Vehicle battery diagnostic system

Abstract

A vehicle battery assessment system (1) is for assessing the remaining life of a battery (11) installed in a vehicle (10). The vehicle battery assessment system (1) comprises an information acquisition section (101), a trained model acquisition section (103), a remaining-life assessment section (105), and an assessment result output section (106). The information acquisition section (101) acquires at least vehicle information about the vehicle, and usage history information about the battery (11). The trained model acquisition section (103) acquires a trained model created in advance by defining at least the vehicle information about the vehicle (10) in which the battery (11) is installed and the usage history information about the battery (11) as explanatory variables, and the remaining life of the battery (11) as a target variable. The remaining-life assessment section (105) assesses the remaining life of the battery (11), on the basis of the vehicle information and the usage history information acquired by the information acquisition section (101), and the trained model. The assessment result output section (106) outputs the assessment result from the remaining-life assessment section (105).

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application is based on Japanese Application No. 2022-79294, filed on May 13, 2022, the contents of which are incorporated herein by reference. [Technical Field]

[0002] The present disclosure relates to a vehicle battery diagnostic system. [Background technology]

[0003] Conventionally, as a configuration for diagnosing and checking the life of a battery mounted on a vehicle, for example, Patent Document 1 discloses a configuration in which the deterioration state of multiple battery units constituting a battery mounted on a vehicle is diagnosed, and based on this, the remaining life of the battery units is displayed on multiple screens so that the vehicle manufacturer, dealer, vehicle user, etc. can each check it. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2010-111276 Summary of the Invention

[0005] The configuration disclosed in Patent Document 1 uses the battery voltage and resistance value of the battery unit to quickly diagnose the deterioration state, but does not take into account vehicle information, including the region in which the vehicle equipped with the battery is used and the vehicle's mileage, and other vehicle usage information. As a result, the accuracy of the deterioration information is low, and the accuracy of remaining life diagnosis is also low. Therefore, there is room for improvement in order to diagnose the remaining life with high accuracy.

[0006] The present disclosure aims to provide a vehicle battery diagnostic system that can diagnose the life of a battery mounted on a vehicle with high accuracy.

[0007] One aspect of the present disclosure is a vehicle battery diagnostic system for diagnosing a remaining life of a battery mounted on a vehicle, comprising: an information acquisition unit that acquires at least vehicle information of the vehicle and usage history information of the battery; a trained model acquisition unit that acquires a trained model that is created in advance using at least vehicle information of the vehicle in which the battery is mounted and usage history information of the battery as explanatory variables and using the remaining life of the battery as a target variable; a remaining life diagnosis unit that diagnoses a remaining life of the battery based on the vehicle information and the usage history information acquired by the information acquisition unit and the trained model; a diagnostic result output unit that outputs a diagnostic result of the life diagnostic unit; a model update unit that updates the trained model based on the information acquired by the information acquisition unit; an update necessity determination unit that determines whether the battery is necessary for updating the trained model; Equipped with 、 The update necessity determination unit determines whether the information acquired by the information acquisition unit is When the information belongs to an unlearned area in the trained model, When the inter-data distance between the data center calculated from information belonging to the learning area in the trained model within a predetermined period from the latest model update and the data center calculated from all information in the learning area is equal to or greater than a reference value, When a difference between data distortion calculated from information belonging to a learning domain in the trained model within a predetermined period from the latest model update and data distortion calculated from all information in the learning domain is equal to or greater than a reference value; and When the difference between the data density calculated from information belonging to the learning area in the trained model within a predetermined period from the latest model update and the data density calculated from all information in the learning area is equal to or greater than a reference value, When at least one of the following is satisfied, it is determined that the trained model needs to be updated. , in a vehicle battery diagnostic system. Another aspect of the present disclosure is a vehicle battery diagnostic system for diagnosing a remaining life of a battery mounted on a vehicle, comprising: an information acquisition unit that acquires at least vehicle information of the vehicle, usage history information of the battery, and battery characteristics of the battery; a trained model acquisition unit that acquires a trained model that is created in advance using at least vehicle information of the vehicle in which the battery is mounted and usage history information of the battery as explanatory variables and using the remaining life of the battery as a target variable; a remaining life diagnosis unit that diagnoses a remaining life of the battery based on the vehicle information and the usage history information acquired by the information acquisition unit and the trained model; a diagnostic result output unit that outputs a diagnostic result of the remaining life diagnostic unit; an association information acquisition unit that compares the battery characteristics acquired by the information acquisition unit from the battery in a state where the battery is mounted on the vehicle with direct battery characteristics acquired by removing the battery from the vehicle and directly measuring the battery, and associates the information acquired by the information acquisition unit with battery identification information of the battery; The present invention relates to a vehicle battery diagnostic system. Yet another aspect of the present disclosure is a vehicle battery diagnostic system for diagnosing a remaining life of a battery mounted in a vehicle, comprising: an information acquisition unit that acquires at least vehicle information of the vehicle and usage history information of the battery; a trained model acquisition unit that acquires a trained model that is created in advance using at least vehicle information of the vehicle in which the battery is mounted and usage history information of the battery as explanatory variables and using the remaining life of the battery as a target variable; a diagnosis possibility determination unit that determines whether or not the remaining life of the battery can be diagnosed; a remaining life diagnosis unit that diagnoses a remaining life of the battery that has been determined to be diagnosable by the diagnosability determination unit based on the vehicle information and the usage history information acquired by the information acquisition unit and the trained model; a diagnostic result output unit that outputs a diagnostic result of the remaining life diagnostic unit; Equipped with The diagnosis feasibility determination unit is in a vehicle battery diagnosis system that determines whether the battery can be diagnosed based on whether the information acquired by the information acquisition unit is within the validity range of the trained model acquired by the trained model acquisition unit.

[0008] In the vehicle battery diagnostic system, the remaining life of a battery installed in a vehicle is diagnosed based on at least information about the vehicle in which the battery is installed, information about the battery's usage history, and a pre-created battery model, thereby enabling the remaining life of the battery to be diagnosed with high accuracy in a short period of time.

[0009] As described above, according to the above aspect, it is possible to provide a vehicle battery diagnostic system that can diagnose the remaining life of a battery mounted on a vehicle with high accuracy.

[0010] Note that the symbols in parentheses in the claims indicate the correspondence with the specific means described in the embodiments described below, and do not limit the technical scope of the present disclosure. [Brief explanation of the drawings]

