Anomaly detection device, anomaly detection method, and anomaly detection program

Abstract

To make it possible to accurately recognize a state of a battery even when abnormality occurs in a device for monitoring the battery.SOLUTION: An abnormality detection device includes: an acquisition unit configured to acquire a current, a voltage, and temperature measured from a battery; an estimation unit configured to estimate an estimation value of a voltage corresponding to the acquired current and temperature by using an estimation model configured to estimate the estimation value of the voltage; a correction unit configured to correct the estimation model by deriving a correction value for making the estimation model to correspond to the battery using the estimated value and an error of the acquired voltage; and a detection unit configured to detect abnormality in the battery using the correction value.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present invention relates to an abnormality detection device, an abnormality detection method, and an abnormality detection program for detecting an abnormality when estimating battery degradation.

Background Art

[0002] Patent Document 1 discloses a power supply system including a main battery that stores electric power to supply power to a drive motor and an auxiliary battery that stores electric power to supply power to auxiliaries, and diagnoses whether there is an abnormality in the main battery and the auxiliary battery.

[0003] Patent Document 2 discloses a power supply system that stops auxiliary system equipment when the power supplied from a DC / DC converter exceeds a predetermined amount, and further stops the auxiliary system battery to diagnose an abnormality of the auxiliary system battery when the supplied power still exceeds the predetermined amount after stopping the auxiliary system equipment.

Prior Art Literature

Patent Literature

[0004]

Patent Document 1

Patent Document 2

Summary of the Invention

Problem to be Solved by the Invention

[0005] In the prior art, a moving body such as a vehicle is equipped with a battery for supplying electric power to equipment, and an abnormality detection device that executes diagnosis for detecting an abnormality of the battery is disclosed.

[0006] Incidentally, batteries installed in mobile vehicles deteriorate over time, so their status is periodically estimated and monitored by a battery monitoring ECU (Electronic Control Unit). However, if an abnormality such as an ECU reset occurs in the mobile vehicle, the sequentially estimated battery status is reset, and there may be a discrepancy between the actual battery status and the estimated battery status.

[0007] Conventional technology requires stopping the power supply from the battery to diagnose its condition, which can make it difficult to immediately detect discrepancies between the actual battery condition and the estimated battery condition. In other words, if a malfunction occurs in the battery monitoring device, there is a risk that the battery condition may not be accurately recognized.

[0008] The present invention aims to provide an abnormality detection device, an abnormality detection method, and an abnormality detection program that can accurately recognize the state of a battery even when an abnormality occurs in a battery monitoring device. [Means for solving the problem]

[0009] The abnormality detection device according to claim 1 includes: an acquisition unit that acquires current, voltage, and temperature measured from a battery; an estimation unit that estimates the voltage corresponding to the acquired current and temperature using an estimation model for estimating the voltage; a correction unit that corrects the estimation model by deriving a correction value to make the estimation model correspond to the battery using the error between the estimated value and the acquired voltage; and using the correction value, The discrepancy between the battery state indicated by the estimation model and the actual battery state installed in the vehicle, resulting from a reset process by a device that monitors the state of the battery installed in the vehicle, is determined by the estimation model. Abnormal as It includes a detection unit that detects, and

[0010] The anomaly detection device according to claim 1 acquires current, voltage, and temperature from the battery, estimates the estimated voltage corresponding to the acquired current and temperature using an estimation model, and derives the error between the estimated voltage and the acquired voltage. The anomaly detection device uses this error to derive a correction value to make the estimation model correspond to the battery, corrects the estimation model, and uses the correction value The discrepancy between the battery state indicated by the estimated model and the actual battery state installed in the vehicle, caused by a reset process by a device that monitors the battery state installed in the vehicle, is considered an anomaly in the estimated model. It detects the battery status, allowing for accurate recognition of the battery's condition.

[0011] The anomaly detection device according to claim 2 is the anomaly detection device according to claim 1, wherein the correction unit derives the correction value using a Kalman filter.

[0012] According to the abnormality detection device described in claim 2, the state of the battery can be accurately estimated.

[0013] The anomaly detection device according to claim 3 is the anomaly detection device according to claim 2, wherein the correction unit uses the Kalman filter to derive the correction value based on the estimated value, and Estimated model If an abnormality is detected, the reference is switched to the voltage and the correction value is derived.

