Secondary battery life prediction method and secondary battery life prediction device

Abstract

To enable a deterioration degree of a secondary battery mounted on a vehicle to be determined with higher accuracy and to predict not only a current deterioration degree but also a future deterioration degree with high accuracy.SOLUTION: A secondary battery life prediction method is provided with a step (ST7) for using information on an object vehicle C and information on a plurality of vehicles to predict a deterioration degree of an object secondary battery mounted on the object vehicle, the information on the object vehicle C includes information on a state of the object secondary battery to be mounted and information showing a use situation of the object vehicle C, the information on the plurality of vehicles includes information on the states of secondary batteries mounted on the respective vehicles and information showing use situations of the respective vehicles, and the information showing the use situation of the object vehicle and the information showing the respective use situations of the plurality of vehicles include at least any one of data about an owner, data about vehicle travel, or data about an environment in which the vehicle travels.SELECTED DRAWING: Figure 4

Description

[Technical Field]

[0001] FIELD OF THE INVENTION The present invention relates to a method and an apparatus for predicting a life span of a secondary battery. [Background technology]

[0002] Patent Document 1 discloses a diagnostic device, a diagnostic method, and a program for deriving the degree of deterioration of a secondary battery based on data collected from all vehicles equipped with secondary batteries. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 7062775 Summary of the Invention [Problem to be solved by the invention]

[0004] However, in the diagnostic device etc. in Patent Document 1, data relating to all vehicles equipped with secondary batteries is used, and therefore there is a possibility that a large amount of data may be included that is not closely related to the secondary battery for which the actual degree of deterioration is to be derived.

[0005] In other words, the degree of deterioration of a secondary battery varies greatly depending on the conditions of use. For example, the conditions of use vary depending on the environment in which the vehicle is usually used (cold, warm, or hot, or rainy or snowy, etc.), driving conditions (plains or mountains, urban areas or highways, etc.), and even the owner's driving style. Therefore, as a large amount of data is used, the accuracy of predicting the degree of deterioration of a secondary battery may decrease.

[0006] The present invention has been made to solve the above-mentioned problems, and an object of the present invention is to provide a method and device for predicting the lifespan of a secondary battery that can more accurately determine the degree of deterioration of a secondary battery installed in a vehicle, and that can accurately predict not only the current degree of deterioration but also the future degree of deterioration. [Means for solving the problem]

[0007] In one embodiment, the method for predicting the lifespan of a secondary battery includes the steps of acquiring information about a target vehicle equipped with a target secondary battery that is the subject of lifespan prediction, acquiring information about a plurality of vehicles equipped with the secondary battery, and predicting the degree of deterioration of the target secondary battery equipped in the target vehicle using the information about the target vehicle and the information about the plurality of vehicles, wherein the information about the target vehicle includes information about the state of the target secondary battery equipped therein and information indicating the usage status of the target vehicle, the information about the plurality of vehicles includes information about the state of the secondary battery equipped in each vehicle and information indicating the usage status of each vehicle, and the information indicating the usage status of the target vehicle and the information indicating the usage status of each of the plurality of vehicles includes at least one of data about the owner, data about the vehicle's operation, or data about the environment in which the vehicle is operated. [Effects of the Invention]

[0008] By adopting such a configuration, the present invention can provide a secondary battery life prediction method and a secondary battery life prediction device that can more accurately determine the degree of deterioration of a secondary battery installed in a vehicle and can accurately predict not only the current degree of deterioration but also the future degree of deterioration. [Brief explanation of the drawings]

[0009] [Figure 1] 1 is a block diagram showing the overall configuration of a life prediction system for a secondary battery according to an embodiment of the present invention; [Figure 2]1 is a block diagram showing an internal configuration of a life prediction device for a secondary battery according to an embodiment of the present invention; [Figure 3] 1 is a table showing an example of information about the state of a secondary battery mounted on a vehicle and information indicating the usage status of the vehicle, which are used in an embodiment of the present invention. [Figure 4] 3 is an explanatory diagram illustrating a process of estimating and predicting a degree of deterioration of a target vehicle in an embodiment of the present invention. FIG. [Figure 5] 1 is a schematic diagram showing an example of distribution information generated based on information about a target vehicle and various information included in information about a plurality of vehicles in an embodiment of the present invention. [Figure 6] 4 is a flowchart showing a flow of a process for predicting a life span of a secondary battery in the embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0010] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. Note that the drawings are schematic and may differ from the actual product. Furthermore, the embodiments of the present invention shown below are merely examples of devices and methods for embodying the technical concept of the present invention, and the technical concept of the present invention does not limit the structure, arrangement, etc. of the components to those described below. The technical concept of the present invention can be modified in various ways within the technical scope defined by the claims.

[0011] 1 is a block diagram showing the overall configuration of a secondary battery life prediction system S according to an embodiment of the present invention. The secondary battery life prediction system S includes a life prediction device 1 connected to a communication network N.

[0012] Furthermore, a plurality of vehicles 2A- (only vehicles 2A to 2C are shown in FIG. 1) are connected to the communication network N, and information relating to the plurality of vehicles is transmitted to the lifespan prediction device 1 via the communication network N. Note that the plurality of vehicles 2A- will hereinafter be collectively referred to as "plurality of vehicles 2" where appropriate.

[0013] Furthermore, a vehicle C (hereinafter, such a vehicle will be referred to as a "target vehicle C") equipped with a target secondary battery for which life expectancy is to be predicted is also connected to the communication network N, and information regarding the target vehicle C is transmitted to the life expectancy prediction device 1 via the communication network N.