[0011] The above and other objects, features and advantages of the present disclosure will become more apparent from the following detailed description taken in conjunction with the accompanying drawings, in which: [Figure 1] FIG. 1 is a block diagram showing the configuration of a vehicle battery diagnostic system according to a first embodiment; [Figure 2] FIG. 2 is a conceptual diagram illustrating a trained model feature space in the first embodiment; [Figure 3] FIG. 3 is a conceptual diagram showing the relationship between collected batteries, estimated rebuilt order placement, and inventory targets in the first embodiment. [Figure 4] FIG. 4 is a flowchart of a preliminary determination process according to the first embodiment; [Figure 5] FIG. 5 is a flowchart of a post-preliminary determination process according to the first embodiment; [Figure 6] FIG. 6 is a flow diagram of a model update process according to the first embodiment; [Figure 7] FIG. 7 is a flow diagram of an inventory update process in the first embodiment. [Figure 8] FIG. 8 is a flow diagram of remaining life assessment processing and rebuild processing in the first embodiment; [Figure 9]FIG. 9 is a flow diagram of the association process in the first embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0012] (Embodiment 1) An embodiment of the vehicle battery diagnostic system will be described with reference to FIGS. 1 to 9. FIG. The vehicle battery diagnostic system 1 of the first embodiment is for diagnosing the remaining life of a battery 11 mounted on a vehicle 10. As shown in Fig. 1, the vehicle battery diagnostic system 1 includes an information acquisition unit 101, a trained model acquisition unit 103, a remaining life diagnosis unit 105, and a diagnosis result output unit 106. The information acquisition unit 101 acquires at least vehicle information about the vehicle and usage history information about the battery 11 . The trained model acquisition unit 103 acquires a trained model that has been created in advance using at least vehicle information of the vehicle 10 equipped with the battery 11 and usage history information of the battery 11 as explanatory variables, and the remaining life of the battery 11 as a target variable. The remaining life assessment unit 105 assesses the remaining life of the battery 11 based on the vehicle information and usage history information acquired by the information acquisition unit 101 and the trained model. The diagnosis result output unit 106 outputs the diagnosis result of the remaining life diagnosis unit 105 .

[0013] The vehicle battery diagnostic system 1 of this embodiment will be described in detail below. 1. Configuration of vehicle battery diagnostic system 1 In the first embodiment, the vehicle battery diagnostic system 1 is constructed from components provided in a dealer 100, a server 200, and a centralized repair shop 300, as shown in Fig. 1. Note that each component is not necessarily limited to being provided in the dealer 100, the server 200, and the centralized repair shop 300 shown in Fig. 1, and each component may be provided in any of the dealer 100, the server 200, and the centralized repair shop 300. Furthermore, all of the elements constituting the vehicle battery diagnostic system 1 are not limited to being provided in the dealer 100, the server 200, and the centralized repair shop 300, and may be provided on a terminal or cloud connected via a communication means.

[0014] The battery 11 to be diagnosed by the vehicle battery diagnostic system 1 is mounted on a vehicle 10 owned by a user. The battery 11 is a rechargeable secondary battery, and may be composed of a single cell or multiple cells. The multiple batteries 11 are electrically connected to each other to form a battery pack 12. Each battery 11 forms a module that can be individually attached and detached from the battery pack 12. There are no restrictions on the type of secondary battery used for the battery 11, and any battery suitable for the vehicle 10 can be used.

[0015] 1-1. About Dealer 100 1, dealer 100 has information acquisition unit 101, diagnosis feasibility determination unit 102, trained model acquisition unit 103, trained model extraction unit 104, remaining life diagnosis unit 105, diagnosis result output unit 106, update necessity determination unit 107, storage necessity determination unit 108, association information acquisition unit 109, battery shipping unit 110, rebuild information request input unit 120, information transmission unit 121, information reception unit 122, information presentation unit 123, rebuild request reception and transmission unit 124, acceptance unit 125, and assembly unit 126. Of these, battery shipping unit 110 can be configured by a shipping system (not shown), information presentation unit 123 can be configured by a display device (not shown), and the other components can be configured by an arithmetic processing unit (not shown).

[0016] The information acquisition unit 101 acquires at least vehicle information about the vehicle 10 and usage history information about the battery 11. In the first embodiment, the information acquisition unit 101 also acquires the battery characteristics of the battery 11. First, the vehicle information is information about the vehicle 10, and examples thereof include the vehicle model, manufacturing date, usage period, mileage, and usage area of ​​the vehicle 10. Note that the manufacturing date does not necessarily have to be specified down to the day, and may be specified down to the year or month. The vehicle information can be acquired from a storage device (not shown) provided in the vehicle 10.

[0017] The usage history information of the battery 11 may include maximum, minimum, average, and cumulative values ​​for a predetermined period, such as the charging and discharging of the battery 11, battery temperature, battery voltage, capacity, and SOC (State of Charge), or the usage period of the device in which the battery 11 is installed. The predetermined period may be any period up to the present, or may be the entire period from the manufacture of the secondary battery to the present. The usage history information of the battery 11 may be acquired from a BMU (Battery Management Unit) provided in the battery 11. The battery characteristics of the battery 11 include information such as the current battery temperature, battery voltage, capacity, SOC, and internal resistance of the battery 11. These may also be acquired from the BMU provided in the battery 11.

[0018] Typically, the information acquisition unit 101 acquires the above information from the battery 11 installed in the vehicle 10 brought by the user to the dealer 100. Therefore, among the battery information acquired from the battery 11, the battery temperature and other battery information do not usually have the same values ​​among the multiple batteries 11 installed in the vehicle 10, and usually includes values ​​that are different from one another.

[0019] Next, the diagnosis possibility determination unit 102 determines whether or not it is possible to diagnose the remaining life of the battery 11 based on the information acquired by the information acquisition unit 101. For example, if the SOC, voltage, or remaining capacity of the battery 11 is below a predetermined value, or if the internal resistance, number of charge / discharge cycles of the battery 11, or the manufacturing date, period of use, or mileage of the vehicle 10 is equal to or greater than a predetermined value, the diagnosis possibility determination unit 102 can determine that it is not possible to diagnose the battery 11.

[0020] The diagnosis feasibility determination unit 102 may also determine whether the battery 11 can be diagnosed based on whether the information acquired by the information acquisition unit 101 is included in a range of validity of a trained model previously set in the trained model acquired by the trained model acquisition unit 103 (described later). The range of validity of the trained model can be set, for example, based on the training data of each trained model. For example, the range of validity of each trained model is defined as a range in which the distance between each data center and the data in the data group of the training data of each trained model is within a predetermined reference value. If the information acquired by the information acquisition unit 101 determines that any of the distances between each data center and the training data of multiple trained models is within the predetermined reference value, the diagnosis is determined to be possible. Otherwise, the diagnosis is determined to be impossible. Note that the diagnosis feasibility determination unit 102 may determine whether the diagnosis can be performed after the trained model extraction unit 104 extracts the trained model, or may perform the determination simultaneously with the trained model extraction unit 104 extracts the trained model.

[0021] Since it is clear that a battery 11 determined by the diagnosability determination unit 102 to be undiagnosable cannot be reused, even without undergoing the diagnosis described below, the computational load can be reduced by excluding the battery 11 before the diagnosis described below is performed. A battery 11 determined to be undiagnosable can be discarded or disassembled to recycle its components.

[0022] The trained model acquisition unit 103 acquires a trained model from the server 200, which will be described later. The trained model uses the above-mentioned information acquired by the information acquisition unit 101 as an explanatory variable and the remaining life of the battery 11 as a target variable. The trained model can be created by machine learning using prepared training data. The trained model can be updated by operating the vehicle battery diagnostic system 1 of this embodiment 1, as will be described later. The trained model acquisition unit 103 can then acquire the latest updated trained model. A plurality of trained models can be prepared, as will be described later, and in this embodiment 1, the trained model acquisition unit 103 acquires a plurality of trained models.

[0023] The trained model extraction unit 104 extracts the optimal trained model from the multiple trained models acquired by the trained model acquisition unit 103. The extraction can be performed based on the information acquired by the information acquisition unit 101. For example, if the battery temperature in the information acquired by the information acquisition unit 101 is lower than a predetermined reference value, a trained model corresponding to the temperature can be extracted. Furthermore, for example, if the average value of the battery temperature history is lower than a predetermined reference value or if the frequency of battery temperature history at low temperatures is high, it can be predicted that the battery 11 was used in a cold region, and the trained model acquisition unit 103 can extract a trained model corresponding to this. Furthermore, if the information acquired by the information acquisition unit 101 indicates that the battery 11 is a specific vehicle model, a trained model corresponding to the specific vehicle model can be extracted. Furthermore, a trained model can be extracted based on the inter-data distance between the information acquired by the information acquisition unit 101 and each data center in the data group of training data for each trained model.