[0014] According to the anomaly detection device described in claim 3, by switching to a higher accuracy standard, it is possible to return to a normal state more quickly.

[0017] Claim 4 The abnormality detection device described in claim 1 to In the anomaly detection device described above, if the correction value is greater than or equal to a predetermined threshold, the detection unit Estimated model Detect anomalies.

[0018] Claim 4 According to the anomaly detection device described, anomalies can be detected even when the battery is in use.

[0019] Claim 5 The abnormality detection device described in the claim 4In the abnormality detection device according to Estimated model an abnormality is detected.

[0020] Claim 5 According to the abnormality detection device described in , even when data indicating an abnormality is accidentally included, Estimated model the presence or absence of an abnormality can be accurately detected.

[0021] Claim 6 The abnormality detection device according to Claim 1 to In the abnormality detection device described in , the correction unit derives, as the correction value, an ion concentration in the battery or a thickness of a solid electrolyte interphase film on an electrode of the battery.

[0022] Claim 6 According to the abnormality detection device described in , the state of the battery can be estimated in consideration of a change in ion concentration in the battery and a growth degree of the solid electrolyte interphase film.

[0025] Claim 7 The abnormality detection method according to , wherein a computer executes processing for: acquiring current, voltage, and temperature measured from a battery; estimating the estimated value of the voltage corresponding to the acquired current and temperature using an estimation model that estimates an estimated value of voltage; deriving a correction value for adapting the estimation model to the battery using the estimated value and an error between the estimated value and the acquired voltage to correct the estimation model; and detecting an abnormality using the correction value. The discrepancy between the battery state indicated by the estimation model and the actual battery state installed in the vehicle, resulting from a reset process by a device that monitors the state of the battery installed in the vehicle, is determined by the estimation model. of the abnormality as detecting, the processing is executed by a computer.

[0026] Claim 7 The abnormality detection method according to , wherein: current, voltage, and temperature are acquired from a battery; an estimated voltage corresponding to the acquired current and temperature is estimated using an estimation model; and an error between the estimated voltage and the acquired voltage is derived. The abnormality detection method derives a correction value for adapting the estimation model to the battery using the error to correct the estimation model, and uses the correction value to The discrepancy between the battery state indicated by the estimated model and the actual battery state installed in the vehicle, caused by a reset process by a device that monitors the battery state installed in the vehicle, is considered an anomaly in the estimated model. It detects the abnormality. In other words, this abnormality detection method allows for accurate recognition of the battery's state.

[0027] Claim 8 The anomaly detection program described herein acquires current, voltage, and temperature measured from the battery, estimates the voltage corresponding to the current and temperature using an estimation model, derives a correction value to adapt the estimation model to the battery using the error between the estimated value and the acquired voltage, corrects the estimation model, and uses the correction value to... The discrepancy between the battery state indicated by the estimation model and the actual battery state installed in the vehicle, resulting from a reset process by a device that monitors the state of the battery installed in the vehicle, is determined by the estimation model. Abnormal as Detect and have the computer perform the necessary processing.

[0028] Claim 8 The computer on which the anomaly detection program described above is executed acquires current, voltage, and temperature from the battery, estimates the voltage corresponding to the acquired current and temperature using an estimation model, and derives the error between the estimated voltage and the acquired voltage. Anomaly detection program This involves using the error to derive a correction value to make the estimated model compatible with the battery, correcting the estimated model, and then using the correction value. The discrepancy between the battery state indicated by the estimated model and the actual battery state installed in the vehicle, caused by a reset process by a device that monitors the battery state installed in the vehicle, is considered an anomaly in the estimated model. It detects the issue. According to this anomaly detection program, the battery status can be accurately recognized. [Effects of the Invention]

[0029] According to the present invention, even if a malfunction occurs in the battery monitoring device, the battery status can be accurately recognized. [Brief explanation of the drawing]

[0030] [Figure 1] This block diagram shows an example of the hardware configuration of the vehicle according to this embodiment. [Figure 2] This is a block diagram showing an example of the functional configuration of the anomaly detection device of this embodiment. [Figure 3] This is a data flow diagram showing an example of the data flow in the process for detecting anomalies in this embodiment. [Figure 4] This graph shows an example of the relationship between charge level and voltage, used to explain the voltage estimation in this embodiment. [Figure 5] This graph shows an example of the relationship between charge level and voltage, used to explain the transition using the correction value of this embodiment. [Figure 6] This flowchart shows an example of the process flow for detecting an anomaly in this embodiment. [Modes for carrying out the invention]

