or indicating devices or circuits
- H02J7/0048—Detection of remaining charge capacity or state of charge [SOC]
- H02J7/0049—Detection of fully charged condition
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J7/00—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries
- H02J7/0047—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries with monitoring or indicating devices or circuits
- H02J7/005—Detection of state of health [SOH]
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J7/00—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries
- H02J7/0069—Charging or discharging for charge maintenance, battery initiation or rejuvenation
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03M—CODING; DECODING; CODE CONVERSION IN GENERAL
- H03M7/00—Conversion of a code where information is represented by a given sequence or number of digits to a code where the same, similar or subset of information is represented by a different sequence or number of digits
- H03M7/30—Compression; Expansion; Suppression of unnecessary data, e.g. redundancy reduction
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E60/00—Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation
- Y02E60/10—Energy storage using batteries
Definitions
- Embodiments of the present invention relate to a storage battery system deterioration control apparatus and method.
- the system is controlled by monitoring measured values such as voltage values, temperature values, and SOC (remaining capacity) values of the storage battery cells.
- the control device incorporated in the storage battery system include a deterioration control device and a charge / discharge control device.
- the deterioration control device plays an important role of estimating the deterioration state of the system based on the actual measurement value.
- Measured battery data is usually treated as usage history data, stored in memory, and stored in a database.
- a data amount of a deterioration model parameter is calculated from use history data at the time of charge / discharge of a storage battery cell, and a deterioration model for estimating a deterioration state is constructed.
- the deterioration model parameter is an element indicating the deterioration state of the storage battery cell, such as a temperature value or an SOC value.
- a deterioration model for the temperature value and the SOC value is constructed by calculating a data amount such as a certain temperature value and how long the storage battery cell stays at the certain SOC value. That is, the data amount of the deterioration model parameter is the amount of deterioration of the storage battery system per unit time, and can represent the deterioration rate of the storage battery system. By integrating the deterioration rate data with time, it is possible to estimate the deterioration state of the storage battery system.
- the deterioration control device estimates the deterioration state of the storage battery system from the constructed deterioration model, and outputs the estimation result to the charge / discharge control device.
- a charge / discharge plan and a storage battery usage pattern are created based on the estimation result obtained from the deterioration control device.
- a charging / discharging control apparatus performs charging / discharging control of a storage battery cell according to these charging / discharging plans and usage patterns. At this time, it is possible to minimize the electricity bill and maximize the storage battery life by taking into account the demand amount prediction and the PV power generation amount prediction.
- Patent Document 1 is a technique for determining the possibility of shortening the battery life from storage battery usage history data and urging charging as necessary.
- patent document 2 the duration and temperature of the high charge state in a storage battery are measured, a table is created from predetermined data, and the deterioration coefficient of a storage battery is calculated with reference to this.
- Patent Documents 3 and 4 as techniques for estimating the deterioration state of a storage battery.
- a linear approximation is performed using the least square method or the like from storage battery usage history data, and the deterioration state of the storage battery is estimated by calculating impedance.
- Patent Document 4 a reference value map is created based on electric power and temperature, a value calculated from this reference value map is compared with an actually measured internal resistance value, and the reference value map is updated to determine the deterioration state of the storage battery. presume.
- the accuracy of deterioration control can be further improved by using a lot of usage history data.
- the storage battery system is configured from a server side such as a cloud and the local system side, and the server side and the local system side are linked. According to such a storage battery system, since a large memory can be provided on the server side, it is possible to cope with an increase in usage history data.
- Patent documents 5-7 etc. are proposed as a degradation control device of such a storage battery system.
- Patent Document 5 the storage battery usage history data is transmitted from the local side to the server using communication, and the lifetime consumption value for each local system is calculated on the server side. Then, the calculation result is transmitted to the local system side, and deterioration control is performed for each local system.
- Patent Document 6 a charger to which a storage battery is attached and a server are connected, the storage battery usage history data is acquired from the charger, this is sent to the server, and the usage history data is converted into a database to store the storage battery. Perform deterioration diagnosis.
- Patent Document 7 storage battery usage history data is recorded in a database via a network, and deterioration of the storage battery is diagnosed using the database.
- Patent Document 8 that collects storage battery usage history data and analyzes the cause of failure
- Patent Document 9 that collects usage history data online and performs data analysis to determine the storage battery usage environment and the life of each storage battery. Etc. are also proposed.
- the conventional storage battery system deterioration control device has the following problems.
- storage battery systems have been used in various places such as ordinary homes, buildings or substations. For this reason, the charging / discharging current value in a use scene changes with uses of a storage battery system, and the capacity
- the C rate is small in a small-scale storage battery system for general home use and the C rate is large in a large-scale storage battery system combined with wind power.
- the deterioration control device of the storage battery system it is desired for the deterioration control device of the storage battery system to create a deterioration model more flexibly and accurately.
- the amount of data becomes even larger. Therefore, in a deterioration control device of a storage battery system, it is required to reduce the usage history data amount used for deterioration control without reducing the accuracy of deterioration control at all.
- Refresh charge / discharge is to charge the battery system from 0% to 100% for the purpose of resetting the capacity value or actually measuring the capacity value, or from 100% to 0%. Is to discharge. Since the data acquired at the time of refresh charge / discharge is for the entire storage battery system, the data amount of the deterioration model parameter calculated based on such data is the data amount of the entire storage battery system. .
- Data regarding the entire storage battery system may not accurately reflect the deterioration state of each storage battery cell.
- the data amount of the deterioration model parameter differs between when the deterioration state of each storage battery cell is uniform or close to that and when the deterioration state of each storage battery cell has a range. This is because the data amount of the entire storage battery system varies due to variations in the deterioration state of each storage battery cell.
- the deterioration control apparatus which eliminates the influence by the dispersion
- Embodiments of the present invention have been proposed to solve the above-described problems, and the purpose thereof is to cope with diversification of applications by accurately learning a deterioration model necessary for deterioration control.
- a storage battery system deterioration control device capable of reducing the amount of data used for deterioration control, and further enabling the construction of a highly accurate deterioration model using data acquired from the entire storage battery system, and Is to provide a method.
- a deterioration control device for a storage battery system having a plurality of storage battery cells comprises the following components (a) to (c).
- a usage history data acquisition unit that acquires usage history data of the storage battery cell.
- B) A learning unit that updates a data amount of a deterioration model parameter indicating a deterioration state of the storage battery cell based on the use history data, learns the deterioration model parameter, and outputs a deterioration rate table as a learning result.
- the use history data and the estimated value are compared to determine whether or not the deterioration model parameter can be learned, and a learning instruction signal is output to the learning unit, and a change amount of the deterioration model parameter is calculated and A learning instruction unit that outputs a change amount to the learning unit.
- the deterioration control method of a storage battery system is also one aspect of the embodiment of the present invention.
- the whole block diagram of a 1st embodiment The figure which shows the example of a deterioration speed table.
- the block diagram of the learning part of 1st Embodiment The example of the capacitance value table and internal resistance value table which are deterioration rate tables.
- the storage battery system is provided with a plurality of storage battery cells 20.
- the storage battery cell 20 is configured to output usage history data A to the deterioration control device 10.
- the usage history data A is data such as a voltage value, a temperature value, a current value, and an SOC value of the storage battery cell 20.
- the deterioration control device 10 is connected to the storage battery cell 20.
- a charge / discharge control device 30 is connected to the deterioration control device 10.
- the charge / discharge control device 30 is a device that creates a charge / discharge plan and a storage battery usage pattern based on the estimation result of the deterioration control device 10 and controls the charge / discharge of the storage battery cell 20.
- the charge / discharge control device 30 obtains a charge / discharge command F from the charge / discharge plan and usage pattern, and outputs this to the storage battery cell 20.
- the storage battery cell 20 that has received the charge / discharge command F from the charge / discharge control device 30 performs charge / discharge according to the charge / discharge command F.
- the degradation control device 10 is a device that receives usage history data A from the storage battery cell 20 and creates a degradation rate table T to estimate the degradation state of the storage battery cell 20.
- the deterioration control device 10 is configured to output the estimation result of the deterioration state of the storage battery cell 20 to the charge / discharge control device 30.
- the deterioration control device 10 is composed of three large blocks. The three blocks are a use history data acquisition unit 1, a learning instruction unit 2, and a learning unit 3. Hereinafter, an outline of the usage history data acquisition unit 1, the learning instruction unit 2, and the learning unit 3 will be described.
- the usage history data acquisition unit 1 is a part that inputs the usage history data A from the storage battery cell 20 and outputs the usage history data A to the learning instruction unit 2 and the learning unit 3.
- the usage history data acquisition unit 1 is provided with a database unit 11.
- the database unit 11 is configured to store the use history data A in a reversible compression manner.
- the usage history data acquisition unit 1 reads out the compressed state of the usage history data A stored in the database unit 11 after restoring it.
- the learning instruction unit 2 includes a determination unit 21 and an estimated value calculation unit 23. Whether the determination unit 21 reads the usage history data A from the usage history data acquisition unit 1, reads the measured capacity B and the deterioration model parameter C from the learning unit 3, compares them, and learns the deterioration rate table T This is a portion that determines whether or not to output a learning instruction signal E.
- the degradation model parameter C is an element indicating the degradation state of the storage battery cell 20 as described in paragraph 0004, and is a temperature value, an SOC value, or the like.
- the measured capacity B will be described in a later paragraph 0035, and the deterioration rate table T will be described in a latter paragraph 0033. Further, when the determination unit 21 determines to learn the deterioration rate table T, the determination unit 21 outputs a learning instruction signal E to the learning unit 3.
- the estimated value calculation unit 23 is a part that calculates the change amount D of the deterioration model parameter C and outputs it to the learning unit 3 when the determination unit 21 determines that the deterioration rate table T is to be learned.
- the change amount D of the deterioration model parameter C is an estimated value of the capacity deterioration amount (usually expressed in%) per storage battery cell 20.
- the learning unit 3 sends the measured capacity B and the deterioration model parameter C to the learning instruction unit 2.
- the learning unit 3 receives the change amount D of the deterioration model parameter C and the learning instruction signal E from the learning instruction unit 2, updates the data amount of the deterioration model parameter C, learns the deterioration model parameter, and performs the deterioration rate.
- the table T is output to the charge / discharge control device 30.
- the deterioration rate table T is a group of data indicating the deterioration rate of the storage battery system, in which the data of the deterioration model parameter C is compiled in a tabular form as a learning result of the learning unit 3.
- An example of the deterioration rate table T is shown in FIG.
- the SOC value of the storage battery is 50%, 40%, 30%, 20%, and the temperature values are 20 ° C., 30 ° C., 40 ° C., 50 ° C.
- There are a table showing the rate of decrease of the capacity value per week in the storage battery system a table showing the rate of decrease of the capacity value per week in the storage battery system in the same situation at 0.5C, and the like.
- the first embodiment is characterized by the configuration of the learning unit 3 among the three blocks in the deterioration control device 10.
- the learning unit 3 includes a current integration unit 31, a constant current capacity detection unit 32, a constant voltage capacity detection unit 33, and a deterioration rate table learning unit 34.
- the current integrating unit 31 calculates the actually measured capacity B according to the use history data A from the use history data acquiring unit 1 and the detection signals G1 and G2 from the detecting units 32 and 33, and the measured capacity B is calculated as the learning instruction unit 2. This is the output part.
- a constant current capacity is calculated by combining the current integrating unit 31 and the constant current capacity detecting unit 32, and a constant voltage capacity is calculated by combining the current integrating unit 31 and the constant voltage capacity detecting unit 33. It has become.
- the actually measured capacity B calculated by the current integrating unit 31 is a value obtained using two kinds of capacity values, that is, a constant current capacity value and a constant voltage capacity value.
- the constant current capacity value and the constant voltage capacity value are capacity values depending on the measurement method, and these two capacities have different physical meanings.
- the constant current capacity value is a capacity value when the storage battery is charged or discharged until it reaches a specific voltage with a constant current. Therefore, the current integrating unit 31 receives the charge or discharge start signal and end signal from the constant current capacity detecting unit 32 as the detection signal G1, and integrates the capacity value therebetween to obtain the constant current capacity value. Yes.