[0024] The remaining life assessment unit 105 assesses the remaining life of a battery 11 determined to be diagnosable by the diagnosis feasibility determination unit 102, based on the information acquired by the information acquisition unit 101 and the trained model. In the first embodiment, the optimal trained model extracted by the trained model extraction unit 104 is used as the trained model. This enables the remaining life to be assessed with higher accuracy. The assessment results of the remaining life assessment unit 105 are stored in an information integration unit 201 of the server 200, which will be described later. Note that the information integration unit 201 may store information for each diagnosed battery 11, and store the above information acquired by the information acquisition unit 101 for that battery 11 together with the remaining life assessment results.

[0025] The diagnostic result output unit 106 outputs the diagnostic result obtained by the remaining life diagnostic unit 105. The output format is not limited, and may be displayed on a display unit, printed on paper, or notified by voice. The output content may include the remaining life as well as other related information. The output content may be different for the user and the dealer. For example, the remaining life of the battery 11 may be displayed to the dealer along with related detailed information, while the user may only be informed of whether the battery 11 needs to be replaced.

[0026] The update necessity determination unit 107 determines whether the battery 11 is necessary for updating the trained model stored in the server 200 (described later). The determination criteria in the update necessity determination unit 107 are not limited, but in the first embodiment, it is determined that the update is necessary when any one of the following four criteria is satisfied:

[0027] The first update criterion is whether the information acquired by the information acquisition unit 101 in the battery 11 belongs to an unlearned region in the latest trained model stored in the trained model storage unit 203 of the server 200. For example, in the trained model shown in FIG. 2, in a feature space defined by feature A and feature B in the training data, it is determined whether the information belongs to an unlearned region L1, which is an unknown range outside the training region Lt, which is a known range in the training data. Here, the training region Lt is within a predetermined distance from the center position Ct in the data space of the training data used in the latest trained model, and the remaining range is the unlearned region L1. The distance between data can be determined based on Mahalanobis distance, Euclidean distance, Manhattan distance, Chebyshev distance, etc.

[0028] The second update criterion is whether or not the inter-data distance between the data center calculated from information belonging to the learning region Lt in the trained model within a predetermined period since the latest model update and the data center calculated from all information in the learning region Lt is equal to or greater than a reference value. For example, as shown in Figure 2, it is determined whether or not the inter-data distance D between the data center C2 in the data space of the data range L2 of N = 100 pieces of training data most recently added to the latest trained model and the data center C of all training data in the latest trained model is equal to or greater than a reference value.

[0029] The third update criterion is whether the difference between the data distortion calculated from information belonging to the learning region Lt of the trained model within a predetermined period since the latest model update and the data distortion calculated from all information in the learning region Lt is equal to or greater than a reference value. For example, as shown in FIG. 2, it is determined whether the difference between the data distortion in the data space of a data range L3 of N=100 pieces of training data most recently added to the latest trained model and the data distortion calculated from all information in the learning region Lt is equal to or greater than a reference value. The data distortion can be calculated, for example, based on the degree of deformation of the outer shape of a data range in the data space where the inter-data distance D from the data center C of all training data is within a reference value.

[0030] The fourth update criterion is whether the difference between the data density calculated from information belonging to the learning region Lt in the trained model within a predetermined period since the latest model update and the data density calculated from all information in the learning region Lt is equal to or greater than a reference value. For example, as shown in FIG. 2, it is determined whether the difference between the data density in the data space of the data ranges L4a and L4b of N=100 training data items most recently added to the latest trained model and the data density calculated from all information in the learning region Lt is equal to or greater than a reference value. The data density can be calculated, for example, as the number of data items per unit area of ​​the target data range in the data space.

[0031] The update necessity determination unit 107 determines that updating of the trained model is necessary when the information acquired by the information acquisition unit 101 satisfies any of the above first to fourth update criteria.

[0032] The storage necessity determination unit 108 determines whether the battery 11 needs to be stored for rebuilding at a centralized repair shop 300 (described later). Rebuilding refers to packing used assembled batteries into used battery packs for reuse. For example, at the centralized repair shop 300, used assembled batteries are disassembled into modules or cells, stratified according to ranks assigned based on the degree of deterioration, etc., and stored in a storage warehouse. Rebuilding can be performed by extracting modules or cells according to the required rank from the storage warehouse, reassembling them, and packing them. Disassembling and sorting used assembled batteries is not necessarily required. Instead, used assembled batteries may be rebuilt by removing the circuit boards, wiring, and housing to form a battery stack, which is an assembly of battery cells or modules, and using the battery stack as is in a new pack.

[0033] The rebuild storage standard, which is the judgment standard in the storage necessity judgment unit 108, is defined based on the inventory target received from the server 200, which will be described later. For example, in the example shown in FIG. 3(a), the relationship between the number of collected batteries 11 and the value of those batteries 11, for example, their capacity, is shown as recovered batteries B. The relationship between the number of batteries estimated to be needed for rebuilding within a predetermined period in the future and their value rank can be shown as estimated rebuild order Rb1. An inventory target Rb2 is set as a number that has a certain degree of leeway corresponding to the estimated rebuild order Rb1. Then, in the relationship between the inventory target Rb2 and collected batteries B, whether or not the battery falls within area B1 surrounded by the inventory target Rb2, collected batteries B, and the horizontal axis becomes the rebuild storage standard in the storage necessity judgment unit 108.

[0034] Based on the storage criteria for rebuilding, batteries 11 that fall within region B1 are determined to need to be stored for rebuilding, while batteries 11 that do not fall within region B1 are determined not to need to be stored. In particular, in FIG. 3(a), among the recovered batteries B, batteries 11 that fall within region B2, which is above the inventory target Rb2 on the page, are determined not to need to be stored for rebuilding because they may become surplus inventory. Note that if there are multiple uses for the batteries 11, the estimated rebuilding order Rb1 may take on a shape with multiple peaks in the graph showing the relationship between value rank and number of batteries. For example, as in the example shown in FIG. 3(b), the estimated rebuilding order Rb1 has two peaks, and the inventory target Rb2 also has two corresponding peaks. In this case, the storage criteria for rebuilding are also determined based on the inventory target Rb2, just as in the case of FIG. 3(a).

[0035] As will be described later, in the first embodiment, the batteries stored in the battery storage unit 306 in the central repair shop 300 are ranked based on the diagnostic results of the batteries 11, and the inventory information includes the number of batteries for each rank. Based on this, the storage standards for rebuilding batteries also stipulate the number of batteries for each rank. In other words, the storage standards for rebuilding batteries are stipulated based on the inventory status in the battery storage unit 306 and the diagnostic results of the batteries 11.

[0036] The association information acquisition unit 109 compares the battery information acquired by the information acquisition unit 101 from the battery 11 while it is mounted on the vehicle with the battery information acquired from the battery 11 while it is removed from the vehicle, confirms that both pieces of battery information were acquired from the same battery 11, and then acquires association information that associates these pieces of information with the battery identification information of the battery 11. The form of the battery identification information is not limited, and can be a number, a symbol, a one-dimensional barcode, a two-dimensional barcode, or a combination of these.

