Excavator management device and excavator

Abstract

A management device for an excavator, which has a learning unit that uses a data set as learning data to learn the relationship between the operation data of the excavator and the degree of abnormality of the excavator. The data set sets the operation data corresponding to the period for determining the presence or absence of an abnormality of the excavator among the operation data indicating the actions of the excavator as input, and sets the information indicating no abnormality as output.

Description

Technical Field

[0001] The present invention relates to a management device for an excavator, a management system for an excavator, a support device for an excavator, and an excavator. Background Art

[0002] Conventionally, a technique has been known in which a plurality of reference data representing operation waveforms during normal operation of a construction machine and verification data composed of time series of characteristic amounts of operation variables of the construction machine are used to determine whether the construction machine is abnormal.

[0003] Prior Art Documents

[0004] Patent Documents

[0005] Patent Document 1: Japanese Unexamined Patent Application Publication No. 2015-83731 Summary of the Invention

[0006] Problems to be Solved by the Invention

[0007] Among them, in recent years, an abnormality determination model that uses the operation data of a construction machine as learning data is sometimes generated, and the presence or absence of an abnormality of the construction machine is determined using this abnormality determination model.

[0008] At this time, the operation data during normal operation of the construction machine is usually set as learning data, but among the operation data during normal operation, inappropriate operation data is sometimes included as learning data.

[0009] Therefore, in view of the above situation, an object is to use learning data within an appropriate range.

[0010] Means for Solving the Problems

[0011] The management device for an excavator according to an embodiment of the present invention includes a learning unit that learns the relationship between the operation data of the excavator and the degree of abnormality of the excavator using a data set as learning data. The data set sets the operation data corresponding to the period for determining the presence or absence of an abnormality of the excavator among the operation data representing the operation of the excavator as input, and sets information indicating no abnormality as output.

[0012] The management system for an excavator according to an embodiment of the present invention is a management system for an excavator having an excavator and a management device for the excavator. The management device includes: an information collection unit that collects operation data representing the operation of the excavator from the excavator; and a learning unit that learns the relationship between the operation data of the excavator and the degree of abnormality of the excavator using a data set as learning data. The data set sets the operation data corresponding to the period for determining the presence or absence of an abnormality of the excavator among the collected operation data as input, and sets information indicating no abnormality as output.

[0013] The support device for an excavator according to an embodiment of the present invention receives the input of the start date of maintenance of the excavator and the completion date of the maintenance of the excavator, and notifies the start date and the completion date to the management device of the excavator having a learning unit. The learning unit uses a data set as learning data to learn the relationship between the operation data of the excavator and the degree of abnormality of the excavator. This data set sets the operation data corresponding to the period for determining the presence or absence of an abnormality of the excavator among the operation data indicating the operation of the excavator as input, and sets the information indicating no abnormality as output.

[0014] The support device for an excavator according to an embodiment of the present invention includes: a start button for inputting the start date of maintenance of the excavator; and a completion button for inputting the completion date of the maintenance of the excavator.

[0015] The excavator according to an embodiment of the present invention is an excavator that communicates with a management device, and has: an input device for inputting the start date of maintenance of the excavator and the completion date of the maintenance of the excavator; and a communication device for sending the operation data indicating the operation of the excavator and the information indicating the start date and the completion date input from the input device to the management device having a learning unit. The learning unit uses a data set as learning data to learn the relationship between the operation data of the excavator and the degree of abnormality of the excavator. This data set sets the operation data corresponding to the period for determining the presence or absence of an abnormality of the excavator as input, and sets the information indicating no abnormality as output.

[0016] The excavator according to an embodiment of the present invention is an excavator that communicates with a management device, and has: a start switch for inputting the start date of maintenance of the excavator; and a completion switch for inputting the completion date of the maintenance of the excavator.

[0017] Advantages of the Invention

[0018] It is possible to use learning data within an appropriate range. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a schematic diagram showing an example of an excavator management system.

[0020] Figure 2 It is a structural diagram showing an example of an excavator management system.

[0021] Figure 3 It is a diagram showing an example of the hardware structure of the management device.

[0022] Figure 4 It is a diagram showing an example of the main information storage unit.

[0023] Figure 5 It is a diagram showing an example of a defective condition information storage unit.

[0024] Figure 6 It is a diagram showing an example of an operation information storage unit.

[0025] Figure 7 It is a diagram showing an example of a vehicle information storage unit.

[0026] Figure 8 It is a diagram explaining the functions of the management device.

[0027] Figure 9 It is a diagram showing an example of the parameters stored in the parameter storage unit.

[0028] Figure 10 It is a flowchart explaining the processing of the management device.

[0029] Figure 11A It is the first diagram explaining the processing of the data range determination unit.

[0030] Figure 11B It is the second diagram explaining the processing of the data range determination unit.

[0031] Figure 12A It is the first diagram showing an output example of the result of abnormality determination.

[0032] Figure 12B It is the second diagram showing an output example of the result of abnormality determination.

[0033] Figure 13 It is a flowchart explaining the processing of the management device of another embodiment. Detailed Embodiment

[0034] (Embodiment)

[0035] Hereinafter, the embodiment will be described with reference to the accompanying drawings. Figure 1 It is a schematic diagram showing an example of an excavator management system.

[0036] The excavator management system SYS of the present embodiment includes an excavator 100, a management device 300, and a support device 400 for the excavator 100. In the excavator management system SYS, the excavator 100, the management device 300, and the support device 400 communicate with each other via a network. The management device 300 manages the excavator 100. The support device 400 supports the operation of the excavator 100. In the following description, the excavator management system SYS will be referred to as the management system SYS.

[0037] The excavator 100 of the present embodiment is an example of a construction machine. The excavator 100 includes: a lower traveling body 1; an upper revolving body 3 rotatably mounted on the lower traveling body 1 via a slewing mechanism 2; a boom 4, an arm 5 and a bucket 6 as an attachment device (working device); and a cab 10.

[0038] The lower traveling body 1 includes, for example, a pair of left and right crawlers, and each crawler travels by being hydraulically driven by traveling hydraulic motors 1A, 1B (refer to Figure 2 ).

[0039] The upper revolving body 3 rotates relative to the lower traveling body 1 by being driven by a slewing hydraulic motor 2A (refer to Figure 2 ).

[0040] The boom 4 is pivotally mounted at the front center of the upper revolving body 3 so as to be able to pitch. The arm 5 is pivotally mounted at the front end of the boom 4 so as to be able to rotate up and down, and the bucket 6 is pivotally mounted at the front end of the arm 5 so as to be able to rotate up and down. The boom 4, the arm 5 and the bucket 6 are respectively hydraulically driven by a boom cylinder 7, an arm cylinder 8 and a bucket cylinder 9.

[0041] The cab 10 is an operating room for an operator (worker) to ride in, and is mounted on the front left side of the upper revolving body 3.

[0042] The excavator 100 can communicate with the management device 300 via, for example, a mobile communication network with a base station as a terminal, a satellite communication network using a communication satellite in the sky, and a predetermined communication network NW including the Internet.

[0043] Moreover, the management device 300 of the present embodiment acquires performance information indicating the performance of the work from the excavator 100.

[0044] The performance information includes: performance information related to the work mode of a predetermined type of work (for example, repetitive work such as excavation work, loading work, finishing work, etc.) (hereinafter, "work mode performance information"); and performance information related to the environmental conditions during work (hereinafter, "environmental condition performance information").

[0045] The operation mode representation includes a series of action types during the idling of the excavator 100 when performing a specified type of operation. Idling refers to when the excavator 100 is not in operation. For example, the operation mode includes the movement trajectories during operations of action elements such as the lower traveling body 1, the upper slewing body 3, the boom 4, the arm 5, and the bucket 6. Also, regarding the operation mode performance information, specifically, it is the detection information of various sensors indicating the performance of the operation mode of the excavator 100 when the excavator 100 actually performs a specified type of operation. And in the environmental conditions, in addition to external environmental conditions such as conditions related to the surrounding environment of the excavator 100, internal environmental conditions such as variable specifications of the excavator 100 (e.g., the length of the arm, the type of bucket, etc.) that affect the operation of the excavator 100 can also be included.

[0046] If the excavator 100 acquires the operation mode performance information and the environmental condition performance information, it sends various information including the operation mode performance information and the environmental condition performance information to the management device 300 (upload).

[0047] In the management system SYS, the management device 300 uses the operation mode performance information and the environmental condition performance information received from the excavator 100 as learning data to create an abnormality determination model for determining the presence or absence of an abnormality in the excavator 100.

[0048] More specifically, the management device 300 determines the operation mode performance information and the environmental condition performance information used in the learning data from the operation mode performance information and the environmental condition performance information when the excavator 100 is operating normally. In the following description, the information including the operation mode performance information and the environmental condition performance information is sometimes referred to as the operation data of the excavator 100.