- the constant voltage capacity value is the total capacity value when the storage battery is charged or discharged until it reaches a constant voltage while gradually decreasing the current value to be charged or discharged. That is, the current integration unit 31 receives the charge or discharge start signal and end signal as the detection signal G2 from the constant voltage capacity detection unit 33 with the constant current I1, and integrates the capacitance value x1 therebetween. Next, the current integrating unit 31 measures the capacity value x2 of the storage battery that is charged or discharged with a constant current I2 that is smaller than the constant current I1 (I1> I2) until a constant voltage is reached again.
- the current integrating unit 31 gradually decreases the current value to be measured, repeatedly measures the capacitance value xn until the current value In becomes sufficiently small, and the sum of the capacitance values (x1 + x2... Xn) is a constant voltage capacity value. It becomes. As described above, the current integrating unit 31 obtains the constant current capacity value and the constant voltage capacity value.
- the current integrating unit 31 outputs a current value-capacitance value set S to the deterioration rate table learning unit 34 in addition to the actually measured capacity B.
- the current value-capacitance value set S is data in which a current value obtained by averaging during charging or discharging when obtaining a constant voltage capacity value and a constant voltage capacity value are combined into one set.
- the deterioration rate table learning unit 34 is connected to the current integration unit 31 and is also connected to the usage history data acquisition unit 1 and the learning instruction unit 2.
- the degradation rate table learning unit 34 receives the current value-capacitance value set S from the current integrating unit 31, the usage history data A from the usage history data acquisition unit 1, the change amount D and the learning instruction signal E from the learning instruction unit 2. Each is to be entered.
- the deterioration rate table learning unit 34 outputs the deterioration model parameter C to the learning instruction unit 2 side, and outputs the learned deterioration rate table T to the charge / discharge control device 30.
- Such a deterioration rate table learning unit 34 is a part that learns the deterioration rate table T according to various input data.
- the learning target of the degradation rate table learning unit 34 is the data amount of the degradation model parameter C on the degradation rate table T.
- the learning is to update the data amount of the deterioration model parameter C based on the change amount D. That is, the deterioration rate table T in which the data amount of the deterioration model parameter C is updated becomes the learning result of the learning unit 3.
- the learning process as described above is realized by performing optimization using a heuristic method algorithm such as a genetic algorithm or an annealing method.
- a deterioration rate table storage unit 35 is connected to the deterioration rate table learning unit 34, and a deterioration rate table calculation unit 36 is connected to the deterioration rate table storage unit 35.
- a charge / discharge control device 30 is connected to the deterioration rate table calculation unit 36.
- the deterioration rate table storage part 35 is a part which memorize
- the deterioration rate table calculation unit 36 extracts the deterioration rate table T from the deterioration rate table storage unit 35 and calculates data included in the two types of deterioration rate tables.
- the two types of deterioration rate tables T are a capacitance value table T1 (left side in FIG. 4) calculated using a constant current capacity value, and an internal resistance value table T2 calculated using a constant voltage capacity value ( The right side of FIG.
- the deterioration rate table calculation unit 36 compares a plurality of internal resistance value tables T2 with each other, and classifies the similarity between the tables T2 into a plurality of groups according to a predetermined criterion. Yes. Specifically, the deterioration rate table calculation unit 36 performs grouping by determining the similarity of the internal resistance value table T2 with respect to the SOC direction using the SOC as a determination factor and the temperature direction using the temperature as a determination factor. .
- the deterioration rate table calculation unit 36 calculates the determination coefficient (R square) of the internal resistance value table T2 for the SOC direction and the temperature direction of the capacitance value table T1, respectively. Based on this determination coefficient, the similarity of the internal resistance value table T2 is determined. This is shown in FIG.
- the number of classifications in the internal resistance value table T2 by the deterioration rate table calculation unit 36 can be set as appropriate.
- the number of classifications may be automatically determined according to a preset criterion, or may be input from the outside by the user. Also good.
- finer deterioration control according to the current value becomes possible.
- the number of classifications is too large, the calculation cost for the control increases. For this reason, considering the balance between the calculation cost and the control accuracy, it is preferable to adopt three classifications as shown in FIG. 6 or two classifications as shown in FIG.
- the classification of the internal resistance value table T2 can be performed using a technique such as the k-means method.
- the deterioration rate table calculation unit 36 determines the internal resistance value that is the center of the group in the internal resistance value table T2 included in each group.
- the table T2 is output to the charge / discharge control device 30.
- the deterioration rate table calculation unit 36 outputs the current values that are the threshold values of the respective classifications to the charge / discharge control device 30 together.
- the current integrating unit 31 inputs the usage history data A from the usage history data acquiring unit 1.
- the constant current capacity detection unit 32 determines whether charging with a constant current is started (S1-1).
- the detection signal G1 is output to the current integration unit 31 (Yes in S1-1).
- the current integrating unit 31 starts integrating the current (S1-2). If the constant current capacity detection unit 32 does not detect the start of charging with a constant current (No in S1-1), the constant current capacity detection unit 32 repeats the determination process for determining whether charging with a constant current has started.
- the constant current capacity detecting unit 32 detects that the charging with the constant current is completed by the complete charging, the constant current capacity detecting unit 32 outputs a detection signal G1 indicating that the charging is completed to the current integrating unit 31, and whether or not the storage battery cell 20 is fully charged. Is confirmed (S1-3). At this time, if the storage battery cell 20 is not fully charged (No in S1-3), the current integration is stopped and the integrated value is reset (S1-4). When the storage battery is fully charged (Yes in S1-3), the current integrating unit 31 receives the detection signal G1 from the constant current capacity detecting unit 32, calculates a constant current capacity value, and sets a current value-capacitance value set. S is output to the deterioration rate table learning unit 34 (S1-5).
- the constant voltage capacity detection unit 33 outputs a detection signal G2 to the current integration unit 31 when detecting that charging with a constant voltage is started (Yes in S1-6). In response to the detection signal G2, the current integrating unit 31 starts integrating the current (S1-7). If the constant current capacity detection unit 32 does not detect the start of charging at a constant voltage (No in S1-6), the constant current capacity detection unit 32 repeats the process of determining whether charging at a constant voltage has started. When the constant voltage capacity detection unit 33 detects that the charging with the constant voltage is completed by the complete charging, it outputs a detection signal G2 to the current integration unit 31 (S1-8). When the charging is not complete (No in S1-8), the current integration unit 31 stops the current integration and resets the integration value (S1-4).
- the current integrating unit 31 receives the detection signal G2 from the constant voltage capacity detection unit 33 and calculates a constant voltage capacity value. Then, the current integrating unit 31 sets the average current value and the capacity value when charging or discharging is performed with a constant voltage as a set, and outputs the current value-capacitance value set S to the deterioration rate table learning unit 34 (S1-10). .
- the deterioration rate table learning unit 34 includes the current value-capacitance value set S output from the current integrating unit 31, the use history data A from the use history data unit 1, the learning instruction signal E and the change amount from the learning instruction unit 2. According to D, the deterioration rate table T corresponding to each capacity is learned (S1-11). Learning of the estimated value in the deterioration rate table T is performed for each current value-capacitance value set S output from the current integrating unit 31, and is repeated for all current value-capacity sets S to which the learning instruction signal E is given. (S1-12). The deterioration rate table storage unit 35 stores a deterioration rate table T calculated for each current value (S1-13).
- the deterioration rate table calculation unit 36 calculates a deterioration rate table using a constant current capacity value as a capacity value table T1 (S1-14). In addition, the deterioration rate table calculation unit 36 calculates the determination coefficient (R square) of the internal resistance value table T2 that is the learning result of the constant voltage capacity value, with respect to the capacitance value table T1 that is the learning result of the constant current capacity value. (S1-15). Further, the deterioration rate table calculation unit 36 classifies the set of internal resistance value tables T2 obtained by using the constant voltage capacity values into several groups (S1-16), and sets the internal resistance at the center of each group. Only the value table T2 is output to the charge / discharge control device 30 (S1-17). As described above, the learning process of the deterioration model by the learning unit 3 is completed.
- the constant current capacity detection unit 32 detects that the discharge by the constant current is started, the detection signal is output to the current integration unit 31, and the detection signal is received.
- the current integrating unit 31 starts integrating the current.
- the constant current capacity detection unit 32 detects that the discharge with the constant current has been completed by the complete discharge, the constant current capacity detection unit 32 outputs a detection signal indicating that the discharge has been completed to the current integration unit 31, and whether or not the storage battery cell 20 has been completely discharged. To check. At this time, if the storage battery cell 20 is not completely discharged, the current integration is stopped and the integrated value is reset.
- the current integrating unit 31 receives the detection signal from the constant current capacity detecting unit 32, calculates the constant current capacity value, and outputs the current value-capacitance value set S to the deterioration rate table learning unit 34. To do.
- the constant voltage capacity detection unit 33 outputs a detection signal to the current integration unit 31 when detecting that the discharge by the constant voltage is started. In response to this detection signal, the current integrating unit 31 starts integrating the current.
- the constant voltage capacity detection unit 33 outputs a detection signal to the current integration unit 31 when detecting that the discharge with the constant voltage is completed by the complete discharge. If it is not complete discharge, the current integration unit 31 stops the current integration and resets the integration value. If the storage battery cell 20 is completely discharged, the current integration unit 31 receives the detection signal from the constant voltage capacity detection unit 33 and calculates a constant voltage capacity value.
- the current integrating unit 31 outputs the current value-capacitance value set S to the deterioration rate table learning unit 34 by setting the average current value and the capacity value when charging or discharging with a constant voltage is performed as a set.
- the learning process of the deterioration rate table learning unit 34 is the same as when charging is started.
- the capacity value of the storage battery system changes depending on the current value when charging and discharging, and the amount of change due to the current value depends on the amount of deterioration of the internal resistance value. For this reason, if the internal resistance value can be stably estimated, it is considered that the accuracy of deterioration control is improved. Therefore, in the first embodiment, the capacitance value table T1 which is a learning result of the constant current capacitance value is calculated, and the internal resistance value is reliably estimated using the internal resistance value table T2 which is efficiently obtained. Therefore, the accuracy of deterioration control is improved.
- the learning unit 3 of the first embodiment uses a constant current capacity value and a constant voltage capacity value in order to output the capacitance value table T1 and the internal resistance value table T2 as learning results. As described above, the constant current capacity value is a capacity value when the storage battery is charged or discharged until a specific voltage is reached at a constant current, so that the measurement is simple and the time required for the measurement is short.
- the constant current capacity value is not only affected by the deterioration of the capacity value of the storage battery system, but also the capacity value changes due to the deterioration effect of the internal resistance value in the storage battery system.
- the internal resistance value of the storage battery system varies depending on the temperature, voltage value, current value, SOC value, and usage of the storage battery system.
- the constant voltage capacity value is measured while gradually decreasing the current value during charging or discharging, it takes a long time to measure.
- the constant voltage capacity value has an advantage that it is not easily affected by deterioration of the internal resistance value.
- the capacitance value table T1 that is the learning result of the constant current capacity value is compared with the internal resistance value table T2 that is the learning result of the constant voltage capacity value, and the internal resistance included in the constant current capacity value.
- the effect of deterioration by value is obtained. That is, according to the first embodiment, it is possible to easily estimate the internal resistance value of the storage battery system, which was difficult to estimate with the prior art, by using the internal resistance value table T2, and the storage battery system. It is possible to carry out the deterioration control with high accuracy.
- the deterioration rate table learning unit 34 classifies a large number of deterioration rate tables T learned by a constant voltage capacity value into several groups. Thereafter, only the table serving as the center of each group is output as the internal resistance value table T2. Thereby, the data amount used for estimation of an internal resistance value can be suppressed significantly. Therefore, it is possible to facilitate control and eliminate the influence of noise and the like.
- the deterioration rate table learning unit 34 groups the internal resistance value table T2 by comparing the similarities of the tables with respect to the SOC direction and the temperature direction, which briefly indicate the deterioration state of the storage battery cell 20. Therefore, the classification work can be performed efficiently, and a quick learning process is possible. Moreover, since the degradation rate table learning unit 34 can specify the number of groups to be classified, the total number of internal resistance value tables T2 that are learning results can be easily adjusted.
- the learning unit 3 of the first embodiment calculates a current value-capacitance value set S that combines the average current value and the constant voltage capacity value when the current integration unit 31 calculates the constant voltage capacity value. Then, learning is repeatedly performed for all current value-capacitance value sets S having the learning instruction signal E. Therefore, fine learning is possible, and the accuracy of deterioration control is improved.
