Information processing device, information processing method, program, and information processing system

Abstract

An information processing device according to the present invention comprises a processing unit. The processing unit: selects a plurality of pieces of common data that are associated with feature information that is the same or similar to designated first feature information from a plurality of pieces of common data that are time-series data that indicate changes in the charging / discharging of a battery and are associated with different feature information that indicates features of the charging / discharging; and synthesizes the selected plurality of pieces of common data to generate first time-series data for the first feature information.

Description

Information processing device, information processing method, program, and information processing system

[0001] The present invention relates to an information processing device, an information processing method, a program, and an information processing system.

[0002] The use of secondary batteries is increasing to stabilize power systems and reduce exhaust gas emissions. Monitoring the degradation state of battery systems is required to prevent sudden failures. For example, methods have been proposed to determine the degradation state from data obtained when special charging and discharging is performed, to calculate an index related to the degradation state using an approximation formula based on the number of charging and discharging cycles (Arrhenius approximation formula), and to determine the degradation state from data obtained during normal operation using an estimation model. The estimation model is trained using training data appropriate for the application, for example.

[0003] Furthermore, for example, when designing a battery system, evaluation indicators for the battery, such as the amount of current that can be used by the battery according to operating conditions and charge / discharge efficiency for carbon footprint purposes, may be calculated.

[0004] International Publication No. 2021 / 044635 Patent No. 6313502 Patent No. 6759466 Patent No. 7214884

[0005] An object of the present invention is to provide an information processing device, an information processing method, a program, and an information processing system that can more efficiently obtain data used for functions such as monitoring the state of a battery and evaluating the battery.

[0006] According to an embodiment, an information processing apparatus includes a processing unit that selects, from a plurality of common data sets each having associated thereto different characteristic information indicating characteristics of charging and discharging, a plurality of common data sets each having associated thereto characteristic information that is the same as or similar to specified first characteristic information, and generates first time-series data for the first characteristic information by combining the selected plurality of common data sets.

[0007] 1 is a block diagram of an information processing system according to an embodiment. FIG. 1 is a diagram showing an example of the configuration of a battery. FIG. 2 is a diagram showing an example of the configuration of a module. FIG. 2 is a diagram showing an example of operation data. FIG. 3 is a diagram showing an example of common data. A flowchart of estimation processing according to a first embodiment. A diagram for explaining an example of common data synthesis. A diagram for explaining an example of feature information for which common data does not exist. A diagram showing data interpolation. A diagram showing an example of a method of data interpolation. A diagram showing an example of generated time series data. A diagram showing an example of generated time series data. A diagram showing an example of an output screen for outputting estimation results. A diagram showing an example of an output screen for displaying warning information. A block diagram of an information processing device according to a second embodiment. A flowchart of evaluation processing according to the second embodiment. A diagram for explaining an example of evaluation processing. A hardware configuration diagram of an information processing device according to an embodiment.

[0008] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of an information processing apparatus according to the present invention will be described in detail below with reference to the accompanying drawings.

[0009] The technology for determining the state of deterioration using the estimation model described above can estimate the state of health (SoH) from data measured during operation, eliminating the need to shut down the battery. This makes it possible to anticipate applications for primary screening to detect abnormalities, making it a necessary technology for monitoring battery systems.

[0010] On the other hand, technologies that use estimation models require the preparation of training data according to the application when deriving the estimation model. For example, estimation of a battery used in an electric vehicle (EV) bus requires training data constructed from charging and discharging patterns simulating an EV bus. Therefore, new data must be acquired according to the application, which increases the time and processing load.

[0011] In the following embodiment, data suited to the characteristics (applications) of the operational data to be estimated is generated (synthesized) from data common to multiple applications (common data). The generated data can be used as training data according to the application. This eliminates the need to prepare training data according to the application in advance. In other words, it becomes possible to more efficiently obtain data (training data) used to monitor the state of the battery.

[0012] The generated data can also be used to calculate battery evaluation indices. This means that data used for battery evaluation can be obtained more efficiently. Evaluation indices include, for example, the amount of current that can be used by the battery depending on the operating conditions, and charge / discharge efficiency for carbon footprint purposes.

[0013] First Embodiment An information processing apparatus according to a first embodiment generates training data used for training an estimation model for estimating a battery state from common data.

[0014] 1 is a block diagram showing an example of the configuration of an information processing system 10 according to this embodiment. As shown in FIG. 1, the information processing system 10 includes an information processing device 100, a battery 200, and a monitoring system 300.

[0015] The information processing device 100 and the battery 200, and the information processing device 100 and the monitoring system 300 are connected via a network. The network connecting the information processing device 100 and the battery 200 and the network connecting the information processing device 100 and the monitoring system 300 may be the same network or different networks. The network may be a wireless network, a wired network, or a mixed network of wireless and wired networks.

[0016] Battery 200 is a battery capable of charging and discharging electrical energy. Battery 200 may have any configuration as long as it is possible to acquire operational data for estimating the state of health. The operational data may include, for example, the voltage, current, temperature, and humidity of each parallel circuit constituting battery 200.