[0049] And the management device 300 of the present embodiment generates an abnormality determination model by learning with the specified operation data as learning data, and determines the presence or absence of an abnormality in the excavator 100.

[0050] In addition, in Figure 1 the example, the number of excavators 100 included in the management system SYS is set to 1, but it is not limited thereto. The number of excavators 100 included in the management system SYS can be arbitrary, and all excavators 100 capable of communicating with the management device 300 can be included in the management system SYS.

[0051] Moreover, the management device 300 of the present embodiment is an information processing device provided at a location geographically separated from the excavator 100. The management device 300 is provided, for example, in a management center or the like outside the work site where the excavator 100 performs operations, and is a server device centered around one or more server computers or the like. At this time, the server device can be the company's own server operated by the operator of the application management system SYS or a related operator related to the operator, or a cloud server.

[0052] The support device 400 of the present embodiment can be, for example, a portable terminal device such as a smartphone or a tablet. For example, if information such as the start date and completion date of the maintenance of the excavator 100 is input, the support device 400 sends the input information to the management device 300.

[0053] Next, with reference to Figure 2 , the management system SYS of the present embodiment will be further described. Figure 2 It is a structural diagram showing an example of the management system of the excavator.

[0054] In addition, in the figure, the mechanical power pipeline is represented by a double line, the high-pressure hydraulic pipeline is represented by a thick solid line, the pilot pipeline is represented by a dotted line, and the electric drive / control pipeline is represented by a thin solid line.

[0055] The hydraulic drive system that hydraulically drives the hydraulic actuators of the excavator 100 of the present embodiment includes an engine 11, a main pump 14, a regulator 14a, and a control valve 17. And, as described above, the hydraulic drive system of the excavator 100 includes hydraulic actuators such as a travel hydraulic motor 1A, a travel hydraulic motor 1B, a swing hydraulic motor 2A, a boom cylinder 7, a bucket cylinder 8, and a dipper cylinder 9 that respectively perform hydraulic drive on the lower travel body 1, the upper swing body 3, the boom 4, the bucket arm 5, and the bucket 6.

[0056] The engine 11 is the main power source in the hydraulic drive system. For example, it is mounted on the rear part of the upper swing body 3. Specifically, the engine 11 rotates at a preset target speed under the control of an engine control device (ECU: Engine Control Unit) 74 described later, and drives the main pump 14 and the pilot pump 15. The engine 11 is, for example, a diesel engine fueled by diesel.

[0057] The regulator 14a controls the discharge amount of the main pump 14. For example, the regulator 14a adjusts the angle ("deflection angle") of the swash plate of the main pump 14 according to a control command from the controller 30.

[0058] For example, like the engine 11, the main pump 14 is mounted on the rear part of the upper swing body 3, and supplies working oil to the control valve 17 through the high-pressure hydraulic pipeline 16. As described above, the main pump 14 is driven by the engine 11. The main pump 14 is, for example, a variable-capacity hydraulic pump. As described above, under the control of the controller 30, the swash plate deflection angle is adjusted by the regulator 14a, thereby adjusting the piston stroke length, and thus the discharge flow rate (discharge pressure) can be controlled.

[0059] The control valve 17 is, for example, mounted on the central part of the upper swing body 3, and is a hydraulic control device that controls the hydraulic drive system according to the operation of the operator on the operating device 26. As described above, the control valve 17 is connected to the main pump 14 via the high-pressure hydraulic pipeline 16, and selectively supplies the working oil supplied from the main pump 14 to the hydraulic actuators (travel hydraulic motors 1A, travel hydraulic motors 1B, swing hydraulic motor 2A, boom cylinder 7, arm cylinder 8, and bucket cylinder 9) according to the operation state of the operating device 26.

[0060] Specifically, the control valve 17 includes a plurality of control valves that control the flow rate and flow direction of the working oil respectively supplied from the main pump 14 to the hydraulic actuators. For example, the control valve 17 includes a control valve corresponding to the boom 4 (boom cylinder 7). And, for example, the control valve 17 includes a control valve corresponding to the arm 5 (arm cylinder 8).

[0061] And, for example, the control valve 17 includes a control valve corresponding to the bucket 6 (bucket cylinder 9). And, for example, the control valve 17 includes a control valve corresponding to the upper swing body 3 (swing hydraulic motor 2A). And, for example, the control valve 17 includes a right travel control valve and a left travel control valve respectively corresponding to the crawlers on the right side and the left side of the lower travel body 1.

[0062] The operating system of the excavator 100 according to the present embodiment includes a pilot pump 15, an operating device 26, and an operation valve 31.

[0063] The pilot pump 15 is, for example, mounted on the rear part of the upper swing body 3, and supplies pilot pressure to the operating device 26 and the operation valve 31 through the pilot pipeline 25. The pilot pump 15 is, for example, a fixed-capacity hydraulic pump, and is driven by the engine 11 as described above.

[0064] The operating device 26 is located near the operator's seat in the cab 10 and serves as an input mechanism for the operator to operate the various operating elements (such as the lower traveling structure 1, upper swing structure 3, boom 4, arm 5, and bucket 6). In other words, the operating device 26 serves as an input mechanism for the operator to operate the hydraulic actuators that drive the various operating elements (i.e., the traveling hydraulic motor 1A, traveling hydraulic motor 1B, swing hydraulic motor 2A, boom cylinder 7, arm cylinder 8, and bucket cylinder 9). The pilot lines on the secondary side of the operating device 26 are connected to the control valves 17.

[0065] Thus, a pilot pressure corresponding to the operating state of the lower traveling body 1, upper swing body 3, boom 4, arm 5, bucket 6, etc. in the operating device 26 can be input to the control valve 17. Therefore, the control valve 17 can drive each hydraulic actuator according to the operating state in the operating device 26.

[0066] The operating valve 31 adjusts the flow area of the pilot line 25 in response to a control command (e.g., a control current) from the controller 30. Thus, the operating valve 31 can use the primary-side pilot pressure supplied from the pilot pump 15 as the initial pressure and output a pilot pressure corresponding to the control command to the secondary-side pilot line.

[0067] The secondary-side port of the operating valve 31 is connected to the left and right pilot ports of the control valve corresponding to the respective hydraulic actuators of the control valve 17, so that a pilot pressure corresponding to a control command from the controller 30 acts on the pilot ports of the control valve. Thus, even when the operator is not operating the operating device 26, the controller 30 can supply the hydraulic fluid discharged from the pilot pump 15 to the corresponding pilot ports of the control valve in the control valve 17 via the operating valve 31, thereby actuating the hydraulic actuators.

[0068] In addition to the operating valve 31, an electromagnetic relief valve may be provided to release excess hydraulic pressure generated within the hydraulic actuator to the hydraulic oil tank. This allows the hydraulic actuator to be actively suppressed in situations such as when the operator operates the operating device 26 excessively. For example, an electromagnetic relief valve may be provided to release excess pressure in the bottom and rod oil chambers of the boom cylinder 7, arm cylinder 8, and bucket cylinder 9 to the hydraulic oil tank.

[0069] The control system of the shovel 100 according to the present embodiment includes a controller 30 , an ECU 74 , a discharge pressure sensor 14 b , an operating pressure sensor 15 a , a display device 40 , an input device 42 , an imaging device 80 , a state detection device S1 , and a communication device T1 .

[0070] The controller 30 performs drive control of the excavator 100. The functions of the controller 30 can be implemented by any hardware, software, or a combination thereof. For example, the controller 30 is centered around a computer including a processor such as a CPU (Central Processing Unit), a memory device such as a RAM (Random Access Memory), a non-volatile auxiliary storage device such as a ROM (Read Only Memory), and various input / output interface devices. The controller 30 realizes various functions, for example, by executing various programs installed on the auxiliary storage device on the CPU.

[0071] For example, the controller 30 performs the following drive control: sets a target rotational speed based on a job mode or the like preset by a prescribed operation by an operator or the like, and makes the engine 11 rotate at a constant speed via the ECU 74 by outputting a control instruction to the ECU 74.

[0072] And, for example, the controller 30 outputs a control instruction to the regulator 14a as needed, and performs so-called total horsepower control or negative control by changing the discharge amount of the main pump 14.

[0073] And, for example, the controller 30 may further have a function of uploading various information related to the excavator 100 to the management device 300 (hereinafter referred to as the "upload function"). Specifically, the controller 30 can send (upload) the job mode performance information and the environmental condition performance information during a prescribed type of operation of the excavator 100 to the management device 300 via the communication device T1.

[0074] The controller 30 includes, for example, an information transmission unit 301 as a functional unit related to the upload function, and the upload function is realized by executing one or more programs installed on the auxiliary storage device or the like on the CPU.

[0075] And, for example, the controller 30 performs control related to the equipment guidance function that guides the manual operation of the excavator 100 performed by the operator via the operation device 26. And the controller 30 can perform control related to the equipment control function that automatically supports the manual operation of the excavator 100 performed by the operator via the operation device 26.

