Remaining life estimation method and guidance method, and guidance system, remaining life estimation device, and guidance device

Abstract

To provide a remaining life estimation method and remaining life estimation device, which enable estimation of remaining life of a maintenance target in consideration of repetitive loading, and to provide a guidance method, guidance system, and guidance device, which enable optimization of maintenance-related guidance in consideration of remaining life of a maintenance target.SOLUTION: A remaining life estimation method is provided, comprising a remaining life estimation step of estimating remaining life of a maintenance target 30 from operation data of the maintenance target 30 subjected to repetitive loading using at least one model 60.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present disclosure relates to a method and apparatus for estimating the remaining life of a maintenance target, as well as a guidance method, guidance apparatus, and guidance system for maintaining the maintenance target.

Background Art

[0002] As a conventional equipment maintenance, Patent Document 1 describes a method of obtaining the size and position of a crack by flaw detection and obtaining the creep life by a remaining life preservation model. Further, Patent Document 2 describes an apparatus for predicting the failure probability due to aging deterioration and the failure rate due to accidental failure using a physical model and a statistical model.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0004] The method described in Patent Document 1 can handle cracks due to local creep rupture, but cannot handle cracks due to vibration fatigue. The apparatus described in Patent Document 2 statistically processes without obtaining individual crack information, so it does not reflect the damage situation due to individual cracks. Further, the apparatus described in Patent Document 2 does not obtain a numerical value of the remaining life. Therefore, it is difficult to determine a specific and optimal maintenance plan in the apparatus described in Patent Document 2. It is required to estimate the remaining life of the maintenance target in consideration of the repeated load causing vibration fatigue.

[0005] Therefore, an object of the present disclosure is to provide a remaining life estimation method and a remaining life estimation device capable of estimating the remaining life of a maintenance object in consideration of repeated loads. Another object of the present disclosure is to provide a guidance method, a guidance system, and a guidance device capable of optimizing guidance regarding maintenance in consideration of the remaining life of a maintenance object.

Means for Solving the Problems

[0006] A remaining life estimation method according to an embodiment of the present disclosure includes a remaining life estimation step of estimating the remaining life of the maintenance object using at least one model from operation data of the maintenance object that receives repeated loads.

[0007] (2) The remaining life estimation method according to (1) above may further include an abnormality detection step of detecting that an abnormality has occurred in the maintenance object based on the operation data of the maintenance object. When it is detected in the abnormality detection step that an abnormality has occurred in the maintenance object, the remaining life estimation step may be executed.

[0008] (3) In the remaining life estimation method according to (1) or (2) above, the operation data of the maintenance object may include data related to at least one of vibrations, accelerations, loads, or sounds of at least a part of the maintenance object.

[0009] (4) In the remaining life estimation method according to any one of (1) to (3) above, the model may receive the operation data of the maintenance object as an input and output an estimation result of the remaining life of the maintenance object.

[0010] (5) In the remaining life estimation method according to any one of (1) to (4) above, the model may include a first model and a second model. The first model may receive the operation data of the maintenance object as an input and output crack information, which is information regarding cracks occurring in at least a part of the maintenance object. The second model may receive the crack information as an input and output an estimation result of the remaining life of the maintenance object.

[0011] (6) In the remaining life estimation method according to (5) above, the crack information may specify at least one of the length, direction, or position of a crack occurring in at least a part of the object to be maintained.

[0012] The guidance method according to an embodiment of the present disclosure includes a maintenance information generation step of generating maintenance information, which is information regarding maintenance of the object to be maintained, based on an estimation result of the remaining life of the object to be maintained obtained by executing the remaining life estimation method according to any one of (1) to (6) above.

[0013] (8) The guidance method according to (7) above may further include an output step of outputting the maintenance information in a predetermined format.

[0014] A guidance system according to an embodiment of the present disclosure includes a remaining life estimation device that executes the remaining life estimation method according to any one of (1) to (6) above, and a guidance device that executes the guidance method according to (7) or (8) above.

[0015] A remaining life estimation device according to an embodiment of the present disclosure executes the remaining life estimation method according to any one of (1) to (6) above.

[0016] A guidance device according to an embodiment of the present disclosure executes the guidance method according to (7) or (8) above.

Advantages of the Invention

[0017] According to the remaining life estimation method and the remaining life estimation device according to the present disclosure, the remaining life of the object to be maintained is estimated in consideration of repeated loads. Further, according to the guidance method, the guidance system, and the guidance device according to the present disclosure, the guidance regarding maintenance is optimized in consideration of the remaining life of the object to be maintained.

Brief Description of the Drawings

[0018]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5A

Figure 5B

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Embodiments for Carrying Out the Invention

[0019] Hereinafter, embodiments of a remaining life estimation method and a remaining life estimation device 10 (see FIG. 1), and a guidance method, a guidance system 1 (see FIG. 1), and a guidance device 20 (see FIG. 1) according to the present disclosure will be described with reference to the drawings. Each drawing is schematic and may be different from the actual one. Further, the following embodiments exemplify an apparatus or method for embodying the technical idea of the present disclosure, and do not specify the configuration to the following. That is, various changes can be made to the technical idea of the present disclosure within the technical scope described in the claims.

[0020] According to the remaining life estimation method and the remaining life estimation device 10 according to the present disclosure, the remaining life of the maintenance target 30 (see FIG. 1) is estimated from the data detected during the actual operation of the maintenance target 30. In the present disclosure, the remaining life of the maintenance target 30 is defined as the operating time or the number of operations until the length of a crack generated in at least a part of the maintenance target 30 progresses to a predetermined ratio or more with respect to dimensions such as the thickness of the crack generation site and the maintenance target 30 can no longer withstand use. The predetermined ratio is determined according to the material, shape, or mechanical characteristics of the crack generation site. The estimation accuracy of the remaining life is improved according to the damage state of the maintenance target 30 including the state of the crack generated in the maintenance target 30 from the data detected during the actual operation of the maintenance target 30.

[0021] According to the guidance method and the guidance device 20 according to the present disclosure, based on the time until the maintenance target 30 reaches damage, which is clarified by the estimation of the remaining life, the maintenance plan can be specified or optimized.

[0022] (Configuration example of the guidance system 1) As shown in FIG. 1, a guidance system 1 according to an embodiment includes a remaining life estimation device 10, a guidance device 20, a detection device 40, and a database 50. The guidance system 1 detects the operation data of the maintenance target 30 by the detection device 40. The operation data of the maintenance target 30 is data related to the actual operation of the maintenance target 30.

[0023] The maintenance target 30 includes a device or facility that receives a repeated load. The maintenance target 30 may be a facility within a factory. The repeated load may be a load applied to a device or facility by repeatedly rolling or transporting a steel plate. The repeated load may include repeatedly receiving an impact load. The repeated load may include a periodically changing load caused by vibration. The repeated load may include repeatedly changing the thermal stress according to a temperature change.

[0024] Repeated loads caused by repeated rolling can occur in rolling rolls or bearings of a mill, etc. Repeated loads caused by repeated impact loads can occur in devices such as sizing presses that intermittently execute pressing. Repeated loads caused by repeated impact loads can also occur in rolling equipment. Repeated loads caused by repeated impact loads can also occur in equipment that is impacted by the dropping of an object. Repeated loads caused by vibration can occur in equipment that is vibrated, such as a motor or a compressor. Repeated loads caused by temperature changes can occur in equipment that repeatedly heats and cools, such as a heating furnace or a heat exchanger.

[0025] In the present disclosure, it is assumed that the object 30 to be maintained is a sizing press for steel plates. The object 30 to be maintained is not limited to the example of the present disclosure and may be various other devices or equipment.

[0026] A sizing press is a device that presses a steel plate in the width direction by narrowing the interval between blocks located on both sides in the width direction of the steel plate. The sizing press moves the blocks along the width direction of the steel plate to control the interval between the blocks in order to narrow the interval between the blocks. The sizing press moves the blocks along the width direction of the steel plate by rotating a crankshaft connected to the blocks via a connecting rod.

