WO2023106166A1 - 解析装置、予測装置、解析方法、予測方法及びプログラム - Google Patents
解析装置、予測装置、解析方法、予測方法及びプログラム Download PDFInfo
- Publication number
- WO2023106166A1 WO2023106166A1 PCT/JP2022/043969 JP2022043969W WO2023106166A1 WO 2023106166 A1 WO2023106166 A1 WO 2023106166A1 JP 2022043969 W JP2022043969 W JP 2022043969W WO 2023106166 A1 WO2023106166 A1 WO 2023106166A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- factor
- data
- prediction
- measurement data
- life
- Prior art date
Links
- 238000004458 analytical method Methods 0.000 title claims abstract description 111
- 238000000034 method Methods 0.000 title claims description 56
- 238000005259 measurement Methods 0.000 claims abstract description 129
- 238000004364 calculation method Methods 0.000 claims abstract description 15
- 230000008859 change Effects 0.000 claims abstract description 12
- 239000011159 matrix material Substances 0.000 claims description 123
- 230000007704 transition Effects 0.000 claims description 55
- 238000012545 processing Methods 0.000 description 30
- 238000010586 diagram Methods 0.000 description 28
- 230000005540 biological transmission Effects 0.000 description 20
- 238000004891 communication Methods 0.000 description 11
- 238000013500 data storage Methods 0.000 description 9
- 230000006870 function Effects 0.000 description 9
- 238000012360 testing method Methods 0.000 description 9
- 239000000284 extract Substances 0.000 description 7
- 230000014759 maintenance of location Effects 0.000 description 7
- 230000007774 longterm Effects 0.000 description 5
- 230000009471 action Effects 0.000 description 4
- 230000010365 information processing Effects 0.000 description 4
- 238000007599 discharging Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000000704 physical effect Effects 0.000 description 3
- 238000005070 sampling Methods 0.000 description 3
- 239000004065 semiconductor Substances 0.000 description 3
- 238000007600 charging Methods 0.000 description 2
- 238000010277 constant-current charging Methods 0.000 description 2
- 230000007423 decrease Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000005401 electroluminescence Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 239000000047 product Substances 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 239000002000 Electrolyte additive Substances 0.000 description 1
- 238000007476 Maximum Likelihood Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000010280 constant potential charging Methods 0.000 description 1
- 239000000470 constituent Substances 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 230000006866 deterioration Effects 0.000 description 1
- 238000003487 electrochemical reaction Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000007773 negative electrode material Substances 0.000 description 1
- 239000007774 positive electrode material Substances 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
Images
Classifications
-
- 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
- G01R31/3842—Arrangements for monitoring battery or accumulator variables, e.g. SoC combining voltage and current measurements
-
- 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]
- G01R31/385—Arrangements for measuring battery or accumulator variables
- G01R31/387—Determining ampere-hour charge capacity or SoC
- G01R31/388—Determining ampere-hour charge capacity or SoC involving voltage measurements
-
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
-
- 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
- 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
-
- 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
- the present invention relates to an analysis device, a prediction device, an analysis method, a prediction method and a program.
- the materials and designs that make up the battery are evaluated by performing charge-discharge cycle tests and measuring the life characteristics of the battery cells. Since the charge-discharge cycle test takes time, a technique is used to predict life characteristics from the measurement results of a short-time charge-discharge cycle test using a mathematical model or machine learning technology.
- Patent Document 1 discloses an invention that evaluates the charge/discharge tendency and deterioration state based on the QV curve or dQ/dV curve of the battery.
- Patent Document 2 discloses an invention that calculates long-term characteristic prediction data by applying an artificial neural network trained using initial characteristic learning data and long-term characteristic learning data to initial characteristic measurement data of a battery.
- US Pat. No. 5,300,005 discloses an invention that uses data-driven predictive modeling to predict life from early cycle discharge voltage curves of battery cells.
- Patent No. 6488105 Japanese Patent Publication No. 2010-539473 JP 2019-113524 A
- the conventional technology has the problem that it is not possible to identify the physical phenomenon that affects the life characteristics of the battery. Unless the physical phenomenon that affects the life characteristics is identified, it is difficult to improve the life characteristics, even if it is possible to predict that there is a problem with the life characteristics.
- the purpose of this disclosure is to present information about factors that affect battery life characteristics.
- the present disclosure has the configuration shown below.
- an acquisition unit configured to acquire life data from cycle measurement data of a target battery; By factorizing the relationship between the voltage and the current capacity calculated from the cycle measurement data of the target battery, factor strength transition data showing changes in the strength of each factor affecting the current capacity and the voltage of each factor a calculation unit configured to calculate factor data representing the relationship with ampacity; an output unit configured to output the lifespan data, the factor data, and the factor intensity transition data; Analysis device comprising.
- [4] By factorizing the relationship between the voltage and the current capacity calculated from the cycle measurement data of the target battery up to a predetermined cycle, the change in the strength of each factor affecting the current capacity up to the predetermined cycle.
- a calculation unit configured to calculate factor strength transition data indicating a generation unit configured to generate a feature amount from the factor strength transition data of each factor;
- a predictor configured to predict life characteristics of the target battery;
- a prediction device comprising a
- the prediction device according to [4], An output unit configured to output a predicted value of the life characteristic and factor data representing a relationship between the voltage and the current capacity of a predetermined factor calculated from the learning cycle measurement data. , prediction device.
- the prediction device configured to acquire, by analyzing the model, contributions to prediction of the life characteristics of the target battery for each of the feature values generated by the generating unit; an identifying unit configured to extract, from among the feature amounts generated by the generating unit, a feature amount in which the contribution mode satisfies a predetermined condition, and to identify a corresponding factor; further comprising The output unit outputs a predicted value of the life characteristic of the target battery, the factor data related to the specified factor, and a contribution mode corresponding to the specified factor, prediction device.
- the computer an acquisition procedure for acquiring life data from the cycle measurement data of the target battery; By factorizing the relationship between the voltage and the current capacity calculated from the cycle measurement data of the target battery, factor strength transition data showing changes in the strength of each factor affecting the current capacity and the voltage of each factor A calculation procedure for calculating factor data representing the relationship with ampacity; an output procedure for outputting the lifespan data, the factor data, and the factor intensity transition data; The analysis method to perform.
- the computer By factorizing the relationship between the voltage and the current capacity calculated from the cycle measurement data of the target battery up to a predetermined cycle, the factor that indicates the change in the strength of each factor that affects the current capacity up to the predetermined cycle. a calculation procedure for calculating intensity transition data; a generation procedure for generating a feature amount from the factor strength transition data of each factor; By inputting the feature amount generated by the generation procedure into a model that has learned the relationship between the feature amount of each factor generated based on the learning cycle measurement data up to the predetermined cycle and the life data, a prediction procedure for predicting the life characteristics of the target battery; Prediction method to perform
- an acquisition procedure for acquiring life data from the cycle measurement data of the target battery By factorizing the relationship between the voltage and the current capacity calculated from the cycle measurement data of the target battery, factor strength transition data showing changes in the strength of each factor affecting the current capacity and the voltage of each factor
- a calculation procedure for calculating factor data representing the relationship with ampacity an output procedure for outputting the lifespan data, the factor data, and the factor intensity transition data; program to run the
- the factor that indicates the change in the strength of each factor that affects the current capacity up to the predetermined cycle By factorizing the relationship between the voltage and the current capacity calculated from the cycle measurement data of the target battery up to a predetermined cycle, the factor that indicates the change in the strength of each factor that affects the current capacity up to the predetermined cycle.
- a calculation procedure for calculating intensity transition data a generation procedure for generating a feature amount from the factor strength transition data of each factor;
- a prediction procedure for predicting the life characteristics of the target battery program to run the
- FIG. 1 is a block diagram showing an example of the overall configuration of a battery life analysis system.
- FIG. 2 is a block diagram showing an example of the hardware configuration of the computer.
- FIG. 3 is a block diagram showing an example of the functional configuration of the battery life analysis system.
- FIG. 4 is a flow chart showing an example of the procedure of the battery life analysis method.
- FIG. 5 is a diagram showing an example of cycle measurement data.
- FIG. 6A is a diagram for explaining a method of obtaining life data based on the discharge capacity retention rate.
- FIG. 6B is a diagram for explaining a method of obtaining life data based on the number of charge/discharge cycles.
- FIG. 7A is a diagram showing an example of a QV curve.
- FIG. 1 is a block diagram showing an example of the overall configuration of a battery life analysis system.
- FIG. 2 is a block diagram showing an example of the hardware configuration of the computer.
- FIG. 3 is a block diagram showing an example of the functional configuration of
- FIG. 7B is a diagram showing an example of a dQ/dV curve.
- FIG. 8 is a diagram for explaining a method of generating a measurement data matrix.