- the learning unit 3 may perform learning by distinguishing between a capacity value during discharging and a capacity value during charging. This is because the capacity value at the time of discharging and the capacity value at the time of charging are not theoretically completely matched due to hysteresis or the like. Whether charging or discharging can stably measure the capacity value depends on the electrode material and the like. Therefore, the learning unit 3 uses the electrode material and the like as a judgment material to determine the capacity value at the time of discharging and the capacity value at the time of charging. By discriminating and learning, it is possible to further improve the estimation accuracy of the degradation model. Further, the learning unit 3 may delete the use history data A held by the learning unit 3 in accordance with the learning instruction signal E from the learning instruction unit 2. Since this usage history data A is no longer needed, the memory capacity can be reduced by deleting it.
- the second embodiment is characterized by the learning unit 3 as in the first embodiment, and the basic configuration is the same as that of the first embodiment. For this reason, the same components as those in the first embodiment are denoted by the same reference numerals, and description thereof is omitted. In the learning unit 3 in the first embodiment, all components are provided on the local side.
- the learning unit 3 although the components of the learning unit 3 are the same, the place where they are arranged is divided into the local side and the server side of the storage battery system.
- the configuration of such a second embodiment is shown in FIG.
- the learning unit of the second embodiment is configured by being divided into a local learning unit 3A and a server learning unit 3B.
- the local learning unit 3A includes a current integrating unit 31, a constant current capacity detecting unit 32, a constant voltage capacity detecting unit 33, and a system information output unit 37.
- the server side learning unit 3B is provided with a deterioration rate table learning unit 34, a deterioration rate table storage unit 35, a deterioration rate table calculation unit 36, and a system information approximation degree determination unit 38.
- the learning units 3A and 3B are connected via a network N.
- the system information output unit 37 of the local learning unit 3A sends information related to the storage battery system to the system information approximation degree determination unit 38 of the server learning unit 3B via the network N. It is supposed to send.
- the system information approximation degree determination unit 38 collects information about the storage battery systems sent from the plurality of local learning units 3A, and determines that the storage battery systems are the same or similar based on the mutual information. It is a part to do.
- the information on the storage battery system includes the manufacturer of the storage battery, the production lot, the environmental information of the storage battery system, that is, the latitude and longitude where the storage battery system is located, the temperature and humidity around the system, the current life of the storage battery system, and the like.
- the system information output unit 37 of the local learning unit 3A transmits information about the storage battery system to the server learning unit 3B via the network N (S2-11). .
- the system information approximation degree determination unit 38 receives information related to the storage battery system.
- the system information approximation degree determination unit 38 collects a plurality of pieces of information related to the storage battery system from the plurality of local learning units 3A. Then, the system information approximation degree determination unit 38 determines whether the information regarding the storage battery system is the same or similar (S2-12).
- the system information determination unit 38 determines that the information regarding the storage battery system is the same or similar in the learning step (S2-13, S2-14) of the deterioration rate table T, the deterioration rate table learning is performed.
- the unit 34 can learn efficiently by referring to the learning result of the similar storage battery system. This is because, when the server-side learning unit 3B determines that the information regarding the storage battery system is the same or similar, it is more desirable to compare the information regarding the storage battery system with the same information regarding the deterioration state.
- the following third to sixth embodiments are characterized by the configuration of the usage history data acquisition unit 1.
- the usage history data acquisition unit 1 is one of the three blocks roughly classified in the deterioration control device 10.
- the configuration other than the usage history data acquisition unit 1 is the same as that of the first embodiment, and a description thereof will be omitted.
- a characteristic parameter detection unit 12 is provided in addition to the database unit 11 described in paragraph 0029.
- a first simulation unit 14A and an encoding unit 15 are connected to the characteristic parameter detection unit 12, and a difference detection unit 13 is connected to the first simulation unit 14A.
- the encoding unit 15 is connected to the difference detection unit 13, and the first communication unit 16A, the database unit 11, the second communication unit 16B, and the decoding unit 17 are sequentially connected to the encoding unit 15. Yes.
- An adder 18 and a second simulation unit 14B are connected to the decoding unit 17.
- the adder 18 is connected to the second simulation unit 14B.
- the learning unit 3 is connected to the adding unit 18.
- the characteristic parameter detection unit 12 is a part that detects the characteristic parameter P of the storage battery cell 20 by inputting and analyzing the usage history data A.
- the characteristic parameter P of the storage battery cell 20 is a parameter that characterizes the behavior of the storage battery cell 20, and is generally a time constant, an internal resistance, or the like.
- the internal resistance includes a plurality of parameters such as a direct current internal resistance, an alternating current internal resistance, a charge side internal resistance, and a discharge side internal resistance.
- a method of taking usage history data A from a storage battery cell 20 that is charged / discharged at random and detecting a characteristic parameter P therefrom a method of detecting a characteristic parameter P therefrom using a specific charge / discharge pattern, etc. is there.
- a method is adopted in which usage history data A is captured and a plurality of characteristic parameters P are detected therefrom during maintenance of the storage battery system, refresh charge / discharge performed at regular intervals, or the like.
- the characteristic parameter detection unit 12 is configured to input a detection trigger R from the encoding unit 15 and execute re-detection of the characteristic parameter P when the amount of encoded data described later is equal to or greater than a threshold value. Further, the characteristic parameter detection unit 12 outputs the current value I and the ambient temperature t of the storage battery cell 20 together with the detected characteristic parameter P to the first simulation unit 14A and the encoding unit 15.
- the first simulation unit 14A takes in the current value I and the ambient temperature t of the storage battery cell 20 and the characteristic parameter P from the characteristic parameter detection unit 12, and simulates the deterioration state of the storage battery cell 20 using these data. Part.
- the first simulation unit 14A calculates a first simulation value M1 including a voltage value, a temperature value, an SOC value, and the like as a simulation result.
- the difference detection unit 13 acquires the first simulation value M1 from the first simulation unit 14A, further acquires the use history data A from the storage battery cell 20, detects the difference value Q between them, It is configured to output to the encoding unit 15.
- the encoding unit 15 receives the current value I, the ambient temperature t, and the characteristic parameter P from the characteristic parameter detection unit 12, and the difference value Q from the difference detection unit 13, respectively. After these are encoded, The data is output to the first communication unit 16A.
- the encoding unit 15 outputs a detection trigger R to the characteristic parameter detection unit 12 when the amount of encoded data becomes equal to or greater than a threshold value. Further, the encoding unit 15 does not output when the data of the characteristic parameter P is the same as the previous time.
- the encoding unit 15 performs entropy encoding such as Huffman encoding when encoding the difference value Q.
- entropy encoding such as Huffman encoding when encoding the difference value Q.
- a short code length is assigned to a small value. That is, the encoding unit 15 assigns a short code length when the first simulation value M1 is close to an actual measurement value such as the usage history data A.
- the codeword table used in the encoding unit 15 is also used in the decoding unit 17.
- the encoding unit 15 does not use entropy encoding but simply performs binary conversion of the value.
- the first communication unit 16A is a means for sending the data encoded by the encoding unit 15 to the database unit 11.
- the database unit 11 stores the encoded data sent from the first communication unit 16A.
- the second communication unit 16B is means for reading data from the database unit 11 and sending it to the decryption unit 17.
- the decoding unit 17 decodes the characteristic parameter P in the encoded data state, the current value I of the storage battery cell 20 and the ambient temperature t, and further the difference value Q, and outputs them to the adding unit 18.
- the second simulation unit 14B takes in the characteristic parameter P, the current value I of the storage battery cell 20 and the ambient temperature t from the decoding unit 17, simulates the deterioration state of the storage battery cell 20, and adds the second simulation value M2. This is a part to be output to the unit 18.
- the adding unit 18 takes in the second simulation value M2 and the decoded difference value Q, adds them together, and outputs them to the learning instruction unit 2 and the learning unit 3.
- the characteristic parameter detector 12 detects the characteristic parameter P as follows. First, the characteristic parameter detection unit 12 inputs the usage history data A such as the voltage value, temperature value, current value, SOC value, etc. of the battery cell from the outside, and analyzes these to determine the characteristic parameter P of the storage battery cell 20. Detect (S3-2).
- the first simulation unit 14A receives the characteristic parameter P, the current value I of the storage battery cell 20 and the ambient temperature t, and simulates the deterioration state of the storage battery cell 20 using these data.
- First simulation values M1 such as values, temperature values, SOC values, etc. are calculated.
- the difference detection unit 13 detects the difference between the usage history data A and the first simulation value M1, and sends the difference value Q to the encoding unit 15 (S3-4).
- the encoding unit 15 encodes the characteristic parameter P, the current value I of the storage battery cell 20, and the ambient temperature t (S3-5). However, the encoding unit 15 does not output the data of the characteristic parameter P when it is the same as the previous time in order to reduce the amount of encoded data. Also, the encoding unit 15 entropy encodes the difference value Q (S3-6).
- the encoding unit 15 determines whether the amount of encoded data is equal to or greater than a threshold (S3-7). An increase in the amount of encoded data in the encoding unit 15 indicates that the simulation result is different from the actual measurement value, that is, the value of the characteristic parameter P tends to be different from the actual value. Therefore, if the amount of encoded data is equal to or greater than the threshold (Yes in S3-7), the encoding unit 15 indicates that the simulation result is different from the actual measurement value, and outputs the detection trigger R to output the characteristic parameter detection unit. 12, re-detection of the characteristic parameter P is executed (S3-8). Thereafter, the database unit 11 records the encoded data from the encoding unit 15 (S3-9). If the encoded data amount is less than the threshold (No in S3-7), the encoding unit 15 records the encoded data without outputting the detection trigger R (S3-9).
- the second communication unit 16B outputs the data read from the database unit 11 to the decryption unit 17.
- the decoding unit 17 uses the codeword table used by the encoding unit 15 to decode the data output from the second communication unit 16B.
- the decryption unit 17 outputs the characteristic parameter P, current I, and ambient temperature t to the second simulation unit 14A, and outputs the difference value Q to the addition unit 18 (S3-10).
- the second simulation unit 14B as in the first simulation unit 14A, the current value I and the ambient temperature t are input, the deterioration state of the storage battery cell 20 is simulated using the characteristic parameter P, and the voltage value and the temperature value are simulated.
- the second simulation value M2 such as the SOC value is output to the adding unit 18 (S3-11).
- the addition unit 18 adds the simulation value M2 output from the second simulation unit 14B and the difference value Q output from the decoding unit 17 to detect a final value.
- This final value is the same as the usage history data A, and this value is output to the learning instruction unit 2 and the learning unit 3 as the usage history data A (S3-12).
- the encoding unit 15 transmits the current value I and the atmosphere. Only the temperature t and each difference value Q are used, and entropy coding is used. Therefore, the amount of data to be encoded becomes very small. In addition, since the encoding unit 15 assigns a short code length to a small value and does not output when the data of the characteristic parameter P is the same as the previous time, the data amount can be reduced.
- the encoding unit 15 when the amount of data to be encoded exceeds a threshold value and the simulation result dissociates from the actual measurement value, the encoding unit 15 outputs a detection trigger R to the characteristic parameter detection unit 12. Then, the characteristic parameter P is detected again. Further, the characteristic parameter detection unit 12 of the third embodiment detects a plurality of characteristic parameters P. According to such 3rd Embodiment, the data which reflected the behavior of the storage battery cell 20 correctly can be taken in.
- the third embodiment when the first simulation value M1 becomes larger than a predetermined value, the encoding process for the characteristic parameter P is simply changed from entropy encoding to binary conversion. According to the third embodiment, it is possible to avoid a significant increase in code length and maintain a small amount of data.
- the fourth embodiment is a modification of the usage history data unit 1 in the third embodiment, and the basic configuration is the same as that of the third embodiment. Therefore, the same components as those in the third embodiment are denoted by the same reference numerals, and the description thereof is omitted.
- the difference detection unit 13 is connected to a quantization unit 19A
- the decoding unit 17 is connected to an inverse quantization unit 19B.
- the quantization unit 19A is a part that quantizes a difference value before encoding and calculates a quantization difference value
- the inverse quantization unit 19B is a part that performs inverse quantization on the decoded quantization difference value. is there.