[0076] The controller 30 includes, for example, a job mode acquisition unit 302 and an equipment guidance unit 303 as functional units related to the equipment guidance function and the equipment control function, and the equipment guidance function and the equipment control function are realized by executing one or more programs installed on the auxiliary storage device or the like on the CPU.

[0077] In addition, a part of the functions of the controller 30 can also be implemented by other controllers (control devices). That is, the functions of the controller 30 can also be implemented in a distributed manner by multiple controllers. For example, the above-described equipment guidance function and equipment control function can also be implemented by dedicated controllers (control devices).

[0078] The ECU 74 controls various actuators (such as fuel injection devices, etc.) of the engine 11 according to the control instructions from the controller 30, so that the engine 11 rotates at a set target speed (set speed) for constant rotation (constant rotation control). At this time, the ECU 74 performs constant rotation control of the engine 11 based on the speed of the engine 11 detected by the engine speed sensor 11a.

[0079] The discharge pressure sensor 14b detects the discharge pressure of the main pump 14. A detection signal corresponding to the discharge pressure detected by the discharge pressure sensor 14b is input to the controller 30.

[0080] As described above, the operation pressure sensor 15a detects the pilot pressure on the secondary side of the operating device 26, that is, the pilot pressure corresponding to the operating state of each action element (hydraulic actuator) in the operating device 26. Detection signals of the pilot pressure corresponding to the operating states of the lower traveling body 1, the upper swing body 3, the boom 4, the arm 5, the bucket 6, etc. in the operating device 26 based on the operation pressure sensor 15a are input to the controller 30.

[0081] The display device 40 is connected to the controller 30 and is arranged at a position that is easily visually recognizable by the operator sitting in the cab 10 under the control of the controller 30, and displays various information images. The display device 40 is, for example, a liquid crystal display or an organic EL (Electroluminescence) display, etc.

[0082] The input device 42 is arranged within the reach of the operator sitting in the cab 10, receives various operations based on the operator, and outputs a signal corresponding to the operation content. For example, the input device 42 is integrated with the display device 40.

[0083] Moreover, the input device 42 of the present embodiment includes a switch 42a operated when starting the maintenance of the excavator 100 and a switch 42b operated when the maintenance is completed. The switch 42a is an example of a maintenance start switch and is operated by an operator or the like performing the maintenance of the excavator 100. The switch 42b is an example of a maintenance completion switch and is operated by an operator or the like performing the maintenance of the excavator 100.

[0084] In addition, switches 42a and 42b of the present embodiment can be displayed on the display of the support device 400, for example. At this time, the user of the support device 400 can operate switches 42a and 42b.

[0085] Regarding the excavator 100 of the present embodiment, for example, by operating switches 42a and 42b at the start and end of maintenance (repair), for example, the start and end of maintenance and other situations can be notified to surrounding operators and the like.

[0086] In the present embodiment, by operating switches 42a and 42b, it is possible to determine the maintenance start date and maintenance completion date described later Figure 5 、 Figure 7 In addition, by operating switches 42a and 42, in addition to determining the date, the time can also be determined together.

[0087] Moreover, in the image captured by the imaging device 80, when it is determined that there is a person within a specified range from the excavator 100 before the actuator operates, even if the operator operates the joystick, the operation of the actuator can be disabled or in a low-speed state. Specifically, when it is determined that there is a person within a specified range from the excavator 100, the door lock valve (not shown) is brought into a locked state, whereby the operation of the actuator can be disabled. In the case of an electric operation device, by invalidating the signal from the controller 30 to the operation control valve, the operation of the actuator can be disabled.

[0088] In the case of using an operation lever of other methods, it is the same in the case of an operation control valve that uses a pilot pressure corresponding to a control command from the controller 30 and applies the pilot pressure to a corresponding pilot port in the control valve. When it is desired to make the operation of the actuator in a low-speed state, the actuator can be made in a low-speed state by reducing the signal from the controller 30 to the operation control valve.

[0089] In this way, if it is determined that the detected object exists within the specified range, even if the operation device is operated, the actuator will not be driven, or it will be driven at a low speed with an output smaller than the input to the operation device.

[0090] Furthermore, in the operation of the operator's joystick, when it is determined that there is a person within a specified range from the excavator 100, the operation of the actuator can be stopped or decelerated without relying on the operator's operation. Specifically, when it is determined that there is a person within a specified range from the excavator 100, the actuator can be stopped by bringing the door lock valve into a locked state.

[0091] In the case of an operating control valve that uses a pilot pressure corresponding to a control instruction from the controller 30 and applies the pilot pressure to a corresponding pilot port of a control valve within the control valve, it is possible to invalidate the signal from the controller 30 to the operating control valve or output a deceleration instruction, thereby preventing the actuator from operating. Also, in the case where the detected object is a truck, control stop is not required.

[0092] Control the actuator to avoid the detected truck. Thus, control the actuator based on the type of the detected object being easily recognizable.

[0093] Moreover, the controller 30 stores the location, time, and action content (such as walking, slewing, etc.) when it is determined that a person exists within a specified range of the excavator. Here, when the controller 30 is in the ON state, if the operator approaches the excavator for maintenance, the controller 30 determines that a person exists. However, this approach at this time is an approach for maintenance based on normal operation.

[0094] Therefore, by the operator operating the switch 42a, the controller 30 can determine that the approach of the operator after the operation is an approach for maintenance, and store by associating the record of the content determined to have detected a person (location, date and time, action content, etc.) with the record of the content that it is an approach for maintenance. In other words, when the controller 30 is in the ON state, if it receives the operation of the switch 42a, the controller 30 determines that the approach of a person within the specified range is an approach for maintenance. Also, the controller 30 stores in a storage device or the like by associating the information of the record indicating the detection of a person approaching within the specified range with the information indicating that it is an approach of a person for maintenance. The information of the record indicating the detection of a person approaching within the specified range includes the location, date, and time of detecting the person approaching, etc.

[0095] The controller 30 of the excavator 100 sends, in a state where the record of the content determined to have detected a person (location, date and time, action content, etc.) is associated with the record of the content that it is an approach for maintenance, to the management device 300. Then, at the end of maintenance, if the operator operates the switch 42b, the controller 30 determines that the maintenance is over, and also determines that the correspondence association between the record of the content determined to have detected a person and the record of the content that it is an approach for maintenance is over. Thus, even if there is a record of the content determined to have detected a person (location, date and time, etc.) in the management device 300, it is possible to grasp that the reason for determining the detection of a person is an approach for maintenance.

[0096] Further, the input device 42 can be provided separately from the display device 40. The input device 42 includes: a touch panel mounted on the display of the display device 40, a knob switch provided at the front end of the lever included in the operating device 26, a button switch provided around the display device 40, a lever, a changeover key, etc. A signal corresponding to the operation content with respect to the input device 42 is input to the controller 30.

[0097] The imaging device 80 images the surroundings of the excavator 100. The imaging device 80 includes: a camera 80F that images the front of the excavator 100, a camera 80L that images the left side of the excavator 100, a camera 80R that images the right side of the excavator 100, and a camera 80B that images the rear of the excavator 100.

[0098] The camera 80F is installed, for example, on the ceiling of the cab 10, that is, inside the cab 10. Further, the camera 80F can also be installed outside the cab 10 such as on the roof of the cab 10 or on the side of the boom 4. The camera 80L is installed at the left end of the upper surface of the upper swing body 3, the camera 80R is installed at the right end of the upper surface of the upper swing body 3, and the camera 80B is installed at the rear end of the upper surface of the upper swing body 3.

[0099] The imaging device 80 (cameras 80F, 80B, 80L, 80R) is respectively, for example, a monocular wide-angle camera having a very wide field of view angle. Further, the imaging device 80 can also be a stereo camera or a distance image camera, etc. The captured image of the surroundings of the excavator 100 based on the imaging device 80 (hereinafter, referred to as "surroundings image") is input to the controller 30.

[0100] The state detection device S1 outputs detection information related to various states of the excavator 100. The detection information output from the state detection device S1 is input to the controller 30.

[0101] For example, the state detection device S1 detects the posture state or the operation state of the attachment. Specifically, the state detection device S1 can detect the pitch angles of the boom 4, the arm 5, and the bucket 6 (hereinafter, referred to as "boom angle", "arm angle", "bucket angle") respectively.

[0102] That is, the state detection device S1 can include a boom angle sensor, an arm angle sensor, and a bucket angle sensor that respectively detect the boom angle, the arm angle, and the bucket angle.

[0103] Moreover, the state detection device S1 can detect the acceleration, angular acceleration, etc. of the boom 4, the arm 5, and the bucket 6. At this time, the state detection device S1 can include, for example, rotary encoders, acceleration sensors, angular acceleration sensors, 6-axis sensors, and IMUs (Inertial Measurement Units) respectively installed on the boom 4, the arm 5, and the bucket 6. Moreover, the state detection device S1 can include cylinder sensors that detect the cylinder position, speed, acceleration, etc. of the boom cylinder 7, the arm cylinder 8, and the bucket cylinder 9 that drive the boom 4, the arm 5, and the bucket 6 respectively.