[0027] By repeatedly performing the operation of pressing the steel plate in the width direction by the sizing press, a repeated load is applied to the crankshaft. When a repeated load is applied to the crankshaft, cracks can occur in the crank bearing used in the crankshaft.

[0028] In the present disclosure, the remaining life estimation device 10 estimates, as the remaining life, the operating time or the number of operations of the sizing press until the sizing press fails due to the occurrence and progression of cracks in the crank bearing used in the crankshaft of the sizing press. That is, in the present disclosure, the remaining life estimation device 10 focuses on the cracks occurring in the crank bearing used in the sizing press in order to estimate the remaining life of the sizing press which is the object 30 to be maintained. The location where the cracks are focused on for estimating the remaining life is not limited to the crank bearing, and may be various other locations or components.

[0029] The guidance system 1 estimates the remaining life of the object 30 to be maintained in consideration of the repetitive load applied to the object 30 to be maintained based on the operation data of the object 30 to be maintained by the remaining life estimation device 10. The remaining life of the object 30 to be maintained may be specified by at least one of the time or the number of times the object 30 to be maintained operates.

[0030] Estimating the remaining life of the object 30 to be maintained while considering the repetitive load is difficult for the following reasons.

[0031] First, the remaining life of the object 30 to be maintained may be correlated with the length or direction of the crack occurring in the object 30 to be maintained, the position of the crack, or the number of cracks. Therefore, the remaining life of the object 30 to be maintained may be estimated using the crack information of the object 30 to be maintained. The crack information of the object 30 to be maintained is information regarding the length or direction of the crack occurring in the object 30 to be maintained, the position of the crack, or the number of cracks. When the remaining life of the object 30 to be maintained is estimated using the crack information, the remaining life of the object 30 to be maintained is estimated via the crack information from the operation data of the object 30 to be maintained, and thus is not directly estimated from the operation data of the object 30 to be maintained.

[0032] Here, the repetitive load applied to the object 30 to be maintained is not necessarily a repetition of a load of a certain intensity. Also, the repetitive load applied to the object 30 to be maintained is not necessarily repeated at a constant cycle. The repetitive load applied to the object 30 to be maintained is determined according to the operating conditions of the object 30 to be maintained. The operating conditions of the object 30 to be maintained may include, for example, when the object 30 to be maintained is a sizing press, the reduction width set to achieve the target width of the slab, the cycle of executing the press, or the load applied when executing the press, etc. Therefore, when estimating the remaining life of the object 30 to be maintained in consideration of the repetitive load applied to the object 30 to be maintained, in order to improve the estimation accuracy of the remaining life, it is necessary to assume the repetitive load applied to the object 30 to be maintained according to the operating conditions of the object 30 to be maintained.

[0033] The physical phenomenon in the creep phenomenon of continuously receiving a constant load is significantly different from the physical phenomenon in the vibration fatigue of receiving a repetitive load. Therefore, the estimation of the remaining life considering the repetitive load is more difficult than the estimation of the remaining life in the state where the creep load is applied. In order to estimate the remaining life of the object 30 to be maintained that receives a repetitive load, it is not possible to apply the remaining life estimation model in the state where the creep load is applied. It is necessary to perform the estimation of the remaining life considering the repetitive load based on a physical theoretical system specialized for vibration fatigue.

[0034] The guidance system 1 generates maintenance information of the object 30 to be maintained by the guidance device 20 based on the estimation result of the remaining life of the object 30 to be maintained. The maintenance information of the object 30 to be maintained is information regarding the maintenance work to be performed on the object 30 to be maintained. The maintenance information of the object 30 to be maintained may include information specifying the timing of performing maintenance work such as inspection, part replacement, or repair of the object 30 to be maintained. The maintenance information of the object 30 to be maintained may include information specifying the work content to be performed as maintenance work.

[0035] Hereinafter, each component of the guidance system 1 will be described.

[0036] <Remaining life estimation device 10> The remaining life estimation device 10 includes an estimation unit 11, a storage unit 12, and a communication unit 13.

[0037] The estimation unit 11 may be configured to include, for example, a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit) in order to control or manage various functions of the remaining life estimation device 10. The estimation unit 11 may realize the functions of the remaining life estimation device 10 by reading and executing a program stored in the storage unit 12.

[0038] The storage unit 12 stores various kinds of information or data used in the remaining life estimation device 10. The storage unit 12 may store, for example, a program executed in the estimation unit 11, or data or processing results used in the processing executed in the estimation unit 11. The storage unit 12 may function as a work memory of the estimation unit 11. The storage unit 12 may be configured to include, for example, a semiconductor memory or the like, but is not limited thereto. The storage unit 12 may be configured as, for example, an internal memory of the estimation unit 11, or may be configured as an electromagnetic recording medium such as a hard disk drive (HDD) accessible from the estimation unit 11. The storage unit 12 may be configured as a non-temporary readable medium. The storage unit 12 may be configured integrally with the estimation unit 11, or may be configured separately from the estimation unit 11.

[0039] The communication unit 13 may be configured to include a communication interface for communicating with other devices such as the guidance device 20, the detection device 40, or the database 50 by wire or wirelessly. The communication interface may be configured to communicate with other devices via a network. The communication unit 13 may be configured to include an input / output port for inputting / outputting data to / from other devices. The communication unit 13 may communicate based on a wired communication standard or may communicate based on a wireless communication standard. The wireless communication standard may include a communication standard for cellular phones such as 4G (4th Generation) or 5G (5th Generation). Also, the wireless communication standard may include IEEE802.11 and Bluetooth (registered trademark), etc. The communication unit 13 may support one or more of these communication standards. The communication unit 13 is not limited to these examples and may communicate with other devices or input / output data based on various standards.

[0040] The remaining life estimation device 10 may be provided with an input device or an output device. The input device may be configured to be the same as or similar to the input device of the guidance device 20 described later. The output device may be configured to be the same as or similar to the output device of the guidance device 20 described later.

[0041] The remaining life estimation device 10 may be realized in an on-premises environment or may be realized using a cloud service. The remaining life estimation device 10 may be realized in a form that mixes an on-premises environment and a cloud service.

[0042] <Guidance device 20> The guidance device 20 includes a control unit 21, a storage unit 22, a communication unit 23, an input unit 24, and an output unit 25.

[0043] The control unit 21 may be configured to include, for example, a CPU or a GPU to control or manage various functions of the guidance device 20. The control unit 21 may realize the functions of the guidance device 20 by reading and executing a program stored in the storage unit 22.

[0044] The storage unit 22 stores various types of information or data used in the guidance device 20. The storage unit 22 may store, for example, a program executed in the control unit 21, or data or processing results used in the processing executed in the control unit 21. The storage unit 22 may function as a work memory of the control unit 21. The storage unit 22 may be configured identically or similarly to the storage unit 12 of the remaining life estimation device 10. The storage unit 22 may be configured integrally with the control unit 21 or separately from the control unit 21.

[0045] The communication unit 23 may include a communication interface for communicating with other devices such as the remaining life estimation device 10 or the database 50 by wire or wirelessly. The communication unit 23 may be configured identically or similarly to the communication unit 13 of the remaining life estimation device 10.

[0046] The input unit 24 may include an input device that receives an input from a user of the guidance system 1. The input device may include, for example, a keyboard or physical keys, or may include a touch panel or touch sensor or a pointing device such as a mouse. The input device is not limited to these examples and may include various other devices.

[0047] The output unit 25 may be configured to output the information acquired from the control unit 21. The output unit 25 may notify the user of the information by outputting visual information such as characters, graphics, or images, either directly or via an external device or the like. The output unit 25 may include a display device, or may be connected to the display device by wire or wirelessly. The display device may include various displays such as, for example, a liquid crystal display. The output unit 25 may also notify the user of the information by outputting auditory information such as sound, either directly or via an external device or the like. The output unit 25 may include an audio output device such as a speaker, or may be connected to the audio output device by wire or wirelessly. The output unit 25 may include a vibration device. The output unit 25 may notify the user of the information not only by visual information, auditory information, or tactile information, but also by outputting information that can be perceived by the user through other senses, either directly or via an external device or the like.