[0014] Note that, as described above, for convenience, the vehicle equipped with the target secondary battery is referred to as a "target vehicle C" to distinguish it from the multiple vehicles 2. However, the target vehicle C is also included in the multiple vehicles 2. That is, for example, if "vehicle 2B" shown in FIG. 1 is designated as the "target vehicle C," the target vehicle C shown in FIG. 1 is included in the multiple vehicles 2, and information about the vehicle is used to determine the degree of deterioration of the secondary battery equipped in vehicle 2B, which is the target vehicle C.

[0015] The life prediction device 1 is a device used to determine the predicted life (degree of deterioration) of a target secondary battery mounted in a target vehicle C, based on information about a plurality of vehicles 2 and information about the target vehicle C. Fig. 2 is a block diagram showing the internal configuration of the secondary battery life prediction device 1 (hereinafter referred to as "life prediction device 1" where appropriate) in an embodiment of the present invention.

[0016] The life prediction device 1 includes a CPU (Central Processing Unit) 11, a ROM (Read Only Memory) 12, a RAM (Random Access Memory) 13, and an input / output interface 14, which are connected via a bus 15. A communication control unit 16, a storage unit 17, a model generation unit 18, an extraction unit 19, and a judgment unit 20 are connected to the input / output interface 14.

[0017] The CPU 11 reads and executes a boot program for starting the life prediction device 1 from the ROM 12 based on an input signal from an input unit (not shown), and reads various operating systems stored in the storage unit 17. The CPU 11 also controls various devices based on input signals from other external devices (not shown in FIG. 2) via the input / output interface 14, for example.

[0018] Furthermore, the CPU 11 is a processing device that reads out programs and data stored in the RAM 13, the memory unit 17, etc., loads them into the RAM 13, and performs processing to determine the degree of deterioration of the target secondary battery installed in the target vehicle C based on the commands of the program read from the RAM 13.

[0019] The communication control unit 16 is a means such as a LAN card or a modem, and is a means that enables the lifespan prediction device 1 to connect to a communication network N such as the Internet or a LAN. Data transmitted and received to and from the communication network N via the communication control unit 16 is transmitted and received as an input signal or an output signal to the CPU 11 via the input / output interface 14 and the bus 15.

[0020] The storage unit 17 is configured with a semiconductor or a magnetic disk, and stores programs and data executed by the CPU 11. That is, it stores information about the vehicles 2 transmitted from the plurality of vehicles 2. The information about the plurality of vehicles 2 may be transmitted periodically from the plurality of vehicles 2, or information stored in the vehicles 2 at any time may be transmitted collectively when the vehicles have finished traveling.

[0021] Here, the information related to the target vehicle C includes information related to the state of the target secondary battery mounted in the target vehicle C and information indicating the usage status of the target vehicle C. Furthermore, the information related to the multiple vehicles 2 includes information related to the state of the secondary battery mounted in each vehicle and information indicating the usage status of each vehicle.

[0022] Furthermore, the information indicating the usage status of the target vehicle C or the plurality of vehicles 2 includes data on the owner, data on the running of the vehicle, and data on the environment in which the vehicle has run. Fig. 3 is a table showing an example of information indicating the usage status of a vehicle and information on the state of a secondary battery mounted on the vehicle, which is used in the embodiment of the present invention.

[0023] As described above, the information indicating the usage status may include all of the data related to the owner, the data related to the vehicle's travel, and the data related to the environment in which the vehicle traveled, or may include one of these types of data (for example, only the data related to the owner). It may also include two of these types of data (for example, the data related to the owner and the data related to the vehicle's travel). That is, it is sufficient for the information indicating the usage status to include at least one of the data related to the owner, the data related to the vehicle's travel, and the data related to the environment in which the vehicle traveled.

[0024] In the table shown in Figure 3, information on multiple vehicles 2 is shown in the columns. Here, the multiple vehicles 2 refer to vehicles equipped with secondary batteries that are widely distributed (in operation). Vehicles that have been scrapped are also included in the multiple vehicles 2. Therefore, all vehicles that have been put on the market so far are included.

[0025] Furthermore, as vehicles equipped with secondary batteries are sold in the future, the sold vehicles will also be included in the plurality of vehicles 2. Note that the table shown in Fig. 3 shows vehicles 2A to 2C shown in Fig. 1 among the plurality of vehicles 2.

[0026] On the other hand, the rows show items that indicate information about the vehicle's usage status and the status of the secondary battery installed in the vehicle. The table has two main items: "Information indicating usage status" and "Information about the status of the secondary battery (represented in the table as "battery status data")."

[0027] Of these, the "information indicating usage status" is information about the attributes of the vehicle user, the movements of the vehicle while it is traveling, and the environment in which the vehicle is traveling. Therefore, the "information indicating usage status" is divided into "owner data," "travel data," and "environmental data" from the top of the table.

[0028] Furthermore, these three data are further divided into smaller categories (subcategories). For example, "owner data" includes categories such as the vehicle owner's "age," "gender," and "driving history," as well as the vehicle's "model." Furthermore, "driving data" is divided into categories such as "speed," "acceleration," and "distance traveled per trip."

[0029] Furthermore, as "environmental data," data relating to the environment in which the vehicle 2 is traveling, such as "outside temperature," "humidity," "barometric pressure," "latitude," "longitude," and "slope," can be acquired.