[0017] Battery 200 may be a battery mounted on a mobile object that operates using electrical energy as a power source. Examples of mobile objects include electric vehicles (EVs), electric buses, trains, next-generation light rail transit (LRT) systems, bus rapid transit (BRT) systems, automated guided vehicles (AGVs), airplanes, and ships. Battery 200 may be a battery component mounted on an electrical device (such as a smartphone or a personal computer), or a battery that supplies and receives power for demand response. Battery 200 may also be a battery for other purposes.

[0018] Battery 200 is charged by a charger located at a charging stand, roadside, parking lot, etc., or a charger connected to an outlet, etc. The power stored in battery 200 may be discharged (reverse power flow) to the power grid via the charger. The method of transmitting power from the charger to battery 200 may be either a contact charging method or a contactless charging method.

[0019] An example configuration of the battery 200 will be described below. The battery 200 is not limited to this example configuration. The battery 200 includes multiple battery panels. The multiple battery panels are connected in series or parallel. Alternatively, the multiple battery panels are connected in series and parallel. The battery may also be a single battery.

[0020] FIG. 2 is a diagram showing an example configuration of battery 200. Battery 200 includes multiple battery panels 201-1 to 201-N (N is an integer of 2 or greater). Each of battery panels 201-1 to 201-N includes multiple modules 202-1 to 202-M (M is an integer of 2 or greater). Because the multiple battery panels 201-1 to 201-N have the same configuration, they will be simply referred to as battery panels 201 when there is no need to distinguish them. Because the multiple modules 202-1 to 202-M have the same configuration, they will be simply referred to as modules 202 when there is no need to distinguish them. Note that the configuration shown in FIG. 2 is just an example, and battery 200 may have any other configuration.

[0021] The modules 202 are connected in series, in parallel, or in series and parallel. In the example of Fig. 2, the number of modules provided on each battery panel 201 is the same, but it is not necessary that they are the same.

[0022] Fig. 3 is a diagram showing an example of the configuration of one module 202. The module 202 includes a plurality of battery cells 203. The module 202 may include a temperature sensor and a CMU (Cell Monitoring Unit). The plurality of battery cells 203 are connected in series, in parallel, or in series and parallel. In the example of Fig. 3, the plurality of battery cells 203 are connected in series and parallel.

[0023] Returning to FIG. 1 , the monitoring system 300 monitors the battery 200 based on information indicating the state of the battery 200 (hereinafter, state information) provided by the information processing device 100. For example, the monitoring system 300 generates screen data to be used for monitoring and displays the generated screen data on a monitor. A user (monitor) can understand the state of the battery 200 being monitored by referring to the screen displayed on the monitor. The monitoring system 300 may control the operation of the battery 200 according to the monitoring results or according to a command from the user.

[0024] The information processing device 100 includes a storage unit 131 , an acquisition unit 101 , an output control unit 102 , an addition unit 103 , a feature calculation unit 111 , a generation unit 112 , a learning unit 121 , and an estimation unit 122 .

[0025] The storage unit 131 stores various types of information used by the information processing device 100. For example, the storage unit 131 stores common data, operation data acquired (input) from the battery 200, the health level estimated by the estimation unit 122, and information (intermediate products) obtained during the estimation.

[0026] The storage unit 131 can be configured from any commonly used storage medium such as a flash memory, a memory card, a RAM (Random Access Memory), a HDD (Hard Disk Drive), or an optical disk.

[0027] The acquisition unit 101 acquires various types of information used in the information processing device 100. For example, the acquisition unit 101 acquires operation data of the battery 200 from the battery 200. The operation data is, for example, time-series data including a measurement time (the time when the data was measured), a voltage, and a current. The operation data may be acquired at regular time intervals (for example, one second) or irregularly. The acquisition unit 101 stores the acquired operation data in the storage unit 131. The operation data may be acquired for each cell, module, battery panel, or battery 200 (multiple battery panels connected to each other). The following description assumes that the operation data is acquired for each battery 200.

[0028] The operation data includes, for example, battery identification information (battery ID), time, charge amount (SoC: State of Charge), power, and temperature information. Current may be acquired instead of power. In this case, the power value may be calculated by calculation from the current value and the voltage value. SoC is an index indicating the charge amount of the battery. For example, SoC is calculated by dividing the amount of power (unit: Wh) or amount of charge (unit: Ah) stored in the battery 200 by the rated capacity (amount of power or amount of charge) of the battery 200. Instead of acquiring the SoC, the SoC may be calculated by integrating the current. FIG. 4 is a diagram showing an example of operation data. As shown in FIG. 4, the operation data includes information on the battery ID, time, SoC, voltage, power, and temperature.

[0029] The operation data may include information indicating the operating state of the battery, such as charging, discharging, system operation (operating), and system non-operating (non-operating). If the health level is measured, the operation data may be associated with the health level value. Whether the battery is charging or discharging may be distinguished by whether the current or power value is positive or negative, or by information indicating whether the battery is charging or discharging (e.g., a charge / discharge flag).

[0030] The output control unit 102 controls the output of various information used in the information processing device 100. For example, the output control unit 102 outputs the estimation result by the estimation unit 122 to the monitoring system 300.