[0048] The guidance device 20 may be implemented in an on-premises environment, or may be implemented using a cloud service. The guidance device 20 may be implemented in a form that mixes an on-premises environment and a cloud service.

[0049] At least some functions of the guidance device 20 may be implemented by the remaining useful life estimation device 10. That is, the remaining useful life estimation device 10 may include at least a part of the guidance device 20.

[0050] <Detection device 40> The detection device 40 may include a sensor attached to the object to be maintained 30. The detection device 40 may include a camera that photographs at least a part of the object to be maintained 30. The detection device 40 may be configured to communicate with a sensor attached to the object to be maintained 30 or a camera that photographs the object to be maintained 30, and acquire a detection result from the sensor or the camera.

[0051] The sensor attached to the object to be maintained 30 may include, for example, a vibration sensor, an acceleration sensor, a load sensor, or an acoustic sensor so as to be able to detect data when the object to be maintained 30 is operating.

[0052] The vibration sensor detects at least a part of the mechanical vibration of the object 30 to be maintained. The vibration sensor may output the frequency spectrum of the vibration. The frequency spectrum of the vibration includes the intensity of the vibration for each frequency component. That is, the vibration sensor may detect the frequency and intensity of the vibration.

[0053] The acceleration sensor detects at least a part of the acceleration of the object 30 to be maintained. The acceleration sensor may detect the acceleration of the vibration. The acceleration sensor may detect the acceleration generated when at least a part of the object 30 to be maintained receives an external force. The sensor attached to the object 30 to be maintained may include a gyro sensor or an inertial sensor in addition to the acceleration sensor. The acceleration sensor may be replaced with an inertial sensor.

[0054] The load sensor detects a load such as tension, compression, shear, bending, or torsion applied to at least a part of the object 30 to be maintained. The load sensor may include a load cell. The load sensor may be configured to detect a static load. The load sensor may be configured to detect a dynamic load including an impact load or a repeated load.

[0055] The acoustic sensor detects sound generated from at least a part of the object 30 to be maintained. The acoustic sensor may include an AE (Acoustic Emission) sensor. The AE sensor can detect acoustic waves in the ultrasonic region. The acoustic sensor may detect the sound pressure level. The acoustic sensor may output the frequency spectrum of the sound. The frequency spectrum of the sound includes the intensity of the sound for each frequency component.

[0056] <Database 50> The database 50 may store data used by the remaining life estimation device 10 to estimate the remaining life of the object 30 to be maintained.

[0057] The database 50 may store data regarding the object 30 to be maintained. The data regarding the object 30 to be maintained is also referred to as the data of the object 30 to be maintained. The data of the object 30 to be maintained may include the operation data of the object 30 to be maintained. The data of the object 30 to be maintained may include the operating conditions of the object 30 to be maintained. The data of the object 30 to be maintained may include maintenance plan data that specifies the maintenance time or the content of the maintenance work of the object 30 to be maintained.

[0058] The data of the object 30 to be maintained may include the material data of the object 30 to be maintained. The material data of the object 30 to be maintained is data that specifies the material of at least one part of the object 30 to be maintained. The material data of the object 30 to be maintained may include data that specifies the name of the material of the part, the shape of the part, or mechanical properties, etc. The mechanical properties may include tensile strength, yield stress, ultimate strength, Young's modulus, Poisson's ratio, or density, etc.

[0059] The database 50 may store a model 60 (see FIGS. 5A or 5B) that outputs an estimated remaining life of the object 30 to be maintained.

[0060] The database 50 may be configured as a part of the storage unit 12 of the remaining life estimation device 10. The database 50 may be configured as a storage device separate from the remaining life estimation device 10. The database 50 may be realized in an on-premises environment or may be realized using a cloud service. The database 50 may be realized in a form that mixes an on-premises environment and a cloud service.

[0061] (Operation example of the remaining life estimation device 10) In the guidance system 1 according to the present embodiment, the remaining life estimation device 10 acquires the operation data of the object 30 to be maintained. The remaining life estimation device 10 may detect whether an abnormality has occurred in the object 30 to be maintained based on the operation data. When the remaining life estimation device 10 detects that an abnormality has occurred in the object 30 to be maintained, the remaining life estimation device 10 may estimate the remaining life of the object 30 to be maintained.

[0062] The remaining life estimation device 10 does not necessarily need to detect whether an abnormality has occurred in the object 30 to be maintained. The remaining life estimation device 10 may estimate the remaining life of the object 30 to be maintained regardless of whether an abnormality has occurred in the object 30 to be maintained. The remaining life estimation device 10 may estimate the remaining life of the object 30 to be maintained at a predetermined timing. The predetermined timing may be set periodically or may be set irregularly. The predetermined timing may also be when a predetermined condition is satisfied. The predetermined condition may be, for example, that an impact equal to or greater than the impact determination threshold is applied to at least a part of the object 30 to be maintained. The impact determination threshold may be appropriately set based on material data of the object 30 to be maintained and the like. The predetermined condition may be, for example, that the temperature change of at least a part of the object 30 to be maintained becomes equal to or greater than the temperature difference determination threshold. The temperature difference determination threshold may be appropriately set based on material data of the object 30 to be maintained and the like.

[0063] Hereinafter, an operation example in which the remaining life estimation device 10 executes a remaining life estimation method including the procedure of the flowchart illustrated in FIG. 2 to estimate the remaining life of the object 30 to be maintained will be described. The remaining life estimation method may be realized as a remaining life estimation program to be executed by the estimation unit 11 of the remaining life estimation device 10. The remaining life estimation program may be stored in a non-transitory computer-readable medium.

[0064] The estimation unit 11 acquires operation data of the object 30 to be maintained from the detection device 40 (step S1). The estimation unit 11 may acquire, as the operation data of the object 30 to be maintained, data such as vibration, acceleration, load, or sound of at least a part of the object 30 to be maintained. That is, the operation data of the object 30 to be maintained may include data related to at least one of vibration, acceleration, load, or sound of at least a part of the object 30 to be maintained.

[0065] When the operation data is acceleration data, as exemplified schematically in FIG. 3, the acceleration data in each of the idling and pressing cases is acquired. During idling, it corresponds to the case where the block moves in an empty state, i.e., with no steel plate positioned between the blocks. During pressing, it corresponds to the case where the block moves with a steel plate positioned between the blocks and presses the steel plate in the width direction. In each of the idling and pressing cases, different acceleration data is acquired depending on whether there is damage to the crank bearing or not.

[0066] Returning to FIG. 2, the estimation unit 11 determines whether it is detected that an abnormality has occurred in the maintenance object 30 based on the operation data of the maintenance object 30 (step S2). Step S2 is also referred to as an abnormality detection step. The abnormality detection step is a procedure for detecting that an abnormality has occurred in the maintenance object 30 based on the operation data of the maintenance object 30.

[0067] The operation data can change according to the state of the maintenance object 30. The estimation unit 11 may determine whether it is detected that an abnormality has occurred in the maintenance object 30 by comparing the operation data with the data in the normal state.

[0068] For example, FIG. 4 shows a graph schematically representing the change over time of the acceleration data. The vertical axis represents acceleration. The horizontal axis represents the date. In the case where the acceleration of the maintenance object 30 gradually increases according to the deterioration of the maintenance object 30, the estimation unit 11 may determine that it is detected that an abnormality has occurred in the maintenance object 30 when the acceleration exceeds a predetermined threshold value. The period until the occurrence of the abnormality is detected is represented as the normal period. The period after the occurrence of the abnormality is detected is represented as the abnormal occurrence period. The abnormal occurrence period includes the period during which a failure has occurred in the maintenance object 30.