- FIG. 9 is a diagram for explaining non-negative matrix factorization.
- FIG. 10 is a diagram for explaining factor data.
- FIG. 11 is a diagram for explaining factor strength transition data.
- FIG. 12 is a block diagram showing an example of the overall configuration of the battery life prediction system.
- FIG. 13 is a block diagram showing an example of the functional configuration of the battery life prediction system.
- FIG. 14 is a flowchart illustrating an example of a processing procedure of predictive model learning processing.
- FIG. 15 is a flowchart illustrating an example of a processing procedure of life characteristic prediction processing.
- FIG. 16 is a diagram showing an example of a prediction result display screen.
- a first embodiment of the present invention is a battery life analysis system that analyzes life characteristics of a battery based on cycle measurement data of the battery.
- the battery life analysis system in the present embodiment includes life data obtained from long-term cycle measurement data (hereinafter also referred to as "analysis cycle measurement data") of a battery to be analyzed (hereinafter also referred to as "analysis target battery”) and Output analysis results including information on factors affecting the life data. Therefore, according to the battery life analysis system of the present embodiment, the user can easily identify the factors affecting the life characteristics of the battery when there is a problem with the life characteristics of the battery to be analyzed. , it becomes possible to lead to actions for improving life characteristics.
- the long-term cycle measurement data is the cycle measurement data when the charge-discharge cycle test is executed until the life characteristics of the battery can be measured.
- the battery life analysis system in this embodiment performs non-negative matrix decomposition on the matrix representing the cycle measurement data, and for each factor that affects the current capacity, the voltage and the factor of each factor.
- a factor data matrix representing the relationship with current capacity and a factor intensity transition data matrix representing changes in the intensity of each factor with charge/discharge cycles are calculated.
- Each factor data included in the factor data matrix reflects an electrochemical reaction specific to the components of the battery. Therefore, by referring to the factor data matrix and the factor intensity transition data matrix, it becomes easy to identify the factors that strongly affect the characteristics of the battery in a certain charge/discharge cycle.
- FIG. 1 is a block diagram showing an example of the overall configuration of a battery life analysis system according to this embodiment.
- the battery life analysis system 1 in this embodiment includes an analysis device 10 and a user terminal 30.
- the analysis device 10 and the user terminal 30 are connected for data communication via a communication network 90 such as a LAN (Local Area Network) or the Internet.
- a communication network 90 such as a LAN (Local Area Network) or the Internet.
- the analysis device 10 is an information processing device such as a PC (Personal Computer), workstation, or server that analyzes battery cycle measurement data in response to a request from the user terminal 30 .
- the analysis device 10 receives analysis cycle measurement data from the user terminal 30 .
- the analysis device 10 analyzes the cycle measurement data for analysis and transmits the analysis result to the user terminal 30 .
- the analysis result includes information on the life characteristics of the battery to be analyzed and factors affecting the life characteristics of the battery to be analyzed.
- the user terminal 30 is an information processing terminal operated by a user, such as a PC, a tablet terminal, or a smartphone.
- the user terminal 30 accepts the input of analysis cycle measurement data and transmits it to the analysis device 10 according to the user's operation. Also, the user terminal 30 receives the analysis result from the analysis device 10 and outputs it to the user.
- the overall configuration of the battery life analysis system 1 shown in FIG. 1 is merely an example, and various system configuration examples are possible depending on the application and purpose.
- the analysis device 10 may be implemented by a plurality of computers, or may be implemented as a cloud computing service.
- the battery life analysis system 1 may be realized by a stand-alone computer that has the functions that the analysis device 10 and the user terminal 30 should have.
- FIG. 2 is a block diagram showing an example of the hardware configuration of computer 500 in this embodiment.
- computer 500 includes CPU (Central Processing Unit) 501, ROM (Read Only Memory) 502, RAM (Random Access Memory) 503, HDD (Hard Disk Drive) 504, input device 505, It has a display device 506 , a communication I/F (Interface) 507 and an external I/F 508 .
- the CPU 501, ROM 502 and RAM 503 form a so-called computer.
- Each piece of hardware of the computer 500 is interconnected via a bus line 509 .
- the input device 505 and the display device 506 may be connected to the external I/F 508 for use.
- the CPU 501 is an arithmetic unit that implements the overall control and functions of the computer 500 by reading programs and data from a storage device such as the ROM 502 or HDD 504 onto the RAM 503 and executing processing.
- the ROM 502 is an example of a non-volatile semiconductor memory (storage device) that can retain programs and data even when the power is turned off.
- the ROM 502 functions as a main storage device that stores various programs, data, etc. necessary for the CPU 501 to execute various programs installed in the HDD 504 .
- the ROM 502 stores boot programs such as BIOS (Basic Input/Output System) and EFI (Extensible Firmware Interface) that are executed when the computer 500 is started, OS (Operating System) settings, network settings, and other data. is stored.
- BIOS Basic Input/Output System
- EFI Extensible Firmware Interface
- the RAM 503 is an example of a volatile semiconductor memory (storage device) that erases programs and data when the power is turned off.
- the RAM 503 is, for example, a DRAM (Dynamic Random Access Memory) or an SRAM (Static Random Access Memory).
- a RAM 503 provides a work area that is expanded when various programs installed in the HDD 504 are executed by the CPU 501 .
- the HDD 504 is an example of a non-volatile storage device that stores programs and data.
- the programs and data stored in the HDD 504 include an OS, which is basic software that controls the entire computer 500, and applications that provide various functions on the OS.
- the computer 500 may use a storage device using flash memory as a storage medium (for example, SSD: Solid State Drive, etc.) instead of the HDD 504 .
- the input device 505 includes a touch panel used by the user to input various signals, operation keys and buttons, a keyboard and mouse, and a microphone for inputting sound data such as voice.
- the display device 506 is composed of a display such as liquid crystal or organic EL (Electro-Luminescence) that displays a screen, a speaker that outputs sound data such as voice, and the like.
- a display such as liquid crystal or organic EL (Electro-Luminescence) that displays a screen
- a speaker that outputs sound data such as voice, and the like.
- a communication I/F 507 is an interface for connecting to a communication network and allowing the computer 500 to perform data communication.
- the external I/F 508 is an interface with an external device.
- the external device includes a drive device 510 and the like.
- a drive device 510 is a device for setting a recording medium 511 .
- the recording medium 511 here includes media such as CD-ROMs, flexible disks, magneto-optical disks, etc. that record information optically, electrically, or magnetically.
- the recording medium 511 may also include a semiconductor memory or the like that electrically records information, such as a ROM or a flash memory. This allows the computer 500 to read from and/or write to the recording medium 511 via the external I/F 508 .
- Various programs installed in the HDD 504 are, for example, a distributed recording medium 511 set in the drive device 510 connected to the external I/F 508, and the various programs recorded in the recording medium 511 are read by the drive device 510. installed by Alternatively, various programs installed in HDD 504 may be installed by being downloaded from another network different from the communication network via communication I/F 507 .
- FIG. 3 is a block diagram showing an example of the functional configuration of the battery life analysis system 1 according to this embodiment.
- the analysis device 10 in this embodiment includes a data reception unit 101, a lifespan acquisition unit 102, a curve generation unit 103, a matrix generation unit 104, a factorization unit 105, and a result transmission unit 106. .
- Each processing unit provided in the analysis device 10 is realized by processing that the CPU 501 is caused to execute by the program developed on the RAM 503 from the HDD 504 shown in FIG.
- the data receiving unit 101 receives analysis cycle measurement data from the user terminal 30 .
- the data receiving section 101 also sends the analysis cycle measurement data received from the user terminal 30 to the life acquisition section 102 and the curve generation section 103 .
- the life acquisition unit 102 acquires life data representing the life characteristics of the battery to be analyzed based on the analysis cycle measurement data received from the data reception unit 101 .
- the lifespan acquisition unit 102 also sends the acquired lifespan data to the result transmission unit 106 .
- the curve generating unit 103 Based on the cycle measurement data for analysis received from the data receiving unit 101, the curve generating unit 103 generates a QV curve representing the relationship between the voltage V and the current capacity Q, or a dQ/dV curve obtained by differentiating the QV curve with respect to the voltage V. Generate. Curve generation section 103 also sends the generated QV curve or dQ/dV curve to matrix generation section 104 .
- the matrix generation unit 104 Based on the QV curve or dQ/dV curve received from the curve generation unit 103, the matrix generation unit 104 generates a measurement data matrix representing the relationship between voltage and current capacity for each charge/discharge cycle. Matrix generation section 104 also sends the generated measurement data matrix to factorization section 105 .
- the factorization section 105 performs non-negative matrix factorization on the measurement data matrix received from the matrix generation section 104 . As a result, factorization section 105 obtains a factor data matrix and a factor intensity transition data matrix. In addition, factorization section 105 sends the factorization result to result transmission section 106 .