[0037] The battery information in the association information may be, for example, the battery voltage of the battery 11. The association information acquisition unit 109 may acquire the association information immediately after the battery is removed from the vehicle 10, or may acquire the association information when a predetermined period of time has elapsed since the battery is removed from the vehicle 10. When acquiring the association information when a predetermined period of time has elapsed, the battery information in the association information may be corrected in consideration of changes in the battery information over time, and the association information may be acquired using the corrected data.

[0038] The battery shipping unit 110 ships the battery 11 to the central repair shop 300, which will be described later. The shipped battery 11 is one that has been determined by the update necessity determination unit 107 to be necessary for updating the trained model, and one that has been determined by the storage necessity determination unit 108 to be necessary for storage for rebuilding.

[0039] Next, the rebuild information request input unit 120 can be used by the user to input information regarding rebuilding the battery pack 12 of the vehicle 10 after checking the output result of the diagnosis result output unit 106, for example, whether or not the battery 11 needs to be replaced. When the rebuild information request is input, the information transmission unit 121 transmits the input result to the central repair shop 300, which will be described later. Then, the information reception unit 122 receives the rebuild information, which will be described later, transmitted from the central repair shop 300, and the received information is presented to the user by the information presentation unit 123. The information can be presented by displaying it on a display unit (not shown), printing it on paper, or notifying it by voice. The rebuild information includes information such as the delivery time and cost of rebuilding. In addition to this, the rebuild information may also include the delivery time and cost of parts required to replace the battery 11.

[0040] If the user requests a rebuilt battery pack after checking the rebuilding information, the rebuilding request acceptance / transmission unit 124 accepts the rebuilding request and transmits the request to the central repair shop 300. The receiving unit 125 accepts the rebuilt battery pack shipped from the central repair shop 300, and an assembly unit 126 incorporates it into the user's battery pack 12. The battery pack 12 is then mounted on the vehicle 10. If the user wishes to replace the battery with a new one rather than a rebuilt one, the rebuilding request acceptance / transmission unit 124 accepts the request and can place an order for a new one via a new product ordering unit 127.

[0041] 1-2. About Server 200 As shown in Fig. 1, server 200 is connected to dealer 100 and a centralized repair shop 300 (described later) via a network line. Server 200 can be provided on the cloud via an internet line. Server 200 includes an information integration unit 201, a model update unit 202, a trained model storage unit 203, an inventory information acquisition unit 204, and an inventory target creation unit 205. These components can be configured using a processing unit (not shown).

[0042] The information integration unit 201 integrates the information acquired by the information acquisition unit 101 of the dealer 100, the diagnosis results of the remaining life diagnosis unit 105, and the diagnosis results of the battery diagnosis unit 305 in the centralized repair shop 300 described below.

[0043] The model update unit 202 updates the trained model based on the information from the information integration unit 201. The updated trained model is stored in the trained model storage unit 203. The trained model can be a model created by machine learning using the information acquired by the information acquisition unit 101 as an explanatory variable and the remaining lifespan as a target variable. The trained model can use a predictive model such as a regression equation, for example, linear regression, LASSO regression, Ridge regression, decision tree, support vector regression, etc. Furthermore, a neural network or XGBoost (eXtreme Gradient Boosting / gradient boosting regression tree) can be configured.

[0044] The inventory information acquisition unit 204 acquires inventory information on batteries 11 stored as batteries 11 for rebuilding in a battery storage unit 306 in a centralized repair shop 300, which will be described later. Then, the inventory target creation unit 205 creates an inventory target based on a rebuild order forecast for a predetermined period. The inventory target can be created, for example, as shown in the example in FIG. 3.

[0045] 1-3.About Centralized Repair Shop 300 Centralized repair shop 300 performs repairs upon receiving requests from multiple dealers 100. Centralized repair shop 300 is equipped with an information receiving unit 301, a rebuilding information creating unit 302, an information transmitting unit 303, a receiving unit 304, a battery diagnosis unit 305, a battery storage unit 306, a rebuilding unit 307, and a rebuilt battery shipping unit 308. Of these, information receiving unit 301, rebuilding information creating unit 302, information transmitting unit 303, receiving unit 304, and battery diagnosis unit 305 can be configured using a processing unit (not shown).

[0046] Information receiving unit 301 receives input information on a rebuild request transmitted from information transmitting unit 121 of dealer 100. Rebuild information creating unit 302 creates rebuild information including the delivery date and cost of the rebuilt battery in accordance with the input information on the rebuild request. The created rebuild information is sent by information transmitting unit 303 to dealer 100 and received by information receiving unit 122.

[0047] The receiving unit 304 receives the battery 11 shipped by the battery shipping unit 110 of the dealer 100. The received battery 11 is diagnosed by the battery diagnosis unit 305. The diagnosis by the battery diagnosis unit 305 is more detailed and includes more diagnostic items than the diagnosis by the remaining life diagnosis unit 105 of the dealer 100. To improve diagnostic accuracy, the battery 11 may be diagnosed after the temperature of the battery 11 is set to a predetermined state. Therefore, it takes more time to obtain diagnostic results than the diagnosis by the remaining life diagnosis unit 105 of the dealer 100. The diagnostic method used by the battery diagnosis unit 305 is not limited, and any method capable of providing a detailed diagnosis may be used. For example, the battery 11 may acquire battery characteristics related to the transition of the battery state over a predetermined voltage range, and the degree of deterioration may be determined based on the battery characteristics or battery characteristic-related values ​​calculated based on the battery characteristics, thereby evaluating and ranking the value of the battery 11.

[0048] The battery storage unit 306 is configured as a warehouse where batteries 11 can be stored. The storage status of the batteries 11 in the battery storage unit 306, i.e., inventory information including the number of stored batteries 11 and battery information about these batteries 11, is transmitted to the server 200 and acquired by the inventory information acquisition unit 204. In the first embodiment, the batteries in the battery storage unit 306 are stored in a state where they are assigned a predetermined ranking based on the diagnosis results of the battery diagnosis unit 305. The above-mentioned inventory information also includes information about the number of batteries for each rank.

[0049] The rebuilding unit 307 receives a rebuilding request from the rebuilding request receiving and transmitting unit 124 of the dealer 100 and rebuilds the battery in accordance with the request. The rebuilt battery is then shipped to the dealer 100 by the rebuilt battery shipping unit 308 and received by the receiving unit 125.

[0050] 2. Control flow of vehicle battery diagnostic system 1 2-1. Pre-determination process Next, regarding the control flow of the vehicle battery diagnostic system 1, first, the preliminary determination process shown in Fig. 4 will be described. In the preliminary determination process, in step S1 shown in Fig. 4, the information acquisition unit 101 of the dealer 100 acquires vehicle information of the vehicle 10, usage history information of the battery 11, and the voltage as a battery characteristic of the battery 11. The acquired information is stored in the information integration unit 201 of the server 200.

[0051] Thereafter, in step S2, the update necessity determination unit 107 determines whether or not the information acquired by the information acquisition unit 101 satisfies the first update criterion. If it is determined that the acquired information satisfies the first update criterion, that is, if it is determined that the acquired information is information that belongs to the unlearned region L1 in the latest trained model stored in the trained model storage unit 203 of the server 200, the process proceeds to Yes in step S2, and it is determined in step S3 that an association process is to be performed, and the pre-determination process is terminated.