[0030] The above-mentioned "information indicating usage status" is not data that directly indicates the status of the secondary batteries installed in the multiple vehicles 2, but it can be said to be a factor that has a significant impact on the degree of deterioration of the secondary batteries. In contrast, "battery status data" is data that indicates the status of the secondary batteries themselves. In the table of Figure 3, six sub-items are listed: "current," "voltage," "resistance," "capacity," "temperature," and "accumulated time."

[0031] In the table of FIG. 3, the details of the detailed items (information indicating the usage status, battery state data, and similarity determination) for vehicles 2A to 2C are omitted.

[0032] Here, the "owner data" is data that can be stored in the storage unit 17 when the vehicle is sold, for example. On the other hand, the "driving data" and "environmental data" are data that are obtained when the vehicle is driven, and are acquired by a computer or the like installed in the vehicle.

[0033] The data may be stored in the computer, or may be set to transmit the data at regular intervals to the lifespan prediction device 1. The lifespan prediction device 1 stores the data received via the communication control unit 16 in the storage unit 17.

[0034] It should be noted that these three sub-items listed as data are merely examples. Therefore, other sub-items may be provided. Furthermore, for example, the "information indicating usage status" is not limited to "owner data" which is data on the person who actually owns the vehicle, but may also include data on a person who is different from the owner but who actually drives the vehicle frequently.

[0035] Similarly to the case of "information indicating usage status," the items listed as sub-items of the "battery status data" are not limited to these items. Furthermore, as a sub-item of the battery status data, data indicating the degree of deterioration of the secondary battery may be directly acquired and stored in the storage unit 17.

[0036] Furthermore, the description here is based on the assumption that the storage unit 17 is provided inside the lifespan prediction device 1. However, the present invention is not limited to this configuration, and for example, a storage device that stores information about the target vehicle C and multiple vehicles 2 may be separately connected to the communication network N.

[0037] The model generation unit 18 generates, by machine learning, a similar vehicle determination model used to extract a vehicle (hereinafter, such a vehicle will be referred to as a "similar vehicle A") that has information indicating a usage situation or battery state data similar to the information indicating a usage situation or battery state data of the target vehicle C. The model generation unit 18 also generates, by machine learning, a prediction model that predicts the future degree of deterioration of the target secondary battery mounted on the target vehicle C.

[0038] That is, when generating the similar vehicle determination model, the model generation unit 18 uses, as input data, information indicating the usage status of the target vehicle C. Information indicating the usage status of all of the multiple vehicles 2 stored in the storage unit 17 is added to this and used as data for classification.

[0039] On the other hand, when the model generation unit 18 generates a prediction model, it uses as input data the information indicating the usage status and battery state data of the target vehicle C. On the other hand, the training data is the information indicating the usage status and battery state data of the similar vehicle A extracted by the similar vehicle determination model.

[0040] To generate the similar vehicle determination model or the predictive model, for example, a machine learning algorithm is used. Examples of machine learning algorithms that can be used to generate the similar vehicle determination model include K-means clustering, hierarchical clustering, DBSCAN, self-organizing maps, and Gaussian mixture models. Examples of machine learning algorithms that can be used to generate the predictive model include neural networks, linear regression, decision tree learning, random forests, support vector regression, and k-nearest neighbor methods.

[0041] The reason for extracting a similar vehicle A that is similar to the target vehicle C will be explained below with reference to Fig. 4. Fig. 4 is an explanatory diagram (graph) illustrating the process of estimating and predicting the degree of deterioration of the target vehicle C in an embodiment of the present invention.

[0042] In the graph shown in Figure 4, the vertical axis represents battery capacity. Secondary batteries gradually deteriorate as they are repeatedly charged and discharged, and their charge capacity decreases. If the fully charged state (capacity) of a new secondary battery is taken as 100%, the chargeable capacity gradually decreases as the secondary battery is used. In other words, the "capacity" here indicates the current state of charge of the secondary battery and what state it will be in the future, based on the state of a new secondary battery (fully charged state).

[0043] In other words, the battery capacity here indicates the degree of deterioration of the secondary battery (SOH: State Of Health). Therefore, the vertical axis represents "battery capacity (SOH)," while the horizontal axis represents the cumulative time the secondary battery has been used.

[0044] Near the origin, the accumulated time is short, so the vehicle equipped with the secondary battery has not been used much, in other words, it is in a state close to that of a new vehicle. At this time, the capacity of the secondary battery is also close to that of a new vehicle. However, as the vehicle is used, the secondary battery undergoes multiple charge and discharge processes, and as the accumulated usage time increases, the battery capacity gradually decreases. In other words, the degree of deterioration of the secondary battery increases.

[0045] In reality, the deterioration state of the secondary battery differs for each vehicle. Therefore, the degree of deterioration varies, for example, there are cases where the battery capacity does not decrease much even with accumulated time, and cases where the battery capacity decreases suddenly, and the deterioration does not necessarily show a gradual curve as shown in the graph of Fig. 4. However, for the sake of explanation, these various states are not shown, and the multiple lines shown in the graph of Fig. 4 are shown to slope generally downward to the right.

[0046] In actuality, some vehicles are used for a long time, while others are not used for long and are scrapped. For secondary batteries installed in scrapped vehicles, in particular, information on their usage status after scrapping is unavailable. Therefore, if the line showing the degree of deterioration of the secondary battery for such a vehicle is plotted as in the graph shown in Figure 4, the line showing the degree of deterioration of the secondary battery will end at a certain point in the accumulated time, i.e., after the vehicle is scrapped.