- the factorization result includes a factor data matrix and a factor intensity transition data matrix.
- the result transmission unit 106 transmits the analysis result of the cycle measurement data for analysis to the user terminal 30 .
- the analysis results include the lifespan data received from the lifespan acquisition unit 102 and the factorization results received from the factorization unit 105 .
- the user terminal 30 in this embodiment has a data input section 301 and a result output section 302 .
- Each processing unit provided in the user terminal 30 is realized by processing that the CPU 501 is caused to execute by the program developed on the RAM 503 from the HDD 504 shown in FIG.
- the data input unit 301 accepts input of cycle measurement data for analysis according to the user's operation. Further, the data input unit 301 transmits the received cycle measurement data for analysis to the analysis device 10 .
- the result output unit 302 receives analysis results from the analysis device 10 . Further, the result output unit 302 outputs the received analysis result to the display device 506 or the like.
- FIG. 4 is a flow chart showing an example of the processing procedure of the battery life analysis method according to this embodiment.
- step S101 the data input unit 301 provided in the user terminal 30 accepts input of cycle measurement data for analysis according to the user's operation.
- the data input unit 301 transmits the received cycle measurement data for analysis to the analysis device 10 .
- the data reception unit 101 receives analysis cycle measurement data from the user terminal 30 . Next, the data receiving section 101 sends the received analytical cycle measurement data to the life acquiring section 102 and the curve generating section 103 .
- FIG. 5 is a diagram showing an example of cycle measurement data in this embodiment.
- the cycle measurement data is obtained by performing a charge-discharge cycle test on a single cell of the battery and recording the sampling time and the physical property values of each cell at each sampling time.
- the sampling time is desirably selected appropriately so as to sufficiently follow changes in physical property values, and does not have to be at regular intervals.
- the physical property values recorded in this embodiment include the voltage applied to the cell, the current flowing through the cell, and the ampacity.
- One cycle in the charge-discharge cycle test always includes a constant-current charging step followed by a constant-current discharging step. At this time, a step of constant voltage charging may be included between the step of constant current charging and the step of constant current discharging. A rest step may also be included between each charge/discharge step.
- step S ⁇ b>102 the lifespan acquisition unit 102 provided in the analysis device 10 receives analysis cycle measurement data from the data reception unit 101 .
- the life acquisition unit 102 acquires life data representing the life characteristics of the battery to be analyzed based on the cycle measurement data for analysis. Lifespan acquisition unit 102 then sends the acquired lifespan data to result transmission unit 106 .
- FIG. 6 is a diagram illustrating a method of acquiring lifespan data.
- the life characteristics in this embodiment are the discharge capacity retention rate (see FIG. 6A) when a predetermined charge-discharge cycle is performed. , or defined as the number of charge/discharge cycles, that is, the cycle life (see FIG. 6(B)) when the discharge capacity retention rate is below a predetermined discharge capacity retention rate.
- the discharge capacity retention rate is the ratio of the discharge capacity measured at each cycle to the discharge capacity measured at the first cycle.
- Figs. 6(A) and 6(B) show an example of acquiring the lifetime characteristics of one cell.
- the life data acquired by the life acquisition unit 102 includes life characteristics acquired for each cell of the battery to be analyzed.
- the curve of the discharge capacity retention rate itself may be used as the life data.
- step S ⁇ b>103 the curve generation unit 103 provided in the analysis device 10 receives analysis cycle measurement data from the data reception unit 101 .
- the curve generator 103 generates a QV curve representing the relationship between the voltage V and the current capacity Q, or a dQ/dV curve obtained by differentiating the QV curve with respect to the voltage V, based on the cycle measurement data for analysis.
- Curve generation section 103 then sends the generated QV curve or dQ/dV curve to matrix generation section 104 . If the internal resistance of each battery cell differs significantly, it is necessary to standardize the voltage by subtracting the value of the IR drop from the voltage.
- FIG. 7 is a diagram showing an example of a QV curve during discharge and a dQ/dV curve during discharge.
- FIG. 7A is an example of a QV curve
- FIG. 7B is an example of a dQ/dV curve.
- a plurality of curves shown in FIG. 7 correspond to each charge/discharge cycle. As shown in FIG. 7, in both the QV curve and the dQ/dV curve, the entire curve changes downward (that is, the discharge capacity decreases) as the charge-discharge cycle progresses. can be seen (see the cycle change arrows in each figure).
- the QV curve and the dQ/dV curve it is known that a specific voltage range corresponds to the battery components (positive and negative electrode active materials, electrolyte additive, etc.).
- the QV curve and the dQ/dV curve are results obtained by superimposing patterns for each component of the battery, and it is known that the mixing amount of each pattern changes as the charge/discharge cycle progresses.
- the QV curve and the dQ/dV curve are decomposed into patterns of components corresponding to each member, and the mixture amount of each component in each charge/discharge cycle is known, the change in battery characteristics in a specific charge/discharge cycle can be affected. This is useful information for identifying the given member. For example, if the discharge capacity of a battery suddenly decreases after a specific charge/discharge cycle, it can be assumed that there was some problem with the material whose mixture amount changed significantly around that cycle. .
- step S ⁇ b>104 the matrix generation unit 104 included in the analysis device 10 receives the QV curve or the dQ/dV curve from the curve generation unit 103 .
- matrix generator 104 generates a measurement data matrix representing the relationship between voltage and current capacity for each charge/discharge cycle based on the QV curve or dQ/dV curve.
- Matrix generation section 104 then sends the generated measurement data matrix to factorization section 105 .
- FIG. 8 is a diagram illustrating a method of generating a measurement data matrix for a QV curve during discharge.
- Each table on the left side shown in FIG. 8 is a matrix summarizing the QV curves during discharge of each charge/discharge cycle for each cell of the battery, with the cycle number as the row and the voltage V value as the column.
- Q 111 in the table is the value of discharge capacity at cell 1, cycle 1, voltage V1 .
- the voltage grids V 1 , V 2 , . choose wisely so that you can follow suit. If the current capacities Q at the voltage grids V 1 , V 2 , .
- the table on the right side shown in FIG. 8 is a matrix obtained by stacking the matrices corresponding to the cells on the left side in order of cell number. This produces a single matrix (measurement data matrix) representing all cycle measurement data corresponding to each cell of the battery. If the capacity of each battery cell is different, it is necessary to standardize such as dividing the current capacity by the design capacity when stacking.
- step S105 the factorization unit 105 included in the analysis device 10 performs non-negative matrix factorization on the measurement data matrix received from the matrix generation unit 104 to calculate a factor data matrix and a factor intensity transition data matrix.
- the factor data matrix represents the relationship between voltage and ampacity for each factor that affects ampacity.
- the factor intensity transition data matrix represents changes in the intensity of each factor with charge/discharge cycles.
- the intensity of each factor is the magnitude of the amount of each factor mixed with respect to the whole.
- Factorization section 105 then sends the factorization result including the factor data matrix and the factor strength transition data matrix to result transmission section 106 .
- FIG. 9 is a diagram illustrating the results of non-negative matrix factorization of the measurement data matrix shown in FIG. 8.
- NMF Non-negative matrix factorization
- Each column of the base matrix is also called a base vector, and each column of the coefficient matrix is also called a coefficient vector.
- the measurement data matrix X can be approximated by the product of the factor data matrix H and the factor intensity transition data matrix W, as shown in FIG.
- the factor data matrix H corresponds to the base matrix
- Each base vector H 1 , H 2 , . . . included in the factor data matrix H is hereinafter referred to as “factor data”.
- each coefficient vector W 1 , W 2 , . . . included in the factor strength transition data matrix W is called “factor strength transition data”.
- Factor data represents the relationship between voltage and ampacity for each member.
- Factor intensity transition data represents changes in intensity of each factor data associated with charge/discharge cycles.
- FIG. 10 is a diagram illustrating the factor data shown in FIG. 9;
- the first row of the table shown in FIG. 10 is a graph plotting the factor data H 11 , H 12 , .
- the second row of the table shown in FIG. 10 is a graph plotting the factor data H 21 , H 22 , .
- the QV curve pattern differs for each factor. Since the specific voltage ranges are derived from the battery components as described above, the factor data patterns correspond to the battery components.
- FIG. 11 is a diagram exemplifying the factor strength transition data shown in FIG.
- a column of the table shown in FIG. 11 corresponds to each cell of the battery and a row corresponds to each factor.
- the first column of the table shown in FIG. 11 represents the transition of the intensity of each factor in the QV curve of cell 1 of the battery with charge/discharge cycles.