[0069] The criteria for detecting the occurrence of an abnormality may be set based on the operation data when an abnormality occurred in the object 30 to be maintained in the past, the magnitude of the damage when the object 30 to be maintained is damaged, the operation rate or operation frequency of the object 30 to be maintained, or the influence of noise or error included in the operation data of the object 30 to be maintained. For example, the criteria for detecting the occurrence of an abnormality may be set such that the numerical value of the operation data is outside the range of ±10% with respect to the numerical value during normal operation. The ratio of the comparison criteria with respect to the normal state is not limited to 10%, and may be various other values. The data or numerical value during normal operation may be set based on the past operation results of the object 30 to be maintained. The data or numerical value during normal operation may be calculated, for example, as the average value of the operation data of the object 30 to be maintained for the past one day. The data or numerical value during normal operation may also be set by the user of the guidance system 1.

[0070] The operation data may change when a crack occurs in the object 30 to be maintained. For example, the rigidity of the part or component where the crack occurs may change due to the occurrence of the crack. In response to the change in rigidity in the object 30 to be maintained, data such as vibration, acceleration, load, or sound obtained from the object 30 to be maintained changes.

[0071] The operation data may include acoustic data generated from the object 30 to be maintained when a crack occurs in the object 30 to be maintained. For example, when a crack occurs in the inner ring part of a bearing, the frequency or sound pressure level of the sound generated when the roller passes through the crack changes. The AE sensor can detect the change in the frequency or sound pressure level of the sound generated when the roller passes through the crack.

[0072] The estimation unit 11 may detect that a crack has occurred in at least a part of the object 30 to be maintained based on the operation data of the object 30 to be maintained. The occurrence of a crack is included in the occurrence of an abnormality. Therefore, the estimation unit 11 may detect that an abnormality has occurred in the object 30 to be maintained based on the operation data of the object 30 to be maintained.

[0073] Based on the position where the operation data serving as the basis for detecting the occurrence of a crack is acquired, the estimation unit 11 may identify the part or component where the crack has occurred, that is, the position where the crack has occurred. When it is detected that a crack has occurred in at least a part of the object 30 to be maintained, the estimation unit 11 may identify the length or direction of the crack based on the operation data of the object 30 to be maintained. The estimation unit 11 may newly generate information identifying the length or direction of the generated crack, or the position of the generated crack, as crack information of the object 30 to be maintained, or may add it to the already generated crack information.

[0074] Returning to FIG. 2, when it is not detected that an abnormality has occurred in the object 30 to be maintained (step S2: NO), the estimation unit 11 returns to the procedure for acquiring the operation data of the object 30 to be maintained in step S1. When it is detected that an abnormality has occurred in the object 30 to be maintained (step S2: YES), the estimation unit 11 may execute the procedures after step S3 in order to estimate the remaining life of the object 30 to be maintained.

[0075] The estimation unit 11 may shorten the interval at which the procedures of steps S1 and S2 are executed so that it can detect the occurrence of an abnormality in the object 30 to be maintained from the occurrence of a crack in a short time. The estimation unit 11 may also execute the procedures of steps S1 and S2 so that it can detect the occurrence of an abnormality in the object 30 to be maintained in real time. Real-time detection may be, for example, detecting within one cycle or several cycles of the repeated load applied to the object 30 to be maintained.

[0076] When it is detected that an abnormality has occurred in the object 30 to be maintained, the estimation unit 11 may estimate the remaining life of the object 30 to be maintained within a predetermined time after the detection. The predetermined time may be appropriately set according to the specifications of the object 30 to be maintained. The predetermined time may be set, for example, based on the average time from when a crack occurs in the object 30 to be maintained until the part where the crack has occurred breaks or becomes unable to withstand use.

[0077] The estimation unit 11 may execute the procedures after step S3 in order to estimate the remaining life of the object 30 to be maintained, without executing the procedures of steps S1 and S2, that is, regardless of whether an abnormality has occurred in the object 30 to be maintained. When the estimation unit 11 estimates the remaining life of the object 30 to be maintained without determining whether an abnormality has occurred in the object 30 to be maintained, the interval for executing the estimation of the remaining life may be appropriately set according to the specifications of the object 30 to be maintained. The estimation unit 11 may execute the estimation of the remaining life of the object 30 to be maintained at an interval shorter than, for example, the average time between failures of the object 30 to be maintained.

[0078] The estimation unit 11 executes the procedures after step S3 in order to estimate the remaining life of the object 30 to be maintained. In the present disclosure, the estimation unit 11 uses the model 60 illustrated in FIG. 5A or FIG. 5B in order to estimate the remaining life of the object 30 to be maintained. The estimation unit 11 may estimate the remaining life of the object 30 to be maintained in other manners.

[0079] The estimation unit 11 acquires the material data of the object 30 to be maintained (step S3). The estimation unit 11 may acquire the material data of the part or component where a crack is presumed to have occurred. The estimation unit 11 may acquire the material data using a material database in which the part or component where a sensor is attached to the object 30 to be maintained is associated with the material data of the part or component. The material data is data necessary for the crack propagation analysis described later. The crack propagation behavior in parts or components of different materials is different from each other.

[0080] The estimation unit 11 inputs the operation data and material data of the object 30 to be maintained into the model 60 (step S4). The estimation unit 11 acquires the estimation result of the remaining life of the object 30 to be maintained output from the model 60 (step S5). Step S5 is also referred to as the remaining life estimation step. The remaining life estimation step is a procedure for estimating the remaining life of the object 30 to be maintained using at least one model 60 from the operation data of the object 30 to be maintained that receives repeated loads. When the abnormality detection step is executed, the remaining life estimation step may be executed when it is detected in the abnormality detection step that an abnormality has occurred in the object 30 to be maintained.

[0081] As illustrated in FIG. 5A, the model 60 may be configured to output an estimated remaining life result of the object 30 to be maintained when operation data of the object 30 to be maintained is input. That is, the model 60 may receive the operation data of the object 30 to be maintained as an input and output an estimated remaining life result of the object 30 to be maintained.

[0082] As illustrated in FIG. 5B, the model 60 may include a first model 61 and a second model 62. The first model 61 may be configured to output an estimated result of crack information, which is information regarding cracks occurring in at least a part of the object 30 to be maintained, when operation data of the object 30 to be maintained is input. That is, the first model 61 receives the operation data of the object 30 to be maintained as an input and outputs crack information, which is information regarding cracks occurring in at least a part of the object 30 to be maintained. The crack information of the object 30 to be maintained is information specifying at least one of the length or direction of the crack occurring in at least a part of the object 30 to be maintained, the position of the crack, or the number of cracks. The second model 62 may be configured to output an estimated result of the remaining life of the object 30 to be maintained when the crack information of the object 30 to be maintained is input. That is, the second model 62 may receive the crack information of the object 30 to be maintained as an input and output an estimated result of the remaining life of the object 30 to be maintained.

[0083] The model 60 may be generated using, for example, the results of simulations based on fracture mechanics so as to be able to estimate the remaining life of the object 30 to be maintained with high accuracy. The simulation may include analyzing the progress of cracks occurring in the object 30 to be maintained based on the operating conditions and crack information of the object 30 to be maintained. In this case, the model 60 may be configured to associate the operation data and the crack information of the object 30 to be maintained and output an estimated result of the remaining life of the object 30 to be maintained based on the prediction of the progress of the crack analyzed from the crack information corresponding to the operation data.

[0084] The model 60 may be configured to associate the operation data of the object 30 to be maintained with the remaining life of the object 30 to be maintained and output an estimation result of the remaining life of the object 30 to be maintained based on the operation data. That is, the model 60 may be configured to output an estimation result of the remaining life without considering the crack information. The model 60 may be a learned model generated by learning using, as learning data or teacher data, data associating the operation data of the object 30 to be maintained with the remaining life of the object 30 to be maintained.

[0085] When the model 60 includes the first model 61 and the second model 62, the first model 61 may be a learned model generated by learning using, as learning data or teacher data, data associating the operation data of the object 30 to be maintained with the crack information of the object 30 to be maintained. The second model 62 may be a learned model generated by learning using, as learning data or teacher data, data associating the crack information of the object 30 to be maintained with the remaining life of the object 30 to be maintained.

[0086] The estimation unit 11 may generate the model 60 by itself or acquire the model 60 from an external device. A method for the estimation unit 11 to generate the model 60 will be described later.