- Quantization is to divide by a certain value, and inverse quantization is to multiply by a certain value that is the opposite. As a result, the value is rounded. For example, when quantization is performed with 5, the values of 1, 2, 3, and 4 are 0, and remain 0 even when inverse decoding is performed.
- the quantization unit 19A quantizes the difference value before encoding (S4-5), calculates the quantized difference value, and outputs this to the encoding unit 15 To do.
- the encoding unit 15 performs entropy encoding on the quantized difference value (S4-7). Also.
- the decoding unit 17 decodes the quantized difference value and outputs it to the inverse quantization unit 19B.
- the inverse quantization unit 19B performs inverse quantization to detect a difference value (S4-11).
- the local use history data acquisition unit 1A includes a characteristic parameter detection unit 12, a difference detection unit 13, a first simulation unit 14A, an encoding unit 15, a first communication unit 16A, and a quantization unit 19A. Yes.
- the first communication unit 16A transmits data to a remote server by wired or wireless communication.
- the server-side usage history data acquisition unit 1B includes a database unit 11, a second communication unit 16B, a decoding unit 17, a second simulation unit 14B, an addition unit 18, and an inverse quantization unit 19B. Yes.
- the sixth embodiment shown in FIG. 17 is a modification of the fifth embodiment, in which the characteristic parameter detection unit 12 of the usage history data acquisition unit 1A provided on the local side has learned from the learning unit 3. There is a feature in that it is configured to take in the deterioration rate table T.
- the data processing of the sixth embodiment is shown in the flowchart of FIG.
- the data processing of the sixth embodiment is basically the same as that of the third embodiment.
- the characteristic parameter detection unit 12 obtains usage history data on the server side. It is determined whether or not the characteristic parameter P is transmitted from the unit 1B, and when the characteristic parameter P is received from the server-side usage history data acquiring unit 1B (S5-3), the characteristic parameter P in the local-side usage history data acquiring unit 1A is received. (S5-4). Thereafter, the replaced characteristic parameter P is used.
- Steps S5-5 to S5-14 in FIG. 18 are the same as steps S3-3 to S3-12 in FIG.
- the seventh to ninth embodiments are characterized by the configuration of the learning instruction unit 2.
- the learning instruction unit 2 is one of the three blocks roughly classified in the deterioration control device 10.
- the configuration other than the learning instruction unit 2 is the same as that of the first embodiment, and a description thereof will be omitted.
- the learning instruction unit 2 according to the seventh embodiment includes a database unit 22 and an estimated value use unit in addition to the determination unit 21 and the estimated value calculation unit 23 described in paragraphs 0030 and 0031.
- a difference detection unit 24, a variation amount detection unit 25, a variation amount display unit 26, a deterioration amount display unit 27, a learning display unit 28, and a storage unit 29 are provided.
- the storage unit 29 is a part for storing the time series data Z.
- the time-series data Z is data in which the difference value Q detected by the estimated value difference detection unit 24 and the variation amount W detected by the variation amount detection unit 25 are set.
- the determination unit 21 analyzes the time-series data Z and determines that the trend of the time-series data Z is non-linear, thereby outputting a learning instruction signal E to the learning unit 3.
- the determination unit 21 is a part that determines whether or not the deterioration rate table T can be learned, as described above.
- the actual value change due to the variation in the deterioration state of the storage battery cell 20 is determined. Whether or not learning is possible is determined while removing the influence of. That is, the determination unit 21 determines whether or not the correlation between the difference value Q and the variation amount W is nonlinear, thereby eliminating the influence of the change in the usage history data A due to the variation amount W. Yes.
- the determination unit 21 is configured to be able to adjust the distribution width ⁇ that serves as a reference for determining whether or not the trend of the time series data Z is nonlinear.
- the non-linear determination of the distribution width ⁇ of the time-series data Z in the determination unit 21 will be described with reference to FIG.
- the time series data Z shows a certain tendency.
- the numbers 1 to 6 indicated by ⁇ are the numbers of the time series data Z.
- the determination unit 21 determines the value of the distribution width ⁇ from the tendency indicated by the time series data Z.
- the value of the distribution width ⁇ is a coefficient necessary for the determination unit 21A to determine that the time series data Z is distributed linearly.
- the value of the distribution width ⁇ is small, the correlation between the variation amount W and the difference value Q is strong.
- the determination unit 21 can easily determine that the trend of the time series data Z is non-linear. In other words, it becomes difficult for the determination unit 21 to determine that the trend of the time-series data Z is linear, and as a result, the learning instruction signal E is frequently output to the learning unit 3.
- the determination unit 21 determines that the trend of the time-series data Z is non-linear because the time-series data Z does not easily protrude from the distribution width ⁇ . In other words, the determination unit 21 can easily determine that the trend of the time-series data Z is non-linear, and the output frequency of the learning instruction signal E to the learning unit 3 decreases.
- the determination unit 21 controls the learning speed according to the output frequency of the learning instruction signal and the deterioration accuracy by adjusting the value of the distribution width ⁇ .
- the determination unit 21 adjusts the value of the distribution width ⁇ according to the situation where the storage battery system is placed.
- the distribution width ⁇ is set by time series numbers 1 to 5. This is an example in which the distribution width ⁇ is set in advance so as to be based on the variation amount W of the first five points.
- the distribution width ⁇ can be changed according to the type and characteristics of the storage battery cell 20.
- the data of time series number 6 is out of the linear region.
- the determination unit 21 issues a learning instruction signal E to the learning unit 3.
- the learning unit 3 learns the deterioration rate table T, the estimated value of the capacity deterioration amount in the storage battery cell 20 so that the difference value Q at the time series number 6 falls within the center value in the linear region.
- Correct ⁇ . This ⁇ is the distance between the median value of the distribution width ⁇ determined to be linear and the difference value Q.
- the estimated value calculation unit 23 is a part that calculates an estimated value ⁇ of the capacity deterioration amount of the storage battery cell 20.
- the calculated estimated value of the capacity deterioration amount is sent to the learning unit 3 and is called the change amount D because it is used to update the deterioration rate table T.
- the estimated value ⁇ it is referred to as it is.
- the estimated value difference detecting unit 24 is a part that detects a difference value Q between the estimated value ⁇ calculated by the estimated value calculating unit 23 and the actually measured value that is the usage history data A of the storage battery cell 20.
- the actual measurement value of the storage battery cell 20 is a capacity value, for example, the capacity value from a fully discharged state to a fully charged state is the actual capacity value. Alternatively, a capacity value from a fully charged state to a completely discharged state may be used as the actually measured capacity value.
- the estimated value difference detection unit 24 inputs such a measured capacity value as the measured capacity B from the learning unit 3.
- the measured capacity value may be either the measured capacity value until full charge or the measured capacity value until complete discharge, but we considered minimizing the amount of degradation when measuring the capacity value. In this case, it is desirable to measure the measured capacity value by charging and the measured capacity value by discharging at least once, calculate the respective deterioration amounts, and adopt the smaller one as the measured capacity value.
- the variation amount detection unit 25 is a portion that detects the variation amount W when the deterioration state of the storage battery cell 20 is viewed from the entire storage battery system, and the variation amount display unit 26 displays the variation amount W on a monitor such as a display. Part. Examples of the variation amount W include the following.
- the database unit 22 is a part that is stored as use history data A and converted into a database.
- the deterioration amount display unit 27 is a part that displays the deterioration amount calculated by the estimated value calculation unit 23 on a monitor such as a display.
- the learning display unit 28 is a part that displays the learning process to the outside using an LED or the like.
- the variation amount detection unit 25 inputs an actual measurement value in the storage battery cell 20 and an actual measurement value of the entire storage battery system from the outside of the learning instruction unit 2, and detects the variation amount W of the storage battery cell 20 from these actual measurement values (S6). -1).
- the variation amount display unit 26 displays the variation amount W on the monitor (S6-2). At this time, the variation amount W indicates the variation amount of the entire storage battery system with respect to a certain deterioration model parameter C.
- the database unit 27 stores the actual measurement value of the storage battery cell 20 as usage history data A (S6-3).
- the usage history data A is an actual measurement value of each storage battery cell 20 in use, and includes a voltage value, a temperature value, a current value, an SOC value, and the like.
- the estimated value calculation unit 23 reads the deterioration model parameter C from the learning unit 3, reads the usage history data A from the database unit 27, and estimates the amount of deterioration per storage battery cell 20 using these data and an arbitrary function.
- the value ⁇ is calculated (S6-4).
- the deterioration amount display unit 27 displays the estimated value ⁇ of the deterioration amount on the monitor (S6-5).
- the estimated value difference detecting unit 24 detects a difference value Q between the estimated value ⁇ of the deterioration amount calculated by the estimated value calculating unit 23 and the actually measured value of the entire storage battery system (S6-6).
- the storage unit 29 stores the difference value Q and the variation amount W as time series data Z as a set (S6-7).
- the determination unit 21 uses this time-series data Z to determine what amount of variation W is present and the actual measurement value of the entire storage battery system has been measured.
- the actual measured value of the entire storage battery system is not only influenced by the deterioration state of each storage battery cell 20, but also the actual measured value of the entire storage battery system 20 changes due to variations in the deterioration state of the storage battery cell 20 in principle. .
- the change in the measured capacity value due to the variation of the storage battery cells 20 is not an irreversible change, but only temporary. Therefore, if the influence of the change in the capacity of the entire storage battery system due to the variation in the deterioration state of the storage battery cell 20 is removed, the remaining change 2 is the difference between the true estimated value and the actual measurement value.
- the determination unit 21 analyzes the trend of the time series data Z that is a set of the difference value Q and the variation amount W (S6-8). Then, the determination unit 21 determines whether or not the trend of the time series data Z is linear (S6-9), and if the determination unit 21 determines that the trend of the time series data Z is linear (S6-9). Yes), the value of the distribution width ⁇ is reset from the trend of the time series data Z (S6-10).
- the determination unit 21 uses the learning instruction signal E and the estimated capacity deterioration amount ⁇ (change amount D in the first embodiment). The data is output to the learning unit 3 (S6-11).
- the determination unit 21A outputs the learning instruction signal E, the learning display unit 28 externally displays that the learning process is to be performed (S6-12).
- the determination unit 21 outputs the learning instruction signal E to the learning unit 3 after removing the influence of the actual measurement value change due to the variation in the deterioration state of the storage battery cell 20.
- the learning unit 3 can learn the deterioration model with high accuracy. Therefore, even when the capacity actually measured in the entire storage battery system is used, there is no influence due to variations in the deterioration state of the storage battery cells 20, and an excellent deterioration model can be constructed.
- the determination unit 21 performs an experiment in advance to determine the threshold value ⁇ , and when the difference value Q detected by the estimated value difference detection unit 24 is greater than or equal to the threshold value ⁇ , the learning instruction signal E It may also be put out. Thereby, the determination part 21 can output the learning instruction signal E efficiently.
- display units such as the variation amount display unit 26, the deterioration amount display unit 27, and the learning display unit 28 display data and processing to the outside, work efficiency associated with deterioration control is improved. To do.
- the local learning instruction unit 2A includes a database unit 22, an estimated value calculation unit 23, a variation amount detection unit 25, a variation amount display unit 26, a deterioration amount display unit 27, and a learning display unit 28. It has been.
- the learning instruction unit 2B on the server side includes a determination unit 21, an estimated value difference detection unit 24, and a storage unit 29. These learning instruction units 2A and 2B are connected via wired or wireless communication means (not shown).
- the variation amount W and the capacity deterioration amount estimated value ⁇ are transmitted as a set to the server side learning instruction unit 2B by wired or wireless communication means.
- a learning instruction signal E and an estimated value change amount D are set as a set from the server-side learning instruction unit 2B, and transmitted to the local learning unit 3A by wired or wireless communication means.
- the eighth embodiment when analyzing the correlation between the variation amount W and the difference value Q, not only the result of one storage battery system but also other storage battery systems having similar configurations are used. It becomes possible to analyze the trends of the two together. Therefore, the number of data sets increases, and it becomes possible to set the distribution width ⁇ and the estimated value ⁇ of the capacity deterioration amount more suitable for the storage battery system. Thereby, the precision of deterioration control improves further.