[0047] On the other hand, since the cumulative time increases each time the vehicle is used and information indicating the vehicle's usage status is updated, the dashed lines shown in the graph in Figure 4 will gradually extend to the right as long as the vehicle continues to be used.

[0048] Therefore, for convenience of explanation, the graph shown in Figure 4 does not take into account the scrapping of vehicles during the cumulative time. That is, the degree of deterioration of all secondary batteries shown by dashed lines in the graph of Figure 4 is shown without any breaks in the cumulative time on the horizontal axis, and all are shown until the same cumulative time has elapsed.

[0049] This decrease in battery capacity is a phenomenon that occurs in all vehicles equipped with secondary batteries. On the other hand, for example, a vehicle that is always driven on weekdays and a vehicle that is only driven on weekends will both gradually lose battery capacity, but the extent of this decrease will be different. In other words, for the same cumulative time, the battery capacity of a secondary battery installed in the former vehicle will decrease and the degree of deterioration will be greater than that of a secondary battery installed in the latter vehicle.

[0050] That is, the degree of deterioration of the secondary batteries installed in each vehicle differs. Therefore, for example, if there are 10,000 vehicles 2 equipped with secondary batteries, it is possible to draw a line on the graph in Figure 4 that indicates the degree of deterioration of the secondary batteries of all 10,000 vehicles. However, since it is not possible to show the degree of deterioration of the secondary batteries of all 2 vehicles equipped with secondary batteries on the graph in Figure 4, the graph in Figure 4 only shows the degree of deterioration of the secondary batteries of five vehicles.

[0051] Specifically, for example, five vehicles are selected from all the vehicles 2 stored in the memory unit 17, and the dashed lines show the changes in battery capacity over cumulative time. The dashed line M shown at the highest position and the dashed line N shown at the lowest position in the graph of Fig. 4 are dashed lines that indicate the degree of deterioration of the secondary batteries in two different vehicles.

[0052] In particular, dashed line M shows the battery capacity of the secondary battery of the vehicle with the most remaining battery capacity at the same cumulative time, and dashed line N shows the battery capacity of the secondary battery of the vehicle with the least remaining battery capacity.

[0053] Therefore, the change in battery capacity of the secondary batteries installed in all of the multiple vehicles 2 is represented between dashed line M and dashed line N. Looking at dashed line M and dashed line N, even if the initial battery capacity is the same, the difference in battery capacity becomes larger as time passes (the secondary battery is used for longer).

[0054] Also, a similar dashed line O is shown directly below the dashed line M shown at the highest position. This dashed line O indicates the degree of deterioration of a secondary battery installed in a vehicle that is included in the plurality of vehicles 2 but is different from the two vehicles described above.

[0055] Next, the graph shown in Figure 4 shows two lines, each of which is a dashed dotted line, indicating the degree of deterioration of the secondary battery. The degrees of deterioration of the secondary batteries indicated by these two dashed dotted lines are the degrees of deterioration of the secondary batteries installed in two different similar vehicles A. Then, at the same cumulative time, dashed dotted line P indicates the battery capacity of the secondary battery of the vehicle with the most remaining battery capacity, and dashed dotted line Q indicates the battery capacity of the secondary battery of the vehicle with the least remaining battery capacity.

[0056] As described above, similar vehicle A is included in multiple vehicles 2, and among these multiple vehicles 2, the information indicating the usage status, etc. is similar to that of target vehicle C. Furthermore, the information indicating the usage status, etc. indicated by similar vehicle A is similar to the information indicating the usage status, etc. indicated by target vehicle C, but the information indicating the usage status, etc. indicated by similar vehicles A are also similar to each other.

[0057] Therefore, with regard to the degree of deterioration of the secondary battery, the difference in battery capacity between the dashed line P shown on the upper side and the dashed line Q shown on the lower side at the same cumulative time is smaller than the difference in battery capacity between the dashed line M and dashed line N, which show the battery capacity extracted from all vehicles.

[0058] That is, in an embodiment of the present invention, when estimating and predicting the degree of deterioration of a target secondary battery installed in a target vehicle, information indicating the usage status of all of the multiple vehicles 2 is not used, but information indicating the usage status of similar vehicles that have similar information indicating their usage status to that of the target vehicle is used. This allows for the use of more narrowed down information, making it possible to more accurately determine the degree of deterioration of the target secondary battery.

[0059] Therefore, a similar vehicle A is extracted from all of the multiple vehicles 2, and the degree of deterioration of the target secondary battery is determined using information indicating the usage status of the similar vehicle A. Details of the process of extracting the similar vehicle A will be described later, but a similar vehicle determination model is used to extract the similar vehicle A.

[0060] The extraction unit 19 uses the similar vehicle determination model generated by the model generation unit 18 to extract a similar vehicle A that is similar to the target vehicle C. Note that, because similar vehicles A differ for each target vehicle C, the model generation unit 18 generates a similar vehicle determination model each time so that a similar vehicle A that matches the target vehicle C can be extracted. The extraction unit 19 also extracts a similar vehicle A for each target vehicle C. Therefore, the similar vehicle determination model and the prediction model described below are both generated for each target vehicle, and are, so to speak, custom-made models.

[0061] As described above, the dashed-dotted line P and the dashed-dotted line Q indicate the cases where the battery capacity is the highest and the lowest remaining for the same cumulative time for multiple similar vehicles A. Therefore, the change in the battery capacity of the target vehicle C over cumulative time should be shown between the dashed-dotted line P and the dashed-dotted line Q.