[0031] This invention describes an abnormality detection device mounted on a vehicle. The abnormality detection device is a device that collects detection results by detecting current, voltage, and temperature from the battery and monitors the state of the battery. In this embodiment, the abnormality detection device is described in a form mounted on a mobile body. However, it is not limited to this. The abnormality detection device may be mounted on a terminal such as a personal computer, or on any device that is mounted together with the battery. In this embodiment, a vehicle is described as the mobile body that is equipped with a battery. However, it is not limited to this. The mobile body may be, for example, an electric bicycle or electric motorcycle, or an aerial vehicle such as a drone or aircraft, or any mobile body that is equipped with a battery.

[0032] (vehicle) As shown in Figure 1, the vehicle 10 according to this embodiment is composed of a battery 12, an on-board unit 20, and each on-board device 22. The battery 12 supplies power to each device mounted on the vehicle 10. The on-board unit 20 is a device that monitors the status of the battery 12.

[0033] The in-vehicle unit 20 consists of a CPU (Central Processing Unit) 20A, ROM (Read Only Memory) 20B, RAM (Random Access Memory) 20C, an in-vehicle communication interface (I / F) 20D, and an input / output interface (I / F) 20E. The CPU 20A, ROM 20B, RAM 20C, in-vehicle communication interface 20D, and input / output interface 20E are interconnected via an internal bus 20F so that they can communicate with each other.

[0034] The CPU20A is the central processing unit, which executes various programs and controls various components. Specifically, the CPU20A reads programs from the ROM20B and executes them using the RAM20C as its working area.

[0035] ROM20B stores various programs and data. In this embodiment, ROM20B stores an abnormality detection program 100 that collects detection results such as current, voltage, and temperature from the battery 12 and detects abnormalities in the battery 12, and a battery model 110 that represents the state of the battery 12. Upon execution of the abnormality detection program 100, the in-vehicle unit 20 executes a process to detect an abnormality in the battery 12. The battery model 110 is a model that estimates the voltage value output by the battery 12 according to the current value and temperature of the battery 12 and the average Li ion concentration of the negative electrode in the battery 12. RAM20C temporarily stores programs or data as a work area.

[0036] The in-vehicle communication interface 20D is an interface for connecting to an ECU (Electronic Control Unit) (not shown) installed in vehicle 10. This interface uses the CAN protocol for communication. The in-vehicle communication interface 20D is connected to an external bus (not shown).

[0037] The input / output interface 20E is an interface for communicating with the in-vehicle equipment 22. The input / output interface 20E is connected to the current sensor 22A, the voltage sensor 22B, and the temperature sensor 22C as part of the in-vehicle equipment 22. The current sensor 22A detects the current output by the battery 12 and outputs a signal indicating the current value to the in-vehicle unit 20 as a result of the detection. The voltage sensor 22B detects the voltage output by the battery 12 and outputs a signal indicating the voltage value to the in-vehicle unit 20 as a result of the detection. The temperature sensor 22C detects the temperature of the battery 12 and outputs a signal indicating the temperature to the in-vehicle unit 20 as a result of the detection.

[0038] As shown in Figure 2, in the in-vehicle device 20 of this embodiment, the CPU 20A functions as an acquisition unit 200, an estimation unit 210, an extraction unit 220, a correction unit 230, and a detection unit 240 by executing the abnormality detection program 100.

[0039] The acquisition unit 200 acquires detection results from the current sensor 22A, the voltage sensor 22B, and the temperature sensor 22C, respectively. Specifically, as shown in Figure 3 as an example, the acquisition unit 200 acquires the current value 300, the voltage value 310, and the temperature 320 from the current sensor 22A, the voltage sensor 22B, and the temperature sensor 22C, respectively.

[0040] The estimation unit 210 uses the acquired current value 300 and temperature 320 to estimate the voltage value output by the battery 12. Hereafter, the voltage value estimated by the estimation unit 210 will be referred to as the "estimated voltage".