- a storage battery system that is likely to vary and a system that is not so mixed may be mixed. That is, in a storage battery system that tends to have a small variation amount W, the reliability of the data set of the variation amount W and the difference value Q is set high, and in a system that tends to have a large variation amount W, the reliability may be set low. Is possible.
- the distribution width ⁇ and the estimated value ⁇ can be set by preferentially using a data set with high reliability, and the deterioration rate table T can be efficiently learned.
- the ninth embodiment also has the learning instruction unit 2 divided and arranged on the local side and the server side, and the basic configuration is the same as that of the seventh embodiment. Therefore, the same components as those in the seventh embodiment are denoted by the same reference numerals, and the description thereof is omitted.
- the learning instruction unit 2 of the ninth embodiment the variation amount detection unit 25, the variation amount display unit 26, the deterioration amount display unit 27, the learning display unit 28, and the local display side are provided. These components are provided on the remote server side.
- the specific contents and numerical values of the information are free and are not limited to specific contents and numerical values.
- the determination of the magnitude of the threshold value, the determination of coincidence / non-coincidence, etc. it is determined whether to include a value as below, or not to include a value as larger, smaller, exceeding or not exceeding. is there. Therefore, for example, depending on the setting of values, even if “more than” is read as “greater than” and “less than” is read as “less than”, it is substantially the same.
- deterioration amount display unit 28 ... learning display unit 29 ... Storage unit 30 ... Charge / discharge control device 31 ... Current integration unit 32 ... Constant current capacity detection unit 33 ... Constant voltage capacity detection unit 34 ... Degradation rate table learning unit 35 ... Degradation rate table storage unit 36 ... Degradation rate table Calculation unit A ... Usage history data B ... Actual capacity C ... Degradation model parameter D ... Change amount E ... Learning instruction signal F ... Charge / discharge commands G1, G2 ... Detection signal P ... Characteristic parameter Q ... Difference value T ... Degradation speed table T1 ... capacitance value table T2 ... internal resistance value table W ... variation amount Z ... time-series data ⁇ ... estimated value of capacity deterioration amount
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Chemical & Material Sciences (AREA)
- Electrochemistry (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Chemical & Material Sciences (AREA)
- Manufacturing & Machinery (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Power Engineering (AREA)
- General Health & Medical Sciences (AREA)
- Automation & Control Theory (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Secondary Cells (AREA)
- Charge And Discharge Circuits For Batteries Or The Like (AREA)
- Tests Of Electric Status Of Batteries (AREA)
Abstract
Description
(課題1)蓄電池システムの用途の多様化への対応
近年、蓄電池システムは、一般家庭やビルもしくは変電所等、さまざまな場所で使用されている。このため、蓄電池システムの用途により、使用場面での充放電電流値が異なっており、用途ごとに使用できる容量が変化する。
蓄電池システムでは蓄電池セルの使用履歴データをメモリに保存するが、1つの蓄電池システムに具備される蓄電池セルは数千~数万個という数にのぼる。さらに、多数の蓄電池システムを1つのサーバで一括して管理するとなると、サーバ側に設けられるメモリは、膨大な量の使用履歴データを保存しなくてはならない。したがって、使用履歴データのデータ量は非常に大きくなり、これらを保存するためのメモリは負荷が増大していた。
蓄電池システムにおいて蓄電池セルの使用履歴データを取得するタイミングは、システムのメンテナンスを行う時や、一定間隔で実行されるリフレッシュ充放電を実施する時が多い。
(a)前記蓄電池セルの使用履歴データを取得する使用履歴データ取得部。
(b)前記使用履歴データに基づき前記蓄電池セルの劣化状態を示す劣化モデルパラメータのデータ量を更新して当該劣化モデルパラメータの学習を行い、学習結果として劣化速度テーブルを出力する学習部。
(c)前記使用履歴データと前記推定値を比較し前記劣化モデルパラメータの学習の可否を判断して学習指示信号を前記学習部に出力すると共に、前記劣化モデルパラメータの変更量を算出して当該変更量を前記学習部に出力する学習指示部。
また、蓄電池システムの劣化制御方法も本発明の実施形態の一態様である。
図1を用いて、本発明に係る実施形態の全体構成について説明する。蓄電池システムには、複数の蓄電池セル20が設けられている。蓄電池セル20は劣化制御装置10に使用履歴データAを出力するようになっている。使用履歴データAとは、蓄電池セル20の電圧値、温度値、電流値、SOC値等のデータである。
劣化制御装置10は、蓄電池セル20から使用履歴データAを入力し、劣化速度テーブルTを作成して蓄電池セル20の劣化状態を推定する装置である。劣化制御装置10は、蓄電池セル20の劣化状態の推定結果を、充放電制御装置30に出力するようになっている。劣化制御装置10には3つの大きなブロックから構成されている。3つのブロックとは、使用履歴データ取得部1、学習指示部2、学習部3である。以下、使用履歴データ取得部1、学習指示部2、学習部3の概要について説明する。
使用履歴データ取得部1は、蓄電池セル20から使用履歴データAを入力し、使用履歴データAを学習指示部2及び学習部3に出力する部分である。使用履歴データ取得部1にはデータベース部11が設置されている。データベース部11は、使用履歴データAを可逆圧縮して保存するように構成されている。使用履歴データ取得部1はデータベース部11に保存した使用履歴データAの圧縮状態を復元してからこれを読み出すようになっている。
学習指示部2には、判断部21と、推定値算出部23とが設けられている。判断部21は、使用履歴データ取得部1から使用履歴データAを読み出すと共に、学習部3から実測容量B及び劣化モデルパラメータCを読み出し、両者を比較して、劣化速度テーブルTの学習を行うか否かを判断して学習指示信号Eを出力する部分である。なお、劣化モデルパラメータCとは、段落0004でも述べたように、蓄電池セル20の劣化状態を示す要素であり、温度値やSOC値等である。また、実測容量Bについては後段の段落0035で、劣化速度テーブルTについては後段の段落0033で述べることにする。さらに、判断部21は、劣化速度テーブルTの学習を行うと判断すると、学習部3に学習指示信号Eを出力するようになっている。
学習部3は、学習指示部2に対し実測容量B及び劣化モデルパラメータCを送るようになっている。また、学習部3は、学習指示部2から劣化モデルパラメータCの変更量D及び学習指示信号Eを受け取り、劣化モデルパラメータCのデータ量を更新して当該劣化モデルパラメータの学習を行い、劣化速度テーブルTを充放電制御装置30に出力するようになっている。
[構成]
第1の実施形態は、劣化制御装置10における3つのブロックのうち、学習部3の構成に特徴がある。図3に示すように、学習部3には、電流積算部31と、一定電流容量検出部32と、一定電圧容量検出部33と、劣化速度テーブル学習部34が設けられている。このうち電流積算部31は、使用履歴データ取得部1からの使用履歴データA及び前記検出部32、33からの検出信号G1、G2に従って実測容量Bを算出し、実測容量Bを学習指示部2に出力する部分である。
図8のフローチャートを参照して、学習部3による劣化モデルの学習処理について説明する。電流積算部31は使用履歴データ取得部1から使用履歴データAを入力する。一定電流容量検出部32は、一定電流による充電が開始されたか否かを判定する(S1-1)。一定電流容量検出部32が、一定電流による充電が開始されたことを検出すると、検出信号G1を電流積算部31に出力する(S1-1のYes)。この検出信号G1を受けて、電流積算部31は電流の積算を開始する(S1-2)。一定電流容量検出部32は、一定電流による充電の開始を検出しなければ(S1-1のNo)、一定電流による充電が開始されたか否かの判定処理を繰り返す。
蓄電池システムの規模等に応じて充放電電流値の大きさは異なるが、低い充放電電流値によるゆっくりした充放電であっても、高い充放電電流値による急速な充放電であっても、同一の劣化速度テーブルTを利用して蓄電池システムの状態を推定しなくてはならないとすると、劣化制御の精度を高くすることは望めない。
[構成]
第2の実施形態は上記第1の実施形態と同じく学習部3に特徴があり、基本的な構成は第1の実施形態と同一である。そのため、上記第1の実施形態と同一の構成要素に関しては同一符号を付して、説明は省略する。上記第1の実施形態における学習部3は全ての構成要素がローカル側に設けられていた。
図10のフローチャートを用いて、第2の実施形態の学習処理について説明する。第2の実施形態の学習処理では、前述の図8に示したフローチャートに対し、S2-11、S2-12を追加した点に特徴がある。このため、図10のS2-1からS2-10までは、図8のS1-1からS1-10までと同様であり、図10のS2-13からS2-19までは、図8のS1-11からS1-17までと同様である。
以上のような第2の実施形態によれば、ローカル側単体の蓄電池システムでは学習できない内容であっても、サーバ側の学習部3Bにて蓄電池システムに関する情報を集め、複数の蓄電池システムにおける相互の近似度を判定して把握することができる。このため、近似度の高い他の蓄電池システムの学習結果を、参照することができる。その結果、サーバ側の学習部3Bでは、劣化速度テーブル学習部34が多様な劣化速度テーブルTを学習して、学習速度が速くなり、推定精度を高めて劣化制御性能がいっそう向上する。
以下の第3~第6の実施形態では、使用履歴データ取得部1の構成に特徴がある。使用履歴データ取得部1とは、劣化制御装置10において大別された3つのブロックのうちの1つである。第3~第6の実施形態においては、使用履歴データ取得部1以外の構成については第1の実施形態と同一であり、説明は省略する。
図11に示すように第3の実施形態に係る使用履歴データ取得部1では、段落0029にて述べたデータベース部11に加えて、特性パラメータ検出部12が設けられている。特性パラメータ検出部12には第1のシミュレーション部14A及び符号化部15が接続され、第1のシミュレーション部14Aには差分検出部13が接続されている。
図13のフローチャートを参照して、第3の実施形態におけるデータ処理について説明する。まず、データベース部11へのデータ書き込み処理について述べる。最初に、特性パラメータ検出部12は検出トリガRがあるか否かを判定する(S3-1)。
以上のような第3の実施形態によれば、第1及び第2のシミュレーション値M1、M2と実測値とが近い値の場合に、符号化部15は、送信するデータは電流値Iと雰囲気温度tと各差分値Qのみとなり、エントロピー符号化を用いている。そのため、符号化するデータ量は非常に小さくなる。しかも、符号化部15は、値の小さいものに短い符号長を割り当てており、特性パラメータPのデータが前回と同じ場合には出力しないので、データ量を削減することができる。
[構成]
第4の実施形態は上記第3の実施形態における使用履歴データ部1の変形例であり、基本的な構成は上記第3の実施形態と同一である。そのため、第3の実施形態と同一の構成要素に関しては同一符号を付して、説明は省略する。