[0052] On the other hand, in step S2, if the update necessity determination unit 107 of the dealer 100 determines that the information acquired by the information acquisition unit 101 does not satisfy the first update criterion, the process proceeds to No in step S2, and in step S4 it is determined whether the information acquired by the information acquisition unit 101 satisfies any one of the second to fourth update criteria described above. If it is determined that the acquired information satisfies any one of the second to fourth update criteria, the process proceeds to Yes in step S4, and it is determined in the above-mentioned step S3 that the association process will be performed, and the preliminary determination process is terminated.

[0053] Furthermore, if it is determined in step S4 that the acquired information does not satisfy any of the second to fourth update criteria, the process proceeds to No in step S4. Then, in step S5, the storage necessity determination unit 108 of the dealer 100 determines whether or not the above-mentioned storage criteria for rebuilding are satisfied. If it is determined that the storage criteria for rebuilding are satisfied, the process proceeds to Yes in step S5, where it is determined in the above-mentioned step S3 that the association process will be performed, and the preliminary determination process ends.

[0054] On the other hand, if it is determined in step S5 that the storage standard for rebuilding is not met, the process proceeds to No in step S5, and it is determined in step S6 that disposal or recycling processing is to be performed, and the preliminary determination processing ends.

[0055] 2-2. Pre-judgment and post-processing Next, the pre-determination post-processing in the control flow of the vehicle battery diagnostic system 1 will be described. As shown in Fig. 5, the pre-determination post-processing involves parallel processing of steps S7 to S8, step S9, and step S10. In step S7, which is the first parallel processing, the battery 11 that was determined to be subjected to association processing in the above-mentioned step S3 is removed from the vehicle, and then the association processing described below is performed. Thereafter, in step S8, the battery is shipped to the central repair shop 300. In step S9, which is the second parallel processing, the battery 11 that was determined to be recycled in the above-mentioned step S6 is removed from the vehicle, and the parts are removed by disassembly or the like, and a recycling process is performed to reuse the parts. In addition, in step S10, which is the third parallel processing, the battery 11 that was determined to be discarded in the above-mentioned step S6 is removed from the vehicle and discarded.

[0056] 2-3.Model update process Next, the model update process in the control flow of the vehicle battery diagnostic system 1 will be described. As shown in Fig. 6, the model update process begins in step S11, when the central repair shop 300 receives a battery 11 from the dealer 100 that satisfies any one of the first to fourth update criteria. Then, in step S12, the battery diagnostic unit 305 of the central repair shop 300 performs a detailed diagnosis of the battery 11. The results of the detailed diagnosis are stored in the information integration unit 201 of the server 200. In the first embodiment, the degree of deterioration of the battery 11 is calculated as the diagnosis result, and the remaining lifespan is diagnosed. Then, in step S13, the model update unit 202 updates the trained model based on the diagnosis result, and the model update process ends.

[0057] 2-4. Inventory update process Next, the inventory update process in the control flow of the vehicle battery diagnostic system 1 will be described. As shown in Figure 7, the model update process begins in step S15, where the central repair shop 300 receives batteries 11 that meet the storage standards for rebuilding from the dealer 100. Then, in step S16, the batteries are stored as rebuilding inventory in the battery storage unit 306 of the central repair shop 300. Then, in step S17, the inventory information acquisition unit 204 of the server 200 acquires inventory information from the battery storage unit 306 and updates the inventory information. The inventory target creation unit 205 creates an inventory target based on the inventory information, updates the above-mentioned storage standards for rebuilding, and ends the inventory update process.

[0058] In addition, for batteries 11 shipped to the central repair shop 300 for the model update processing shown in Figure 6, it may be determined whether or not they satisfy the storage criteria for rebuilding in step S5 shown in Figure 4, and if they satisfy the storage criteria for rebuilding, the inventory update processing shown in Figure 7 may be applied and they may be stored in the battery storage unit 306 as inventory for rebuilding as necessary.

[0059] In addition, in step S15, the batteries 11 that meet the storage criteria for rebuilding and are shipped to the central repair shop 300 may also be subjected to a detailed diagnosis by the remaining life diagnosis unit 105, and the batteries 11 may be stored in the battery storage unit 306 with a ranking based on the results of the detailed diagnosis.

[0060] 2-5. Remaining life assessment and rebuilding The following describes the remaining life assessment process and rebuild process in the control flow of the vehicle battery diagnostic system 1. In step S1 shown in Fig. 4, the information acquisition unit 101 of the dealer 100 acquires vehicle information of the vehicle 10, usage history information of the battery 11, and battery characteristics of the battery 11, and then proceeds to step S20 in Fig. 8 as indicated by the symbol A.

[0061] Then, in step S20, the diagnosis possibility determination unit 102 determines whether the battery 11 can be diagnosed based on the above information acquired by the information acquisition unit 101. If it is determined that the battery 11 cannot be diagnosed, the process proceeds to No in step S20, and the flow ends.

[0062] On the other hand, if it is determined in step S20 that the battery 11 can be diagnosed, the process proceeds to Yes in step S20. Then, in step S21, the trained model extraction unit 104 extracts an optimal trained model.

[0063] Then, in step S22, the remaining life diagnosis unit 105 of the dealer 100 diagnoses the remaining life of the battery 11 when it is installed in the vehicle, based on the information acquired by the information acquisition unit 101 and the optimal trained model. Then, in step S23, the diagnosis result output unit 106 outputs the diagnosis result to a display unit (not shown) and also transmits it to the information integration unit 201 of the server 200.

[0064] Thereafter, in step S24, the rebuild information request input unit 120 of the dealer 100 determines whether or not the user has requested rebuild information that will serve as information for deciding whether or not to order a rebuilt product. If rebuild information has been requested, the process proceeds to Yes in step S24, and in step S25, the rebuild information creation unit 302 of the central repair shop 300 creates rebuild information, and the information transmission unit 303 transmits the rebuild information to the dealer 100. The rebuild information also includes information on the delivery time and cost when a new battery is used in place of the rebuilt product. The information reception unit 122 of the dealer 100 then receives the rebuild information, and the information presentation unit 123 presents the rebuild information to the user. The user can decide whether or not to proceed with rebuilding by taking into account the delivery time, cost, etc. of the rebuilt battery contained in the rebuild information.

[0065] Then, in step S26, the rebuild request acceptance / transmission unit 124 of the dealer 100 determines whether or not a rebuild request has been input by the user, indicating an intention to perform rebuilding. If a rebuild request has been input by the user, the process proceeds to Yes in step S26, and the rebuild request is transmitted to the central repair shop 300. When the rebuild request is received by the rebuilding unit 307 of the central repair shop 300, in step S27, a battery that meets the rebuild request is extracted from the plurality of batteries stored in the battery storage unit 306 at the central repair shop 300, and a rebuilt battery is created based on this. Alternatively, a plurality of rebuilt batteries of various ranks may be created in advance by combining the plurality of batteries determined to be stored in the battery storage unit 306, and these may be stored in the battery storage unit 306, and one that meets the rebuild request may be extracted from these stored rebuilt batteries. Thereafter, in step S28, the rebuilt battery that has been created or extracted is shipped from the central repair shop 300 to the dealer 100 by the rebuilt battery shipping unit 308.