[0104] Moreover, for example, the state detection device S1 detects the fuselage, that is, it detects the posture state of the lower traveling body 1 and the upper slewing body 3. Specifically, the state detection device S1 can detect the inclination state of the fuselage with respect to the horizontal plane. At this time, the state detection device S1 can include, for example, an inclination sensor that is installed on the upper slewing body 3 and detects the inclination angles of the upper slewing body 3 around two axes in the front-rear direction and the left-right direction (hereinafter referred to as "front-rear inclination angle" and "left-right inclination angle").

[0105] Moreover, for example, the state detection device S1 detects the slewing state of the upper slewing body 3. Specifically, the state detection device S1 detects the slewing angular velocity or slewing angle of the upper slewing body 3. At this time, the state detection device S1 can include, for example, a gyro sensor, a resolver, a rotary encoder, etc. installed on the upper slewing body 3. That is, the state detection device S1 can include a slewing angle sensor that detects the slewing angle, etc. of the upper slewing body 3.

[0106] Moreover, for example, the state detection device S1 detects the acting state of the force acting on the excavator 100 through the attachment device. Specifically, the state detection device S1 can detect the working pressure (cylinder pressure) of the hydraulic actuator. At this time, the state detection device S1 can include pressure sensors that detect the pressures in the rod-side oil chambers and the bottom-side oil chambers of the boom cylinder 7, the arm cylinder 8, and the bucket cylinder 9.

[0107] Moreover, for example, the state detection device S1 can include sensors that detect the displacement of the valve stem of the control valve in the control valve 17. Specifically, the state detection device S1 can include a boom valve stem displacement sensor that detects the displacement of the boom valve stem. Moreover, the state detection device S1 can include an arm valve stem displacement sensor that detects the displacement of the arm valve stem.

[0108] Moreover, the state detection device S1 may include a bucket valve stem displacement sensor for detecting the displacement of the bucket valve stem. Moreover, the state detection device S1 may include a swing valve stem displacement sensor for detecting the displacement of the swing valve stem. Moreover, the state detection device S1 may include a right travel valve stem displacement sensor and a left travel valve stem displacement sensor for detecting the displacements of the right travel valve stem and the left travel valve stem that respectively constitute the right travel control valve and the left travel control valve.

[0109] Moreover, for example, the state detection device S1 detects the position of the excavator 100, the orientation of the upper swing body 3, etc. At this time, the state detection device S1 may include, for example, a GNSS (Global Navigation Satellite System) compass, a GNSS sensor, an azimuth sensor, etc. installed on the upper swing body 3.

[0110] The communication device T1 communicates with an external device via the communication network NW. The communication device T1 is, for example, a mobile communication module corresponding to a mobile communication standard such as LTE (Long Term Evolution), 4G (4th Generation), 5G (5th Generation), or a satellite communication module for connecting to a satellite communication network.

[0111] The information transmission unit 301 transmits the operation mode performance information and the environmental condition performance information of the excavator 100 during a specified type of operation to the management device 300 via the communication device T1.

[0112] In the operation mode performance information transmitted by the information transmission unit 301, for example, it includes various detection information input from the state detection device S1.

[0113] Moreover, in the environmental condition performance information transmitted by the information transmission unit 301, for example, it includes the surrounding image of the excavator 100 input from the imaging device 80. Moreover, in the environmental condition performance information transmitted by the information transmission unit 301, it may also include the internal environmental conditions of the excavator 100, for example, information related to variable specifications such as the large capacity bucket specification, the long arm specification, the quick coupler specification, etc.

[0114] Regarding the information transmission unit 301, for example, it sequentially determines whether a pre-specified object type of operation is being performed. If it is determined that the object type of operation is being performed, the operation mode performance information (i.e., various detection information input from the state detection device S1) and the environmental condition information (i.e., the surrounding image of the excavator 100 input from the imaging device 80) during the period of performing this operation are associated and recorded in an internal memory or the like.

[0115] At this time, it is also possible to store in the internal memory in such a way that the date and time information related to the start and end of the work of the object type, as well as the position information of the excavator 100 during the work, are further associated with the set of work mode performance information and environmental condition performance information. At this time, the date and time information can be obtained, for example, from a prescribed timekeeping mechanism (e.g., RTC (Real Time Clock)) inside the controller 30. Then, at a prescribed time such as when the power supply of the excavator 100 is turned off (stopped), the information transmission unit 301 transmits the set of the recorded work mode performance information and environmental condition performance information to the management device 300 through the communication device T1. Also, the information transmission unit 301 can transmit the set of the recorded work mode performance information and environmental condition performance information to the management device 300 through the communication device T1 after each work of the object type is completed.

[0116] In addition, in the environmental condition performance information, instead of or in addition to the imaging device 80, it may include detection information detected by other sensors mounted on the excavator 100. For example, it may be in such a way that other sensors such as a millimeter-wave radar and a LIDAR (Light Detecting and Ranging) are mounted on the excavator 100, and the detection information of these distance sensors is included in the environmental condition performance information. The same applies to the current environmental condition information described later.

[0117] Also, weather information may be included in the environmental condition performance information. The weather information can include, for example, detection information such as a rain drop sensor and an illuminance sensor that can be included in the state detection device S1. And the information transmission unit 301 can transmit only the work mode performance information to the management device 300.

[0118] Also, the information transmission unit 301 can sequentially upload the detection information of the state detection device S1 or the surrounding image of the excavator 100 based on the imaging device 80 to the management device 300 through the communication device T1. At this time, the management device 300 can extract the information during the work of the object type from the information uploaded by the excavator 100 and generate the work mode performance information and the environmental information.

[0119] When performing a specified type of operation, the operation mode acquisition unit 302 acquires from the management device 300 an operation mode (optimal operation mode) that is most suitable for the current environmental conditions related to a specified target index. For example, based on a specified operation (hereinafter referred to as an "acquisition request operation") performed by the operator on the input device 42, the operation mode acquisition unit 302 sends, via the communication device T1, a signal (acquisition request signal) requesting the acquisition of an operation mode including information related to the current environmental conditions of the excavator 100 (hereinafter referred to as "current environmental condition information") to the management device 300.

[0120] Thereby, the management device 300 can provide the excavator 100 with an optimal operation mode that conforms to the current environmental conditions of the excavator 100. The current environmental condition information includes, for example, the latest surrounding image of the excavator 100 based on the imaging device 80.

[0121] In addition, the current environmental condition information may also include the internal environmental conditions of the excavator 100, for example, information related to variable specifications such as a large-capacity bucket specification, a long arm specification, and a quick coupler specification. Further, the current environmental condition information may also include detection information such as a rain drop sensor or an illuminance sensor that may be included in the state detection device S1, that is, weather information. Then, the operation mode acquisition unit 302 acquires information related to the operation mode that is sent from the management device 300 according to the acquisition request signal and received by the communication device T1.

[0122] The equipment guidance unit 303 performs control related to the equipment guidance function and the equipment control function. That is, the equipment guidance unit 303 supports the operation of various action elements (the lower traveling body 1, the upper slewing body 3, and the attachment device including the boom 4, the arm 5, and the bucket 6) performed by the operator through the operation device 26.

[0123] For example, when the operator operates the arm 5 through the operation device 26, the equipment guidance unit 303 can automatically operate at least one of the boom 4 and the bucket 6 so that a preset target design surface (hereinafter simply referred to as the "design surface") coincides with the front end (for example, the cutting edge or the back surface) of the bucket 6.

[0124] Moreover, the equipment guidance unit 303 can also automatically operate the arm 5 regardless of the operation state of the operation device 26 for operating the arm 5. That is, the equipment guidance unit 303 can trigger the attachment device to perform a preset action based on the operation of the operator on the operation device 26.

[0125] More specifically, the equipment guidance unit 303 obtains various information from the state detection device S1, the imaging device 80, the communication device T1, the input device 42, and the like. And, for example, the equipment guidance unit 303 calculates the distance between the bucket 6 and the design surface based on the obtained information. Then, based on the calculated distance between the bucket 6 and the design surface and the like, the equipment guidance unit 303 appropriately controls the operation valve 31, and by individually and automatically adjusting the pilot pressure acting on the control valve corresponding to the hydraulic actuator, each hydraulic actuator can be automatically operated.