[0031] The adding unit 103 adds new common data to the storage unit 131. For example, the adding unit 103 adds data obtained by associating the characteristic information (characteristic amount) calculated by the characteristic calculation unit 111 with the operational data acquired by the acquisition unit 101 as new common data to the common data stored in the storage unit 131. Note that the function of adding common data is not essential. That is, the information processing device 100 does not need to include the adding unit 103.

[0032] Here, the common data will be described. The common data is time-series data (charge / discharge data) that indicates changes in charge and discharge and is common to multiple applications. The common data is, for example, data including operation data and a health level obtained when an experiment is conducted on charging and discharging the battery 200 so that the value of the characteristic information becomes discrete. Different pieces of characteristic information are associated with each other in the common data. For example, the common data is associated with the value of the characteristic information at the time of the experiment. The health level is estimated from the operation data during the experiment, for example. For the estimation, an estimation model that has been trained in advance to input operation data and output the health level may be used. It is desirable to conduct the experiment so that the value of the health level also becomes discrete.

[0033] 1 shows one battery 200, but the battery may include a plurality of batteries 200. In this case, the common data may be defined for each of the plurality of batteries 200 (for each type of battery).

[0034] 5 is a diagram showing an example of common data when the charge rate and the discharge rate are used as characteristic information. The charge rate and the discharge rate are sometimes collectively referred to as the C rate. The C rate is calculated, for example, by normalizing (dividing) the current value by the rated current, or by normalizing (dividing) the power value calculated by multiplying the voltage and the current by the rated power value.

[0035] 5 shows that the operating data obtained when a charging / discharging experiment is conducted so that the charging rate and discharging rate are the same is used as common data. For example, operating data obtained when the charging rate and discharging rate are both 0.1C, 1 / 3C, 2 / 3C, 1C, 2C, and 3C are used as multiple common data. Hereinafter, common data obtained when the charging rate and discharging rate are both 0.1C may be referred to as common data with a C rate of 0.1C. The same applies to cases where the C rate is other values.

[0036] Operational data when the charge rate value and the discharge rate value are different is generated by synthesizing (combining) multiple common data. For example, operation data simulating application PA, in which the charge rate is 3C and the discharge rate is 1C, is generated from data during charging among the common data with a C rate of 3C and data during discharging among the common data with a C rate of 1C. Similarly, operation data simulating application PB, in which the charge rate is 1C and the discharge rate is 0.1C, is generated from data during charging among the common data with a C rate of 1C and data during discharging among the common data with a C rate of 0.1C. The same applies to other combinations of charge rate and discharge rate.

[0037] 1 , the feature calculation unit 111 calculates feature information of the operational data from the operational data. For example, the feature calculation unit 111 calculates feature information of the operational data (first feature information) using the operational data of the battery 200 (first battery) whose state is to be estimated.

[0038] The characteristic information may be any information, for example, data indicating the characteristics of charging and discharging of the battery 200. For example, the characteristic information may be some or all of the statistical values ​​of measurement values ​​such as power, charge rate, discharge rate, and temperature. The statistical values ​​may be, for example, average values ​​and values ​​indicating the degree of variation (such as variance). The characteristic information may be calculated for the entire operation data, or may be calculated separately for charging and discharging.

[0039] In addition to the above, the characteristic information may further include the following information: A cycle of switching between charging and discharging Profile information of the battery 200, such as whether calendar deterioration or cycle deterioration has occurred.

[0040] The profile information may be provided as an input or may be estimated from historical operational data.

[0041] The generation unit 112 generates time-series data DA (first time-series data) for feature information FA designated as the target for data generation by combining a plurality of common data associated with feature information identical or similar to feature information FA. For example, the generation unit 112 selects a plurality of common data associated with feature information identical or similar to feature information FA from the plurality of common data stored in the storage unit 131, and generates time-series data DA by combining the selected plurality of common data. For example, feature information calculated by the feature calculation unit 111 is designated as feature information FA.

[0042] The learning unit 121 uses the time-series data DA generated using the feature information FA to learn an estimation model for estimating the state (health level, etc.) of the battery 200. The time-series data DA is stored, for example, in association with the health level. The learning unit 121 learns (constructs, generates) an estimation model for estimating the health level using the health level associated with the time-series data DA as a target variable and variables calculated from the time-series data DA as explanatory variables.

[0043] The estimation model and the learning method using the estimation model may be any model and method, but for example, the same models and methods as those disclosed in Patent Documents 3 and 4 may be used.

[0044] The estimation unit 122 estimates the state of the battery 200 using the learned estimation model.

[0045] At least some of the above units (the acquisition unit 101, the output control unit 102, the addition unit 103, the feature calculation unit 111, the generation unit 112, the learning unit 121, and the estimation unit 122) may be realized by a single processing unit. The above units are realized, for example, by one or more processors. For example, the above units may be realized by software, such as by having a processor such as a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit) execute a program. The above units may be realized by hardware, such as a processor with a dedicated IC (Integrated Circuit). The above units may be realized by a combination of software and hardware. When multiple processors are used, each processor may realize one of the units, or two or more of the units.

[0046] The information processing device 100 may be physically configured as a single device or may be physically configured as multiple devices. For example, the information processing device 100 may be built in a cloud environment. Furthermore, each unit within the information processing device 100 may be distributed across multiple devices.