[0087] The estimation unit 11 may estimate the remaining life of the object 30 to be maintained from the operation data of the object 30 to be maintained based on the correspondence between the operation data of the object 30 to be maintained and the remaining life of the object 30 to be maintained without using the model 60 illustrated in FIG. 5A. The correspondence between the operation data of the object 30 to be maintained and the remaining life of the object 30 to be maintained may be obtained by experiments or numerical analysis.

[0088] The correspondence between the operation data of the object 30 to be maintained and the remaining life of the object 30 to be maintained may be obtained by associating the past operation data of the object 30 to be maintained with the time from the detection of an abnormality in the object 30 to be maintained until the object 30 to be maintained fails (see FIG. 4).

[0089] The correspondence relationship between the operation data of the object 30 to be maintained and the remaining life of the object 30 to be maintained may be specified as a table. The table specifying the correspondence relationship between the operation data of the object 30 to be maintained and the remaining life of the object 30 to be maintained may be generated in a database format. A table representing the correspondence relationship between the operation data of the object 30 to be maintained and the remaining life of the object 30 to be maintained may be used as the model 60.

[0090] The model 60 may be generated by executing learning using, as learning data or teacher data, data associating the past operation data of the object 30 to be maintained with the time from the detection of an abnormality in the object 30 to the failure of the object 30 to be maintained.

[0091] The estimation unit 11 may estimate the crack information of the object 30 to be maintained from the operation data of the object 30 to be maintained based on the correspondence relationship between the operation data of the object 30 to be maintained and the crack information of the object 30 to be maintained, without using the first model 61 illustrated in FIG. 5B. The correspondence relationship between the operation data of the object 30 to be maintained and the crack information of the object 30 to be maintained may be obtained by experiments or numerical analysis.

[0092] The correspondence relationship between the operation data of the object 30 to be maintained and the crack information of the object 30 to be maintained may be obtained, for example, by numerical analysis using a press dynamic model. In the press dynamic model, the peaks in the frequency spectrum of the acoustic data obtained from the object 30 to be maintained may be analyzed for the case without damage and the case with damage, respectively. The crack information of the object 30 to be maintained may be generated from the peaks in the frequency spectrum of the acoustic data.

[0093] The correspondence relationship between the operation data of the object 30 to be maintained and the crack information of the object 30 to be maintained may be specified as a table. The table specifying the correspondence relationship between the operation data of the object 30 to be maintained and the crack information of the object 30 to be maintained may be generated in a database format. A table representing the correspondence relationship between the operation data of the object 30 to be maintained and the crack information of the object 30 to be maintained may be used as the first model 61.

[0094] The estimation unit 11 may estimate the remaining life of the object 30 to be maintained from the crack information of the object 30 to be maintained based on the correspondence between the crack information of the object 30 to be maintained and the remaining life of the object 30 to be maintained without using the second model 62 illustrated in FIG. 5B. The correspondence between the crack information of the object 30 to be maintained and the remaining life of the object 30 to be maintained may be obtained by experiment or numerical analysis.

[0095] The correspondence between the crack information of the object 30 to be maintained and the remaining life of the object 30 to be maintained may be obtained by numerical analysis using, for example, a crack growth analysis model. The crack growth analysis model may be a static model or a dynamic model. In the crack growth analysis model, how the crack specified by the crack information progresses when the object 30 to be maintained operates may be analyzed. The operating time or the number of operations until the crack progresses to the extent that the object 30 to be maintained can no longer withstand use may be calculated as the remaining life of the object 30 to be maintained.

[0096] The correspondence between the crack information of the object 30 to be maintained and the remaining life of the object 30 to be maintained may be specified as a table. The table specifying the correspondence between the crack information of the object 30 to be maintained and the remaining life of the object 30 to be maintained may be generated in a database format. The table representing the correspondence between the crack information of the object 30 to be maintained and the remaining life of the object 30 to be maintained may be used as the second model 62.

[0097] The estimation unit 11 may estimate the remaining life of the object 30 to be maintained from the operation data of the object 30 to be maintained by combining the correspondence between the operation data of the object 30 to be maintained and the crack information of the object 30 to be maintained and the correspondence between the crack information of the object 30 to be maintained and the remaining life of the object 30 to be maintained.

[0098] The estimation unit 11 may use a table instead of the model 60, the first model 61, or the second model 62 so as to shorten the time required to estimate the remaining life of the object 30 to be maintained from the operation data of the object 30 to be maintained. The estimation unit 11 may use a simplified version of the model 60, the first model 61, or the second model 62. By shortening the time from the acquisition of the operation data of the object 30 to be maintained to the estimation of the remaining life of the object 30 to be maintained, the remaining life can be estimated in real time.

[0099] The model 60 illustrated in FIG. 5A may be configured such that the material data of the object 30 to be maintained is further input. The model 60 may be a learned model generated by learning using, as learning data or teacher data, data associating the operation data of the object 30 to be maintained, the material data of the object 30 to be maintained, and the remaining life of the object 30 to be maintained.

[0100] The first model 61 or the second model 62 illustrated in FIG. 5B may be configured such that the material data of the object 30 to be maintained is further input. The first model 61 may be a learned model generated by learning using, as learning data or teacher data, data associating the operation data of the object 30 to be maintained, the material data of the object 30 to be maintained, and the crack information of the object 30 to be maintained. The second model 62 may be a learned model generated by learning using, as learning data or teacher data, data associating the material data of the object 30 to be maintained, the crack information of the object 30 to be maintained, and the remaining life of the object 30 to be maintained.

[0101] The material data of the object 30 to be maintained may be incorporated into the model 60, or the first model 61 or the second model 62. When the material data of the object 30 to be maintained is incorporated into the model 60, or the first model 61 or the second model 62, the estimation unit 11 may, in the procedure of step S3, acquire the model 60, or the first model 61 or the second model 62 in which appropriate material data is incorporated instead of acquiring the material data.

[0102] The estimation unit 11 may estimate the remaining life of the object 30 to be maintained based on the correspondence relationship among the operation data of the object 30 to be maintained, the material data of the object 30 to be maintained, and the remaining life of the object 30 to be maintained, without using the model 60. The correspondence relationship among the operation data of the object 30 to be maintained, the material data of the object 30 to be maintained, and the remaining life of the object 30 to be maintained may be obtained in advance by experiments or numerical analysis. The correspondence relationship among the operation data of the object 30 to be maintained, the material data of the object 30 to be maintained, and the remaining life of the object 30 to be maintained may be specified as a table. A table representing the correspondence relationship among the operation data of the object 30 to be maintained, the material data of the object 30 to be maintained, and the remaining life of the object 30 to be maintained may be used as the model 60.

[0103] Regardless of whether the estimation unit 11 uses the model 60 or not, when it detects that an abnormality has occurred in the object 30 to be maintained from the operation data of the object 30 to be maintained, it may specify the position where a crack is presumed to have occurred based on the position where the operation data serving as the basis for the detection was acquired. The estimation unit 11 may acquire the material data of the component corresponding to the position where a crack is presumed to have occurred in the object 30 to be maintained as the material data of the object 30 to be maintained. The estimation unit 11 may perform fatigue propagation analysis in at least one pattern and estimate the expansion of the crack presumed to have occurred in the object 30 to be maintained. The estimation unit 11 may estimate the operation time or the number of operations of the object 30 to be maintained until the length of the expanded crack exceeds the critical length as the remaining life of the object 30 to be maintained. The critical length is determined according to the material or shape of the component.

[0104] The estimation unit 11 may determine whether to estimate the operating time of the object 30 to be maintained as the remaining life or to estimate the number of operations as the remaining life according to the operating rate of the object 30 to be maintained. For example, when the object 30 to be maintained is a device or facility that operates stably at a high operating rate at regular intervals, the correlation between the operating time and the number of operations of the object 30 to be maintained becomes high. In this case, the estimation unit 11 may estimate the operating time of the object 30 to be maintained as the remaining life. On the other hand, when the object 30 to be maintained is a device or facility that operates irregularly at a low operating rate, the correlation between the operating time and the number of operations of the object 30 to be maintained becomes low. In this case, the estimation unit 11 may estimate the number of operations of the object 30 to be maintained as the remaining life.