[0041] The estimation unit 210 includes a battery model 110 that represents the state of the battery 12. The battery model 110 takes a current value 300 and a temperature 320 as inputs and outputs an estimated voltage 330 that the battery 12 outputs. Here, the state of the battery 12 is, for example, the relationship between the charge level and the open-circuit voltage in the battery 12 shown in Figure 4. The battery model 110 sets the positive electrode charging curve 400 and the negative electrode charging curve 410 according to the average Li (lithium) ion concentration of the negative electrode, the current value 300, and the temperature 320. The voltage output by the battery 12 is estimated according to the open-circuit potential (OCP) of the positive electrode and the open-circuit potential of the negative electrode at a given charge level. Here, the charge level according to this embodiment is the ratio of the current charge capacity to the full charge capacity at each of the positive and negative electrodes.

[0042] The derivation unit 220 derives the voltage error using the estimated voltage 330 estimated by the estimation unit 210 and the voltage value 310 of the battery 12 acquired by the acquisition unit 200.

[0043] The correction unit 230 uses the derived voltage error to derive a correction value 340 for the battery model 110 and corrects the battery model 110. Specifically, the correction unit 230 uses an extended Kalman filter (EKF) to derive a correction value 340 for the average Li ion concentration in the battery model 110 and corrects the battery model 110.

[0044] For example, if the average Li ion concentration at the negative electrode of battery 12 decreases, an error occurs between the voltage value 310 output by battery 12 and the estimated voltage 330 estimated by battery model 110. The correction unit 230 uses the error between the voltage value 310 and the estimated voltage 330 to derive a correction value 340 for the average Li ion concentration at the negative electrode of battery 12, and corrects battery model 110. As an example, as shown in Figure 5, battery model 110 shifts the current charge rate in the negative electrode charging curve 410 according to the correction value 340, and shifts the positive electrode charging curve 400 to correspond to the charge rate in the shifted negative electrode charging curve 410, thereby setting the positive electrode charging curve 420.

[0045] Furthermore, if the detection unit 240 (described later) does not detect an abnormality (the battery model 110 is in a normal state), the correction unit 230 sets the estimated voltage 330 to the true value and derives the correction value 340 based on the estimated voltage 330. If the detection unit 240 (described later) detects an abnormality (the battery model 110 is in an abnormal state), the correction unit 230 sets the voltage value 310 to the true value and derives the correction value 340 based on the voltage value 310. In other words, the correction unit 230 switches the reference (true value) according to the state of the battery model 110 and derives the correction value 340.

[0046] The detection unit 240 detects abnormalities in the battery model 110 using the correction value 340 derived by the correction unit 230. Specifically, the detection unit 240 detects an abnormality if the correction value 340 is greater than or equal to a predetermined threshold, as this indicates that the battery model 110 is in an abnormal state. When an abnormality is detected by the detection unit 240, the correction unit 230 switches the reference (true value) to the voltage value 310 and derives the correction value 340.

[0047] Here, normally, the Li-ion concentration of the battery 12 currently installed in the vehicle 12 fluctuates gradually, so the correction value 340 for the battery model 110 is similarly derived to a small value corresponding to the fluctuation in Li-ion concentration, and the battery model 110 is corrected. However, if the battery state indicated by the battery model 110 is reset (initialized) due to a reset process of the ECU (onboard unit 20), the state of the battery 12 indicated by the initialized battery model 110 and the state of the currently installed battery 12 will diverge. As a result, the voltage error between the voltage value 310 and the estimated voltage 330 will increase, and the value of the derived correction value 340 will increase. In other words, if the correction value 340 is above a predetermined threshold, the detection unit 240 detects an abnormality, indicating that the state of the battery 12 indicated by the battery model 110 and the state of the currently installed battery 12 are diverging.

[0048] (Control flow) The processing flow performed by the in-vehicle unit 20 of this embodiment will be explained using the flowchart in Figure 6. The processing in the in-vehicle unit 20 is performed by the CPU 20A of the in-vehicle unit 20, which functions as an acquisition unit 200, an estimation unit 210, an extraction unit 220, a correction unit 230, and a detection unit 240. The process for detecting an abnormality in the battery 12 shown in Figure 6 is executed, for example, when an instruction to detect an abnormality in the battery 12 is input.

[0049] In step S100, the CPU 20A sets the "estimated voltage" as the initial value, which is the average Li ion concentration of the negative electrode related to the battery model 110, and the reference (true value) when deriving the correction value 340.

[0050] In step S101, the CPU 20A obtains a current value of 300, a voltage value of 310, and a temperature of 320 from the battery 12.