[0047] Next, the estimation process performed by the information processing device 100 according to the first embodiment will be described. Fig. 6 is a flowchart showing an example of the estimation process according to the first embodiment. The estimation process includes a process of learning an estimation model using training data generated from common data, and a process of estimating the state of the battery 200 using the learned estimation model.

[0048] The acquisition unit 101 acquires operation data of the battery 200 whose state is to be estimated (step S101). The feature calculation unit 111 calculates feature information of the acquired operation data (step S102).

[0049] The generation unit 112 generates charge / discharge data by combining the common data according to the characteristic information (step S103). This makes it possible to generate time-series data (charge / discharge data) that indicates changes in charge / discharge that are suited to the characteristic information of the operation data. The generation process by the generation unit 112 will be described in detail later.

[0050] The learning unit 121 learns (constructs) an estimation model using the generated charge / discharge data (step S104). The estimation unit 122 estimates the state of health using the learned estimation model (step S105). The output control unit 102 outputs the estimation result by the estimation unit 122 (step S106) and ends the estimation process.

[0051] Next, the details of the generation process will be described. In the following, a case where the characteristic information is a C rate (charge rate, discharge rate) will be described, but the same procedure can be applied to other characteristic information.

[0052] For example, suppose that charge / discharge data for application PA is specified to be generated with a charge rate of 3 C and a discharge rate of 1 C. In this case, the generation unit 112 generates operation data simulating application PA by combining data during charging among the common data with a C rate of 3 C and data during discharging among the common data with a C rate of 1 C.

[0053] When the common data includes the SoC, the generator 112 combines multiple pieces of common data using the SoC. For example, the generator 112 determines, among the common data with a C rate of 3C, data from when the SoC increases from the minimum value to the maximum value SoC_Max as charging data. The generator 112 determines, among the common data with a C rate of 1C, data from when the SoC decreases from a value equal to SoC_Max to the minimum value as discharging data. The generator 112 generates operational data simulating application PA by combining the charging data with the discharging data. Note that the time included in the generated operational data may be changed, for example, so that the charging data follows the discharging data.

[0054] When combining the charging data and the discharging data as described above, it is desirable to acquire the charging data from a fully discharged state, and the discharging data from a fully charged state.

[0055] If the common data does not include the SoC, the generator 112 may estimate the SoC from the integrated current value or the voltage value and use the estimated SoC. The SoC estimation method may be any conventional method, such as a method using a Kalman filter.

[0056] Fig. 7 is a diagram illustrating an example of combining common data. The upper part of Fig. 7 shows an example of a graph showing charge / discharge characteristics for each of three common data. The graph showing charge / discharge characteristics in Fig. 7 is data (QV data) showing the relationship between SoC and voltage. The C rates of the three common data are 0.1C, 1C, and 3C, respectively.

[0057] For the purpose PA, the generation unit 112 selects, from the three pieces of common data, the common data with a C rate of 3C and the common data with a C rate of 1C. In the example of Fig. 7, the common data with a C rate of 3C corresponds to the graph 701, and the common data with a C rate of 1C corresponds to the graph 702.

[0058] The generation unit 112 generates operation data for application PA by combining the data during charging among the common data with a C rate of 3 C and the data during discharging among the common data with a C rate of 1 C. Graph 711 at the bottom of Figure 7 corresponds to the graph corresponding to the generated operation data.

[0059] If there is no common data of feature information matching the value of the feature information FA for the specified purpose, the generator 112 generates the time-series data DA by interpolation, including at least one of interpolation and extrapolation, using a plurality of common data. The common data used for interpolation corresponds to the common data associated with feature information similar to the feature information FA.

[0060] For example, as shown in FIG. 8 , assume that a specific PC is designated with a charge rate of 1.5C and a discharge rate of 0.2C. In this case, the generation unit 112 generates data for charging at a C rate of 1.5C by interpolating data for charging among the common data with a C rate of 1C and data for charging among the common data with a C rate of 2C. The generation unit 112 also generates data for discharging at a C rate of 0.2C by interpolating data for discharging among the common data with a C rate of 0.1C and data for discharging among the common data with a C rate of 1 / 3C. The generation unit 112 then combines the generated data for charging and discharging to generate data as operating data simulating the specific PC. FIG. 9 illustrates the data interpolation process.

[0061] 10 is a diagram showing an example of a specific data interpolation method. As shown in Fig. 10, the generator 112 prepares a graph in which the horizontal axis represents SoC (or time) and the vertical axis represents a desired item such as voltage or current.

[0062] Graph 1011 corresponds to data when charging at a C rate of 2 C. Graph 1012 corresponds to data when charging at a C rate of 1 C. Note that the dashed and solid lines are shown for convenience of explanation, and it is not necessary to obtain curves equivalent to the dashed and solid lines.

[0063] The generation unit 112 calculates an interior division point 1003 between a point 1001 included in the graph 1011 and a corresponding point 1002 included in the graph 1012. For example, when interpolating data of a C rate of 1.5C from data of C rates 2C and 1C, the generation unit 112 calculates the point that divides the points 1001 and 1002 internally in the ratio (2-1.5):(1.5-1) as the interior division point 1003. The generation unit 112 performs this process for each point included in the graphs 1011 and 1012. As a result, an interpolation result such as that shown in the graph 1013 of FIG. 10 is obtained.