[0105] When the estimation unit 11 detects that an abnormality has occurred in the object 30 to be maintained, it may estimate the remaining life of the object 30 to be maintained within a predetermined time after the detection. The predetermined time may be appropriately set according to the specifications of the object 30 to be maintained. The predetermined time may be set, for example, based on the average time from when a crack occurs in the object 30 to be maintained until the crack generation site breaks or becomes unable to withstand use.

[0106] Returning to FIG. 2, the estimation unit 11 generates and outputs maintenance information based on the estimation result of the remaining life of the object 30 to be maintained (step S6). Step S6 includes a maintenance information generation step of generating maintenance information and an output step of outputting the maintenance information. The maintenance information generation step is a procedure for generating maintenance information, which is information regarding the maintenance of the object 30 to be maintained, based on the estimation result of the remaining life of the object 30 to be maintained. The output step is a procedure for outputting the maintenance information. After executing the procedure of step S6, the estimation unit 11 ends the execution of the flowchart in FIG. 2.

[0107] The estimation unit 11 may output the estimation result of the remaining life to the guidance device 20 without generating maintenance information. As will be described later, the maintenance information may be generated and output by the guidance device 20. The estimation unit 11 may generate the maintenance information and output the maintenance information to the guidance device 20 without outputting it to the user in the remaining life estimation device 10.

[0108] As described above, according to the remaining life estimation device 10 according to the present disclosure, the remaining life of the maintenance target 30 is estimated in consideration of the repeated load.

[0109] As operation data of the maintenance target 30, data such as vibration, acceleration, load, or sound of at least a part of the maintenance target 30 may be used. When data such as vibration, acceleration, load, or sound of at least a part of the maintenance target 30 is used as operation data of the maintenance target 30, even if it is difficult to detect the occurrence of an abnormality such as a crack in the maintenance target 30 by visual inspection of the maintenance target 30 or by photographing the maintenance target 30 with a camera, the occurrence of an abnormality such as a crack is detected, and the remaining life of the maintenance target 30 is estimated in response to the detection of the occurrence of the abnormality such as a crack.

[0110] Further, when the estimation unit 11 uses data such as vibration, acceleration, load, or sound of at least a part of the maintenance target 30 as operation data of the maintenance target 30, the operation data can be acquired during the operation of the maintenance target 30. As a result, the estimation unit 11 can detect the occurrence of an abnormality such as a crack in the maintenance target 30 without stopping the operation of the maintenance target 30, and can predict the remaining life of the maintenance target 30.

[0111] The physical quantity acquired as operation data is preferably a physical quantity that propagates on or inside the surface of the maintenance target 30. The physical quantity that propagates on or inside the surface of the maintenance target 30 can be detected at a position away from the position where an abnormality such as a crack has occurred. That is, the estimation unit 11 can capture a change in the physical quantity associated with the occurrence of an abnormality such as a crack at a position away from the position where the abnormality such as a crack has occurred by acquiring the physical quantity that propagates on or inside the surface of the maintenance target 30 as operation data.

[0112] For example, even when it is not possible to directly attach a sensor to a position where an abnormality such as a crack may occur due to reasons such as insufficient installation space, the estimation unit 11 captures vibrations propagating through the maintenance target 30 with a sensor attached to a position away from the position where an abnormality such as a crack may occur, thereby discovering an abnormality such as a crack occurring at a position other than the position where the sensor is directly attached and predicting the remaining life.

[0113] Also, even when it is difficult to predict at which position an abnormality such as a crack will occur, by attaching sensors at appropriately set positions at regular or irregular intervals and acquiring operation data from a wide range of the maintenance target 30, it may be possible to discover the occurrence of an abnormality such as a crack. As described above, when acquiring a physical quantity propagating through the surface or inside of the maintenance target 30 as operation data, in order to discover the occurrence of an abnormality such as a crack, it is not necessary to directly arrange a sensor at the position where the abnormality such as a crack occurs.

[0114] As operation data of the maintenance target 30 for generating the model 60 or the first model 61, data of one type of physical quantity among vibration, acceleration, load, or sound may be used. As operation data of the maintenance target 30 for generating the model 60 or the first model 61, data combining two or more types of physical quantities among vibration, acceleration, load, or sound may also be used. When data combining two or more types of physical quantities is used as operation data of the maintenance target 30, the model 60 or the first model 61 may be generated by executing learning using learning data in which the remaining life or crack information is associated with the data combining two or more types of physical quantities. For example, when data combining two physical quantities of load and vibration is used, a state without abnormality may be associated with data in which both the load and vibration increase to the same extent, and a state with abnormality may be associated with data in which only the vibration increases while the load remains constant.

[0115] The data used to generate the model 60 or the first model 61 may include auxiliary conditions in addition to the operation data of the object 30 to be maintained. The auxiliary conditions may be conditions obtained indirectly without directly obtaining from or directly measuring the object 30 to be maintained. The auxiliary conditions may include the ambient temperature of the object 30 to be maintained. The auxiliary conditions may also include operating conditions such as line speed or slab shape. For example, when data combining the temperature as the operation data of the object 30 to be maintained and the line speed as the operating conditions as the auxiliary conditions is used, a state without abnormality may be associated with the data in which both the temperature and the line speed are increasing, and a state with abnormality may be associated with the data in which only the temperature is increasing while the line speed is constant.

[0116] (Operation example of the guidance device 20) In the guidance system 1 according to the present embodiment, the guidance device 20 may acquire the estimation result of the remaining life of the object 30 to be maintained from the remaining life estimation device 10. The guidance device 20 may generate and output maintenance information of the object 30 to be maintained based on the estimation result of the remaining life of the object 30 to be maintained.

[0117] Hereinafter, an operation example in which the guidance device 20 executes a guidance method including the procedure of the flowchart illustrated in FIG. 6 to generate and output maintenance information of the object 30 to be maintained will be described. The guidance method may be realized as a guidance program to be executed by the control unit 21 of the guidance device 20. The guidance program may be stored in a non-temporary computer-readable medium.

[0118] The control unit 21 acquires the estimation result of the remaining life of the object 30 to be maintained from the remaining life estimation device 10 (step S11).

[0119] The control unit 21 determines whether the maintenance target 30 reaches its remaining life until the next maintenance of the maintenance target 30 (step S12). When the control unit 21 determines that the maintenance target 30 reaches its remaining life until the next maintenance of the maintenance target 30 (step S12: YES), it proceeds to the procedure of step S13. When the control unit 21 determines that the maintenance target 30 does not reach its remaining life until the next maintenance of the maintenance target 30 (step S12: NO), it skips the procedure of step S13 and proceeds to the procedure of step S14.

[0120] When the remaining life of the maintenance target 30 is represented by the operating time, the control unit 21 may determine that the maintenance target 30 reaches its remaining life until the next maintenance of the maintenance target 30 when the time from when the remaining life is estimated to the next maintenance time is shorter than the estimated result of the remaining life. The next maintenance time is a time determined based on the maintenance plan of the maintenance target 30.

[0121] When the remaining life of the maintenance target 30 is represented by the number of operations, the control unit 21 may calculate the time when the maintenance target 30 reaches its remaining life based on the average operation interval and the remaining life of the maintenance target 30. When the time when the maintenance target 30 reaches its remaining life comes earlier than the next maintenance time, the control unit 21 may determine that the maintenance target 30 reaches its remaining life until the next maintenance of the maintenance target 30.

[0122] When the control unit 21 determines that the maintenance target 30 reaches its remaining life until the next maintenance of the maintenance target 30 (step S12: YES), it outputs an alarm as maintenance information of the maintenance target 30 (step S13). That is, when the maintenance target 30 reaches its remaining life until the next maintenance time of the maintenance target 30, the control unit 21 may issue an alarm as maintenance information. The alarm may be output as a sound of a buzzer. The alarm may be output as the lighting of a lamp or a display on a display. The output mode of the alarm is not limited to these.