[0062] In the graph of FIG. 4, the solid line R indicates the change in battery capacity over time in the subject vehicle C, i.e., the degree of deterioration of the subject secondary battery. In the solid line R, one end R0 indicates the battery capacity at the start of use of the subject secondary battery. As described above, at the start of use, even if charging and discharging processes are performed, the chargeable capacity does not differ significantly from the original charge capacity. From this point on, the battery capacity gradually decreases.

[0063] On the other hand, the other end R1 of the solid line R indicates the battery capacity when determining the degree of deterioration of the target secondary battery. More specifically, as will be described later, this indicates the degree of deterioration (estimated SOH) of the secondary battery estimated at this point in time.

[0064] When estimating the battery capacity, which indicates the degree of deterioration of the target secondary battery, the battery state data of the similar vehicle A extracted by the similar vehicle determination model is used. By using this battery state data of the similar vehicle A, the degree of deterioration of the target secondary battery can be estimated with greater accuracy.

[0065] The degree of degradation of the estimated secondary battery is then predicted to determine how it will change in the future. A prediction model is used to predict the degree of degradation of the target secondary battery. The prediction model is generated by model generation unit 18, and determination unit 20 uses this prediction model to predict the degree of degradation of the target secondary battery.

[0066] The prediction model is generated by the model generation unit 18 using the information indicating the usage status of similar vehicle A extracted by the similar vehicle determination model as described above and the battery state data as training data. Here, the graph in Fig. 4 shows a dotted line S indicating the degree of deterioration predicted using information indicating the usage status of all vehicles, etc., and a dotted line T indicating the degree of deterioration predicted using information indicating the usage status of similar vehicle A, etc.

[0067] In the graph shown in Figure 4, the dotted line S is represented as "prediction S from information on all vehicles," while the dotted line T is represented as "prediction T from information on similar vehicles."

[0068] Comparing dotted line S and dotted line T, both dotted line S and dotted line T are positioned between dashed dotted line P and dashed dotted line Q, which indicate the degree of deterioration of the vehicle secondary battery extracted as similar vehicle A. On the other hand, dotted line S is positioned higher than dotted line T and is closer to dashed dotted line P. In contrast, dotted line T is positioned approximately midway between dashed dotted line P and dashed dotted line Q.

[0069] Looking at dotted line S, it appears to be roughly flat from the estimated SOH, which indicates the current degree of deterioration of subject vehicle C. This is unnatural, as it indicates that the degree of deterioration of the subject secondary battery will not decrease even if subject vehicle C is used in the same way as before.

[0070] As described above, when predicting the degree of deterioration of a secondary battery using information indicating the usage status of multiple vehicles 2, although there is a large amount of information that can be referenced, the information also includes information about vehicles that are used in ways that have significantly different information indicating their usage status from that of the target vehicle C. For this reason, it is thought that the accuracy of predicting the degree of deterioration of the target secondary battery installed in the target vehicle C is not very high.

[0071] This can also be seen from the large difference between the case of the least degree of deterioration (dashed line M) and the case of the most degree of deterioration (dashed line N) shown in the graph of Fig. 4. The dotted line S is shown in a position close to the dashed-dotted line P because the determination unit 20 made the determination using a prediction model generated using information indicating the usage status of multiple vehicles 2 and battery state data as training data.

[0072] On the other hand, when looking at the prediction of the future degree of degradation of the target secondary battery as indicated by the dotted line T, the capacity of the target secondary battery will decrease in a natural manner from the degree of degradation up to now as indicated by the solid line. The prediction of the degree of degradation of the target secondary battery as indicated by the dotted line T is made by the determination unit 20 using a prediction model generated using information indicating the usage status of similar vehicle A and battery state data as training data.

[0073] In other words, as described above, when checking the degree of deterioration of the secondary battery installed in the target vehicle C, rather than using information from multiple vehicles 2 equipped with secondary batteries, it is possible to more accurately calculate the degree of deterioration of the secondary battery by extracting a similar vehicle A from all vehicles whose information indicating usage status is similar to that of the target vehicle C, and then using the information indicated by the similar vehicle A.

[0074] When the model generation unit 18 generates a similar vehicle determination model, as described above, information indicating the usage status of the target vehicle C is used as input data, and information indicating the usage status of the multiple vehicles 2 is added to this and used as data. However, the amount of information indicating the usage status of the multiple vehicles 2 is enormous. When generating a similar vehicle determination model for each target vehicle C, generating a similar vehicle determination model using such an enormous amount of information each time would require a significant allocation of calculation resources in, for example, the lifespan prediction device 1.

[0075] Therefore, for example, the following processing can be performed: Fig. 5 is a schematic diagram showing an example of distribution information generated based on various information included in information about a target vehicle C and information about a plurality of vehicles 2 in the embodiment of the present invention.

[0076] The horizontal axis indicates the "range of use of each data item." Here, "each data item" refers to a subitem in the information indicating the usage status. For example, this refers to "outside temperature" or "humidity" in the "environmental data" column in the table of Figure 3. The vertical axis indicates the "cumulative time."

[0077] That is, for example, if "outside temperature" is applied as the "usage range of each data" on the horizontal axis, the outside temperature during vehicle usage will differ for multiple vehicles 2 depending on the usage conditions of each vehicle. For example, when a vehicle is used in a low-latitude region, the outside temperature tends to be high. On the other hand, when a vehicle is used in a high-latitude region, the outside temperature tends to be low.