[0064] If the C rate for the specified purpose is outside the range of the C rates of the common data (0.1C to 3C in the example of FIG. 5 ), the generator 112 generates time-series data DA by extrapolating from two pieces of common data corresponding to C rates close to the C rate for the specified purpose. For example, the generator 112 calculates the external division point between a point included in one piece of common data and a corresponding point included in the other piece of common data.

[0065] Since calculations by extrapolation (external division points) may result in large errors, it is desirable to output a warning indicating that the calculation was extrapolated. For this reason, the generation unit 112, for example, stores extrapolation information indicating that extrapolation has been performed in association with the generated data. When the extrapolation information is associated with the data, the output control unit 102 may output output information (such as a warning) indicating that the data was generated by extrapolation when outputting the data. The extrapolation information may include a value (such as a flag) indicating whether or not extrapolation has been performed, as well as information indicating which item (such as charge rate, discharge rate, or temperature) was extrapolated.

[0066] As described above, the generation unit 112 generates the time series data DA by combining the time series data portion indicating changes in charging among the common data having the same or similar charging rates included in the specified characteristic information FA, and the time series data portion indicating changes in discharging among the common data having the same or similar discharging rates included in the characteristic information FA.

[0067] If data that is more detailed (having a larger number of samples) than the common data is required, the generating unit 112 may use an interpolation technique such as spline interpolation.

[0068] The time-series data DA generated by interpolation or extrapolation can be used as teacher data when the learning unit 121 learns an estimation model. The estimation model learned using the time-series data DA generated by extrapolation may be stored in association with extrapolation information. This allows the output control unit 102 to output, for example, information (such as warning information) indicating whether or not the time-series data generated by extrapolation was used when outputting an estimation result based on the estimation model.

[0069] The present invention may be configured such that interpolation or extrapolation is not performed, and common data corresponding to feature information close to (similar to) the feature information for the specified application is used as training data. Furthermore, while the feature information has been the C rate up until now, it may also include temperature.

[0070] In the following, an example will be described in which a charge rate of 1.5 C, a discharge rate of 0.8 C, and a temperature of 22° C. In addition to the same charge rate and discharge rate as above, the common data is also acquired at temperatures of 20° C. and 25° C.

[0071] In this case, the generation unit 112 selects two common data items corresponding to feature information items that sandwich the value of the feature information item for the specified purpose within a narrower range, and generates data by combining the two selected common data items. When multiple feature information items are used, the generation unit 112 generates the same number of combined data items as the number of combinations of the multiple feature information items. For example, when three pieces of feature information items, namely, charge rate, discharge rate, and temperature, are used, eight pieces of data items corresponding to 2 x 2 x 2 = 8 combinations are generated. The learning unit 121 uses the data items generated in this way, corresponding to the number of combinations, as training data for learning the estimation model.

[0072] In the above example, because the charge rate of the specified use is 1.5C, the generation unit 112 selects common data with C rates of 1C and 2C. Furthermore, because the discharge rate of the specified use is 0.8C, the generation unit 112 selects common data with C rates of 0.5C and 1C. Furthermore, because the temperature of the specified use is 22 degrees, the generation unit 112 selects common data with temperatures of 20 degrees and 25 degrees. The generation unit 112 selects one of two values ​​from each piece of characteristic information, combines it with the other piece of characteristic information, and generates data that combines the charging data and the discharging data for each combination.

[0073] Fig. 11 is a diagram showing an example of time-series data generated in this way. In the example of Fig. 11, the following eight pattern combinations are used: Charge rate 2C, discharge rate 1C, temperature 20°C Charge rate 2C, discharge rate 1C, temperature 25°C Charge rate 2C, discharge rate 0.5C, temperature 20°C Charge rate 2C, discharge rate 0.5C, temperature 25°C Charge rate 1C, discharge rate 1C, temperature 20°C Charge rate 1C, discharge rate 1C, temperature 25°C Charge rate 1C, discharge rate 0.5C, temperature 20°C Charge rate 1C, discharge rate 0.5C, temperature 20°C Charge rate 1C, discharge rate 0.5C, temperature 25°C

[0074] For each of the eight combinations, the generation unit 112 generates data by combining two pieces of common data. The generated data is used by the learning unit 121 to learn an estimation model. The test data is data used for estimation by the estimation model. For example, the test data in FIG. 11 is data obtained under conditions (applications) of a charge rate of 1.5 C, a discharge rate of 0.8 C, and a temperature of 22 degrees.

[0075] The test data may be data obtained at different temperatures during charging and discharging. For example, the test data may be data obtained under conditions (applications) where the charging rate is 1.5 C, the discharging rate is 0.8 C, the charging temperature is 26°C, and the discharging temperature is 22°C. In such a case, the generation unit 112 generates data in the same manner as described above, treating the charging temperature and the discharging temperature as different characteristic information.

[0076] Fig. 12 is a diagram showing an example of time-series data generated in this case. In the example of Fig. 12, a total of 16 pattern combinations are used, which are four combinations of charge rate (2C, 1C) and discharge rate (1C, 0.5C) and four combinations of temperature during charge (28°C, 24°C) and temperature during discharge (24°C, 20°C).