第4の実施形態のデータ処理について図15のフローチャートを用いて説明する。本実施形態のデータ処理は、基本的に図13に示した上記第3の実施形態のデータ処理と同様であるが、差分値Qの量子化及び逆量子化を行う(S4-5、S4-12)点に特徴がある。これ以外のステップは、図13に示したステップと同様である。
以上のような第4の実施形態によれば、量子化部19Aによって、符号化する前の差分値を量子化するので、データベース部11への書き込み情報を欠落させて、データ量を削減することが可能である。一方で、第4の実施形態においては、符号化の種類が減らせるため、符号長も激減させることができる。この実施形態では不可逆符号化なので元の値を完全に再現することはできないが、その分、使用履歴データAのデータ量を激減させることができるといったメリットがある。
[構成]
図16に示すように、第5の実施形態では、使用履歴データ取得部1の構成要素がローカル側の使用履歴データ取得部1Aとサーバ側の使用履歴データ取得部1Bとに分割配置されている。ローカル側の使用履歴データ取得部1Aには、特性パラメータ検出部12、差分検出部13、第1のシミュレーション部14A、符号化部15、第1の通信部16A及び量子化部19Aが配置されている。
以上のような第5の実施形態によれば、サーバ側の使用履歴データ取得部1Bにデータベース部11を設けたので、このデータベース部11に、サーバ側の大規模なデータサーバを使用することができ、大きな使用履歴データAを記録することができる。したがって、サーバ側で多くの蓄電池システムの劣化診断を統計的に処理する際などで、ローカル側とサーバ側の間で通信する使用履歴データAを大幅に削減することが可能である。
[構成]
図17に示す第6の実施形態は、前記第5の実施形態の変形例であって、ローカル側に設けた使用履歴データ取得部1Aの特性パラメータ検出部12が、学習部3からの学習結果である劣化速度テーブルTを取り込むように構成した点に特徴がある。
第6の本実施形態のデータ処理を図18のフローチャートに示す。第6の実施形態のデータ処理は、基本的に上記第3の実施形態と同様であるが、図18のS5-3に示すように、特性パラメータ検出部12では、サーバ側の使用履歴データ取得部1Bから特性パラメータPが送信されてきたかを判定し、サーバ側の使用履歴データ取得部1Bから特性パラメータPが受信すると(S5-3)、ローカル側の使用履歴データ取得部1Aにおける特性パラメータPと置き換える(S5-4)。それ以降、この置き換えた特性パラメータPを用いる。図18のS5-5からS5-14までは、図13のS3-3からS3-12と同様である。
このような第6の実施形態によれば、ローカル側の使用履歴データ取得部1Aにおける特性パラメータPの値として、ローカル側で検出したものを使うだけでなく、外部であるサーバ側の使用履歴データ取得部1Bから送信されてきた特性パラメータPの値を使用することができる。したがって、ローカル側の使用履歴データAのデータ量を大幅に削減することが可能である。
第7~第9の実施形態は、学習指示部2の構成に特徴がある。学習指示部2は劣化制御装置10において大別された3つのブロックのうちの1つである。第7~第9の実施形態では、学習指示部2以外の構成については第1の実施形態と同一であり、説明は省略する。
[構成]
図19に示すように、第7の実施形態に係る学習指示部2には、段落0030、0031にて述べた判断部21と推定値算出部23に加えて、データベース部22と、推定値用差分検出部24と、ばらつき量検出部25と、ばらつき量表示部26と、劣化量表示部27と、学習表示部28と、記憶部29とが設けられている。
2.蓄電池セル間の最大SOCと最小SOC値との差
3.蓄電池セル間の最大温度値と最小温度値との差
4.以上のいずれかの差に依存する値
5.蓄電池システムにおける電圧値の分布を示す標準偏差
6.蓄電池システムにおけるSOC値の分布を示す標準偏差
7.蓄電池システムにおける温度値の分布を示す標準偏差
図21のフローチャートを参照して、第7の実施形態におけるデータ処理について説明する。ばらつき量検出部25は、学習指示部2の外部から、蓄電池セル20における実測値と、蓄電池システム全体の実測値を入力し、これらの実測値から蓄電池セル20のばらつき量Wを検出する(S6-1)。ばらつき量表示部26はばらつき量Wをモニタに表示する(S6-2)。このとき、ばらつき量Wは、ある劣化モデルパラメータCに関して蓄電池システム全体のばらつき量を示している。
以上のような第7の実施形態によれば、判断部21は、蓄電池セル20の劣化状態のばらつきによる実測値変化の影響を取り除いた上で、学習指示信号Eを学習部3に出力するため、学習部3は劣化モデルの学習を高い精度で実施することができる。したがって、蓄電池システム全体で実測した容量を使用した場合でも、蓄電池セル20の劣化状態のばらつきによる影響がなく、優れた劣化モデルの構築が可能である。
[構成]
第8の実施形態は基本的な構成は第7の実施形態と同一であるため、上記第7の実施形態と同一の構成要素に関しては同一符号を付して、説明は省略する。前記第7の実施形態の学習指示部2は全ての構成要素がローカル側に設けられていたが、図22に示す第8の実施形態の学習指示部2は、ローカル側の学習指示部2Aと、サーバ側の学習指示部2Bとから構成されている。
第8の実施形態では、ローカル側の学習指示部2Aからは、ばらつき量W、容量劣化量の推定値γがセットで、サーバ側の学習指示部2Bに有線もしくは無線の通信手段により送信され、サーバ側の学習指示部2Bからは学習指示信号Eと推定値の変更量Dとがセットになって、有線もしくは無線の通信手段によりローカル側の学習部3Aに送信される。
[構成]
第9の実施形態も第8の実施形態と同様、上記学習指示部2をローカル側とサーバ側とに分割配置したものであり、基本的な構成は第7の実施形態と同一である。そのため、上記第7の実施形態と同一の構成要素に関しては同一符号を付して、説明は省略する。図23に示すように、第9の実施形態の学習指示部2では、ばらつき量検出部25、ばらつき量表示部26、劣化量表示部27及び学習表示部28、ローカル側に具備され、それ以外の構成要素は遠隔にあるサーバ側に具備される構成である。
第9の実施形態では、上記第8の実施形態の作用及び効果に加えて、データベース部22がサーバ側にあるので、ローカル側での使用履歴データA等を保存するためのメモリ量を削減することができるといった独自の作用及び効果がある。
なお、上記の実施形態は、本明細書において一例として提示したものであって、発明の範囲を限定することを意図するものではなく、その他の様々な形態で実施されることが可能である。また、発明の範囲を逸脱しない範囲で、種々の省略や置き換え、変更を行うことも可能である。これらの実施形態やその変形例は、発明の範囲や要旨に含まれると同様に、請求の範囲に記載された発明とその均等の範囲に含まれるものである。
2、2A、2B…学習指示部
3、3A、3B…学習部
10…劣化制御装置
11、22…データベース部
12…特性パラメータ検出部
13…差分検出部
14A…第1のシミュレーション部
14B…第2のシミュレーション部
15…符号化部
16A…第1の通信部
16B…第2の通信部
17…復号化部
18…加算部
19A…量子化部
19B…逆量子化部
20…蓄電池セル
21…判断部
23…推定値算出部
24…推定値用差分検出部
25…ばらつき量検出部
26…ばらつき量表示部
27…劣化量表示部
28…学習表示部
29…記憶部
30…充放電制御装置
31…電流積算部
32…一定電流容量検出部
33…一定電圧容量検出部
34…劣化速度テーブル学習部
35…劣化速度テーブル記憶部
36…劣化速度テーブル算出部
A…使用履歴データ
B…実測容量
C…劣化モデルパラメータ
D…変更量
E…学習指示信号
F…充放電指令
G1、G2…検出信号
P…特性パラメータ
Q…差分値
T…劣化速度テーブル
T1…容量値テーブル
T2…内部抵抗値テーブル
W…ばらつき量
Z…時系列データ
γ…容量劣化量の推定値
Claims (28)
- 複数の蓄電池セルを有する蓄電池システムの劣化制御装置において、
前記蓄電池セルの使用履歴データを取得する使用履歴データ取得部と、
前記使用履歴データに基づき前記蓄電池セルの劣化状態を示す劣化モデルパラメータのデータ量を更新して当該劣化モデルパラメータの学習を行い、学習結果として劣化速度テーブルを出力する学習部と、
前記使用履歴データと前記推定値を比較し前記劣化モデルパラメータの学習の可否を判断して学習指示信号を前記学習部に出力すると共に、前記蓄電池セルの劣化モデルパラメータの変更量を算出して当該変更量を前記学習部に出力する学習指示部を備えたことを特徴とする蓄電池システムの劣化制御装置。 - 前記学習部は、
一定電流で特定の電圧になるまで充電もしくは放電した一定電流容量値を算出する一定電流容量値算出部と、
一定電流で蓄電池を充電もしくは放電して一定電圧に達した後、電流値を小さくした一定電流で充電もしくは放電を再び一定電圧に達するまで行い、電流値の漸減と一定電圧に達するまでの充電もしくは放電を繰り返して、電流値が十分に小さくなった時までの容量値を積算して一定電圧容量値を算出する一定電圧容量値算出部と、
前記一定電圧容量値を求めた時の電流値と前記一定電圧容量値を組み合わせた電流値-容量値セットを算出する電流-容量セット算出部を備えたことを特徴とする請求項1に記載の蓄電池システムの劣化制御装置。 - 前記学習部は、前記一定電流容量値を用いた容量値テーブルと、前記一定電圧容量値を用いた内部抵抗値テーブルを算出する劣化速度テーブル算出部を備えたことを特徴とする請求項2に記載の蓄電池システムの劣化制御装置。
- 前記劣化速度テーブル算出部は、予め設定された判断基準に沿ってテーブルの相似性を相互に比較し前記内部抵抗値テーブルを複数のグループに分類することを特徴とする請求項3に記載の蓄電池システムの劣化制御装置。
- 前記劣化速度テーブル算出部は、前記蓄電池セルの残量を判定要因としたSOC方向と、温度を判定要因とした温度方向についてテーブルの相似性を比較することを特徴とする請求項4に記載の蓄電池システムの劣化制御装置。
- 前記劣化速度テーブル算出部は、分類するグループ数を指定可能であることを特徴とする請求項4又は5に記載の蓄電池システムの劣化制御装置。
- 前記劣化速度テーブル算出部は、分類するグループにおいて同一のグループ内に前記内部抵抗テーブルが複数存在するとき、同一のグループに含まれる複数の前記内部抵抗テーブル群の中から1つの前記内部抵抗テーブルを選択することを特徴とする請求項4~6のいずれか1項に記載の蓄電池システムの劣化制御装置。
- ローカルシステムと、サーバと、これらローカルシステム及びサーバ間でデータ通信を行う通信部とからなる蓄電池システムの劣化制御装置であって、
前記使用履歴データ取得部、前記学習部及び前記学習指示部のうちの少なくとも1つは、各部分に含まれる構成要素を、前記ローカルシステム側と前記サーバ側に分割して配置することを特徴とする請求項1~7のいずれか1項に記載の蓄電池システムの劣化制御装置。 - 前記使用履歴データを入力して可逆圧縮して保存し、圧縮したデータを復元してから出力するデータベース部を備えたことを特徴とする請求項1~8のいずれか1項に記載の蓄電池システムの劣化制御装置。
- 前記使用履歴データ取得部は、
前記蓄電池セルの特性パラメータを検出する特性パラメータ検出部と、
前記特性パラメータを用いて前記蓄電池セルの劣化状態をシミュレートして第1のシミュレーション値を算出する第1のシミュレーション部と、
前記蓄電池セルの使用履歴データに基づいて当該蓄電池セルの劣化状態の実測値を算出する実測値算出部と、
前記第1のシミュレーション値と前記実測値との差分値を検出する差分検出部と、
前記差分値及び前記特性パラメータを符号化する符号化部と、
前記符号化部にて符号化されたデータを復号化して前記差分値及び前記特性パラメータを読み出す復号化部と、
前記復号化部で読み出した前記特性パラメータを用いて蓄電池の劣化状態をシミュレートし第2のシミュレーション値を算出する第2のシミュレーション部と、
前記第2のシミュレーション値と復号化した前記差分値とを足し合わせる加算部を備えたことを特徴とする請求項1~9のいずれか1項に記載の蓄電池システムの劣化制御装置。 - 前記特性パラメータを複数備えたことを特徴とする請求項10に記載の蓄電池システムの劣化制御装置。
- 前記符号化部は、符号化量が閾値を超えた際に前記特性パラメータ検出部に前記特性パラメータを再検出させる検出トリガを出力することを特徴する請求項10又は11に記載の蓄電池システムの劣化制御装置。
- 前記符号化部は、前記差分値をエントロピー符号化することを特徴とする請求項10~12のいずれか1項に記載の蓄電池システムの劣化制御装置。
- 前記符号化部は、前記差分値の小さいものに短い符号を割り当ててエントロピー符号化を行うことを特徴とする請求項13に記載の蓄電池システムの劣化制御装置。
- 前記符号化部は、エントロピー符号化を行わなかった場合と比べて符号長が短くなった場合に限りエントロピー符号化を行うことを特徴とする請求項14に記載の蓄電池システムの劣化制御装置。
- 前記符号化部は、前記第1のシミュレーション値が所定値よりも大きくなると、前記特性パラメータに対する符号化処理をエントロピー符号化から単純に値を2値変換に変更することを特徴する請求項15に記載の蓄電池システムの劣化制御装置。
- 前記符号化部は符号化の前に量子化処理を行い、量子化処理時の閾値を符号化して出力し、
前記復号化部は復号時に前記閾値を取り出し、復号化後に逆量子化を行い、値を取り出すことを特徴する請求項16に記載の蓄電池システムの劣化制御装置。 - 前記特性パラメータ検出部とは別に、前記第1のシミュレーション部に対し前記特性パラメータを出力する特性パラメータ出力部を備えたことを特徴する請求項10~17のいずれか1項に記載の蓄電池システムの劣化制御装置。
- 前記学習指示部は、
前記蓄電池セルの劣化状態を蓄電池システム全体から見たときのばらつき量を検出するばらつき量検出部と、
前記蓄電池セルの劣化量の推定値を算出する推定値算出部と、
前記推定値算出部が算出した推定値と、前記蓄電池セルの実測値との差分値を検出する推定値用差分検出部と、
前記差分値と前記ばらつき量とのセットを時系列データとし、当該時系列データの相関性を分析して学習可否を判断する判断部を備えたことを特徴とする請求項1~18のいずれか1項に記載の蓄電池システムの劣化制御装置。 - 前記推定値用差分検出部に用いる前記実測値は、蓄電池システムが完全に放電した状態から完全に充電した状態までの容量、あるいは蓄電池システムが完全に充電した状態から完全に放電した状態までの容量を算出することを特徴とする請求項19に記載の蓄電池システムの劣化制御装置。
- 前記学習指示部は、前記差分値が予め設定された閾値以上であるとき、前記学習指示信号を前記学習部に出力することを特徴する請求項19又は20に記載の蓄電池システムの劣化制御装置。
- 前記判断部は、前記ばらつき量と前記差分値との相関性が線形であるか、あるいは非線形であるかを判断し、前記相関性が非線形であると判断した場合には前記学習指示信号を前記学習部に出力することを特徴とする請求項19~21のいずれか1項に記載の蓄電池システムの劣化制御装置。
- 前記判断部は、前記相関性が予め設定された閾値で示される分布の幅に収まるのであれば、当該相関性が線形であると判断することを特徴とする請求項22に記載の蓄電池システムの劣化制御装置。
- 前記使用履歴データを保存するデータベース部を備え、当該データベース部は、前記学習部が前記劣化推定値の学習を行った後は学習に使用された前記使用履歴データを削除することを特徴とする請求項19~23のいずれか1項に記載の蓄電池システムの劣化制御装置。