[0078] Furthermore, even for the same vehicle, the outside temperature may be the same or different. Furthermore, if a vehicle that is normally used at a low altitude is moved to a higher altitude, the outside temperature will also change. Therefore, based on information indicating the usage status of multiple vehicles 2, the cumulative usage time at each outside temperature is calculated. The results are then converted into distribution information as shown in Figure 5.

[0079] As a result, for example, the distribution of outside air temperature is shown to be roughly divided into three parts, as shown in the schematic diagram of Figure 5. The distribution of outside air temperature for target vehicle C is also shown, and is indicated by a solid line in the approximate center of the schematic diagram of Figure 5.

[0080] The model generation unit 18 converts the numerical information for each sub-item shown in FIG. 3 into distribution information as shown in FIG. 5. Then, a similar vehicle determination model is generated based on the distribution information for each sub-item. In this way, when generating a similar vehicle determination model, by inserting a process of converting the information into distribution information, the calculation load can be reduced compared to generating a similar vehicle determination model from information that is only numerical. The reduction in calculation load leads to more rapid estimation and prediction of the degree of deterioration of the target secondary battery.

[0081] [Operation] Next, the flow of estimating and predicting the degree of deterioration of a target secondary battery mounted on a target vehicle C will be described below with reference to Fig. 6. Fig. 6 is a flowchart showing the flow of processing for predicting the life span of a secondary battery in an embodiment of the present invention.

[0082] First, information indicating the usage status of the target vehicle C is acquired (ST1). Note that, here, it is assumed that the target vehicle C equipped with the target secondary battery whose degree of degradation is to be estimated and predicted has been selected in advance.

[0083] Furthermore, the process of estimating and predicting the degree of deterioration of the secondary battery can be performed at various times, such as when the target vehicle C is brought to a dealer or a sales shop, or when the owner of the target vehicle C performs the estimation and prediction himself / herself. In the latter case, the estimation and prediction may be performed by accessing the lifespan prediction device 1 from a mobile device such as a personal computer or a smartphone.

[0084] Next, information indicating the usage status of the plurality of vehicles 2 is acquired (ST2). In FIG. 6, the plurality of vehicles 2 is represented as "all vehicles." Information on all the vehicles and information on the target vehicle C are acquired from the storage unit 17. Alternatively, the information may be acquired from, for example, a server connected to the communication network N of the secondary battery life prediction system S.

[0085] Here, information on all vehicles is obtained after information on target vehicle C, but the order in which information indicating usage status is obtained may vary, or the information may be obtained in parallel.

[0086] Based on the acquired information indicating the usage status, the model generation unit 18 generates a similar vehicle determination model for extracting a similar vehicle A that is similar to the target vehicle C (ST3). Note that, as described using Fig. 5, for example, the model generation unit 18 may convert the information indicating the usage status of all vehicles, etc. into distribution information as appropriate, and then perform the process of generating a similar vehicle determination model.

[0087] The generated similar vehicle determination model is used by the extraction unit 19 to extract a similar vehicle A based on information indicating the usage status of all vehicles, etc. (ST4). Then, the determination unit 20 estimates the current battery capacity (the degree of deterioration of the target secondary battery: SOH) of the target vehicle C based on the extracted information of the similar vehicle A, i.e., the battery state data of the similar vehicle A (ST5).

[0088] Furthermore, model generation unit 18 generates a prediction model for predicting the future degree of degradation of the target secondary battery of target vehicle C, using information on the degree of degradation of the target secondary battery estimated by determination unit 20 and information on similar vehicle A (ST6). Determination unit 20 further predicts the future degree of degradation of the target secondary battery of target vehicle C, using the prediction model generated by model generation unit 18 (ST7).

[0089] By performing processing in this manner, the life prediction device 1 is able to more accurately determine the degree of deterioration of the secondary battery installed in the vehicle, and is also able to accurately predict not only the current degree of deterioration but also the future degree of deterioration.

[0090] The above explanation has been given on the assumption that the secondary battery life prediction system S is provided with a life prediction device 1 as a device for estimating and predicting the degree of deterioration of the secondary battery. However, without providing such a device, a device simply equipped with the function of a server may take on the role of the life prediction device 1.

[0091] Furthermore, although not described as an element constituting the lifespan prediction device 1, for example, an input unit configured with an input device such as a keyboard or dial used to input information about the target vehicle may be provided. Also, a display unit such as a liquid crystal display may be provided to display the results when the degree of deterioration of the target secondary battery is estimated and predicted.

[0092] Furthermore, up to this point, the explanation has been given on the assumption that information indicating the usage status of multiple vehicles, including the target vehicle, is stored in the memory unit of the lifespan prediction device via the communication network N. In other words, information indicating the usage status of multiple vehicles is collected in the lifespan prediction device, and the lifespan prediction device estimates and predicts the degree of deterioration of the target secondary battery.

[0093] However, instead of using this method, for example, each vehicle may be equipped with a device or function that corresponds to a lifespan prediction device, in which case the degree of deterioration of the secondary battery installed in the target vehicle is estimated and predicted.

[0094] Furthermore, the target vehicle may directly acquire information indicating the usage status of the vehicle from a vehicle in its vicinity (regardless of whether it is moving or not). If such a method can be adopted, it may be useful in extracting similar vehicles.

[0095] For example, when considering environmental data, it is highly likely that vehicles encountered while a target vehicle is traveling are used in the same environment as the target vehicle. Therefore, it is highly likely that such vehicles are similar to the target vehicle. Therefore, by identifying such vehicles as similar vehicles in advance, it is possible to improve the accuracy and extraction speed of the process when extracting similar vehicles.