- the 1st row, 1st column of the table shown in FIG. 11 shows factor strength transition data W 111 , W 112 , . 1 is a graph plotted as intensity;
- the intensity of factor 1 in cell 1 of the battery does not change significantly as the charge and discharge cycles progress, but the intensity of factor 2 does not change as the charge and discharge cycles progress. It can be seen that it is declining.
- the second column of the table shown in FIG. 11 the same can be said for cell 2 as for cell 1, but it can be seen that the initial strength of factor 2 is less than cell 1.
- the number of factors in the non-negative matrix factorization must be the minimum number while sufficiently reducing the error when approximating the measurement data matrix X by the product of the factor data matrix H and the factor intensity transition data matrix W. not.
- Bayes information criterion, Akaike information criterion, Malinowski's IND function, and maximum likelihood estimation can be used as methods for determining the optimum number of factors.
- step S104 The processing procedure from step S104 to step S105 for the QV curve during discharge has been described above.
- similar processing can be performed by appropriately replacing the above description and FIGS. 8 to 10 .
- step S ⁇ b>106 the result transmission unit 106 included in the analysis device 10 receives lifespan data from the lifespan acquisition unit 102 . Also, the result transmission unit 106 receives the factorization result from the factorization unit 105 . Then, the result transmitting unit 106 transmits the analysis results including the lifespan data and the factorization results to the user terminal 30 .
- the result output unit 302 receives the analysis result from the analysis device 10 . Then, the result output unit 302 outputs the received analysis result to the display device 506 or the like.
- the battery life analysis system 1 in this embodiment analyzes battery cycle measurement data and outputs the analysis results to the user.
- the analysis results include information about the life characteristics of the battery and factors affecting the life characteristics of the battery. Therefore, according to the battery life analysis system 1 of the present embodiment, when there is a problem with the life characteristics of the battery, the user can easily identify the members that affect the life characteristics, thereby improving the life characteristics. It is possible to connect to actions for
- a second embodiment of the present invention is a battery life prediction system that predicts the life characteristics of a battery based on cycle measurement data of the battery.
- the battery life prediction system in the present embodiment uses long-term cycle measurement data (hereinafter also referred to as “learning cycle measurement data”) of a battery to be learned (hereinafter also referred to as “learning target battery”) to create a prediction model.
- Life of the prediction target battery is calculated from the initial cycle measurement data (hereinafter also referred to as "prediction cycle measurement data") of the prediction target battery (hereinafter also referred to as the "prediction target battery”) by learning and using the prediction model. Predict properties.
- the battery life prediction system outputs, to the user, the predicted value of the life characteristic of the battery to be predicted, as well as information on the factors affecting the life characteristic of the battery. Therefore, when it is predicted that there is a problem with the life characteristics of the battery to be predicted, the user can easily identify the factors affecting the life characteristics of the battery, and can find ways to improve the life characteristics. It can be linked to action.
- the initial cycle measurement data is the cycle measurement data when the charge/discharge cycle test is executed from the start to the predetermined number of predicted execution cycles.
- the predicted number of execution cycles may be any number as long as it is shorter than the expected cycle life, and may be set to 100 cycles, for example.
- FIG. 12 is a block diagram showing an example of the overall configuration of the battery life prediction system according to this embodiment.
- the battery life prediction system 2 in this embodiment includes a prediction device 20 and a user terminal 30.
- the prediction device 20 and the user terminal 30 are connected for data communication via a communication network 90 such as a LAN (Local Area Network) or the Internet.
- a communication network 90 such as a LAN (Local Area Network) or the Internet.
- the prediction device 20 is an information processing device such as a PC (Personal Computer), workstation, or server that predicts battery life characteristics in response to a request from the user terminal 30 .
- the prediction device 20 receives learning cycle measurement data and prediction cycle measurement data from the user terminal 30 .
- the prediction device 20 learns a prediction model for predicting battery life characteristics based on the learning cycle measurement data.
- the prediction device 20 uses the prediction model to predict life characteristics from the cycle measurement data for prediction, and transmits the prediction result to the user terminal 30 .
- the prediction result includes the predicted value of the life characteristics of the battery to be predicted and information on factors affecting the life characteristics of the battery.
- the user terminal 30 is an information processing terminal operated by a user, such as a PC, a tablet terminal, or a smartphone.
- the user terminal 30 accepts inputs of learning cycle measurement data and prediction cycle measurement data and transmits them to the prediction device 20 according to the user's operation. Also, the user terminal 30 receives the prediction result from the prediction device 20 and outputs it to the user.
- the overall configuration of the battery life prediction system 2 shown in FIG. 12 is an example, and various system configuration examples are possible depending on the application and purpose.
- the prediction device 20 may be implemented by a plurality of computers or may be implemented as a cloud computing service.
- the prediction device 20 and the user terminal 30 may be realized by a stand-alone computer having the functions that the prediction device 20 and the user terminal 30 should have.
- the prediction device 20 and the user terminal 30 in this embodiment are realized by a computer 500 as shown in FIG. 2, for example.
- FIG. 13 is a block diagram showing an example of the functional configuration of the battery life prediction system 2 in this embodiment.
- the prediction device 20 in this embodiment includes a data reception unit 201, a lifespan acquisition unit 202, a curve generation unit 203, a matrix generation unit 204, a factorization unit 205, a feature amount generation unit 206, A model learning unit 207 , a model storage unit 208 , a lifespan prediction unit 209 , a model analysis unit 210 , a factor identification unit 211 , a result transmission unit 212 and a factor data storage unit 213 are provided.
- Each processing unit (excluding the model storage unit 208 and the factor data storage unit 213) provided in the prediction device 20 is realized by processing that is executed by the CPU 501 from the HDD 504 shown in FIG. .
- the model storage unit 208 and the factor data storage unit 213 included in the prediction device 20 are implemented using, for example, the HDD 504 shown in FIG.
- the data receiving unit 201 receives learning cycle measurement data and prediction cycle measurement data from the user terminal 30 .
- the data receiving unit 201 also sends the received learning cycle measurement data to the life acquisition unit 202 and the curve generation unit 203 .
- the data receiving section 201 sends the received prediction cycle measurement data to the curve generating section 203 .
- the life acquisition unit 202 acquires life data representing battery life characteristics based on the learning cycle measurement data received from the data reception unit 201 .
- the lifespan acquisition unit 202 also sends the acquired lifespan data to the model learning unit 207 .
- the curve generator 203 generates a QV curve or a dQ/dV curve based on the learning cycle measurement data received from the data receiver 201 . Also, the curve generator 203 generates a QV curve or a dQ/dV curve based on the cycle measurement data for prediction received from the data receiver 201 . Further, curve generation section 203 sends the generated QV curve or dQ/dV curve to matrix generation section 204 .
- the matrix generation unit 204 Based on the QV curve or dQ/dV curve received from the curve generation unit 203, the matrix generation unit 204 generates a measurement data matrix representing the relationship between voltage and current capacity for each charge/discharge cycle. Matrix generation section 204 also sends the generated measurement data matrix to factorization section 205 .
- the factorization section 205 performs non-negative matrix factorization on the measurement data matrix received from the matrix generation section 204 . As a result, the factorization unit 205 obtains a factor data matrix and a factor intensity transition data matrix. The factorization unit 205 stores the factor data matrix in the factor data storage unit 213 when performing non-negative matrix factorization on the learning cycle measurement data. On the other hand, when the factorization unit 205 performs non-negative matrix factorization on the prediction cycle measurement data, the factor data matrix stored in the factor data storage unit 213 is used, and the factor strength transition data matrix is: Generated by inverting the stored factor data matrix against the measurement data matrix. Furthermore, in either case, the factorization unit 205 sends the factorization result to the feature quantity generation unit 206 and the result transmission unit 212 .
- the factorization result includes a factor data matrix and a factor intensity transition data matrix.
- the factor data storage unit 213 stores the factor data matrix generated by the factorization unit 205 based on the learning cycle measurement data.
- the feature amount generation unit 206 extracts predetermined feature amounts from the factor strength transition data matrix received from the factorization unit 205 .
- the feature amount generation unit 206 sends the extracted feature amount to the model learning unit 207 .
- the feature amount generation unit 206 sends the extracted feature amount to the life prediction unit 209 .
- the model learning unit 207 adds the lifespan data received from the lifespan acquisition unit 202 to the feature quantity received from the feature quantity generation unit 206 . Thereby, the model learning unit 207 generates learning data. The model learning unit 207 also uses the generated learning data to learn a prediction model that uses the feature quantity as the explanatory variable and the life span data as the objective variable. Furthermore, the model learning unit 207 stores the learned prediction model in the model storage unit 208 .
- the model storage unit 208 stores a prediction model that has learned the relationship between the feature amount and the life data, generated by the model learning unit 207 based on the cycle measurement data.