[0126] In the operation valve 31, for example, it includes a boom proportional valve corresponding to the boom 4 (boom cylinder 7). And, in the operation valve 31, for example, it includes a stick proportional valve corresponding to the stick 5 (stick cylinder 8). And, in the operation valve 31, for example, it includes a bucket proportional valve corresponding to the bucket 6 (bucket cylinder 9). And, in the operation valve 31, for example, it includes a swing proportional valve corresponding to the upper swing body 3 (swing hydraulic motor 2A). And, in the operation valve 31, for example, it includes a right travel proportional valve and a left travel proportional valve corresponding to the crawlers on the right and left sides of the lower traveling body 1 respectively.

[0127] For example, in order to support the excavation operation, the equipment guidance unit 303 can automatically extend and retract at least one of the boom cylinder 7, the stick cylinder 8, and the bucket cylinder 9 according to the opening and closing operations of the stick 5 with respect to the operating device 26.

[0128] The excavation operation is an operation of excavating the ground along the design surface using the tip of the bucket 6. Regarding the equipment guidance unit 303, for example, when the operator manually operates the operating device 26 in the closing direction of the stick 5 (hereinafter, referred to as "stick closing operation"), at least one of the boom cylinder 7 and the bucket cylinder 9 is automatically extended and retracted.

[0129] And, for example, in order to support the trimming operation of the slope or the horizontal plane, the equipment guidance unit 303 can automatically extend and retract at least one of the boom cylinder 7, the stick cylinder 8, and the bucket cylinder 9. The trimming operation, for example, includes an operation of pulling the bucket 6 toward the front along the design surface while pressing the back surface of the bucket 6 against the ground.

[0130] Regarding the equipment guidance unit 303, for example, when the operator manually operates the operating device 26 in the stick closing operation, at least one of the boom cylinder 7 and the bucket cylinder 9 is automatically extended and retracted. Thereby, it is possible to move the bucket 6 along the completed slope or horizontal plane, that is, the design surface, while pressing the back surface of the bucket 6 against the slope (slope) or the horizontal plane before completion with a specified pressing force.

[0131] Further, the equipment guiding unit 303 can also automatically rotate the slewing hydraulic motor 2A to align the upper slewing structure 3 with the design surface. At this time, the equipment guiding unit 303 can align the upper slewing structure 3 with the design surface by operating a specified switch included in the operation input device 42. Further, the equipment guiding unit 303 can also only operate the specified switch to align the upper slewing structure 3 with the design surface and start the equipment control function.

[0132] Further, for example, when the equipment guiding unit 303 performs a specified type of operation (such as excavation operation, loading operation, finishing operation, etc.), it controls according to the operation performed by the operator on the operation device 26 so that the actions of at least a part of the attachment device, the upper slewing structure 3, and the lower traveling body 1 conform to the operation mode (optimal operation mode) obtained by the operation mode acquisition unit 302.

[0133] Thereby, the operator can make the actions of the excavator 100 consistent with the specified target index without depending on the proficiency related to the operation of the excavator 100. For example, the operation mode output from the management device 300 with a relatively high evaluation of the operation speed and most suitable for the current environmental conditions of the excavator 100.

[0134] Further, the equipment guiding unit 303 can, according to the optimal operation mode, control the actions of the excavator 100 while displaying the actions of the excavator 100 corresponding to the optimal operation mode on the display device 40 for the operator. For example, when the equipment guiding unit 303 controls the actions of the excavator 100 according to the optimal operation mode, a dynamic image of the simulation test result corresponding to the optimal operation mode is displayed on the display device 40. Thereby, the operator can perform the operation while confirming the content of the actual operation mode through the dynamic image of the display device 40.

[0135] The management device 300 of the present embodiment includes a main information storage unit 310, a malfunction information storage unit 320, an action information storage unit 330, a vehicle information storage unit 340, an information collection unit 350, a data range determination unit 360, a learning unit 370, and an abnormality determination unit 380.

[0136] The main information storage unit 310 stores the main information for determining the excavator 100. The malfunction information storage unit 320 stores the malfunction information related to the malfunction of the excavator 100. The action information storage unit 330 stores the action information related to the actions of the excavator 100 sent from the excavator 100. The action information includes the operation data of the excavator 100. The vehicle information storage unit 340 stores the vehicle information associating the main information, the malfunction information, and the action information for each excavator 100. A detailed description of each of the above storage units will be given later.

[0137] The information collection unit 350 of the present embodiment collects the information stored in the main information storage unit 310, the defect information storage unit 320, and the operation information storage unit 330, and stores it in each storage unit. Further, the information collection unit 350 associates the collected information items to form vehicle information, and stores it in the vehicle information storage unit 340. Specifically, the information collection unit 350 communicates with a system managed by an enterprise selling the excavator 100 or the like, or a system managed by an enterprise performing maintenance on the excavator 100 or the like, to collect the main information or the defect information. Further, the information collection unit 350 collects operation information from the excavator 100. Then, the information collection unit 350 stores the vehicle information associating the main information, the defect information, and the operation information in the vehicle information storage unit 340.

[0138] Based on the vehicle information stored in the vehicle information storage unit 340 and the period for determining the presence or absence of an abnormality in the excavator 100, the data range determination unit 360 determines the range of the operation data included in the operation information stored in the operation information storage unit 330 to be the learning data input to the learning unit 370.

[0139] Based on the learning data, the learning unit 370 learns the relationship between the operation data of the excavator 100 and the presence or absence of an abnormality in the excavator 100. Specifically, based on the operation data (learning data) in a non-abnormal state, the learning unit 370 generates an abnormality determination model associating the input operation data with the presence or absence of an abnormality in the excavator 100.

[0140] Based on the output of the abnormality determination model, the abnormality determination unit 380 determines the presence or absence of an abnormality in the excavator 100 and outputs the determination result.

[0141] Hereinafter, the management device 300 of the present embodiment will be further described. Figure 3 It is a diagram showing an example of the hardware configuration of the management device.

[0142] The management device 300 of the present embodiment is a computer including an input device 311, an output device 312, a drive device 313, an auxiliary storage device 314, a memory device 315, an arithmetic processing device 316, and an interface device 317 that are respectively connected to each other via a bus B.

[0143] The input device 311 is a device for inputting various information, and can be realized by, for example, a keyboard or a pointing device. The output device 312 is for outputting various information, and can be realized by, for example, a display. The interface device 317 includes a LAN card or the like and is used for connecting to a network.

[0144] The program that implements the information collection unit 350, data range determination unit 360, learning unit 370, and anomaly determination unit 380 of the management device 300 is at least a part of the various programs that control the management device 300. For example, the program can be provided through the allocation of the storage medium 318 or downloading from the network. The storage medium 318 that records the program can use various types of storage media, such as optical, electrical, or magnetic recording information storage media like CD-ROM, flexible disk, magneto-optical disk, and semiconductor memories like ROM and flash memory that electrically record information.

[0145] Moreover, if the storage medium 318 that records the program is set on the drive device 313, the program is installed in the auxiliary storage device 314 from the storage medium 318 via the drive device 313. The program downloaded from the network can be installed in the auxiliary storage device 314 via the interface device 317.

[0146] The auxiliary storage device 314 implements each storage unit and the like of the management device 300, and while storing the programs installed in the management device 300, it also stores various necessary files, data, etc. based on the management device 300. The memory device 315 reads and stores the programs from the auxiliary storage device 314 when the management device 300 is started. And the arithmetic processing device 316 implements various processes as described later according to the programs stored in the memory device 315.

[0147] Next, refer to Figures 4 to 7 to describe each storage unit of the management device 300. Figure 4 is a diagram showing an example of the main information storage unit.

[0148] The main information stored in the main information storage unit 310 of the present embodiment includes, as information items, the body identification number, delivery date, and delivery location. The item "body identification number" is associated with other items.

[0149] The main information of the present embodiment is information including the values of the items "body identification number", "delivery date", and "delivery location". The main information of the present embodiment is, for example, information managed by the sales company of the excavator 100 and is collected and stored in the main information storage unit 310 by the information collection unit 350.

[0150] The value of the item "body identification number" is identification information used to determine the excavator 100. The value of the item "delivery date" indicates the date of delivery to an enterprise that purchases or leases the excavator 100. And the value of the item "delivery date" can be the date of moving into the job site where the excavator 100 is used for work.

[0151] The value of the item "Delivery Location" indicates the location of the work site where the operation of the excavator 100 is carried out. Also, the value of the item "Delivery Location" may also indicate the location of the company that purchases or leases the excavator 100. Additionally, the main information storage unit 310 may also include location information such as latitude and longitude indicating the location of the work site where the operation based on the excavator 100 is carried out.

[0152] Figure 5 It is a diagram showing an example of the defect situation information storage unit. The defect situation information stored in the defect situation information storage unit 320 has, as information items, a body identification number, a defect occurrence date, and a repair completion date, and the item "body identification number" is associated with other items.

[0153] The defect situation information of the present embodiment is information including the values of the items "body identification number", "defect occurrence date", "repair start date", and "repair completion date". The defect situation information of the present embodiment is, for example, information managed by a company or the like that manages the maintenance etc. of the excavator 100, and is collected by the information collection unit 350 and stored in the defect situation information storage unit 320.