[0096] In such a case, the similar vehicle does not have to be included in all of the plurality of vehicles, and therefore the similar vehicle does not have to be extracted from all of the plurality of vehicles.

[0097] Furthermore, with regard to the process of extracting similar vehicles, we have explained a method of extracting similar vehicles by returning information indicating the usage status of multiple vehicles to distribution information. However, this method is not limited to this. For example, it is also possible to compare the target vehicle with information indicating the usage status of the target vehicle, calculate the similarity for each of multiple vehicles, and determine whether the similarity exceeds a preset threshold. In this case, vehicles determined to be similar are extracted as similar vehicles.

[0098] Furthermore, the information indicating the usage status of multiple vehicles has been explained above on the assumption that it is stored in the storage unit 17. However, the content of the information indicating the usage status is updated as the vehicles are used.

[0099] Therefore, for example, the memory unit 17 constantly updates information indicating the usage status of multiple vehicles, so that the information used by the model generation unit 18 to generate a model can always be kept up to date.

[0100] Alternatively, the storage unit 17 may first construct a database of information indicating the usage status of multiple vehicles, and then the model generation unit 18 may perform processing to reflect updated content as needed.

[0101] [Effects of the Example] (1) A method for predicting the lifespan of a secondary battery according to an embodiment of the present invention includes the steps of acquiring information about a target vehicle equipped with a target secondary battery for which lifespan is to be predicted, acquiring information about a plurality of vehicles equipped with secondary batteries, and predicting the degree of deterioration of the target secondary battery equipped in the target vehicle using the information about the target vehicle and the information about the plurality of vehicles, wherein the information about the target vehicle includes information about the state of the target secondary battery equipped therein and information indicating the usage status of the target vehicle, the information about the plurality of vehicles includes information about the state of the secondary batteries equipped in each vehicle and information indicating the usage status of each vehicle, and the information indicating the usage status includes at least one of data about the owner, data about the vehicle's driving, or data about the environment in which the vehicle is driven.

[0102] When estimating and predicting the degree of deterioration of a target secondary battery, not only information on the state of the secondary battery but also information indicating the usage status of each vehicle is used. By using this information to predict the lifespan of a secondary battery, it is possible to more accurately determine the degree of deterioration of the secondary battery installed in a vehicle, and to accurately predict not only the current degree of deterioration but also the future degree of deterioration.

[0103] (2) In the method for predicting the lifespan of a secondary battery described in (1) above, after the step of acquiring information about the target vehicle and information about a plurality of vehicles, the method includes the steps of generating a similar vehicle determination model using the acquired information about the target vehicle and information about the plurality of vehicles, and extracting similar vehicles that are similar to the target vehicle from among the plurality of vehicles using the generated similar vehicle determination model, and is characterized in that in the step of predicting the degree of deterioration of the target secondary battery installed in the target vehicle, the degree of deterioration of the target secondary battery is predicted using the extracted information about the similar vehicles.

[0104] In this way, the accuracy of life prediction can be improved by using only information about vehicles (similar vehicles) that are close to the target vehicle, rather than information about all vehicles.

[0105] (3) In the method for predicting the lifespan of a secondary battery described in (2) above, the similar vehicle determination model is generated by machine learning. By using machine learning in this way, it is possible to handle large amounts of information and improve the accuracy of estimation and prediction.

[0106] (4) In the method for predicting the lifespan of a secondary battery in (2) or (3) above, the step of predicting the degree of deterioration of a target secondary battery mounted on a target vehicle, which is executed after the step of extracting similar vehicles, includes a step of estimating the current degree of deterioration of the target secondary battery mounted on the target vehicle based on information about the extracted similar vehicles. By using information about the similar vehicles in this way, the current degree of deterioration of the target secondary battery can be estimated with greater accuracy.

[0107] (5) In the method for predicting the lifespan of a secondary battery in (4) above, after the step of estimating the current degree of deterioration of the target secondary battery installed in the target vehicle, the method includes the steps of generating a prediction model for predicting the future degree of deterioration of the target secondary battery installed in the target vehicle based on information about similar vehicles and information about the estimated current degree of deterioration, and predicting the future degree of deterioration of the target secondary battery using the generated prediction model.

[0108] When predicting the future degree of deterioration of the target secondary battery installed in the target vehicle, by using information about similar vehicles, a more accurate prediction can be made than when using information about all vehicles.

[0109] (6) In the method for predicting the lifespan of a secondary battery in (5) above, the prediction model is generated by machine learning. By using machine learning for the prediction model, it is possible to handle large amounts of information and improve the accuracy of predictions of the degree of deterioration of the secondary battery.

[0110] (7) In any of the above (2) to (7), the method for predicting the lifespan of a secondary battery includes, after the step of acquiring information about the target vehicle and information about a plurality of vehicles, a step of converting the acquired information about the target vehicle and information about the plurality of vehicles into distribution information indicating the cumulative time for each piece of data contained in the information, and the similar vehicle determination model is generated based on the distribution information.

[0111] Information indicating vehicle usage status is expressed as numerical values. However, if information indicating the usage status of all vehicles is used as is, particularly when extracting similar vehicles to a target vehicle from all vehicles, it may be necessary to allocate a large amount of the computational resources of the lifespan prediction device. This could result in inconveniences such as a heavy processing load, slow performance of the lifespan prediction device, and a long time until results are available. Therefore, by converting information indicating usage status into distribution information in advance, not only can accuracy be expected to improve, but computational costs can also be reduced and processing can be accelerated.