[0123] The control unit 21 acquires data of the object to be maintained 30 (step S14). The data of the object to be maintained 30 may include information specifying parts or components of the object to be maintained 30. The data of the object to be maintained 30 may include, for example, the degree of damage when the object to be maintained 30 fails. The degree of damage may be determined according to the importance of the object to be maintained 30 within the factory. The degree of damage may be determined according to the magnitude of the loss when the object to be maintained 30 fails. The loss when the object to be maintained 30 fails may include the opportunity loss caused by the stoppage of the object to be maintained 30. The loss when the object to be maintained 30 fails may include the cost required to repair or replace the object to be maintained 30.

[0124] The control unit 21 generates and displays a maintenance priority map as maintenance information of the object to be maintained 30 (step S15). After executing the procedure of step S15, the control unit 21 ends the execution of the flowchart in FIG. 6.

[0125] As illustrated in FIG. 7, the maintenance priority map may be a two-dimensional map having an axis of remaining life and an axis of degree of damage of the object to be maintained 30. The axis of remaining life is configured such that the shorter the remaining life of the object to be maintained 30, the more it is plotted to the right. The axis of remaining life may be represented as the reciprocal of the length of the remaining life of the object to be maintained 30. The axis of remaining life may be replaced with an axis representing the height of the damage risk of the object to be maintained 30. The axis of degree of damage is configured such that the greater the degree of damage when the object to be maintained 30 fails, the more it is plotted upward.

[0126] The maintenance priority map may be divided into a plurality of regions that divide the remaining life and the degree of damage. A priority order for maintaining the object to be maintained 30 may be set for each region. For example, the priority order of the upper right region indicating a short remaining life and a large degree of damage may be set to the first place. As the remaining life becomes longer or the degree of damage becomes smaller, the priority order set for each region may be lowered.

[0127] The control unit 21 may plot and display, on the maintenance priority map, data in which the estimated remaining life result of the object 30 to be maintained is associated with the data of the object 30 to be maintained. The user of the guidance system 1 can confirm the priority order for performing maintenance on the object 30 to be maintained according to which area of the maintenance priority map the data is plotted in.

[0128] As maintenance information, the control unit 21 may output the estimated remaining life result itself of the object 30 to be maintained. As maintenance information, the control unit 21 may output a proposal for a maintenance plan for the object 30 to be maintained. As maintenance information, the control unit 21 may output a comparison result between the estimated remaining life result of the object 30 to be maintained and the maintenance time. The comparison result between the estimated remaining life result of the object 30 to be maintained and the maintenance time may be output as a ratio of the length of time until the maintenance time to the length of the remaining life of the object 30 to be maintained. As maintenance information, the control unit 21 may output the past failure history of the object 30 to be maintained, or the scale of damage when the object 30 to be maintained fails. As maintenance information, the control unit 21 may output the number of operating years of the object 30 to be maintained.

[0129] The procedures from step S12 to S15 described above are also referred to as the maintenance information generation step. The maintenance information generation step is a procedure for generating maintenance information, which is information regarding the maintenance of the object 30 to be maintained, based on the estimated remaining life result of the object 30 to be maintained obtained by executing the remaining life estimation method.

[0130] The control unit 21 may output the maintenance information generated in the maintenance information generation step. The control unit 21 may output the maintenance information in a predetermined format. The predetermined format may include a map format such as the above-described maintenance priority map. The control unit 21 may output the maintenance information to a predetermined output destination. The predetermined output destination may be the output unit 25 of the guidance device 20. The predetermined output destination may also be an output device different from the guidance device 20.

[0131] As described above, according to the guidance device 20 according to the present embodiment, guidance regarding the maintenance of the object to be maintained 30 is optimized in consideration of the remaining life of the object to be maintained 30.

[0132] (Operation example of generating the model 60) The remaining life estimation device 10 may generate the model 60 by itself. Hereinafter, an operation example in which the estimation unit 11 of the remaining life estimation device 10 executes a model generation method including the procedure of the flowchart illustrated in FIG. 8 to generate the model 60 will be described. The model generation method may be realized as a model generation program to be executed by the estimation unit 11. The model generation program may be stored in a non-transitory computer-readable medium.

[0133] The estimation unit 11 acquires the operation data of the object to be maintained 30 (step S21). The estimation unit 11 acquires the material data of the object to be maintained 30 (step S22).

[0134] The estimation unit 11 associates the operation data and material data of the object to be maintained 30 with the crack information of the object to be maintained 30 (step S23). The crack information may be obtained by experiment or numerical analysis.

[0135] The estimation unit 11 associates the crack information of the object to be maintained 30 with the remaining life of the object to be maintained 30 (step S24). The remaining life of the object to be maintained 30 may be set based on the past operation record of the object to be maintained 30.

[0136] The estimation unit 11 associates the operation data and material data of the object to be maintained 30 with the remaining life of the object to be maintained 30 (step S25). The estimation unit 11 generates a model 60 from the data associating the operation data and material data with the remaining life in the procedure of step S25 (step S26). The estimation unit 11 may generate the model 60 by performing learning with the data associating the operation data and material data with the remaining life as learning data or teacher data. The estimation unit 11 may generate, as the model 60, a table or database-type model that specifies the relationship associating the operation data and material data with the remaining life. After executing the procedure of step S26, the estimation unit 11 ends the execution of the flowchart of FIG. 8.

[0137] The estimation unit 11 may execute a model generation method including the procedure of the flowchart illustrated in FIG. 9 to generate a model 60 that outputs the remaining life of the object to be maintained 30 in response to the input of the operation data of the object to be maintained 30.

[0138] The estimation unit 11 acquires the history of the operation data of the object to be maintained 30 or the data of the change over time (step S31). The history of the operation data of the object to be maintained 30 or the data of the change over time may include the results of measuring the operation data of the object to be maintained 30 at each of a plurality of time points.

[0139] The estimation unit 11 acquires the time point when the object to be maintained 30 for which the operation data has been acquired becomes unusable (step S32). The time point when the object to be maintained 30 becomes unusable may include the time point when the object to be maintained 30 reaches destruction, the time point when the object to be maintained 30 malfunctions, or the time point when the operation of the object to be maintained 30 stops, etc.

[0140] The estimation unit 11 associates the history of the operation data or the data of the time change with the time until the maintenance target object 30 becomes unusable, that is, the actual remaining life value at each time point (step S33). When the history of the operation data or the data of the time change includes the values of the operation data of the maintenance target object 30 at each of a plurality of time points, the estimation unit 11 may associate the actual remaining life value with the value of the operation data at each time point. For example, the estimation unit 11 may associate that the value of the operation data of the maintenance target object 30 is A with the remaining life of the maintenance target object 30 being B.

[0141] The estimation unit 11 calculates the change rate of the value of the operation data at each time point for the history of the operation data or the data of the time change of the maintenance target object 30, and may associate the actual remaining life value with the change rate of the value of the operation data at each time point. For example, the estimation unit 11 may associate that the change rate of the value of the operation data of the maintenance target object 30 is X with the remaining life of the maintenance target object 30 being Y. The change rate may be calculated as the change in the value of the operation data per unit time. The unit time may be set to 1 second, 1 minute, 1 hour, 1 day, etc. The unit time may also be set as an interval of a certain operating time of the maintenance target object 30. The change rate may be calculated, for example, as the slope at each time point of a graph of the time change of the operation data with the operating time on the horizontal axis and the value of the operation data on the vertical axis. When the value of the operation data at time T1 is P1 and the value of the operation data at time T2 is P2, the change rate of the value of the operation data in the time from time T1 to time T2 may be calculated as (P2 - P1) / (T2 - T1).

[0142] The estimation unit 11 generates a model 60 from data associating the actual remaining life value of the object to be maintained 30 with the history of operation data or data on the change over time (step S34). The estimation unit 11 may generate the model 60 by performing learning using learning data or teacher data in which the actual remaining life value is associated with the history of operation data or data on the change over time. The estimation unit 11 may generate, as the model 60, a table or a database-type model that specifies the relationship associating the history of operation data or data on the change over time with the actual remaining life value. After executing the procedure of step S34, the estimation unit 11 ends the execution of the flowchart of FIG. 9. The procedure of the flowchart of FIG. 9 may be executed by a model generation device different from the remaining life estimation device 10.