[0066] Then, at dealer 100, after receiving unit 125 receives the rebuilt battery, in step S29, assembly unit 126 of dealer 100 forms battery pack 12 using the rebuilt battery and assembles it into vehicle 10. This ends the processing flow.

[0067] On the other hand, if there is no rebuilding request input from the user in step S26, the process proceeds to No in step S26, and in step S30 the rebuilding request acceptance / transmission unit 124 determines whether or not the user desires replacement with a new battery. If the user desires replacement with a new battery, the process proceeds to Yes in step S30, and in step S31 a new battery ordering unit 127 of the dealer 100 places an order for a new battery with a battery manufacturer (not shown). After the new battery arrives at the dealer 100, it is accepted by the receiving unit 125, and the battery pack 12 is formed using the new battery by the assembly unit 126 and assembled into the vehicle 10. This ends the processing flow. Also, if the user does not desire replacement with a new battery in step S30, the process proceeds to No in step S30, and in step S32 it is determined that the battery 11 will not be replaced, and the processing flow ends.

[0068] If the user has not requested rebuild information in step S24, the process proceeds to No in step S24, and it is determined in step S2 that the battery 11 will not be replaced, and this processing flow ends.

[0069] In this embodiment 1, a battery stored in the battery storage unit 306 in response to a rebuild request from a user is used, but instead, the battery may be initially installed in the user's vehicle 10, sent to a central repair shop 300, and partially replaced and repaired, before being returned to the user as a rebuilt product.

[0070] 2-6. Association process Next, the association processing in step S7 shown in Fig. 4 will be described with reference to Fig. 9. First, in step S22 shown in Fig. 8, the remaining life diagnosis unit 105 of the dealer 100 diagnoses the remaining life of the battery 11 while the battery 11 is installed in the vehicle, and stores the diagnosis result in the information integration unit 201 of the server 200. After that, as indicated by the symbol B, the process proceeds to step S40 in Fig. 9. Then, in step S40, the association information acquisition unit 109 of the dealer 100 acquires battery identification information for the battery 11 removed from the vehicle 10.

[0071] Then, in step S41, the battery voltage of the battery 11 after removal is directly measured and obtained as the direct battery characteristic. After that, in step S42, in order to take into account the change in battery voltage as a direct battery characteristic that occurs over time from when the battery 11 is removed from the vehicle 10 until the measurement of the battery voltage, the battery voltage as a direct battery characteristic is corrected according to the elapsed time. Note that if the elapsed time is shorter than a predetermined reference value, the change in battery voltage that occurs over time may be ignored and step S42 may not be performed.

[0072] Then, in step S43, the battery voltage before removal and after removal or correction are compared to determine whether they match. If it is determined that they match, proceed to Yes in step S43, and in step S44, associate the information acquired before removal from the vehicle 10 with the information on the battery 11 after removal, and end the flow.

[0073] On the other hand, if it is determined in step S43 that the battery voltage before removal and the battery voltage after removal or correction do not match, the information acquired before removal from vehicle 10 does not match the information on battery 11 after removal, and the two are not associated, so the process returns to step S41, where the battery voltage after removal is acquired again and corrected in step S42. Then, in step S43, the corrected battery voltage after removal is compared with the battery voltage before removal of a battery 11 other than the battery previously compared, and it is determined whether the two match. Based on the result of this comparison, step S44 or step S41 and subsequent steps are performed again as described above.

[0074] In this embodiment, the association process in step S7 is performed at the dealer 100, but this is not limited to this, and the process may be performed at the centralized repair shop 300 after the battery 11 is shipped from the dealer 100 to the centralized repair shop 300.

[0075] Next, the effects of the vehicle battery diagnostic system 1 of this embodiment 1 will be described in detail. The vehicle battery diagnostic system 1 of this embodiment 1 diagnoses the remaining life of the battery 11 mounted on the vehicle 10 based on at least information about the vehicle 10 on which the battery 11 is mounted, information about the battery usage history, and a trained model created in advance. This makes it possible to diagnose the remaining life of the battery 11 with high accuracy in a short time.

[0076] Furthermore, in the first embodiment, a diagnosis possibility determination unit 102 is provided that determines whether the remaining life of the battery 11 can be diagnosed, and the remaining life diagnosis unit 105 diagnoses the remaining life of the battery 11 that has been determined to be diagnosable by the diagnosis possibility determination unit 102. This makes it possible to avoid diagnosing the remaining life of the battery 11 for excessively deteriorated or damaged batteries, thereby improving the efficiency of the calculation process.

[0077] Furthermore, in the first embodiment, the trained model acquisition unit 103 acquires a plurality of trained models created in advance. The trained model acquisition unit 103 includes a trained model extraction unit 104 that extracts at least one from the plurality of trained models acquired by the trained model acquisition unit 103 based on the information acquired by the information acquisition unit 101. Furthermore, the remaining life assessment unit 105 uses the trained model extracted by the trained model extraction unit 104 as the trained model. This allows the remaining life assessment unit 105 to diagnose the remaining life based on the optimal trained model corresponding to the information acquired by the information acquisition unit 101, thereby improving the accuracy of the diagnosis.

[0078] Furthermore, in the first embodiment, a model update unit 202 is provided that updates the trained model based on the information acquired by the information acquisition unit 101. As a result, even if the deterioration trend of the battery 11 changes, the trained model is updated accordingly, so that the remaining life can be diagnosed with higher accuracy.

[0079] Furthermore, in the first embodiment, an update necessity determination unit 107 is provided that determines whether or not the battery 11 is required for updating the trained model. This allows detailed diagnosis to be performed only on the battery 11 that requires updating, thereby improving the efficiency of the calculation process.

[0080] In addition, in this embodiment 1, the update necessity determination unit 107 determines that the trained model needs to be updated when at least one of the following conditions is met: the information acquired by the information acquisition unit belongs to an untrained area of ​​the trained model; the information belongs to the training area of ​​the trained model, and the data distance between the data center calculated from information within a predetermined period since the latest model update and the data center calculated from all information in the training area is equal to or greater than a reference value; the information belongs to the training area of ​​the trained model, and the difference between the data distortion calculated from information within a predetermined period since the latest model update and the data distortion calculated from all information in the training area is equal to or greater than a reference value; or the information belongs to the training area of ​​the trained model, and the difference between the data density calculated from information within a predetermined period since the latest model update and the data density calculated from all information in the training area is equal to or greater than a reference value. This makes it possible to optimize the extraction of batteries 11 required for training the trained model, thereby further improving the accuracy of remaining life diagnosis.

[0081] Furthermore, in the first embodiment, a storage necessity determination unit 108 is provided that determines whether or not the battery 11 should be stored in the inventory for rebuilding of the battery 11. This makes it possible to prevent surplus batteries from being stored for rebuilding.

[0082] Furthermore, in the first embodiment, the storage necessity determination unit 108 determines whether or not the battery 11 is required for rebuilding based on an inventory target calculated according to inventory information of batteries stored as rebuilding inventory. This makes it possible to efficiently acquire and store batteries with battery characteristics that are in short supply in the rebuilding inventory, while also preventing excess inventory from being held, thereby optimizing inventory management.

[0083] Furthermore, in this embodiment 1, the batteries stored in the battery storage unit 306 in the central repair shop 300 are ranked based on the diagnostic results of the batteries 11, and the inventory information includes the number of batteries for each rank. The rebuild storage standards, which are the criteria for determining whether or not a battery should be stored in the rebuild stock, also stipulate the number of batteries for each rank; in other words, the rebuild storage standards are stipulated based on the inventory status in the battery storage unit 306 and the diagnostic results of the batteries 11. As a result, battery inventory management can be performed taking into account the rank of the batteries 11, further optimizing inventory management.