[0077] In the case of a configuration in which interpolation (interpolation, extrapolation) is not performed and the C rate for the specified purpose is outside the range of C rates for the common data, the generator 112 uses a combination of two common data corresponding to a C rate close to the C rate for the specified purpose. In this case, the generator 112 also stores extrapolation information indicating, for example, that the data is outside the range (i.e., that the conditions are the same as those for extrapolation) in association with the generated data.

[0078] As described above, the generated time-series data DA can be used as training data for learning an estimation model, for example. The learned estimation model can be used to estimate an index indicating the state of the battery 200, such as the health level.

[0079] Next, an example of a process for outputting the estimation results of the estimation model by the output control unit 102 will be described. Fig. 13 is a diagram showing an example of an output screen for outputting the estimation results. Fig. 13 shows an example of an output screen displaying the estimation results of the health levels of multiple batteries (batteries B1, B2, and B3).

[0080] 14 is a diagram showing an example of an output screen that further displays warning information indicating that time-series data DA calculated by extrapolation has been used. As shown in FIG. 14 , the output screen includes warning information 1401. The warning information 1401 is, for example, in a display format (such as a pop-up screen) that appears when a cursor approaches, but any other display format may be used. For example, the warning information may be displayed on a screen separate from the output screen. A button (such as an ON or OFF button) may be provided for specifying whether or not to display the warning information, and the display and non-display of the warning information may be switched depending on the button being pressed.

[0081] Whether the calculation was performed by extrapolation can be determined by, for example, referring to extrapolation information stored in association with the estimation model. For example, the output control unit 102 outputs warning information in response to the estimation result obtained by the estimation model associated with the extrapolation information.

[0082] In this way, in the first embodiment, data suited to the characteristics (applications) of the operational data to be estimated is generated from common data common to multiple applications, and the generated data can be used to train the estimation model, thereby making it possible to more efficiently obtain data (teaching data) to be used for monitoring the state of the battery.

[0083] Second Embodiment An information processing device according to a second embodiment uses data generated from common data to calculate a battery evaluation index. The evaluation index can be used, for example, to evaluate what type of battery is desirable to use when designing a battery system. The following description will be given taking as an example a case in which an information processing system 10 includes multiple types of batteries 200, and an evaluation index is calculated for each of the multiple types of batteries 200. In the following description, the type of battery 200 may also be referred to as a cell type.

[0084] 15 is a block diagram showing an example of the configuration of an information processing system 10-2 including an information processing device 100-2 according to the second embodiment. The configurations of the battery 200 and the monitoring system 300 are the same as those in the first embodiment, and therefore the same reference numerals are used and the description thereof will be omitted.

[0085] As shown in FIG. 15, the information processing device 100-2 includes an acquisition unit 101, an output control unit 102, an addition unit 103, a feature calculation unit 111, a generation unit 112, an index calculation unit 123-2, and an evaluation unit 124-2.

[0086] The second embodiment differs from the first embodiment in that an index calculation unit 123-2 and an evaluation unit 124-2 are provided instead of the learning unit 121 and the estimation unit 122. Other configurations and functions are the same as those in FIG. 1, which is the block diagram of the information processing device 100 of the first embodiment, and therefore the same reference numerals are used and the description thereof will be omitted here.

[0087] When the battery 200 includes a plurality of cell types, the generating unit 112 generates the time-series data DA for each of the plurality of cell types.

[0088] The index calculation unit 123-2 calculates an index of the battery of the cell type corresponding to the characteristic information FA, using the time-series data DA generated by the generation unit 112. The calculated index may be any index, but may be, for example, the chargeable / dischargeable capacity [Ah] under different conditions and the charge / discharge efficiency for carbon footprint.

[0089] Of the chargeable and dischargeable charge amounts, the chargeable capacity is calculated, for example, from the difference between the maximum SoC (SoC at which the upper limit voltage is reached) and the minimum SoC during charging. The discharge capacity is calculated from the difference between the maximum SoC and the minimum SoC (SoC at which the lower limit voltage is reached) during discharging. Under specific conditions, for example, when determining the capacity when a cell with an SoC of 80% is charged and discharged at 1C at 22 degrees, the index calculation unit 123-2 can estimate the chargeable capacity and discharge capacity by creating a composite curve for these conditions. The index calculation unit 123-2 can also calculate the charge and discharge efficiency using a similar method.

[0090] When batteries 200 of multiple cell types are provided, index calculation unit 123-2 calculates an index for each of the multiple cell types using multiple time-series data DA generated for each of the multiple cell types.

[0091] The evaluation unit 124-2 evaluates batteries of multiple cell types using the calculated multiple indexes.

[0092] Next, the evaluation process performed by the information processing device 100-2 according to the second embodiment will be described with reference to Fig. 16. Fig. 16 is a flowchart showing an example of the evaluation process according to the second embodiment. The evaluation process includes a process of calculating an index using data generated from common data, and a process of evaluating batteries of each cell type using the calculated index.