- the lifespan prediction unit 209 calls the prediction model stored in the model storage unit 208, and inputs the feature quantity received from the feature quantity generation unit 206 into the prediction model. Thereby, the life prediction unit 209 predicts the life characteristics of the battery to be predicted. The life prediction unit 209 also sends the predicted value of the life characteristics to the result transmission unit 212 .
- the model analysis unit 210 analyzes the prediction models stored in the model storage unit 208, and obtains the mode of contribution to prediction of lifespan characteristics for each feature amount.
- the model analysis unit 210 also sends the contribution mode information to the factor identification unit 211 .
- the contribution mode information includes a contribution mode related to each feature amount.
- the factor identification unit 211 Based on the contribution mode information received from the model analysis unit 210, the factor identification unit 211 extracts feature quantities whose contribution modes satisfy predetermined conditions. Further, the factor identifying unit 211 identifies factors corresponding to the extracted feature amounts. Furthermore, the factor identification unit 211 sends important factor information to the result transmission unit 212 .
- the important factor information includes factor data corresponding to the specified factor and the contribution mode of the feature quantity corresponding to the factor.
- the result transmission unit 212 transmits the prediction result of the life characteristics of the battery to be predicted to the user terminal 30 .
- the prediction result includes the predicted value of lifespan characteristics received from the lifespan prediction unit 209 and the important factor information received from the factor identification unit 211 .
- the user terminal 30 in this embodiment has a data input section 301 and a result output section 302 .
- the data input unit 301 accepts input of learning cycle measurement data and prediction cycle measurement data according to the user's operation.
- the data input unit 301 also transmits the received learning cycle measurement data and prediction cycle measurement data to the prediction device 20 .
- the result output unit 302 receives prediction results from the prediction device 20 . Further, the result output unit 302 outputs the received prediction result to the display device 506 or the like.
- the battery life prediction method in this embodiment includes prediction model learning processing and life characteristic prediction processing.
- FIG. 14 is a flowchart showing an example of the processing procedure of prediction model learning processing in this embodiment.
- step S201 the data input unit 301 provided in the user terminal 30 accepts input of learning cycle measurement data according to the user's operation.
- the data input unit 301 transmits the received learning cycle measurement data to the prediction device 20 .
- the data receiving section 201 receives learning cycle measurement data from the user terminal 30 . Next, the data receiving section 201 sends the received learning cycle measurement data to the life acquisition section 202 and the curve generation section 203 .
- step S ⁇ b>202 the life acquisition unit 202 provided in the prediction device 20 receives learning cycle measurement data from the data reception unit 201 .
- the lifespan acquisition unit 202 acquires lifespan data representing lifespan characteristics of the learning target battery based on the learning cycle measurement data.
- the lifespan acquisition unit 202 then sends the acquired lifespan data to the model learning unit 207 .
- step S203 the curve generation unit 203 included in the prediction device 20 receives the learning cycle measurement data from the data reception unit 201.
- the curve generator 203 generates a QV curve or a dQ/dV curve based on the learning cycle measurement data.
- Curve generation section 203 then sends the generated QV curve or dQd/dV curve to matrix generation section 204 .
- step S204 the matrix generation unit 204 included in the prediction device 20 receives the QV curve or the dQ/dV curve from the curve generation unit 203.
- matrix generation section 204 generates a measurement data matrix based on the QV curve or the dQ/dV curve.
- Matrix generation section 204 then sends the generated measurement data matrix to factorization section 205 .
- step S205 the factorization unit 205 included in the prediction device 20 performs non-negative matrix factorization on the measured data matrix received from the matrix generation unit 204 to calculate a factor data matrix and a factor intensity transition data matrix.
- factorization section 205 stores the factor data matrix in factor data storage section 213 .
- the factorization unit 205 sends the factorization result including the factor data matrix and the factor intensity transition data matrix to the feature amount generation unit 206 .
- step S ⁇ b>206 the feature amount generation unit 206 provided in the prediction device 20 receives the factorization result from the factorization unit 205 .
- the feature amount generation unit 206 extracts a predetermined feature amount from the factor strength transition data matrix included in the factorization result.
- the feature quantity generation unit 206 then sends the extracted feature quantity to the model learning unit 207 .
- the feature amount generation unit 206 extracts feature amounts from factor strength transition data up to a predetermined number of prediction execution cycles.
- the feature quantity is the final strength, the final slope, etc. when the factor strength transition data up to the predicted number of execution cycles is plotted on a graph as shown in FIG.
- the feature quantity includes a feature quantity corresponding to each cell of the learning target battery.
- step S ⁇ b>207 the model learning unit 207 included in the prediction device 20 receives feature amounts from the feature amount generation unit 206 .
- the model learning unit 207 also receives lifespan data from the lifespan acquisition unit 202 .
- the model learning unit 207 generates learning data by adding lifespan data to the feature amount.
- the model learning unit 207 uses the generated learning data to learn a prediction model that uses the feature quantity as the explanatory variable and the life span data as the objective variable.
- the model learning unit 207 then stores the learned prediction model in the model storage unit 208 .
- FIG. 15 is a flow chart showing an example of the processing procedure of life characteristic prediction processing in this embodiment.
- step S201 the data input unit 301 provided in the user terminal 30 accepts input of predictive cycle measurement data according to the user's operation.
- the data input unit 301 transmits the received cycle measurement data for prediction to the prediction device 20 .
- the data receiving section 201 receives prediction cycle measurement data from the user terminal 30 . Next, the data receiving section 201 sends the received prediction cycle measurement data to the curve generating section 203 .
- steps S203 and S204 is the same as the prediction model learning processing.
- step S205 the factorization unit 205 included in the prediction device 20 performs non-negative matrix factorization on the measured data matrix received from the matrix generation unit 204 to calculate a factor data matrix and a factor intensity transition data matrix.
- the factor data matrix stored in the factor data storage unit 213 is used.
- the factor intensity transition data matrix is generated by multiplying the measured data matrix by the inverse matrix of the factor data matrix.
- the factorization unit 205 sends the factorization result including the factor data matrix and the factor intensity transition data matrix to the feature amount generation unit 206 .
- step S206 is the same as the prediction model learning processing. However, the feature quantity generation unit 206 sends the extracted feature quantity to the life prediction unit 209 .
- step S ⁇ b>209 the life prediction unit 209 provided in the prediction device 20 calls the prediction model stored in the model storage unit 208 and receives the feature amount from the feature amount generation unit 206 .
- the life prediction unit 209 predicts the life characteristics of the prediction target battery by inputting the received feature quantity into the called prediction model.
- the life prediction unit 209 then sends the predicted value of the life characteristics to the result transmission unit 212 .
- step S210 the model analysis unit 210 included in the prediction device 20 analyzes the prediction models stored in the model storage unit 208, and obtains the contributions to the prediction of lifespan characteristics for each feature amount.
- the model analysis unit 210 also sends contribution mode information including the contribution mode for each feature amount to the factor identification unit 211 .
- the model analysis unit 210 analyzes the prediction model by, for example, referring to regression coefficients and Shapley value analysis.
- contribution mode information which is information representing the mode of contribution to the predicted value, is obtained for each feature amount used for prediction. Since each feature amount corresponds to each factor, it can be said that the contribution mode obtained by the analysis represents the contribution mode for each factor.
- the contribution mode desirably contains quantitative information about the positive and negative effects on the predicted value when the feature value takes a certain value.
- step S ⁇ b>211 the factor identification unit 211 included in the prediction device 20 receives contribution mode information from the model analysis unit 210 .
- the factor identifying unit 211 extracts a feature amount whose contribution mode satisfies a predetermined condition based on the received contribution mode information.
- the factor identification unit 211 identifies factors corresponding to the extracted feature amounts.
- the factor identification unit 211 then sends important factor information about the identified factors to the result transmission unit 212 .
- the important factor information includes factor data corresponding to the extracted feature amount, its contribution mode, and the specified factor.
- Predetermined conditions for the factor identification unit 211 to extract the feature amount are, for example, that the contribution mode is equal to or greater than a predetermined value, that the contribution mode is equal to or less than a predetermined value, or that the absolute value of the contribution mode is greater than a predetermined value. , a predetermined number from the top when the contribution modes are sorted in descending order, or any other conditions based on the contribution modes.
- step S ⁇ b>212 the result transmission unit 212 provided in the prediction device 20 receives the predicted value of life characteristics from the life prediction unit 209 . Also, the result transmission unit 212 receives important factor information from the factor identification unit 211 . Then, the result transmitting unit 212 transmits to the user terminal 30 the prediction result including the predicted value of the life characteristic and the important factor information.
- the result output unit 302 receives the prediction result from the prediction device 20 . Then, the result output unit 302 outputs the received prediction result to the display device 506 or the like.
- FIG. 16 is a diagram showing an example of a prediction result display screen for the user terminal 30 to display the prediction result.