[0084] Furthermore, in the present embodiment 1, the information acquisition unit 101 further acquires the battery characteristics of the battery 11 as the above information. This allows the remaining life to be diagnosed taking the battery characteristics into consideration, thereby further improving the accuracy of the diagnosis.

[0085] Furthermore, in the first embodiment, the battery 11 further includes an associated information acquisition unit 109 that compares the battery characteristics acquired by the information acquisition unit 101 from the battery 11 while it is mounted on the vehicle 10 with the direct battery characteristics acquired by removing the battery 11 from the vehicle 10 and measuring it directly, and associates the information acquired by the information acquisition unit 101 with the battery identification information of the battery 11. This makes it possible to ensure that the information acquired when the battery 11 is mounted on the vehicle 10, where it is difficult to physically identify the battery 11, is information about the battery 11, preventing battery mix-ups and improving the reliability of the diagnostic results.

[0086] Furthermore, in the first embodiment, the association information acquisition unit 109 uses, as the direct battery characteristics, the battery characteristics acquired by directly measuring the battery 11, corrected based on the time elapsed from when the battery 11 was removed from the vehicle 10 until when the direct measurement was performed. This makes it possible to improve the accuracy of association even when using battery characteristics acquired some time after the battery was removed from the vehicle 10, prevent battery mix-up, and improve the reliability of the diagnosis results.

[0087] As described above, according to the first embodiment, it is possible to provide a vehicle battery diagnostic system 1 that can diagnose the remaining life of the battery 11 mounted on the vehicle 10 with high accuracy.

[0088] The present disclosure is not limited to the above-described embodiments, and can be applied to various embodiments without departing from the spirit of the present disclosure.

[0089] The features of the present disclosure are as follows: [Section 1] A vehicle battery diagnostic system (1) for diagnosing the remaining life of a battery mounted on a vehicle (10), comprising: an information acquisition unit (101) that acquires at least vehicle information of the vehicle and usage history information of the battery; a trained model acquisition unit (103) that acquires a trained model created in advance using at least vehicle information of the vehicle equipped with the battery and usage history information of the battery as explanatory variables and the remaining life of the battery as a target variable; a remaining life diagnosis unit (105) that diagnoses the remaining life of the battery based on the vehicle information and the usage history information acquired by the information acquisition unit and the trained model; a diagnostic result output unit (106) that outputs a diagnostic result of the remaining life diagnostic unit; A vehicle battery diagnostic system comprising: [Section 2] a diagnosis possibility determination unit (102) for determining whether the remaining life of the battery can be diagnosed; Item 2. The vehicle battery diagnostic system according to item 1, wherein the remaining life diagnostic unit diagnoses the remaining life of the battery that has been determined to be diagnosable by the diagnosis feasibility determination unit. [Section 3] The trained model acquisition unit acquires a plurality of trained models created in advance, a trained model extraction unit (104) that extracts at least one of the trained models acquired by the trained model acquisition unit based on the information acquired by the information acquisition unit; 3. The vehicle battery diagnostic system according to claim 1 or 2, wherein the remaining life assessment unit uses the trained model extracted by the trained model extraction unit as the trained model. [Section 4] 4. The vehicle battery diagnostic system according to claim 1, further comprising a model update unit (202) that updates the trained model based on the information acquired by the information acquisition unit. [Section 5] 5. The vehicle battery diagnostic system according to claim 4, further comprising an update necessity determination unit (107) that determines whether the battery requires updating of the trained model. [Section 6] The update necessity determination unit determines whether the information acquired by the information acquisition unit is When the information belongs to an unlearned area in the trained model, When the inter-data distance between the data center calculated from information belonging to the learning area in the trained model within a predetermined period from the latest model update and the data center calculated from all information in the learning area is equal to or greater than a reference value, When a difference between data distortion calculated from information belonging to a learning domain in the trained model within a predetermined period from the latest model update and data distortion calculated from all information in the learning domain is equal to or greater than a reference value; and When the difference between the data density calculated from information belonging to the learning area in the trained model within a predetermined period from the latest model update and the data density calculated from all information in the learning area is equal to or greater than a reference value, 6. The vehicle battery diagnostic system according to claim 5, wherein the system determines that the trained model needs to be updated when at least one of the following conditions is satisfied: [Section 7] 7. The vehicle battery diagnostic system according to any one of items 1 to 6, further comprising a storage necessity determination unit (108) that determines whether the battery should be stored in a battery rebuild inventory. [Section 8] The vehicle battery diagnostic system described in item 7, wherein the storage necessity determination unit determines whether the battery needs to be rebuilt based on an inventory target calculated according to inventory information of the batteries stored as the rebuilding inventory. [Section 9] 9. The vehicle battery diagnostic system according to any one of items 1 to 8, wherein the information acquisition unit further acquires battery characteristics of the battery. [Section 10] 10. The vehicle battery diagnostic system according to claim 9, further comprising an association information acquisition unit (109) that compares the battery characteristics acquired by the information acquisition unit from the battery while mounted on the vehicle with direct battery characteristics acquired by removing the battery from the vehicle and directly measuring the battery, and associates the information acquired by the information acquisition unit with battery identification information of the battery. [Section 11] 11. The vehicle battery diagnostic system according to claim 10, wherein the association information acquisition unit uses, as the direct battery characteristics, battery characteristics acquired by directly measuring the battery, corrected based on the elapsed time from when the battery was removed from the vehicle until when the direct measurement was performed.

[0090] Although the present disclosure has been described with reference to the embodiments, it is understood that the present disclosure is not limited to the embodiments or structures. The present disclosure also encompasses various modifications and modifications within the scope of equivalents. In addition, various combinations and forms, as well as other combinations and forms including only one element, more than one element, or less than one element, are also within the scope and spirit of the present disclosure.

Claims

1. A vehicle battery diagnostic system (1) for diagnosing the remaining life of a battery mounted on a vehicle (10), comprising: an information acquisition unit (101) that acquires at least vehicle information of the vehicle and usage history information of the battery; a trained model acquisition unit (103) that acquires a trained model created in advance using at least vehicle information of the vehicle equipped with the battery and usage history information of the battery as explanatory variables and the remaining life of the battery as a target variable; a remaining life diagnosis unit (105) that diagnoses the remaining life of the battery based on the vehicle information and the usage history information acquired by the information acquisition unit and the trained model; a diagnostic result output unit (106) that outputs a diagnostic result of the remaining life diagnostic unit; a model update unit (202) that updates the trained model based on the information acquired by the information acquisition unit; an update necessity determination unit (107) that determines whether the battery is necessary for updating the trained model; The update necessity determination unit determines whether the information acquired by the information acquisition unit is When the information belongs to an unlearned area in the trained model, When the inter-data distance between the data center calculated from information belonging to the learning area in the trained model within a predetermined period from the latest model update and the data center calculated from all information in the learning area is equal to or greater than a reference value, When a difference between data distortion calculated from information belonging to a learning domain in the trained model within a predetermined period from the latest model update and data distortion calculated from all information in the learning domain is equal to or greater than a reference value; and When the difference between the data density calculated from information belonging to the learning area in the trained model within a predetermined period from the latest model update and the data density calculated from all information in the learning area is equal to or greater than a reference value, and determining that the trained model needs to be updated when at least one of the following conditions is satisfied.