[0143] When the model 60 is generated based on the history of operation data or data on the change over time of the object to be maintained 30, the estimation unit 11 may acquire the history of operation data or data on the change over time of the object to be maintained 30 for which the remaining life is to be predicted, input the data to the model 60, and acquire the remaining life of the object to be maintained 30 output from the model 60.

[0144] The estimation unit 11 may execute a model generation method including the procedure of the flowchart illustrated in FIG. 10 to generate a first model 61 that outputs crack information regarding cracks occurring in the object to be maintained 30 in response to the input of operation data of the object to be maintained 30.

[0145] The estimation unit 11 acquires the history of operation data or data on the change over time of the object to be maintained 30 (step S41). The history of operation data or data on the change over time of the object to be maintained 30 may include the results of measuring the operation data of the object to be maintained 30 at each of a plurality of time points. The operation data of the object to be maintained 30 may include the position information of the sensor that measured the operation data.

[0146] The estimation unit 11 acquires the crack information of the object to be maintained 30 at the time when the operation data of the object to be maintained 30 is measured (step S42). The crack information may include the presence or absence of crack occurrence or the crack occurrence position. The crack information may include the crack length due to the progression of the generated crack.

[0147] The presence or absence of cracks, the crack generation position, or the crack length may be measured by inspection such as visual inspection or ultrasonic flaw detection. The crack length may be calculated by numerical analysis of the generated crack.

[0148] When the estimation unit 11 acquires crack information at a time point different from the measurement time point of the operation data of the object 30 to be maintained, the estimation unit 11 calculates the presence or absence of cracks or the time change of the crack length at the measurement time point of the operation data of the object 30 to be maintained by interpolating the presence or absence of cracks or the time change of the crack length, and may acquire it as crack information at the measurement time point of the operation data of the object 30 to be maintained.

[0149] The estimation unit 11 associates the crack information of the object 30 to be maintained with the history of the operation data or the data of the time change (step S43). When the history of the operation data or the data of the time change includes the values of the operation data of the object 30 to be maintained at each of a plurality of time points, the estimation unit 11 may associate the crack information with the values of the operation data at each time point. The estimation unit 11 may associate the crack information at the time point when the value of the operation data exceeds the reference value with the value of the operation data.

[0150] The estimation unit 11 calculates the change rate of the value of the operation data at each time point for the history of the operation data or the data of the time change of the object 30 to be maintained, and may associate the crack information with the change rate of the value of the operation data at each time point. The estimation unit 11 may associate the crack information at the time point when the change rate of the value of the operation data exceeds the reference change rate, that is, the time point when the value of the operation data changes rapidly, with the change rate of the value of the operation data. The estimation unit 11 may associate the crack information at the time point when the change rate itself of the value of the operation data changes with the change rate of the value of the operation data.

[0151] The estimation unit 11 generates a first model 61 from data in which crack information of the object to be maintained 30 is associated with the history of operation data or data on time change (step S44). The first model 61 may estimate the presence or absence of crack generation or the crack length based on the value or change rate of the history of operation data or data on time change. The first model 61 may estimate the crack generation position based on the position information of the sensor that measured the operation data.

[0152] The estimation unit 11 may generate the first model 61 by performing learning using learning data or teacher data in which crack information of the object to be maintained 30 is associated with the history of operation data or data on time change. The estimation unit 11 may generate, as the first model 61, a table or database type model that specifies the relationship associating the history of operation data or data on time change with the crack information of the object to be maintained 30. After executing the procedure of step S44, the estimation unit 11 ends the execution of the flowchart of FIG. 10. The procedure of the flowchart of FIG. 10 may be executed by a model generation device different from the remaining life estimation device 10.

[0153] When the first model 61 is generated based on the history of operation data or data on time change of the object to be maintained 30, the estimation unit 11 acquires the history of operation data or data on time change of the object to be maintained 30 for which the remaining life is to be predicted and inputs it to the first model 61, inputs the crack information of the object to be maintained 30 output from the first model 61 to the second model 62, and may acquire the remaining life of the object to be maintained 30 output from the second model 62.

[0154] The first model 61 may be configured to statistically calculate crack information by applying data science techniques. In this case, the second model 62 may be configured to estimate the crack propagation by numerical simulation from the crack information calculated by the first model 61 and predict the remaining life. The second model 62 may be configured as a database type model in which the results of estimating the crack propagation by numerical simulation from the crack information are previously databaseized.

[0155] (Another Embodiment of the Guidance System 1) The guidance device 20 may generate and output maintenance information based on the estimated remaining life of the object 30 to be maintained as described above. The guidance device 20 may obtain the maintenance information from the remaining life estimation device 10 and only execute the output of the maintenance information. When the guidance device 20 generates the maintenance information, the remaining life estimation device 10 may output only the estimated result of the remaining life without generating the maintenance information.

[0156] In addition to the estimation unit 11 that estimates the remaining life, the remaining life estimation device 10 may further include a maintenance information generation unit that generates maintenance information. Regardless of whether the maintenance information is generated by the remaining life estimation device 10 or the guidance device 20, it may be stored in the database 50. The maintenance information may be stored in a dedicated maintenance information database.

[0157] Although the embodiments of the present disclosure have been described based on the drawings and examples, it should be noted that those skilled in the art can make various modifications or alterations based on the present disclosure. Therefore, it should be noted that these modifications or alterations are included in the scope of the present disclosure. For example, the functions included in each component or each step can be rearranged so as not to be logically inconsistent, and a plurality of components or steps can be combined into one or divided. The embodiments according to the present disclosure can also be realized as a program executed by a processor included in the device or a storage medium storing the program. It should be understood that these are also included in the scope of the present disclosure.

Explanation of Reference Numerals

[0158] 1 Guidance system 10 Remaining life estimation device (11: Estimation unit, 12: Storage unit, 13: Communication unit) 20 Guidance device (21: Control unit, 22: Storage unit, 23: Communication unit, 24: Input unit, 25: Output unit) 30 Object to be maintained 40 Detection device 50 Database 60 Model (61: First model, 62: Second model)

Claims

1. A remaining life estimation method comprising a remaining life estimation step of estimating a remaining life of a maintenance object using at least one model from operational data of the maintenance object that is subjected to repeated loads.

2. The method further includes an abnormality detection step of detecting an occurrence of an abnormality in the maintenance object based on the operation data of the maintenance object, When it is detected in the abnormality detection step that an abnormality has occurred in the maintenance object, the remaining life estimation step is executed. The remaining life estimation method according to claim 1 .

3. The remaining life estimation method according to claim 1 , wherein the operational data of the maintenance object includes data on at least one of vibration, acceleration, load, and sound of at least a part of the maintenance object.

4. The remaining life estimation method according to claim 1 , wherein the model receives operational data of the maintenance object as an input, and outputs an estimation result of the remaining life of the maintenance object.

5. the models include a first model and a second model; The first model receives operation data of the maintenance object as an input, and outputs crack information that is information about a crack occurring in at least a part of the maintenance object; The remaining life estimation method according to claim 1 , wherein the second model receives the crack information as an input and outputs an estimation result of the remaining life of the maintenance object.

6. The remaining life estimation method according to claim 5 , wherein the crack information identifies at least one of a length, a direction, and a position of a crack occurring in at least a part of the maintenance object.

7. A guidance method including a maintenance information generation step of generating maintenance information, which is information regarding the maintenance of the maintenance object, based on an estimation result of the remaining life of the maintenance object obtained by executing the remaining life estimation method according to any one of claims 1 to 6.

8. 8. The guidance method according to claim 7, further comprising an output step of outputting the maintenance information in a predetermined format.

9. A guidance system comprising: a remaining life estimation device that executes the remaining life estimation method according to any one of claims 1 to 6; and a guidance device that executes the guidance method according to claim 7.

10. A remaining life estimation device that executes the remaining life estimation method according to any one of claims 1 to 6.

11. A guidance device that performs the guidance method according to claim 7.

Images