[0154] The value of the item "defect occurrence date" indicates the date when a malfunction occurs in the excavator 100. More specifically, the value of the item "defect occurrence date" may, for example, also indicate the date when the operator of the excavator 100 notices the occurrence of a malfunction and notifies the maintenance company etc. of the occurrence of the malfunction. Also, the value of the item "defect occurrence date" may, for example, also indicate the date when a malfunction of the excavator 100 is confirmed by an operator etc. who performs the maintenance of the excavator 100.

[0155] The value of the item "repair start date" indicates the date when the repair (maintenance) of the excavator 100 starts. Specifically, the value of the item "repair start date" indicates the date when an operator etc. who performs the maintenance of the excavator 100 operates the switch 42a.

[0156] The value of the item "repair completion date" indicates the date when the repair (maintenance) of the excavator 100 is completed. Specifically, the value of the item "repair completion date" indicates the date when an operator etc. who performs the maintenance of the excavator 100 operates the switch 42b.

[0157] In this way, the management device 300 of the present embodiment manages by associating the start date and completion date of maintenance with the defect occurrence date in this manner, enabling the manager etc. of the management system SYS to grasp the period from the occurrence of a defect to the start of maintenance, or the number of days required for maintenance, etc.

[0158] Figure 6This is a diagram showing an example of the operation information storage unit. The operation information stored in the operation information storage unit 330 of the present embodiment includes, as information items, the airframe identification number, acquisition date and time, operation data, and presence or absence of repair work. The item "airframe identification number" is associated with other items.

[0159] The operation information of the present embodiment is information including the values of the items "airframe identification number", "acquisition date and time", "operation data", and "type of work". The operation information of the present embodiment is information transmitted through the information transmission unit 301 of the excavator 100 via the communication device T1, received by the information collection unit 350, and stored in the operation information storage unit 330.

[0160] The value of the item "acquisition date and time" indicates the date and time when the operation data is acquired. The value of the item "operation data" is information including the operation mode performance information and environmental condition performance information transmitted from the excavator 100. In other words, the value of the item "operation data" is information including various detection information output from the state detection device S1 and is information indicating the operation of the excavator 100.

[0161] The value of the item "type of work" indicates the type of work that the excavator 100 is performing. Specifically, the value of the item "type of work" indicates whether the work that the excavator 100 is performing is normal work or repair work. The repair work of the excavator 100 is, for example, work (operation) performed during maintenance, etc. of the excavator 100.

[0162] Figure 7 This is a diagram showing an example of the vehicle information storage unit.

[0163] The information collection unit 350 of the present embodiment collects main information, defect information, and operation information for each airframe identification number from the main information storage unit 310, the defect information storage unit 320, and the operation information storage unit 330, respectively, and stores the vehicle information associated with this information in the vehicle information storage unit 340. Therefore, the vehicle information can also be referred to as information generated by the information collection unit

[0164] The vehicle information of the present embodiment includes, as information items, the airframe identification number, delivery date, delivery location, defect occurrence date, repair completion date, acquisition date and time, operation data, and type of work. The item "airframe identification number" is associated with other items.

[0165] The vehicle information of the present embodiment is information including the values of the items "airframe identification number", "delivery date", "delivery location", "defect occurrence date", "repair start date", "repair completion date", "acquisition date and time", "operation data", and "type of work".

[0166] In Figure 7 the example of, it can be known that the excavator 100 identified by the fuselage identification number "XL-0029" was delivered to XX City, XX County on July 31, 2018, had a malfunction on August 10, 2019, started maintenance on August 11, 2019, and completed maintenance on August 15, 2019. Also, it can be known that the excavator 100 identified by the fuselage identification number "XL-0029" obtained the operation data during normal operation on August 20, 2019.

[0167] Next, referring to Figure 8 , the functions of the management device 300 of the present embodiment will be described. Figure 8 It is a diagram for explaining the functions of the management device.

[0168] The management device 300 of the present embodiment includes a storage unit 305, an information collection unit 350, a data range determination unit 360, a learning unit 370, and an abnormality determination unit 380.

[0169] The storage unit 305 is implemented by, for example, an auxiliary storage device 314 or a storage device 315. Also, the information collection unit 350, the data range determination unit 360, the learning unit 370, and the abnormality determination unit 380 are implemented by the arithmetic processing device 316 reading and executing programs stored in the storage device 315 and the like.

[0170] The information collection unit 350 of the present embodiment collects main information, malfunction information, and operation information, and stores them in the main information storage unit 310, the malfunction information storage unit 320, and the operation information storage unit 330 respectively. Also, the information collection unit 350 generates vehicle information associating the main information, the malfunction information, and the operation information for each fuselage identification number of the excavator 100, and stores it in the vehicle information storage unit 340.

[0171] The data range determination unit 360 determines the range of the learning data to be used for learning by the learning unit 370 in the operation data. Hereinafter, the data range determination unit 360 will be described.

[0172] The data range determination unit 360 includes an input reception unit 361, a parameter storage unit 362, a reference mark assignment unit 363, an exclusion range determination unit 364, a selection range determination unit 365, and a learning data output unit 366.

[0173] The input receiving unit 361 receives various inputs with respect to the management device 300. Specifically, the input receiving unit 361 receives the input of information indicating the period for performing the abnormality determination described later. The information indicating the period for performing the abnormality determination represents the date for executing the abnormality determination process by the abnormality determination unit 380. In the following description, the date for executing the abnormality determination process may sometimes be referred to as the execution date.

[0174] The parameter storage unit 362 stores the parameters referred to by the exclusion range determination unit 364 and the adoption range determination unit 365. The parameters stored by the parameter storage unit 362 can be arbitrarily set by the manager of the management system SYS or the like. The detailed description of the parameters stored by the parameter storage unit 362 will be described later.

[0175] The reference mark assignment unit 363 refers to the vehicle information, determines the first exclusion range that cannot be learning data in the operation data, and assigns a mark based on the first exclusion range to the operation data.

[0176] The exclusion range determination unit 364 determines the second exclusion range based on the parameters stored in the parameter storage unit 362 and the first exclusion range.

[0177] That is, the reference mark assignment unit 363 and the exclusion range determination unit 364 of the present embodiment are examples of range determination units that determine the range to be excluded from the learning data in the operation information included in the vehicle information.

[0178] The adoption range determination unit 365 refers to the parameters stored in the parameter storage unit 362 and the second exclusion range, and determines the range of the operation data to be adopted in the learning data based on the period for performing the abnormality determination received by the input receiving unit 361.

[0179] In the following description, the operation data within the range determined by the adoption range determination unit 365 may sometimes be referred to as learning data. In other words, the learning data is the operation data acquired during the period determined by the adoption range determination unit 365.

[0180] The learning data output unit 366 outputs the learning data to the learning unit 370.

[0181] The learning unit 370 learns with the learning data as input and generates an abnormality determination model 375. That is, the learning unit 370 uses the operation data corresponding to the period for determining the presence or absence of an abnormality of the excavator 100 as input, and uses the data set with the information indicating no abnormality as output as the learning data to generate the abnormality determination model 375. In other words, the learning unit 370 uses the operation data in the normal state and the data set with the information indicating no abnormality as output to generate an abnormality determination model that associates the input operation data with the abnormality degree of the excavator 100.

[0182] Further, if the input is operation data that is the object of the abnormality determination process, the learning unit 370 uses this operation data as the input to the abnormality determination model 375 and obtains an index value indicating the presence or absence of an abnormality in the excavator 100. In the following description, the operation data that is the input to the abnormality determination model 375 is sometimes referred to as determination data.

[0183] The learning unit 370 of the present embodiment can use machine learning as the learning method, for example. As machine learning methods, deep learning, autoencoders, SVM (support vector machine), etc. can be cited.

[0184] The abnormality determination unit 380 determines the range of the operation data used in the abnormality determination with reference to the information and parameters indicating the period for performing the abnormality determination, and inputs the determined operation data as determination data to the learning unit 370.

[0185] Further, the abnormality determination unit 380 determines whether there is an abnormality in the excavator 100 based on the index value output from the learning unit 370. Specifically, for example, when the index value output from the learning unit 370 is greater than a specified threshold value, the abnormality determination unit 380 can determine that an abnormality exists.

[0186] In the present embodiment, regarding the determination by the abnormality determination unit 380, for example, it can be executed in the warm-up mode, the automatic regeneration mode, or the manual regeneration mode, and further, it can be executed when the supercharger of the engine 11 is in the cooling operation (in the turbine cooling mode), etc.

[0187] In addition, in the present embodiment, the range of the operation data used in the learning data can represent the range on the time axis, and the range used in the determination data can represent the quantity of the operation data.

[0188] Specifically, the learning data can be a set of operation data acquired during the period determined by the acquisition range determination unit 365. Also, the determination data can be a set of operation data including a specified quantity of operation data acquired before the date of performing the abnormality determination.