- 前記実測値、前記推定値、前記学習部の学習結果、前記差分値、前記ばらつき量、前記時系列データのうち、少なくとも1つを表示する表示部を備えたことを特徴とする請求項19~24のいずれか1項に記載の蓄電池システムの劣化制御装置。
- 蓄電池セルの使用履歴データを取得する使用履歴データ取得ステップと、
前記使用履歴データに基づき前記蓄電池セルの劣化状態を示す劣化モデルパラメータのデータ量を更新して当該劣化モデルパラメータの学習を行い、学習結果として劣化速度テーブルを出力する学習ステップと、
前記使用履歴データと前記推定値を比較し前記劣化モデルパラメータの学習の可否を判断して学習指示信号を出力すると共に、前記劣化モデルパラメータの変更量を算出して当該変更量を出力する学習指示ステップ、を含むことを特徴とする蓄電池システムの劣化制御方法。 - 前記使用履歴データ取得ステップは、
前記蓄電池セルの特性パラメータを検出する特性パラメータ検出ステップと、
前記特性パラメータを用いて前記蓄電池セルの劣化状態をシミュレートして第1のシミュレーション値を算出する第1のシミュレーションステップと、
前記蓄電池セルの使用履歴データに基づいて当該蓄電池セルの劣化状態の実測値を算出する実測値算出ステップと、
前記第1のシミュレーション値と前記実測値との差分値を検出する差分検出ステップと、
前記差分値及び前記特性パラメータを符号化する符号化ステップと、
符号化されたデータを復号化して前記差分値及び前記特性パラメータを読み出す復号化ステップと、
前記特性パラメータを用いて蓄電池の劣化状態をシミュレートし第2のシミュレーション値を算出する第2のシミュレーションステップと、
前記第2のシミュレーション値と復号化した前記差分値とを足し合わせる加算ステップを含むことを特徴とする請求項26記載の蓄電池システムの劣化制御方法。 - 前記学習指示ステップは、
前記蓄電池セルの劣化状態を蓄電池システム全体から見たときのばらつき量を検出するばらつき量検出ステップと、
前記蓄電池セルの劣化量の推定値を算出する推定値算出ステップと、
前記推定値算出ステップにて算出した推定値と、前記蓄電池セルの実測値との差分値を検出する差分値検出ステップと、
前記差分値と前記ばらつき量とのセットを時系列データとし、当該時系列データの相関性を分析して学習可否を判断する判断ステップを含むことを特徴とする請求項26又は27に記載の蓄電池システムの劣化制御方法。
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2016529099A JP6370902B2 (ja) | 2014-06-24 | 2015-02-24 | 蓄電池システムの劣化制御装置及びその方法 |
US15/321,906 US20170294689A1 (en) | 2014-06-24 | 2015-02-24 | Battery-system-deterioration control device, and method thereof |
KR1020177000668A KR101927644B1 (ko) | 2014-06-24 | 2015-02-24 | 축전지 시스템의 열화 제어 장치 및 그 방법 |
EP15811538.6A EP3163669B1 (en) | 2014-06-24 | 2015-02-24 | Degradation control device of storage battery system and method for same |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2014129699 | 2014-06-24 | ||
JP2014-129699 | 2014-06-24 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2015198631A1 true WO2015198631A1 (ja) | 2015-12-30 |
Family
ID=54937734
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2015/055154 WO2015198631A1 (ja) | 2014-06-24 | 2015-02-24 | 蓄電池システムの劣化制御装置及びその方法 |
Country Status (5)
Country | Link |
---|---|
US (1) | US20170294689A1 (ja) |
EP (1) | EP3163669B1 (ja) |
JP (1) | JP6370902B2 (ja) |
KR (1) | KR101927644B1 (ja) |
WO (1) | WO2015198631A1 (ja) |
Cited By (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2018147194A1 (ja) * | 2017-02-07 | 2018-08-16 | 日本電気株式会社 | 蓄電池制御装置、充放電制御方法、及び記録媒体 |
JP2019125482A (ja) * | 2018-01-16 | 2019-07-25 | 株式会社Gsユアサ | 監視装置、監視方法及びコンピュータプログラム |
WO2019142711A1 (ja) * | 2018-01-16 | 2019-07-25 | 株式会社Gsユアサ | 監視装置、監視方法、コンピュータプログラム、劣化判定方法、劣化判定装置、及び劣化判定システム |
WO2020027203A1 (ja) * | 2018-07-31 | 2020-02-06 | 本田技研工業株式会社 | 推定システム、推定装置、推定方法、プログラム、及び記憶媒体 |
JP2020092598A (ja) * | 2020-01-30 | 2020-06-11 | 株式会社東芝 | 充電パターン作成装置、充電制御装置、充電パターン作成方法、プログラム、及び蓄電システム |
WO2020137914A1 (ja) * | 2018-12-28 | 2020-07-02 | 株式会社Gsユアサ | データ処理装置、データ処理方法、及びコンピュータプログラム |
WO2020255557A1 (ja) * | 2019-06-17 | 2020-12-24 | 日置電機株式会社 | 電池劣化診断システム、診断処理装置、測定装置及びプログラム |
US20210091583A1 (en) * | 2018-06-08 | 2021-03-25 | Panasonic Intellectual Property Management Co., Ltd. | Battery management system and battery management method |
US10996282B2 (en) | 2018-03-20 | 2021-05-04 | Gs Yuasa International Ltd. | Abnormality factor determination apparatus, degradation determination apparatus, computer program, degradation determining method, and abnormality factor determining method |
US11527900B2 (en) * | 2019-04-18 | 2022-12-13 | Lg Display Co., Ltd. | Apparatus and method for managing a battery based on degradation determination |
JP7276928B1 (ja) | 2022-03-29 | 2023-05-18 | 東洋システム株式会社 | 二次電池容量推定システム |
CN117388716A (zh) * | 2023-12-11 | 2024-01-12 | 四川长园工程勘察设计有限公司 | 基于时序数据的电池组故障诊断方法、系统及存储介质 |
JP7482088B2 (ja) | 2021-07-27 | 2024-05-13 | 株式会社Tmeic | 電動機巻線劣化状態調査システム |
WO2024166450A1 (ja) * | 2023-02-09 | 2024-08-15 | 株式会社豊田中央研究所 | 情報処理システム、情報処理方法および情報処理プログラム |
JP7596244B2 (ja) | 2021-09-22 | 2024-12-09 | 株式会社日立製作所 | 電池診断装置、プログラムおよび電池診断方法 |
JP7625805B2 (ja) | 2019-08-30 | 2025-02-04 | 株式会社Gsユアサ | 推定装置及び推定方法 |
JP7658464B2 (ja) | 2021-08-13 | 2025-04-08 | エルジー エナジー ソリューション リミテッド | 電池管理装置およびそれを含む電池検査システム |
Families Citing this family (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP6409208B1 (ja) * | 2017-03-29 | 2018-10-24 | 株式会社Gsユアサ | 蓄電量推定装置、蓄電モジュール、蓄電量推定方法、及びコンピュータプログラム |
US12081047B2 (en) | 2018-04-03 | 2024-09-03 | Apple Inc. | Power system with battery charging control |
JP6881428B2 (ja) * | 2018-12-28 | 2021-06-02 | 横河電機株式会社 | 学習装置、推定装置、学習方法、推定方法、学習プログラム、および推定プログラム |
KR102354112B1 (ko) * | 2019-03-26 | 2022-01-24 | 서강대학교산학협력단 | 인공 지능에 기반하여 배터리의 상태를 추정하는 장치 및 방법 |
JP6651063B1 (ja) * | 2019-06-12 | 2020-02-19 | 三菱電機株式会社 | 充放電制御装置および充放電制御方法 |
KR20210041511A (ko) | 2019-10-07 | 2021-04-15 | 삼성에스디아이 주식회사 | 배터리의 건강 상태를 추정하는 방법 및 장치 |
JP7151692B2 (ja) * | 2019-11-15 | 2022-10-12 | トヨタ自動車株式会社 | 電池の充電方法および充電システム |
JP7413741B2 (ja) * | 2019-12-03 | 2024-01-16 | 株式会社Gsユアサ | 蓄電池データ蓄積装置、コンピュータプログラム及び蓄電池データ蓄積方法 |
JP7413806B2 (ja) * | 2020-02-06 | 2024-01-16 | トヨタ自動車株式会社 | バッテリ劣化判定装置、バッテリ劣化判定方法、及びバッテリ劣化判定プログラム |
CN116134659A (zh) * | 2020-07-29 | 2023-05-16 | 松下电器(美国)知识产权公司 | 信息处理装置、程序以及信息处理方法 |
CN112215253B (zh) * | 2020-08-28 | 2025-03-07 | 国网吉林省电力有限公司电力科学研究院 | 一种基于降噪自编码的锂电池储能系统监测方法及系统 |
CN113189505B (zh) * | 2021-03-26 | 2022-06-28 | 深圳市磐锋精密技术有限公司 | 一种基于大数据的手机电池状态监测系统 |
KR20230005534A (ko) * | 2021-07-01 | 2023-01-10 | 주식회사 엘지에너지솔루션 | 전지셀의 벤트 발생시점 예측 시스템 및 예측방법 |
CN113619446B (zh) * | 2021-10-13 | 2022-03-04 | 中兴通讯股份有限公司 | 一种输出控制方法、控制单元及计算机可读存储介质 |
CN114583301B (zh) * | 2022-04-29 | 2022-08-23 | 国网浙江省电力有限公司电力科学研究院 | 基于安全特征参量表征体系的电站热失控预警方法及系统 |
CN118572840B (zh) * | 2024-08-01 | 2024-11-12 | 东莞市锂智慧能源有限公司 | 一种电池智能管理系统 |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2004101188A (ja) * | 2002-09-04 | 2004-04-02 | Sanyo Electric Co Ltd | 電池の満充電容量を検出する方法 |
JP2007205878A (ja) * | 2006-02-01 | 2007-08-16 | Ntt Facilities Inc | 二次電池容量推定システム、プログラム及び二次電池容量推定方法 |
JP2013044598A (ja) * | 2011-08-23 | 2013-03-04 | Toyota Motor Corp | 二次電池の劣化管理システム |
JP2013115863A (ja) * | 2011-11-25 | 2013-06-10 | Honda Motor Co Ltd | バッテリ管理システム |
Family Cites Families (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP3197439B2 (ja) | 1994-11-08 | 2001-08-13 | トヨタ自動車株式会社 | 電気自動車用電池の管理装置 |
JPH08136629A (ja) * | 1994-11-11 | 1996-05-31 | Kyushu Electric Power Co Inc | 蓄電池寿命診断装置 |
US5869949A (en) * | 1996-10-02 | 1999-02-09 | Canon Kabushiki Kaisha | Charging apparatus and charging system for use with an unstable electrical power supply |
TW535308B (en) | 2000-05-23 | 2003-06-01 | Canon Kk | Detecting method for detecting internal state of a rechargeable battery, detecting device for practicing said detecting method, and instrument provided with said |
JP2002154617A (ja) | 2000-11-20 | 2002-05-28 | Tosho Inc | 薬品収納装置 |
JP4794760B2 (ja) | 2001-07-04 | 2011-10-19 | パナソニック株式会社 | 電池パック |
JP4263859B2 (ja) | 2001-10-09 | 2009-05-13 | 古河電池株式会社 | 蓄電池の保守管理方法 |
JP4009537B2 (ja) | 2003-01-15 | 2007-11-14 | 松下電器産業株式会社 | 充電制御装置、電池管理システム、電池パック、及びそれらによる二次電池の劣化判定方法 |
JP4570991B2 (ja) | 2005-03-14 | 2010-10-27 | 富士重工業株式会社 | バッテリ管理システム |