[0051] In step S102, the CPU 20A estimates the estimated voltage 330 using the acquired current value 300 and temperature 320. Here, the estimated voltage 330 is estimated by the battery model 110.

[0052] In step S103, the CPU 20A uses the acquired voltage value 310 and the estimated voltage 330 to derive the voltage error.

[0053] In step S104, the CPU 20A uses the acquired voltage value 310 and the estimated voltage 330 to derive a correction value 340 for the average Li ion concentration at the negative electrode in the battery model 110.

[0054] In step S105, the CPU 20A determines whether the correction value 340 is greater than or equal to the threshold. If the correction value 340 is greater than or equal to the threshold (step S105: YES), the CPU 20A proceeds to step S106. On the other hand, if the correction value 340 is not greater than or equal to the threshold (the correction value 340 is less than the threshold) (step S105: NO), the CPU 20A proceeds to step S108.

[0055] In step S106, the CPU 20A determines whether or not the "estimated voltage" is set as the reference (true value) when deriving the correction value 340. If the "estimated voltage" is set as the reference (true value) (step S106: YES), the CPU 20A proceeds to step S107. On the other hand, if the "estimated voltage" is not set as the reference (true value) (the "voltage value" is set as the reference (true value)) (step S106: NO), the CPU 20A proceeds to step S110.

[0056] In step S107, CPU20A switches and sets the reference (true value) for deriving the correction value 340 to "voltage value".

[0057] In step S108, the CPU 20A determines whether or not a "voltage value" is set as the reference (true value) when deriving the correction value 340. If a "voltage value" is set as the reference (true value) (step S108: YES), the CPU 20A proceeds to step S109. On the other hand, if a "voltage value" is not set as the reference (true value) (i.e., an "estimated voltage" is set as the reference (true value)) (step S108: NO), the CPU 20A proceeds to step S110.

[0058] In step S109, CPU20A switches and sets the reference (true value) for deriving the correction value 340 to "estimated voltage".

[0059] In step S110, the CPU 20A corrects the battery model 110 using a correction value of 340.

[0060] In step S111, the CPU 20A determines whether or not to terminate the process of detecting an abnormality in the battery 12. If the process of detecting an abnormality in the battery 12 is terminated (step S111: YES), the CPU 20A terminates the process of detecting an abnormality in the battery 12. On the other hand, if the process of detecting an abnormality in the battery 12 is not terminated (step S111: NO), the CPU 20A proceeds to step S101 and obtains the current value 300, the voltage value 310, and the temperature 320.

[0061] As described above, according to this embodiment, even if a malfunction occurs in the battery monitoring device, the battery status can be accurately recognized.

[0062] In the above embodiment, a method for deriving a correction value for the average Li ion concentration of the negative electrode was described. However, the invention is not limited to this. A correction value for the average Li ion concentration of the positive electrode may also be derived, or the thickness of the solid electrolyte interface (SEI) layer coating formed on the electrode may be derived. When deriving the thickness of the solid electrolyte interface (SEI) layer coating, the battery model 110 sets the charging curve 400 for the positive electrode and the charging curve 410 for the negative electrode according to the thickness of the solid electrolyte interface (SEI) layer coating, the current value 300, and the temperature 320, and estimates the estimated voltage 330.

[0063] Furthermore, in the above embodiment, a configuration was described in which the detection unit 240 switches the criterion once it detects that the correction value 340 is equal to or greater than a threshold. However, it is not limited to this. The criterion may also be switched if the correction value 340 is detected to be equal to or greater than a threshold multiple times. For example, if the number of times the correction value 340 is equal to or greater than a threshold exceeds a predetermined number within a predetermined period after the detection that the correction value 340 is equal to or greater than a threshold exceeds a predetermined number, an abnormality may be detected and the criterion may be switched. Also, if the percentage of times the derived correction value 340 is equal to or greater than a threshold exceeds a predetermined percentage within the above period, the criterion may be switched.

[0064] In addition, the various processes that the CPU 20A reads and executes in the above embodiment may be executed by various processors other than the CPU. Examples of such processors include PLDs (Programmable Logic Devices) such as FPGAs (Field-Programmable Gate Arrays) whose circuit configuration can be changed after manufacturing, and dedicated electrical circuits that are processors with circuit configurations specifically designed to execute specific processes, such as ASICs (Application Specific Integrated Circuits). Furthermore, each of the above processes may be executed by one of these various processors, or by a combination of two or more processors of the same or different types (for example, multiple FPGAs, and a combination of a CPU and an FPGA). More specifically, the hardware structure of these various processors is an electrical circuit that combines circuit elements such as semiconductor elements.