[0112] (8) A secondary battery life prediction device according to an embodiment of the present invention includes a model generation unit that generates a similar vehicle determination model using information about a target vehicle equipped with a target secondary battery that is the subject of life prediction and information about a plurality of vehicles equipped with secondary batteries; an extraction unit that uses the generated similar vehicle determination model to extract similar vehicles that are similar to the target vehicle from among the plurality of vehicles; and a judgment unit that predicts the degree of deterioration of the secondary battery equipped in the target vehicle using the information about the extracted similar vehicles, wherein the information about the target vehicle includes information about the state of the target secondary battery equipped therein and information indicating the usage status of the target vehicle, the information about the plurality of vehicles includes information about the state of the secondary batteries equipped in each vehicle and information indicating the usage status of each vehicle, and the information indicating the usage status includes at least one of data about the owner, data about the vehicle's driving, or data about the environment in which the vehicle is driven.

[0113] When estimating and predicting the degree of deterioration of a target secondary battery, not only information about the state of the secondary battery but also information indicating the usage status of each vehicle is used. Furthermore, the lifespan of the secondary battery is predicted using information about vehicles (similar vehicles) whose information indicating usage status is similar to that of the target vehicle. By using such a lifespan prediction device, it is possible to more accurately determine the degree of deterioration of the secondary battery installed in the vehicle, and to accurately predict not only the current degree of deterioration but also the future degree of deterioration. [Explanation of symbols]

[0114] 1···Lifespan prediction device, 2···Vehicle, 11···CPU, 12···ROM, 13···RAM, 14···Input / output interface, 15···Bus, 16···Communication control unit, 17···Memory unit, 18···Model generation unit, 19···Extraction unit, 20···Judgment unit, A···Similar vehicle, C···Target vehicle

Claims

1. acquiring information about a target vehicle equipped with a target secondary battery that is a target of life prediction; acquiring information about a plurality of vehicles equipped with secondary batteries; predicting a degree of deterioration of the target secondary battery mounted on the target vehicle using information about the target vehicle and information about the plurality of vehicles; the information about the target vehicle includes information about the state of the target secondary battery installed therein and information indicating a usage status of the target vehicle, and the information about the plurality of vehicles includes information about the state of the secondary battery installed in each vehicle and information indicating a usage status of each vehicle, A method for predicting the lifespan of a secondary battery, characterized in that the information indicating the usage status of the target vehicle and the information indicating the usage status of each of the multiple vehicles includes at least one of data regarding the owner, data regarding the vehicle's driving, or data regarding the environment in which the vehicle is driven.

2. After the step of acquiring information about the target vehicle and information about the plurality of vehicles, generating a similar vehicle determination model using the acquired information about the target vehicle and information about the plurality of vehicles; and extracting a similar vehicle that is similar to the target vehicle from among the plurality of vehicles using the generated similar vehicle determination model, The secondary battery life prediction method described in claim 1, characterized in that in the step of predicting the degree of deterioration of the target secondary battery installed in the target vehicle, the degree of deterioration of the target secondary battery is predicted using information about the extracted similar vehicle.

3. 3. The method for predicting the life of a secondary battery according to claim 2, wherein the similar vehicle determination model is generated by machine learning.

4. a step of predicting a degree of deterioration of the target secondary battery mounted on the target vehicle, which is executed after the step of extracting the similar vehicle, A method for predicting the lifespan of a secondary battery as described in claim 2 or claim 3, characterized in that it includes a step of estimating the current degree of deterioration of the target secondary battery installed in the target vehicle based on the information regarding the extracted similar vehicle.

5. After the step of estimating the current degree of deterioration of the target secondary battery mounted in the target vehicle, generating a prediction model that predicts a future degree of deterioration of the target secondary battery mounted in the target vehicle based on information about the similar vehicle and information about the estimated current degree of deterioration; predicting a future degree of deterioration of the target secondary battery using the generated prediction model; 5. The method for predicting the life of a secondary battery according to claim 4, further comprising:

6. The method for predicting a lifespan of a secondary battery according to claim 5 , wherein the prediction model is generated by machine learning.

7. After the step of acquiring information about the target vehicle and information about the plurality of vehicles, converting the acquired information about the target vehicle and the acquired information about the plurality of vehicles into distribution information indicating a cumulative time for each piece of data included in the information; 7. The method for predicting the life of a secondary battery according to claim 2, wherein the similar vehicle determination model is generated based on the distribution information.

8. a model generation unit that generates a similar vehicle determination model using the acquired information about the target vehicle equipped with the target secondary battery that is the subject of life prediction and information about a plurality of vehicles equipped with the secondary battery; an extraction unit that extracts a similar vehicle that is similar to the target vehicle from among the plurality of vehicles using the generated similar vehicle determination model; a determination unit that predicts a degree of deterioration of the secondary battery mounted on the target vehicle using the extracted information about the similar vehicle, the information about the target vehicle includes information about the state of the target secondary battery installed therein and information indicating a usage status of the target vehicle, and the information about the plurality of vehicles includes information about the state of the secondary battery installed in each vehicle and information indicating a usage status of each vehicle, A secondary battery life prediction device characterized in that the information indicating the usage status of the target vehicle and the information indicating the usage status of each of the multiple vehicles includes at least one of data regarding the owner, data regarding the vehicle's driving, or data regarding the environment in which the vehicle was driven.

Images