[0093] Steps S201 to S203 are the same as steps S101 to S103 in the information processing apparatus 100 of the first embodiment, and therefore a description thereof will be omitted.

[0094] The index calculation unit 123-2 calculates an index using the generated time-series data (charge / discharge data) (step S204). The evaluation unit 124-2 evaluates the cell type using the calculated index (step S205). The output control unit 102 outputs the evaluation result by the evaluation unit 124-2 (step S206) and ends the evaluation process.

[0095] Fig. 17 is a diagram for explaining an example of the evaluation process, which is an example of evaluating a cell type suitable for a specific application used at a charge rate of 3C and a discharge rate of 0.4C.

[0096] First, common data is prepared for each of multiple cell types. Figure 17 shows the common data for two cell types, TA and TB. In this embodiment, the discharge capacity of each of the two cell types, TA and TB, is calculated when new and when the cell is in a deteriorated state (health level 80%). By outputting the calculated index (discharge capacity), a designer, for example, can quantitatively determine which cell type is suitable for the application being evaluated.

[0097] In this way, the information processing device of the second embodiment generates data that matches the characteristics (application) of the operation data to be estimated from common data that is common to multiple applications, and the generated data can be used to calculate the battery evaluation index, thereby making it possible to more efficiently obtain data to be used for the battery evaluation function.

[0098] As described above, according to the first and second embodiments, data corresponding to a desired application can be generated by using common data common to all applications, rather than preparing data for each application, which makes it possible to more efficiently obtain data used for functions such as monitoring the battery state and evaluating the battery.

[0099] Next, the hardware configuration of the information processing apparatus according to the first or second embodiment will be described with reference to Fig. 18. Fig. 18 is an explanatory diagram showing an example of the hardware configuration of the information processing apparatus according to the first or second embodiment.

[0100] The information processing device of the first or second embodiment includes a control device such as a CPU 51, a storage device such as a ROM (Read Only Memory) 52 or a RAM 53, a communication I / F 54 that connects to a network and communicates, and a bus 61 that connects each part.

[0101] The programs executed by the information processing apparatus of the first or second embodiment are provided in advance in the ROM 52 or the like.

[0102] The program executed by the information processing device of the first or second embodiment may be configured to be provided as a computer program product by being recorded in an installable or executable file format on a computer-readable recording medium such as a CD-ROM (Compact Disk Read Only Memory), a flexible disk (FD), a CD-R (Compact Disk Recordable), or a DVD (Digital Versatile Disk).

[0103] Furthermore, the program executed by the information processing device of the first or second embodiment may be stored on a computer connected to a network such as the Internet and provided by being downloaded via the network. Also, the program executed by the information processing device of the first or second embodiment may be provided or distributed via a network such as the Internet.

[0104] The program executed by the information processing device of the first or second embodiment can cause a computer to function as each part of the information processing device described above. In this computer, the CPU 51 can read the program from a computer-readable storage medium onto a main storage device and execute the program.

[0105] Configuration examples of embodiments are described below. (Configuration Example 1) An information processing device including a processing unit that generates first time series data for specified first feature information by selecting, from a plurality of common data to which different feature information indicating characteristics of the charging and discharging is associated, a plurality of common data to which feature information identical or similar to specified first feature information is associated, and combining the selected plurality of common data. (Configuration Example 2) The information processing device according to Configuration Example 1, wherein the processing unit generates the first time series data by at least one of interpolation and extrapolation using the selected plurality of common data. (Configuration Example 3) The information processing device according to Configuration Example 2, wherein, when the first time series data is generated by extrapolation, the processing unit outputs output information indicating that the first time series data was generated by extrapolation. (Configuration Example 4) The information processing device of any one of Configuration Examples 1 to 3, wherein the processing unit calculates the first feature information using operation data of the first battery, learns an estimation model for estimating a state of the first battery using the first time-series data generated using the calculated first feature information, and estimates the state of the first battery using the learned estimation model. (Configuration Example 5) The information processing device of any one of Configuration Examples 1 to 4, wherein the feature information includes a charge rate, a discharge rate, and an average temperature. (Configuration Example 6) The information processing device of Configuration Example 5, wherein the processing unit generates the first time-series data by combining, among the common data for which the charge rate included in the first feature information is the same or similar, a portion of time-series data indicating a change in charge, and among the common data for which the discharge rate included in the first feature information is the same or similar, a portion of time-series data indicating a change in discharge. (Configuration Example 7) The information processing device of any one of Configuration Examples 1 to 6, wherein the feature information includes a degree of variation in charge rate and a degree of variation in discharge rate. (Configuration Example 8) The information processing device according to any one of configuration examples 1 to 7, wherein the characteristic information includes an average value of the temperature during charging and an average value of the temperature during discharging.(Configuration Example 9) The information processing device of any one of Configuration Examples 1 to 8, wherein the characteristic information includes profile information of the battery. (Configuration Example 10) The information processing device of any one of Configuration Examples 1 to 9, wherein the processing unit adds data in which the characteristic information of the battery is associated with operation data of the battery to the common data. (Configuration Example 11) The information processing device of any one of Configuration Examples 1 to 10, wherein the processing unit calculates an index of a type of battery corresponding to the first characteristic information using the generated first time-series data. (Configuration Example 12) The information processing device of Configuration Example 11, wherein the common data is defined for each of a plurality of types of battery, and the processing unit generates the first time-series data for each of the plurality of types, calculates the index for each of the plurality of types using the plurality of first time-series data generated for each of the plurality of types, and evaluates the plurality of types of batteries using the calculated indexes. (Configuration Example 13) An information processing method executed by an information processing device, comprising: a step of selecting, from a plurality of common data which are time-series data showing changes in charging and discharging of a battery and which are associated with mutually different feature information showing characteristics of the charging and discharging, a plurality of common data which are associated with feature information identical or similar to specified first feature information, and synthesizing the selected plurality of common data, thereby generating first time-series data for the first feature information. (Configuration Example 14) A program for causing a computer to execute: a step of selecting, from a plurality of common data which are time-series data showing changes in charging and discharging of a battery and which are associated with mutually different feature information showing characteristics of the charging and discharging, a plurality of common data which are associated with feature information identical or similar to specified first feature information, and synthesizing the selected plurality of common data, thereby generating first time-series data for the first feature information.(Configuration Example 15) An information processing system comprising: a battery; and a processing unit that selects, from a plurality of common data that are time series data showing changes in charging and discharging of the battery and that are associated with mutually different feature information that shows characteristics of the charging and discharging, a plurality of common data that are associated with feature information that is identical to or similar to specified first feature information, and that synthesizes the selected plurality of common data to generate first time series data for the first feature information.