2. A vehicle battery diagnostic system as described in claim 1, wherein the information acquisition unit further acquires the battery characteristics of the battery.

3. A vehicle battery diagnostic system as described in claim 2, further comprising an association information acquisition unit (109) that compares the battery characteristics acquired by the information acquisition unit from the battery while it is mounted on the vehicle with direct battery characteristics acquired by removing the battery from the vehicle and measuring it directly, and associates the information acquired by the information acquisition unit with the battery identification information of the battery.

4. A vehicle battery diagnostic system (1) for diagnosing the remaining life of a battery installed in a vehicle (10), comprising: an information acquisition unit (101) that acquires at least vehicle information of the vehicle, usage history information of the battery, and battery characteristics of the battery; a trained model acquisition unit (103) that acquires a trained model created in advance using at least vehicle information of the vehicle equipped with the battery and usage history information of the battery as explanatory variables and the remaining life of the battery as a target variable; a remaining life diagnosis unit (105) that diagnoses the remaining life of the battery based on the vehicle information and the usage history information acquired by the information acquisition unit and the trained model; a diagnostic result output unit (106) that outputs a diagnostic result of the remaining life diagnostic unit; an association information acquisition unit (109) that compares the battery characteristics acquired by the information acquisition unit from the battery in a state where the battery is mounted on the vehicle with direct battery characteristics acquired by removing the battery from the vehicle and directly measuring the battery, and associates the information acquired by the information acquisition unit with battery identification information of the battery; A vehicle battery diagnostic system comprising:

5. A vehicle battery diagnostic system as described in claim 4, comprising a model update unit (202) that updates the trained model based on the information acquired by the information acquisition unit.

6. A vehicle battery diagnostic system as described in Claim 5, comprising an update necessity determination unit (107) that determines whether the battery is necessary for updating the trained model.

7. A vehicle battery diagnostic system as described in any one of claims 3 to 6, wherein the association information acquisition unit uses, as the direct battery characteristics, battery characteristics obtained by directly measuring the battery, corrected based on the elapsed time from when the battery was removed from the vehicle to when it was directly measured.

8. A diagnosis possibility determination unit (102) for determining whether or not the remaining life of the battery can be diagnosed, 7. The vehicle battery diagnostic system according to claim 1, wherein the remaining life diagnostic unit diagnoses the remaining life of the battery that has been determined as diagnosable by the diagnosis possibility determining unit.

9. A diagnosis possibility determination unit (102) for determining whether or not the remaining life of the battery can be diagnosed, 8. The vehicle battery diagnostic system according to claim 7, wherein the remaining life diagnostic unit diagnoses the remaining life of the battery determined to be diagnosable by the diagnosability determining unit.

10. The trained model acquisition unit acquires a plurality of trained models created in advance, a trained model extraction unit (104) that extracts at least one of the trained models acquired by the trained model acquisition unit based on the information acquired by the information acquisition unit; The vehicle battery diagnostic system according to any one of claims 1 to 6, wherein the remaining life assessment unit uses the trained model extracted by the trained model extraction unit as the trained model.

11. The trained model acquisition unit acquires a plurality of trained models created in advance, a trained model extraction unit (104) that extracts at least one of the trained models acquired by the trained model acquisition unit based on the information acquired by the information acquisition unit; The vehicle battery diagnostic system according to claim 7 , wherein the remaining life assessment unit uses the trained model extracted by the trained model extraction unit as the trained model.

12. A vehicle battery diagnostic system (1) for diagnosing the remaining life of a battery installed in a vehicle (10), comprising: an information acquisition unit (101) that acquires at least vehicle information of the vehicle and usage history information of the battery; a trained model acquisition unit (103) that acquires a trained model created in advance using at least vehicle information of the vehicle equipped with the battery and usage history information of the battery as explanatory variables and the remaining life of the battery as a target variable; a diagnosis possibility determination unit (102) for determining whether the remaining life of the battery can be diagnosed; a remaining life diagnosis unit (105) that diagnoses the remaining life of the battery determined to be diagnosable by the diagnosability determination unit based on the vehicle information and the usage history information acquired by the information acquisition unit and the trained model; a diagnostic result output unit (106) that outputs a diagnostic result of the remaining life diagnostic unit; Equipped with The diagnosis feasibility determination unit determines whether the battery can be diagnosed based on whether the information acquired by the information acquisition unit is within the validity range of the trained model acquired by the trained model acquisition unit. A vehicle battery diagnosis system.

13. The trained model acquisition unit acquires a plurality of trained models created in advance, a trained model extraction unit (104) that extracts at least one of the trained models acquired by the trained model acquisition unit based on the information acquired by the information acquisition unit; The vehicle battery diagnostic system according to claim 12, wherein the remaining life assessment unit uses the trained model extracted by the trained model extraction unit as the trained model.

14. A vehicle battery diagnostic system as described in Claim 12, comprising a model update unit (202) that updates the trained model based on the information acquired by the information acquisition unit.

15. A vehicle battery diagnostic system as described in Claim 14, comprising an update necessity determination unit (107) that determines whether the battery is necessary for updating the trained model.

16. The update necessity determination unit determines whether the information acquired by the information acquisition unit is When the information belongs to an unlearned area in the trained model, When the inter-data distance between the data center calculated from information belonging to the learning area in the trained model within a predetermined period from the latest model update and the data center calculated from all information in the learning area is equal to or greater than a reference value, When a difference between data distortion calculated from information belonging to a learning domain in the trained model within a predetermined period from the latest model update and data distortion calculated from all information in the learning domain is equal to or greater than a reference value; and When the difference between the data density calculated from information belonging to the learning area in the trained model within a predetermined period from the latest model update and the data density calculated from all information in the learning area is equal to or greater than a reference value, The vehicle battery diagnostic system according to claim 15, wherein it is determined that the trained model needs to be updated when at least one of the following conditions is satisfied:

17. A vehicle battery diagnostic system as described in any one of claims 12 to 16, wherein the information acquisition unit further acquires battery characteristics of the battery.

18. A vehicle battery diagnostic system as described in claim 17, further comprising an association information acquisition unit (109) that compares the battery characteristics acquired by the information acquisition unit from the battery while it is mounted on the vehicle with direct battery characteristics acquired by removing the battery from the vehicle and measuring it directly, and associates the information acquired by the information acquisition unit with the battery identification information of the battery.

19. A vehicle battery diagnostic system as described in Claim 18, wherein the association information acquisition unit uses, as the direct battery characteristics, battery characteristics obtained by directly measuring the battery, corrected based on the elapsed time from when the battery was removed from the vehicle to when it was directly measured.

20. A vehicle battery diagnostic system described in any one of claims 12 to 16, wherein the validity range of the trained model is the range in which the distance between the data center in the data group of training data for the trained model is within a predetermined standard value.

21. A vehicle battery diagnostic system as described in Claim 17, wherein the validity range of the trained model is the range in which the distance between the data center in the data group of training data for the trained model is within a predetermined standard value.

22. A vehicle battery diagnostic system as described in Claim 18, wherein the validity range of the trained model is the range in which the distance between the data center in the data group of training data for the trained model is within a predetermined standard value.

23. A vehicle battery diagnostic system as described in Claim 19, wherein the validity range of the trained model is the range in which the distance between the data center in the data group of training data for the trained model is within a predetermined standard value.

Images