- the prediction result display screen 1000 in this embodiment has a predicted lifespan display column 1001, an important factor display column 1010, and a contribution mode display column 1020.
- the predicted life span display field 1001 displays predicted values of life characteristics included in the prediction result.
- the important factor display column 1010 displays the factor data included in the important factor information in descending order of contribution mode.
- the contribution mode display field 1020 displays the contribution mode included in the important factor information.
- the contribution mode display field 1020 displays the feature quantity and the contribution mode corresponding to each factor data included in the important factor information. For example, in the contribution mode display field 1020 in FIG. 16, for the feature amount A and the feature amount B corresponding to the important factor ⁇ , the value of the feature amount and the plot of the contribution to the predicted life by taking the value are displayed. . In the plot, learning data are indicated by circles, and prediction target batteries are indicated by stars. The plot of the feature amount A shows the mode of contribution that the larger the value of the feature amount, the longer the life span. On the other hand, the plot of the feature amount B shows a contribution mode opposite to that of the feature amount A, that is, the larger the value of the feature amount, the longer the life span.
- the user can obtain the predicted value of the life characteristic of the battery to be predicted, the factor data related to the factor affecting the prediction result, and how the factor contributes to the prediction result. It is possible to obtain information indicating whether the Since the battery component can be identified from the pattern indicated by the factor data, the user can guess the component causing the problem, such as a short predicted cycle life, and take corrective measures. It can be used as a reference for consideration.
- the battery life prediction system 2 uses a prediction model learned using cycle measurement data for learning to predict life characteristics of a battery to be predicted from the cycle measurement data for prediction. At this time, the battery life prediction system 2 outputs to the user a prediction result including information on factors affecting life characteristics. Therefore, according to the battery life prediction system 2 of the present embodiment, when it is predicted that there is a problem with the life characteristics of the battery to be predicted, the user can easily identify the members that affect the life characteristics. , it becomes possible to lead to actions for improving life characteristics.
- the battery life prediction system 2 in this embodiment can predict the life characteristics from the cycle measurement data of the charge/discharge cycle test up to the predicted execution cycle number, so that the cycle measurement data used for prediction can be obtained in a short time. can be done. Therefore, it is possible to shorten the time required to analyze and improve the life characteristics of the battery to be predicted.
- the battery life prediction system 2 in the present embodiment analyzes the contribution mode of each feature value from the learned prediction model, and presents the predicted value of the life characteristic and the factor data related to the factors that strongly affect the prediction result. can do. Therefore, the user can preferentially consider members that strongly affect the service life characteristics, and can quickly identify the cause of problems in the service life characteristics.
- the factorization unit 105 and the factorization unit 205 are examples of calculation units.
- the model analysis unit 210 is an example of an acquisition unit.
- the result transmission unit 106 and the result transmission unit 212 are examples of output units.
- processing circuit means a processor programmed by software to perform each function, such as a processor implemented by an electronic circuit, or a processor designed to perform each function described above.
- ASIC Application Specific Integrated Circuit
- DSP digital signal processor
- FPGA field programmable gate array
- analysis device 10 or prediction device 20 includes multiple computing devices, such as server clusters. Multiple computing devices are configured to communicate with each other over any type of communication link, including a network, shared memory, etc., to perform the processes disclosed herein.
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Chemical & Material Sciences (AREA)
- Electrochemistry (AREA)
- General Chemical & Material Sciences (AREA)
- Computer Hardware Design (AREA)
- Manufacturing & Machinery (AREA)
- Quality & Reliability (AREA)
- Power Engineering (AREA)
- Secondary Cells (AREA)
- Charge And Discharge Circuits For Batteries Or The Like (AREA)
Abstract
Description
前記対象バッテリーのサイクル測定データから算出される電圧と電流容量との関係を因子分解することで、前記電流容量に影響を与える各因子の強度の変化を示す因子強度推移データ及び各因子の電圧と電流容量との関係を表す因子データを算出するように構成された算出部と、
前記寿命データ、前記因子データ及び前記因子強度推移データを出力するように構成された出力部と、
を備える解析装置。
前記電圧と前記電流容量との関係は、前記サイクル測定データから生成されるQV曲線又はdQ/dV曲線に基づいて算出される、
解析装置。
前記算出部は、非負値行列因子分解により前記因子分解を行う、
解析装置。
各因子の前記因子強度推移データから、特徴量を生成するように構成された生成部と、
前記所定サイクルまでの学習用サイクル測定データに基づいて生成された各因子の特徴量と、寿命データとの関係を学習したモデルに、前記生成部により生成された特徴量を入力することで、前記対象バッテリーの寿命特性を予測するように構成された予測部と、
を備える予測装置。
前記寿命特性の予測値と、前記学習用サイクル測定データから算出された所定の因子の前記電圧と前記電流容量との関係を表す因子データと、を出力するように構成された出力部をさらに備える、
予測装置。
前記モデルを解析することで、前記生成部により生成された特徴量それぞれについて、前記対象バッテリーの前記寿命特性の予測に対する寄与態様を取得するように構成された取得部と、
前記生成部により生成された特徴量のうち、前記寄与態様が所定の条件を満たす特徴量を抽出し、対応する因子を特定するように構成された特定部と、
をさらに備え、
前記出力部は、前記対象バッテリーの寿命特性の予測値と、前記特定した因子に関する前記因子データと、前記特定した因子に対応する寄与態様と、を出力する、
予測装置。
対象バッテリーのサイクル測定データから寿命データを取得する取得手順と、
前記対象バッテリーのサイクル測定データから算出される電圧と電流容量との関係を因子分解することで、前記電流容量に影響を与える各因子の強度の変化を示す因子強度推移データ及び各因子の電圧と電流容量との関係を表す因子データを算出する算出手順と、