[0106] Although several embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These novel embodiments can be embodied in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their modifications are included within the scope and spirit of the invention, and are also included in the scope of the invention and its equivalents as defined in the claims.

[0107] REFERENCE SIGNS LIST 1 Information processing system 100 Information processing device 101 Acquisition unit 102 Output control unit 103 Addition unit 111 Feature calculation unit 112 Generation unit 121 Learning unit 122 Estimation unit 123-2 Index calculation unit 124-2 Evaluation unit 131 Storage unit 200 Battery 201 Battery panel 202 Module 203 Battery cell 300 Monitoring system

Claims

1. An information processing device comprising: a processing unit that selects, from a plurality of common data sets to which mutually different feature information indicating characteristics of the charging and discharging is associated, a plurality of common data sets to which feature information identical or similar to specified first feature information is associated, and generates first time series data for the first feature information by synthesizing the selected plurality of common data sets.

2. The information processing device according to claim 1, wherein the processing unit generates the first time series data by at least one of interpolation and extrapolation using the selected plurality of common data.

3. The information processing device according to claim 2, wherein the processing unit, when the first time series data is generated by extrapolation, outputs output information indicating that the first time series data has been generated by extrapolation.

4. The information processing device of claim 1, wherein the processing unit calculates the first characteristic information using operating data of the first battery, learns an estimation model for estimating the state of the first battery using the first time series data generated using the calculated first characteristic information, and estimates the state of the first battery using the learned estimation model.

5. The information processing device according to claim 1, wherein the characteristic information includes a charge rate, a discharge rate, and a temperature.

6. The information processing device according to claim 5, wherein the processing unit generates the first time series data by combining a portion of time series data indicating changes in charging among the common data having the same or similar charging rates included in the first feature information, and a portion of time series data indicating changes in discharging among the common data having the same or similar discharging rates included in the first feature information.

7. The information processing device according to claim 1, wherein the characteristic information includes a degree of variation in a charging rate and a degree of variation in a discharging rate.

8. The information processing device according to claim 1, wherein the characteristic information includes an average temperature during charging and an average temperature during discharging.

9. The information processing device according to claim 1, wherein the characteristic information includes profile information of the battery.

10. The information processing device according to claim 1, wherein the processing unit adds data in which characteristic information of the battery is associated with operation data of the battery to the common data.

11. The information processing device according to claim 1, wherein the processing unit calculates an index of the type of battery corresponding to the first characteristic information by using the generated first time-series data.

12. The information processing device according to claim 11, wherein the common data is defined for each of a plurality of battery types, and the processing unit generates the first time series data for each of the plurality of types, calculates the index for each of the plurality of types using the plurality of first time series data generated for each of the plurality of types, and evaluates the plurality of types of batteries using the calculated indexes.

13. An information processing method executed by an information processing device, comprising a step of selecting, from a plurality of common data to which different feature information indicating characteristics of the charging and discharging is associated, a plurality of common data to which feature information identical or similar to specified first feature information is associated, and generating first time series data for the first feature information by synthesizing the selected plurality of common data.

14. A program for causing a computer to execute the steps of: selecting, from a plurality of common data sets to which different characteristic information indicating characteristics of the charging and discharging is associated, a plurality of common data sets to which characteristic information identical or similar to specified first characteristic information is associated, and generating first time series data for the first characteristic information by synthesizing the plurality of selected common data sets.

15. An information processing system comprising: a battery; and a processing unit that generates first time series data for specified first feature information by selecting, from a plurality of common data to which mutually different feature information indicating characteristics of the charging and discharging is associated, a plurality of common data to which feature information identical or similar to specified first feature information is associated, and synthesizing the selected plurality of common data. The common data is time series data indicating changes in charging and discharging of the battery.

Images