[0065] Furthermore, in the above embodiment, each program was described as being pre-stored (installed) on a computer-readable non-temporary recording medium. For example, the abnormality detection program 100 in the in-vehicle unit 20 is pre-stored in ROM 20B. However, the program is not limited to this, and each program may be provided in a form recorded on a non-temporary recording medium such as a CD-ROM (Compact Disc Read Only Memory), DVD-ROM (Digital Versatile Disc Read Only Memory), and USB (Universal Serial Bus) memory. Alternatively, the program may be provided in a form downloaded from an external device via a network.

[0066] The processing flow described in the above embodiment is just one example, and unnecessary steps may be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose. [Explanation of Symbols]

[0067] 10 vehicles 12 batteries 20 Onboard equipment 20A CPU 20B ROM 20C RAM 20D In-vehicle communication interface 20E Input / Output Interface 20F Internal Bus 22 In-vehicle equipment 22A Current Sensor 22B Voltage Sensor 22C Temperature Sensor 100 Anomaly Detection Programs 110 Battery Model 200 Acquisition Department 210 Estimation Department 220 Derivation part 230 Correction Unit 240 Detection unit 300 Current Value 310 Voltage Value 320 temperature 330 Estimated Voltage 340 Correction Value Charging curves of the positive electrode at 400 and 420. 410 Negative electrode charging curve

Claims

1. An acquisition unit that acquires current, voltage, and temperature measured from the battery, An estimation unit that estimates the voltage corresponding to the acquired current and temperature using an estimation model for estimating voltage estimates, A correction unit that uses the estimated value and the error in the acquired voltage to derive a correction value to make the estimated model correspond to the battery and correct the estimated model, A detection unit detects, using the correction value, the discrepancy between the battery state indicated by the estimated model and the actual battery state installed in the vehicle, resulting from a reset process by a device that monitors the state of the battery installed in the vehicle, as an abnormality in the estimated model. An anomaly detection device equipped with the following features.

2. The correction unit, The correction value is derived using a Kalman filter. An anomaly detection device according to claim 1.

3. The correction unit, Using the Kalman filter, the correction value is derived based on the estimated value. If an anomaly is detected in the estimation model, the reference is switched to the voltage and the correction value is derived. An anomaly detection device according to claim 2.

4. The detection unit, If the correction value is greater than or equal to a predetermined threshold, an anomaly in the estimation model is detected. An anomaly detection device according to claim 1.

5. The detection unit, If the number of times the correction value is equal to or greater than the threshold exceeds a predetermined number within a predetermined period, an anomaly in the estimation model is detected. An anomaly detection device according to claim 4.

6. The correction unit, The correction value is derived from the ion concentration in the battery or the thickness of the solid electrolyte interface layer coating of the electrodes in the battery. An anomaly detection device according to claim 1.

7. The current, voltage, and temperature measured from the battery are acquired. Using an estimation model for estimating voltage, the estimated values ​​of the voltage corresponding to the acquired current and temperature are estimated. Using the estimated value and the error in the acquired voltage, a correction value is derived to make the estimation model compatible with the battery, and the estimation model is corrected. Using the aforementioned correction value, the discrepancy between the battery state indicated by the estimation model and the actual battery state installed in the vehicle, resulting from a reset process by a device that monitors the state of the battery installed in the vehicle, is detected as an anomaly in the estimation model. An anomaly detection method performed by a computer.

8. The current, voltage, and temperature measured from the battery are acquired. Using an estimation model for estimating voltage, the estimated values ​​of the voltage corresponding to the acquired current and temperature are estimated. Using the estimated value and the error in the acquired voltage, a correction value is derived to make the estimation model compatible with the battery, and the estimation model is corrected. Using the aforementioned correction value, the discrepancy between the battery state indicated by the estimation model and the actual battery state installed in the vehicle, resulting from a reset process by a device that monitors the state of the battery installed in the vehicle, is detected as an anomaly in the estimation model. An anomaly detection program that causes a computer to execute a process.

Images