前記寿命データ、前記因子データ及び前記因子強度推移データを出力する出力手順と、
を実行する解析方法。
所定サイクルまでの対象バッテリーのサイクル測定データから算出される電圧と電流容量との関係を因子分解することで、前記電流容量に影響を与える各因子の強度の、前記所定サイクルまでの変化を示す因子強度推移データを算出する算出手順と、
各因子の前記因子強度推移データから、特徴量を生成する生成手順と、
前記所定サイクルまでの学習用サイクル測定データに基づいて生成された各因子の特徴量と、寿命データとの関係を学習したモデルに、前記生成手順により生成された特徴量を入力することで、前記対象バッテリーの寿命特性を予測する予測手順と、
を実行する予測方法。
対象バッテリーのサイクル測定データから寿命データを取得する取得手順と、
前記対象バッテリーのサイクル測定データから算出される電圧と電流容量との関係を因子分解することで、前記電流容量に影響を与える各因子の強度の変化を示す因子強度推移データ及び各因子の電圧と電流容量との関係を表す因子データを算出する算出手順と、
前記寿命データ、前記因子データ及び前記因子強度推移データを出力する出力手順と、
を実行させるためのプログラム。
所定サイクルまでの対象バッテリーのサイクル測定データから算出される電圧と電流容量との関係を因子分解することで、前記電流容量に影響を与える各因子の強度の、前記所定サイクルまでの変化を示す因子強度推移データを算出する算出手順と、
各因子の前記因子強度推移データから、特徴量を生成する生成手順と、
前記所定サイクルまでの学習用サイクル測定データに基づいて生成された各因子の特徴量と、寿命データとの関係を学習したモデルに、前記生成手順により生成された特徴量を入力することで、前記対象バッテリーの寿命特性を予測する予測手順と、
を実行させるためのプログラム。
本発明の第1実施形態は、バッテリーのサイクル測定データに基づいて、当該バッテリーの寿命特性を解析するバッテリー寿命解析システムである。本実施形態におけるバッテリー寿命解析システムは、解析対象とするバッテリー(以下、「解析対象バッテリー」とも呼ぶ)の長期サイクル測定データ(以下、「解析用サイクル測定データ」とも呼ぶ)から取得した寿命データ及び当該寿命データに影響を与えた要因に関する情報を含む解析結果を出力する。したがって、本実施形態におけるバッテリー寿命解析システムによれば、ユーザは、解析対象バッテリーの寿命特性に問題がある場合に、当該バッテリーの寿命特性に影響を与えている要因を特定することが容易になり、寿命特性を改善するためのアクションに繋げることが可能となる。
まず、本実施形態におけるバッテリー寿命解析システムの全体構成を、図1を参照しながら説明する。図1は、本実施形態におけるバッテリー寿命解析システムの全体構成の一例を示すブロック図である。
次に、本実施形態におけるバッテリー寿命解析システム1のハードウェア構成を、図2を参照しながら説明する。
本実施形態における解析装置10及びユーザ端末30は、例えばコンピュータにより実現される。図2は、本実施形態におけるコンピュータ500のハードウェア構成の一例を示すブロック図である。
続いて、本実施形態におけるバッテリー寿命解析システムの機能構成を、図3を参照しながら説明する。図3は本実施形態におけるバッテリー寿命解析システム1の機能構成の一例を示すブロック図である。
図3に示されているように、本実施形態における解析装置10は、データ受信部101、寿命取得部102、曲線生成部103、行列生成部104、因子分解部105及び結果送信部106を備える。
図3に示されているように、本実施形態におけるユーザ端末30は、データ入力部301及び結果出力部302を備える。
次に、本実施形態におけるバッテリー寿命解析システム1が実行するバッテリー寿命解析方法の処理手順を説明する。図4は本実施形態におけるバッテリー寿命解析方法の処理手順の一例を示すフローチャートである。
本実施形態におけるバッテリー寿命解析システム1は、バッテリーのサイクル測定データを解析し、解析結果をユーザに対して出力する。解析結果には、バッテリーの寿命特性及び当該バッテリーの寿命特性に影響を与える要因に関する情報が含まれる。したがって、本実施形態におけるバッテリー寿命解析システム1によれば、ユーザは、バッテリーの寿命特性に問題がある場合に、寿命特性に影響を与える部材を特定することが容易になり、寿命特性を改善するためのアクションに繋げることが可能となる。
本発明の第2実施形態は、バッテリーのサイクル測定データに基づいて、当該バッテリーの寿命特性を予測するバッテリー寿命予測システムである。本実施形態におけるバッテリー寿命予測システムは、学習対象とするバッテリー(以下、「学習対象バッテリー」とも呼ぶ)の長期サイクル測定データ(以下、「学習用サイクル測定データ」とも呼ぶ)を用いて予測モデルを学習し、当該予測モデルを用いて、予測対象とするバッテリー(以下、「予測対象バッテリー」とも呼ぶ)の初期サイクル測定データ(以下、「予測用サイクル測定データ」とも呼ぶ)から予測対象バッテリーの寿命特性を予測する。このとき、バッテリー寿命予測システムは、予測対象バッテリーの寿命特性の予測値と共に、当該バッテリーの寿命特性に影響を与える要因に関する情報をユーザに対して出力する。したがって、ユーザは、予測対象バッテリーの寿命特性に問題があると予測される場合に、当該バッテリーの寿命特性に影響を与えている要因を特定することが容易になり、寿命特性を改善するためのアクションに繋げることが可能となる。
まず、本実施形態におけるバッテリー寿命予測システムの全体構成を、図12を参照しながら説明する。図12は、本実施形態におけるバッテリー寿命予測システムの全体構成の一例を示すブロック図である。
次に、本実施形態におけるバッテリー寿命予測システム2のハードウェア構成を説明する。本実施形態における予測装置20及びユーザ端末30は、例えば図2に示したようなコンピュータ500により実現される。
続いて、本実施形態におけるバッテリー寿命予測システムの機能構成を、図13を参照しながら説明する。図13は本実施形態におけるバッテリー寿命予測システム2の機能構成の一例を示すブロック図である。
図13に示されているように、本実施形態における予測装置20は、データ受信部201、寿命取得部202、曲線生成部203、行列生成部204、因子分解部205、特徴量生成部206、モデル学習部207、モデル記憶部208、寿命予測部209、モデル解析部210、因子特定部211、結果送信部212、及び因子データ記憶部213を備える。
図13に示されているように、本実施形態におけるユーザ端末30は、データ入力部301及び結果出力部302を備える。
次に、本実施形態におけるバッテリー寿命予測システム2が実行するバッテリー寿命予測方法の処理手順を説明する。本実施形態におけるバッテリー寿命予測方法は、予測モデル学習処理及び寿命特性予測処理からなる。
図14は本実施形態における予測モデル学習処理の処理手順の一例を示すフローチャートである。
図15は本実施形態における寿命特性予測処理の処理手順の一例を示すフローチャートである。
本実施形態におけるバッテリー寿命予測システム2は、学習用サイクル測定データを用いて学習した予測モデルを用いて、予測用サイクル測定データから予測対象バッテリーの寿命特性を予測する。このとき、バッテリー寿命予測システム2は、寿命特性に影響を与える要因に関する情報を含む予測結果を、ユーザに対して出力する。したがって、本実施形態におけるバッテリー寿命予測システム2によれば、ユーザは、予測対象バッテリーの寿命特性に問題があると予測される場合に、寿命特性に影響を与える部材を特定することが容易になり、寿命特性を改善するためのアクションに繋げることが可能となる。
上記各実施形態において、因子分解部105及び因子分解部205は算出部の一例である。モデル解析部210は取得部の一例である。結果送信部106及び結果送信部212は出力部の一例である。
2 バッテリー寿命予測システム
10 解析装置
20 予測装置
30 ユーザ端末
101、201 データ受信部
102、202 寿命取得部
103、203 曲線生成部
104、204 行列生成部
105、205 因子分解部
206 特徴量生成部
207 モデル学習部
208 モデル記憶部
209 寿命予測部
210 モデル解析部
211 因子特定部
106、212 結果送信部
213 因子データ記憶部
301 データ入力部
302 結果出力部
Claims (10)
- 対象バッテリーのサイクル測定データから寿命データを取得するように構成された取得部と、
前記対象バッテリーのサイクル測定データから算出される電圧と電流容量との関係を因子分解することで、前記電流容量に影響を与える各因子の強度の変化を示す因子強度推移データ及び各因子の電圧と電流容量との関係を表す因子データを算出するように構成された算出部と、
前記寿命データ、前記因子データ及び前記因子強度推移データを出力するように構成された出力部と、
を備える解析装置。 - 請求項1に記載の解析装置であって、
前記電圧と前記電流容量との関係は、前記サイクル測定データから生成されるQV曲線又はdQ/dV曲線に基づいて算出される、
解析装置。 - 請求項2に記載の解析装置であって、
前記算出部は、非負値行列因子分解により前記因子分解を行う、
解析装置。 - 所定サイクルまでの対象バッテリーのサイクル測定データから算出される電圧と電流容量との関係を因子分解することで、前記電流容量に影響を与える各因子の強度の、前記所定サイクルまでの変化を示す因子強度推移データを算出するように構成された算出部と、
各因子の前記因子強度推移データから、特徴量を生成するように構成された生成部と、
前記所定サイクルまでの学習用サイクル測定データに基づいて生成された各因子の特徴量と、寿命データとの関係を学習したモデルに、前記生成部により生成された特徴量を入力することで、前記対象バッテリーの寿命特性を予測するように構成された予測部と、
を備える予測装置。 - 請求項4に記載の予測装置であって、
前記寿命特性の予測値と、前記学習用サイクル測定データから算出された所定の因子の前記電圧と前記電流容量との関係を表す因子データと、を出力するように構成された出力部をさらに備える、
予測装置。 - 請求項5に記載の予測装置であって、
前記モデルを解析することで、前記生成部により生成された前記特徴量それぞれについて、前記対象バッテリーの前記寿命特性の予測に対する寄与態様を取得するように構成された取得部と、
前記生成部により生成された前記特徴量のうち、前記寄与態様が所定の条件を満たす前記特徴量を抽出し、対応する因子を特定するように構成された特定部と、
をさらに備え、
前記出力部は、前記対象バッテリーの前記寿命特性の予測値と、前記特定した因子に関する前記因子データと、前記特定した因子に対応する前記寄与態様と、を出力する、
予測装置。 - コンピュータが、
対象バッテリーのサイクル測定データから寿命データを取得する取得手順と、
前記対象バッテリーのサイクル測定データから算出される電圧と電流容量との関係を因子分解することで、前記電流容量に影響を与える各因子の強度の変化を示す因子強度推移データ及び各因子の電圧と電流容量との関係を表す因子データを算出する算出手順と、
前記寿命データ、前記因子データ及び前記因子強度推移データを出力する出力手順と、
を実行する解析方法。 - コンピュータが、
所定サイクルまでの対象バッテリーのサイクル測定データから算出される電圧と電流容量との関係を因子分解することで、前記電流容量に影響を与える各因子の強度の、前記所定サイクルまでの変化を示す因子強度推移データを算出する算出手順と、
各因子の前記因子強度推移データから、特徴量を生成する生成手順と、
前記所定サイクルまでの学習用サイクル測定データに基づいて生成された各因子の特徴量と、寿命データとの関係を学習したモデルに、前記生成手順により生成された特徴量を入力することで、前記対象バッテリーの寿命特性を予測する予測手順と、
を実行する予測方法。 - コンピュータに、
対象バッテリーのサイクル測定データから寿命データを取得する取得手順と、
前記対象バッテリーのサイクル測定データから算出される電圧と電流容量との関係を因子分解することで、前記電流容量に影響を与える各因子の強度の変化を示す因子強度推移データ及び各因子の電圧と電流容量との関係を表す因子データを算出する算出手順と、
前記寿命データ、前記因子データ及び前記因子強度推移データを出力する出力手順と、
を実行させるためのプログラム。 - コンピュータに、
所定サイクルまでの対象バッテリーのサイクル測定データから算出される電圧と電流容量との関係を因子分解することで、前記電流容量に影響を与える各因子の強度の、前記所定サイクルまでの変化を示す因子強度推移データを算出する算出手順と、
各因子の前記因子強度推移データから、特徴量を生成する生成手順と、
前記所定サイクルまでの学習用サイクル測定データに基づいて生成された各因子の特徴量と、寿命データとの関係を学習したモデルに、前記生成手順により生成された特徴量を入力することで、前記対象バッテリーの寿命特性を予測する予測手順と、
を実行させるためのプログラム。
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US18/555,121 US20240201265A1 (en) | 2021-12-07 | 2022-11-29 | Analyzing device, predicting device, analyzing method, predicting method, and program |