KR100756837B1 (ko) * | 2005-06-30 | 2007-09-07 | 주식회사 엘지화학 | 배터리 상태 추정 방법 및 장치 |
JP5039980B2 (ja) | 2005-11-14 | 2012-10-03 | 日立ビークルエナジー株式会社 | 二次電池モジュール |
JP5228322B2 (ja) | 2006-08-30 | 2013-07-03 | トヨタ自動車株式会社 | 蓄電装置の劣化評価システム、車両、蓄電装置の劣化評価方法およびその劣化評価方法をコンピュータに実行させるためのプログラムを記録したコンピュータ読取可能な記録媒体 |
JP4649682B2 (ja) * | 2008-09-02 | 2011-03-16 | 株式会社豊田中央研究所 | 二次電池の状態推定装置 |
JP5368038B2 (ja) * | 2008-09-11 | 2013-12-18 | ミツミ電機株式会社 | 電池状態検知装置及びそれを内蔵する電池パック |
JP5497594B2 (ja) | 2010-09-10 | 2014-05-21 | 関西電力株式会社 | 蓄電装置を用いたアンシラリーサービス提供装置 |
JP2013247726A (ja) * | 2012-05-24 | 2013-12-09 | Toshiba Corp | 蓄電池劣化制御装置 |
-
2015
- 2015-02-24 WO PCT/JP2015/055154 patent/WO2015198631A1/ja active Application Filing
- 2015-02-24 JP JP2016529099A patent/JP6370902B2/ja not_active Expired - Fee Related
- 2015-02-24 EP EP15811538.6A patent/EP3163669B1/en active Active
- 2015-02-24 US US15/321,906 patent/US20170294689A1/en not_active Abandoned
- 2015-02-24 KR KR1020177000668A patent/KR101927644B1/ko active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2004101188A (ja) * | 2002-09-04 | 2004-04-02 | Sanyo Electric Co Ltd | 電池の満充電容量を検出する方法 |
JP2007205878A (ja) * | 2006-02-01 | 2007-08-16 | Ntt Facilities Inc | 二次電池容量推定システム、プログラム及び二次電池容量推定方法 |
JP2013044598A (ja) * | 2011-08-23 | 2013-03-04 | Toyota Motor Corp | 二次電池の劣化管理システム |
JP2013115863A (ja) * | 2011-11-25 | 2013-06-10 | Honda Motor Co Ltd | バッテリ管理システム |
Non-Patent Citations (1)
Title |
---|
See also references of EP3163669A4 * |
Cited By (29)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPWO2018147194A1 (ja) * | 2017-02-07 | 2020-01-09 | 日本電気株式会社 | 蓄電池制御装置、充放電制御方法、及びプログラム |
WO2018147194A1 (ja) * | 2017-02-07 | 2018-08-16 | 日本電気株式会社 | 蓄電池制御装置、充放電制御方法、及び記録媒体 |
JP2019125482A (ja) * | 2018-01-16 | 2019-07-25 | 株式会社Gsユアサ | 監視装置、監視方法及びコンピュータプログラム |
WO2019142711A1 (ja) * | 2018-01-16 | 2019-07-25 | 株式会社Gsユアサ | 監視装置、監視方法、コンピュータプログラム、劣化判定方法、劣化判定装置、及び劣化判定システム |
US11677253B2 (en) | 2018-01-16 | 2023-06-13 | Gs Yuasa International Ltd. | Monitoring device, monitoring method, computer program, deterioration determination method, deterioration determination device, and deterioration determination system |
CN111602284A (zh) * | 2018-01-16 | 2020-08-28 | 株式会社杰士汤浅国际 | 监视装置、监视方法、计算机程序、劣化判断方法、劣化判断装置、以及劣化判断系统 |
US10996282B2 (en) | 2018-03-20 | 2021-05-04 | Gs Yuasa International Ltd. | Abnormality factor determination apparatus, degradation determination apparatus, computer program, degradation determining method, and abnormality factor determining method |
US12237704B2 (en) * | 2018-06-08 | 2025-02-25 | Panasonic Intellectual Property Management Co., Ltd. | Battery management system and battery management method |
US20210091583A1 (en) * | 2018-06-08 | 2021-03-25 | Panasonic Intellectual Property Management Co., Ltd. | Battery management system and battery management method |
JPWO2020027203A1 (ja) * | 2018-07-31 | 2021-08-26 | 本田技研工業株式会社 | 推定システム、推定装置、推定方法、プログラム、及び記憶媒体 |
WO2020027203A1 (ja) * | 2018-07-31 | 2020-02-06 | 本田技研工業株式会社 | 推定システム、推定装置、推定方法、プログラム、及び記憶媒体 |
JP7217277B2 (ja) | 2018-07-31 | 2023-02-02 | 本田技研工業株式会社 | 推定システム、推定装置、推定方法、プログラム、及び記憶媒体 |
JPWO2020137914A1 (ja) * | 2018-12-28 | 2020-07-02 | ||
WO2020137914A1 (ja) * | 2018-12-28 | 2020-07-02 | 株式会社Gsユアサ | データ処理装置、データ処理方法、及びコンピュータプログラム |
JP7390310B2 (ja) | 2018-12-28 | 2023-12-01 | 株式会社Gsユアサ | データ処理装置、データ処理方法、及びコンピュータプログラム |
US11527900B2 (en) * | 2019-04-18 | 2022-12-13 | Lg Display Co., Ltd. | Apparatus and method for managing a battery based on degradation determination |
WO2020255557A1 (ja) * | 2019-06-17 | 2020-12-24 | 日置電機株式会社 | 電池劣化診断システム、診断処理装置、測定装置及びプログラム |
JP7625805B2 (ja) | 2019-08-30 | 2025-02-04 | 株式会社Gsユアサ | 推定装置及び推定方法 |
JP2020092598A (ja) * | 2020-01-30 | 2020-06-11 | 株式会社東芝 | 充電パターン作成装置、充電制御装置、充電パターン作成方法、プログラム、及び蓄電システム |
JP7482088B2 (ja) | 2021-07-27 | 2024-05-13 | 株式会社Tmeic | 電動機巻線劣化状態調査システム |
JP7658464B2 (ja) | 2021-08-13 | 2025-04-08 | エルジー エナジー ソリューション リミテッド | 電池管理装置およびそれを含む電池検査システム |
JP7596244B2 (ja) | 2021-09-22 | 2024-12-09 | 株式会社日立製作所 | 電池診断装置、プログラムおよび電池診断方法 |
WO2023188582A1 (ja) * | 2022-03-29 | 2023-10-05 | 東洋システム株式会社 | 二次電池容量推定システム |
JP2023146043A (ja) * | 2022-03-29 | 2023-10-12 | 東洋システム株式会社 | 二次電池容量推定システム |
JP7276928B1 (ja) | 2022-03-29 | 2023-05-18 | 東洋システム株式会社 | 二次電池容量推定システム |
WO2024166450A1 (ja) * | 2023-02-09 | 2024-08-15 | 株式会社豊田中央研究所 | 情報処理システム、情報処理方法および情報処理プログラム |
JP7652200B2 (ja) | 2023-02-09 | 2025-03-27 | 株式会社豊田中央研究所 | 情報処理システム、情報処理方法および情報処理プログラム |
CN117388716B (zh) * | 2023-12-11 | 2024-02-13 | 四川长园工程勘察设计有限公司 | 基于时序数据的电池组故障诊断方法、系统及存储介质 |
CN117388716A (zh) * | 2023-12-11 | 2024-01-12 | 四川长园工程勘察设计有限公司 | 基于时序数据的电池组故障诊断方法、系统及存储介质 |
Also Published As
Publication number | Publication date |
---|---|
EP3163669A4 (en) | 2017-11-22 |
JPWO2015198631A1 (ja) | 2017-05-25 |
KR20170016956A (ko) | 2017-02-14 |
KR101927644B1 (ko) | 2018-12-10 |
EP3163669A1 (en) | 2017-05-03 |
EP3163669B1 (en) | 2019-10-30 |
US20170294689A1 (en) | 2017-10-12 |
JP6370902B2 (ja) | 2018-08-15 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP6370902B2 (ja) | 蓄電池システムの劣化制御装置及びその方法 | |
Park et al. | LSTM-based battery remaining useful life prediction with multi-channel charging profiles | |
CN114072684A (zh) | 在服务器侧表征能重复充电的电池 | |
CN109637576A (zh) | 一种基于支持向量回归的闪存寿命预测方法 | |
CN113646949B (zh) | 蓄电池剩余价值确定系统 | |
CN103942453A (zh) | 一种针对非技术性损失的智能用电异常检测方法 | |
CN114548509A (zh) | 一种多能源系统多类型负荷联合预测方法及系统 | |
CN110082682A (zh) | 一种锂电池荷电状态估计方法 | |
CN118330468A (zh) | 一种基于生成式机器学习的退役电池健康状态快速估计方法 | |
CN113966520A (zh) | 用于促进对来自工业自动化控制系统或电力系统的数据进行存储的方法和设备 | |
KR20230028017A (ko) | 시변 부하전류 상황에 강인한 리튬 이온 배터리의 모델 기반 충전 상태 추정 장치 및 방법 | |
Sun et al. | Scalar quantization for relative error | |
CN115219935B (zh) | 新能源设备的健康状况测评方法、系统、装置及介质 | |
CN109815534A (zh) | 一种基于决策树算法的闪存寿命预测方法及装置 | |
CN117272845B (zh) | 空压站能耗水平评估方法、装置及设备 | |
CN112183644B (zh) | 指标稳定性的监控方法、装置、计算机设备及介质 | |
CN119047893B (zh) | 轨道平顺状态的时-频域联合评估方法、装置、设备及介质 | |
Negri et al. | Temperature-Based Accuracy Estimation on Instrument Transformers via Neural Networks and Microcontrollers | |
CN119738750A (zh) | 配电变压器的寿命预测方法、装置与电子设备 | |
CN116736962B (zh) | 一种多功能教育pc的电源管理方法、装置、设备及介质 | |
CN118691125B (zh) | 电力负荷预测方法、装置、设备及存储介质 | |
CN119223271B (zh) | 一种基于惯性器件的温度补偿方法及系统 | |
CN112613632B (zh) | 功率预测模型建立方法、预测方法、装置及电子设备 | |
CN103745130A (zh) | 无线传感器环境下多属性事件的预测方法 | |
CN117574229A (zh) | 基于ResNeSt的工业用户碳排放量估算方法及装置 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 15811538 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
ENP | Entry into the national phase |
Ref document number: 20177000668 Country of ref document: KR Kind code of ref document: A |
|
REEP | Request for entry into the european phase |
Ref document number: 2015811538 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2015811538 Country of ref document: EP |
|
ENP | Entry into the national phase |
Ref document number: 2016529099 Country of ref document: JP Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 15321906 Country of ref document: US |