[0189] Next, Figure 9 the parameters stored in the parameter storage unit 362 will be described. Figure 9 is a diagram showing an example of the parameters stored in the parameter storage unit.

[0190] The parameter storage unit 362 of the present embodiment can store, for example, by associating the name and value of the parameter.

[0191] In Figure 9In the example of , the parameter storage unit 362 stores ○○ days as the period during which an abnormal situation occurs, ×× days as the period of standby after maintenance, and ○× days as the buffer period.

[0192] The period during which an abnormal situation occurs is a parameter set considering the possibility of an abnormal situation occurring in the operation of the excavator 100 before the day when the abnormal situation occurs. Specifically, for example, how many days it takes from when an abnormal situation occurs in the excavator 100 until the operator notices the abnormal situation and notifies a company performing maintenance of the occurrence of the abnormal situation. At this time, before the day when the abnormal situation occurs, an abnormal situation has already occurred in the operation of the excavator 100, and the operation data during this period is not suitable as learning data.

[0193] The period of standby after maintenance is a parameter set considering the possibility that the operation of the excavator 100 just after maintenance is different from the normal operation, or the possibility that the actual maintenance completion date is inconsistent with the maintenance completion date input to the system. Specifically, for example, sometimes after inputting the maintenance completion date as a situation where temporary maintenance has been completed, it is found that there are parts that need additional maintenance. At this time, the operation of the excavator 100 just after inputting the maintenance completion date may be different from the normal operation, and the operation data during this period is not suitable as learning data.

[0194] The buffer period is a parameter set considering separating the determination data from the learning data. For example, when the period for obtaining the operation data used as determination data is close to the period for obtaining the operation data used in the learning data, the learning data and the determination data may become similar data.

[0195] At this time, since the abnormality determination model 375 is generated based on learning data similar to the determination data, even if it is assumed that there is an abnormality in the operation of the excavator 100, it may be set to a normal state, sometimes hindering the improvement of the accuracy of detecting abnormalities.

[0196] In the parameter storage unit 362 of the present embodiment, parameters set considering the above situations are stored.

[0197] Next, with reference to Figure 10 , the processing of the management device �00 of the present embodiment will be described. Figure 10 is a flowchart for explaining the processing of the management device.

[0198] The management device 300 of the present embodiment determines whether it has received the input of the execution date and the airframe identification information through the input receiving unit 361 of the data range determination unit 360 (step S1001). In addition, in the present embodiment, only the airframe identification number of the excavator 100 that is the object of the abnormality determination process can be input.

[0199] In step S1001, when the input of the execution date and the airframe identification information is not received, the management device 300 stands by until the input of the execution date is received.

[0200] In step S100, when the input of the execution date and the airframe identification information is received, the data range determination unit 360 retrieves the vehicle information storage unit 340 using the airframe identification number (step S1002).

[0201] Next, the data range determination unit 360, through the reference mark assignment unit 363, referring to the vehicle information obtained as the retrieval result in step S1002, determines the first exclusion range of the operation data that cannot be learning data in the vehicle information (step S1003). In addition, the reference mark assignment unit 363 assigns a mark based on the first exclusion range to the operation data.

[0202] Next, the data range determination unit 360, through the exclusion range determination unit 364, determines the second exclusion range based on the bad condition occurrence period, the post - repair standby period, and the first exclusion range stored in the parameter storage unit 362 (step S1004).

[0203] Next, the data range determination unit 360, through the adoption range determination unit 365, determines the range of the operation data to be adopted as learning data based on the number of operation data set as the determination data and the buffer period stored in the parameter storage unit 362 (step S1005).

[0204] Next, the data range determination unit 360, through the learning data output unit 366, extracts the operation data within the range determined in step S1005 as learning data and outputs the learning data to the learning unit 370 (step S1006).

[0205] Next, the learning unit 370 of the management device 300 generates an abnormality determination model 375 based on the learning data (step S1007).

[0206] Next, the management device 300, through the abnormality determination unit 380, extracts the operation data that becomes the determination data from the vehicle information storage unit 340 based on the execution date and inputs it to the abnormality determination model 375 (learning unit 370) (step S1008).

[0207] Next, the abnormality determination unit 380 outputs the output of the abnormality determination model 375 together with the information indicating the range of the operation data used as learning data (step S1009), and ends the process.

[0208] Hereinafter, with reference to Figure 11A 、 Figure 11B ,the processing of the data range determination unit 360 of the present embodiment will be further described.Figure 11A It is the first diagram illustrating the processing of the data range determination unit, and is a diagram for explaining the first exclusion range and the second exclusion range. Figure 11B It is the second diagram illustrating the processing of the data range determination unit, and is a diagram for explaining the range of the learning data.

[0209] In Figure 11A 、 Figure 11B 's example, it shows the case where, as a result of retrieving vehicle information stored in the vehicle information storage unit 340 using the airframe identification number received by the input receiving unit 361, vehicle information including operation data for the period from time ts to time te is extracted. The time period can represent a date. Therefore, in Figure 11A 、 Figure 11B 's example, it shows that operation data for the period from date ts to date te is extracted.

[0210] In the data range determination unit 360 of the present embodiment, when the delivery date t0 is included in the extracted vehicle information, the reference mark assigning unit 363 assigns a mark indicating that the period k1 from time ts to the delivery date t0 is a period during which operation data is not used as learning data.

[0211] The operation data acquired during the period k1 is information indicating the operation of the excavator 100 before delivery to the delivery location. In other words, the operation data acquired during the period k1 is information indicating an operation performed in an environment different from the delivery location. Therefore, in the present embodiment, the operation data acquired during the period k1 is not used in the learning data for anomaly determination after being moved into the delivery location.

[0212] Here, the reference mark assigning unit 363 assigns the mark "0" to the period k1, which is the period before the delivery date and indicates that operation data is not used as learning data.

[0213] Moreover, in the present embodiment, when the value of the item "type of work" included in the vehicle information is "repair work", such operation data is not used as learning data. Therefore, the reference mark assigning unit 363 also assigns the mark "0", which indicates a period during which operation data is not used as learning data, to the period when the type of work is "repair work".

[0214] Furthermore, when the occurrence date t1 of a defect and the repair completion date t2 are included in the extracted vehicle information, the reference mark assigning unit 363 assigns the mark "2" to the period k2 from the occurrence date t1 of the defect to the repair completion date t2, which is the repair period of the defect and indicates a period during which operation data is not used as learning data.

[0215] In the present embodiment, the period during which the reference mark is given by the reference mark giving unit 363 is set as the first exclusion range. Therefore, the period k1 before the delivery of the excavator 100 and the period k2 as the repair period of the defective condition of the excavator 100 are the first exclusion ranges in which the operation data obtained during this period cannot be learning data.

[0216] Moreover, the reference mark giving unit 363 of the present embodiment gives the mark "1", and this mark indicates that the period other than the periods in which the marks "0" and "2" are given is the period in which the operation data can be learning data.

[0217] Next, the data range determination unit 360 determines the second exclusion range through the exclusion range determination unit 364, referring to the defective condition occurrence period and the standby period after repair stored in the parameter storage unit 362.

[0218] Specifically, the exclusion range determination unit 364 determines the date t3 by tracing back the number of days set as the defective condition occurrence period from the defective condition occurrence date t1, sets the period k3 (first period) from the date t3 to the defective condition occurrence date t1 as the second exclusion range, and gives the mark "3" indicating the second exclusion range.

[0219] Moreover, the exclusion range determination unit 364 determines the date t4 after the number of days set as the standby period after repair has passed from the repair completion date t2, sets the period k4 (second period) from the repair completion date t2 to the date t4 as the second exclusion range, and gives the mark "3" indicating the second exclusion range.

[0220] In this way, in the present embodiment, the second exclusion range is set before the defective condition occurrence date included in the vehicle information and after the repair completion date.

[0221] Therefore, according to the present embodiment, even when the date when the defective condition actually occurs is inconsistent with the date when the operator of the excavator 100 reports the occurrence of the defective condition, etc., it is possible to suppress inappropriate operation data as learning data from being adopted as learning data.

[0222] Moreover, according to the present embodiment, even when the date when the operation of the excavator 100 actually proceeds in the same manner as before the occurrence of the defective condition is inconsistent with the date when the repair of the excavator 100 is completed, it is possible to suppress inappropriate operation data as learning data from being adopted as learning data.

[0223] In Figure 11A 's example, the operation data obtained during the period from the date t0 to the date t3 and the operation data obtained during the period from the date t4 to the date te become data that can be learning data. That is, in the present embodiment, the operation data obtained during the period given the mark "1" can be learning data.

[0224] Next, the data range determination unit 360 determines the period during which the operation data is used as learning data during the period when the mark "1" is given, by using the range determination unit 365.