CN202280036875.XA CN117441108A (zh) | 2021-12-07 | 2022-11-29 | 解析装置、预测装置、解析方法、预测方法和程序 |
JP2023517407A JP7298802B1 (ja) | 2021-12-07 | 2022-11-29 | 解析装置、予測装置、解析方法、予測方法及びプログラム |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2021198746 | 2021-12-07 | ||
JP2021-198746 | 2021-12-07 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2023106166A1 true WO2023106166A1 (ja) | 2023-06-15 |
Family
ID=86730295
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2022/043969 WO2023106166A1 (ja) | 2021-12-07 | 2022-11-29 | 解析装置、予測装置、解析方法、予測方法及びプログラム |
Country Status (4)
Country | Link |
---|---|
US (1) | US20240201265A1 (ja) |
JP (1) | JP7298802B1 (ja) |
CN (1) | CN117441108A (ja) |
WO (1) | WO2023106166A1 (ja) |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2018036131A (ja) * | 2016-08-31 | 2018-03-08 | 株式会社日産アーク | 構造複合体の状態推定方法及びシステム |
JP2018084549A (ja) * | 2016-11-25 | 2018-05-31 | 本田技研工業株式会社 | 二次電池の状態推定装置及び二次電池の状態推定方法 |
JP6488105B2 (ja) | 2014-10-28 | 2019-03-20 | 株式会社東芝 | 蓄電池評価装置及び方法 |
JP2019113524A (ja) | 2017-10-17 | 2019-07-11 | ザ ボード オブ トラスティーズ オブ ザ レランド スタンフォード ジュニア ユニバーシティー | リチウムイオン電池の容量低下と寿命予測のためのデータ駆動モデル |
JP2019160072A (ja) * | 2018-03-15 | 2019-09-19 | 沖電気工業株式会社 | 情報処理装置、情報処理方法、及びプログラム |
JP2020046420A (ja) * | 2018-09-14 | 2020-03-26 | トヨタ自動車株式会社 | 二次電池システムおよび二次電池の劣化状態推定方法 |
US20210325470A1 (en) * | 2020-04-13 | 2021-10-21 | Samsung Electronics Co., Ltd. | Battery management system and method for determining active material content in electrode of battery |
JP2022085385A (ja) * | 2020-11-27 | 2022-06-08 | 株式会社東芝 | 電池の劣化判定装置、電池の管理システム、電池搭載機器、電池の劣化判定方法、及び、電池の劣化判定プログラム |
-
2022
- 2022-11-29 WO PCT/JP2022/043969 patent/WO2023106166A1/ja active Application Filing
- 2022-11-29 CN CN202280036875.XA patent/CN117441108A/zh active Pending
- 2022-11-29 US US18/555,121 patent/US20240201265A1/en active Pending
- 2022-11-29 JP JP2023517407A patent/JP7298802B1/ja active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP6488105B2 (ja) | 2014-10-28 | 2019-03-20 | 株式会社東芝 | 蓄電池評価装置及び方法 |
JP2018036131A (ja) * | 2016-08-31 | 2018-03-08 | 株式会社日産アーク | 構造複合体の状態推定方法及びシステム |
JP2018084549A (ja) * | 2016-11-25 | 2018-05-31 | 本田技研工業株式会社 | 二次電池の状態推定装置及び二次電池の状態推定方法 |
JP2019113524A (ja) | 2017-10-17 | 2019-07-11 | ザ ボード オブ トラスティーズ オブ ザ レランド スタンフォード ジュニア ユニバーシティー | リチウムイオン電池の容量低下と寿命予測のためのデータ駆動モデル |
JP2019160072A (ja) * | 2018-03-15 | 2019-09-19 | 沖電気工業株式会社 | 情報処理装置、情報処理方法、及びプログラム |
JP2020046420A (ja) * | 2018-09-14 | 2020-03-26 | トヨタ自動車株式会社 | 二次電池システムおよび二次電池の劣化状態推定方法 |
US20210325470A1 (en) * | 2020-04-13 | 2021-10-21 | Samsung Electronics Co., Ltd. | Battery management system and method for determining active material content in electrode of battery |
JP2022085385A (ja) * | 2020-11-27 | 2022-06-08 | 株式会社東芝 | 電池の劣化判定装置、電池の管理システム、電池搭載機器、電池の劣化判定方法、及び、電池の劣化判定プログラム |
Also Published As
Publication number | Publication date |
---|---|
JP7298802B1 (ja) | 2023-06-27 |
US20240201265A1 (en) | 2024-06-20 |
CN117441108A (zh) | 2024-01-23 |
JPWO2023106166A1 (ja) | 2023-06-15 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP7317484B2 (ja) | リチウムイオン電池の容量低下と寿命予測のためのデータ駆動モデル | |
Lucu et al. | Data-driven nonparametric Li-ion battery ageing model aiming at learning from real operation data–Part A: Storage operation | |
JP2022514992A (ja) | バッテリ診断システム、バッテリ診断方法、及び記憶媒体 | |
CN114371409B (zh) | 电池状态预测模型的训练方法、电池状态预测方法及装置 | |
Wang et al. | Near-real-time parameter estimation of an electrical battery model with multiple time constants and SOC-dependent capacitance | |
CN111695301A (zh) | 电池电荷状态的预测方法及预测装置、存储介质、设备 | |
WO2014185163A1 (ja) | 電池状態推定装置、電池状態管理システム、電池、電池状態推定方法、及び、プログラム | |
Sun et al. | A sequential capacity estimation for the lithium-ion batteries combining incremental capacity curve and discrete Arrhenius fading model | |
CN113657661A (zh) | 一种企业碳排放预测方法、装置、计算机设备和存储介质 | |
KR20180134336A (ko) | 전기 열 배터리 모델의 생성 및 사용을 위한 시스템 및 방법 | |
WO2023184824A1 (zh) | 电池簇荷电状态的估计方法及系统、电子设备及存储介质 | |
WO2024109760A1 (zh) | 一种电池寿命预测方法、预测系统及计算机设备 | |
Che et al. | Battery health prognostic with sensor-free differential temperature voltammetry reconstruction and capacity estimation based on multi-domain adaptation | |
CN111753461B (zh) | 潮汐水位修正方法、目标余水位获取方法、装置和设备 | |
US20230258733A1 (en) | Predicting aging of batteries | |
Qian et al. | State of health estimation of lithium-ion battery using energy accumulation-based feature extraction and improved relevance vector regression | |
CN116338470A (zh) | 电池健康评估方法、装置、电子设备及可读存储介质 | |
CN113487019A (zh) | 电路故障诊断方法、装置、计算机设备和存储介质 | |
Fan et al. | Data-driven identification of lithium-ion batteries: A nonlinear equivalent circuit model with diffusion dynamics | |
Kwon et al. | Joint state-of-health and remaining-useful-life prediction based on multi-level long short-term memory model prognostic framework considering cell voltage inconsistency reflected health indicators | |
Ren et al. | High precision implicit function learning for forecasting supercapacitor state of health based on Gaussian process regression | |
WO2023106166A1 (ja) | 解析装置、予測装置、解析方法、予測方法及びプログラム | |
WO2020255794A1 (ja) | 開発支援装置、開発支援方法、および状態検知方法 | |
CN117330987A (zh) | 基于时间的电池健康状态评估的方法、系统、介质和设备 | |
CN114664392B (zh) | 电化学参数预测方法、装置、电子设备及可读存储介质 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 2023517407 Country of ref document: JP |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 22904097 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 18555121 Country of ref document: US |
|
WWE | Wipo information: entry into national phase |
Ref document number: 202280036875.X Country of ref document: CN |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2022904097 Country of ref document: EP |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
ENP | Entry into the national phase |
Ref document number: 2022904097 Country of ref document: EP Effective date: 20240708 |