[0225] First, the range determination unit 365 refers to the parameter storage unit 362 to determine the date t6 of the period k5 that is set to be traced back to the buffer period from the execution date t5 of the abnormality determination process received by the input reception unit 361.

[0226] In addition, in the present embodiment, a specified amount of data including the operation data acquired on the execution date t5 is acquired as determination data input to the abnormality determination model 375. In other words, the determination data is a set of a specified amount of operation data acquired near the execution date t5.

[0227] Therefore, in the present embodiment, the period k5 is preferably longer than the period for acquiring the determination data.

[0228] In addition, the buffer period can be set according to the type of abnormality. For example, when determining the presence or absence of an abnormality manifested as a sudden change in the operation data, the buffer period can be the same as the period for acquiring the determination data. And, for example, when determining the presence or absence of an abnormality manifested as a slow change in the operation data, the buffer period is preferably longer than the period for acquiring the determination data.

[0229] Next, the range determination unit 365 determines the period before the date t6, that is, the period during which the mark "1" is given and the acquired operation data becomes equal to or more than the specified amount. And the range determination unit 365 sets the operation data acquired during the determined period as learning data. Here, the specified amount represents an amount sufficient as learning data. The number of operation data used as this learning data can be set in advance.

[0230] In Figure 11B 's example, as the period before the date t6, that is, the period during which the mark "1" is given, there is the period k6 from the date t4 to the date t6. However, in Figure 11B 's example, the number of operation data acquired during the period k6 is less than the specified amount, and the data amount is not sufficient as learning data.

[0231] Thus, the range determination unit 365 determines the period before the date t6, that is, the period before the date t3 that becomes the period during which the mark "1" is given. At this time, the range determination unit 365 can determine the period before the date t3, that is, the period during which the total of the number of operation data acquired during the period k6 becomes equal to or more than the specified amount. In Figure 11BIn the example, as the period during which the marked "1" is given and the acquired operation data becomes a specified number or more, in addition to the period k6, the period k7 from the date t7 to the date t3 is also determined.

[0232] Therefore, the data range determination unit 360 outputs the operation data groups acquired during the period k6 and the period k7 respectively as learning data to the learning unit 370. The learning unit 370 generates an abnormality determination model 375 based on this learning data.

[0233] In this way, according to the present embodiment, it is possible to determine the range of appropriate learning data in the operation data collected from the excavator 100. In other words, according to the present embodiment, the learning unit 370 can use appropriate learning data (operation data) to learn the relationship between the operation data and the presence or absence of abnormalities in the excavator.

[0234] Next, an output example of the result of the abnormality determination process based on the abnormality determination unit 380 will be described. The abnormality determination unit 380 outputs a group of operation data of a specified number including the operation data acquired on the execution date t5 as determination data to the abnormality determination model 375, and outputs a result of determining the presence or absence of an abnormality based on this output.

[0235] Figure 12A is the first figure showing an output example of the result of the abnormality determination, Figure 12B is the first figure showing an output example of the result of the abnormality determination. Figure 12A 、 Figure 12B The screens 121A and 121B shown in each of them are displayed on the display device 40 included in the management device 300, for example.

[0236] The screen 121A includes display areas 122, 123, 124, 125, and 126. In the display area 122, a message indicating the result of the abnormality determination process and information indicating the execution date of the abnormality determination process are displayed.

[0237] In the display area 123, for example, a moving image captured by the imaging device 80 of the excavator 100 on the execution date is displayed.

[0238] In the display areas 124 and 125, for example, the determination result of the presence or absence of an abnormality based on the abnormality determination unit 380 is displayed. Regarding the determination result of the presence or absence of an abnormality, for example, it can be determined based on an index value output from the abnormality determination model 375 or the like. For example, in the present embodiment, when the index value output from the abnormality determination model 375 is greater than a specified threshold value or the like, it can be determined that an abnormality exists. In the following description, the index value output from the abnormality determination model 375 may sometimes be referred to as the degree of abnormality.

[0239] The display area 124 can display, for example, a graph showing changes in the abnormality degree, and the display area 125 can display, for example, a message indicating the abnormality degree. The display area 126 displays operation buttons and the like for reproducing a moving image or the like of a part where the abnormality degree exceeds a specified threshold value in the display area 123.

[0240] Moreover, in the display area 124, in the graph showing changes in the abnormality degree, a mark 124a indicating a learning period in which the operation data is adopted as learning data and a mark 124b indicating an evaluation period in which the operation data is adopted as determination-use data can be displayed. Further, in the display area 124, the marks 124a and 124b have different display modes respectively.

[0241] In the screen 121A, it can be seen that the change in the abnormality degree during the learning period indicated by the mark 124a is smaller than the change in the abnormality degree during the evaluation period indicated by the mark 124b, and the operation data used during learning is normal operation data.

[0242] Figure 12B The screen 124B shown includes display areas 127 and 128. The display area 127 displays information indicating a period during which the operation data adopted in acquiring the learning data is obtained. The display area 128 displays information indicating a period excluded from the period during which the learning data is acquired. In other words, in the display area 128, information indicating a period during which the acquired operation data cannot become learning data is displayed.

[0243] Thus, in the present embodiment, as the learning data used in generating the abnormality determination model 375, the period during which the operation data is used is displayed. Therefore, according to the present embodiment, it is possible to visualize the period during which the learning data is acquired, and it is possible to enable an operator who performs maintenance or the like of the excavator 100 to deepen the understanding of the determination result of the presence or absence of an abnormality.

[0244] Moreover, in the present embodiment, the presence or absence of an abnormality is determined using the abnormality determination model 375 based on learning data within an appropriate range, so the reliability of the determination result can be improved.

[0245] Moreover, in the present embodiment, the operation data collected from the excavator 100 that is the determination object of the presence or absence of an abnormality is used as the learning data, but it is not limited thereto. Regarding the learning data, for example, it can be adopted from the operation data collected from other excavators other than the excavator 100 that is the determination object of the presence or absence of an abnormality. At this time, the other excavator can be of the same equipment type as the excavator 100 that is the determination object of the presence or absence of an abnormality.

[0246] Further, in the present embodiment, the construction machine that acquires operation data as learning data is set as the excavator 100, but it is not limited thereto. The present embodiment can also be used for abnormality determination of construction machines other than the excavator 100.

[0247] Further, in the present embodiment, it is set that the abnormality determination model 375 outputs an index value indicating the presence or absence of an abnormality, and the abnormality determination unit 380 determines the presence or absence of an abnormality based on the index value, but it is not limited thereto.

[0248] In the present embodiment, the abnormality determination model 375 may also perform determination of the presence or absence of an abnormality.

[0249] Further, in the present embodiment, the index value output by the abnormality determination model 375 can be output as a determination result of the presence or absence of an abnormality. In this case, the determination process of the abnormality determination unit 380 based on the index value is not required.

[0250] (Another Embodiment)

[0251] Hereinafter, with reference to the accompanying drawings, another embodiment will be described. In the present embodiment, the data range determination unit 360 compares the distributions of the operation data that is set as learning data and that is acquired before the operation data acquired in the first exclusion range and the second exclusion range, and the operation data acquired after the operation data acquired in the first exclusion range and the second exclusion range.

[0252] In the following description, the operation data acquired before the operation data acquired in the first exclusion range and the second exclusion range may sometimes be referred to as first operation data, and the operation data acquired after the operation data acquired in the first exclusion range and the second exclusion range may sometimes be referred to as second operation data.

[0253] In other words, in the present embodiment, the operation data acquired during period k7 may sometimes be referred to as first operation data, and the operation data acquired during period k6 may sometimes be referred to as second operation data.

[0254] In the present embodiment, as a result of comparing the distributions of the first operation data and the second operation data, when the respective distributions are different, the first operation data is excluded from the learning data.

[0255] Next, with reference to Figure 13 , the operation of the management device 300 of the present embodiment will be described. Figure 13 It is a flowchart for explaining the processing of the management device in another embodiment.

[0256] Figure 13 The processing from step S1301 to step S1305 of Figure 10The processing from step S1001 to step S1005 is the same, so the description thereof is omitted.

[0257] In Figure 13 In step S1305, if the range of the operation data to be used as learning data is determined, the range determination unit 365 compares the distribution of the first operation data and the distribution of the second operation data in the operation data to be used as learning data (step S1306). In addition, the distribution of the operation data can be represented by, for example, a feature amount of the operation data.

[0258] Next, the range determination unit 365 determines whether the distributions of the two are similar (step S1307). Specifically, the range determination unit 365 obtains the similarity degree or the like of the two, and when the similarity degree is equal to or higher than a specified threshold value, it can be determined that the distribution of the first operation data is similar to the distribution of the second operation data.

[0259] In step S1307, when it is determined that they are similar, the management device 300 proceeds to step S1309 described later.

[0260] In step S1307, when it is determined that they are not similar, the range determination unit 365 excludes the first operation data from the learning data, sets the second operation data as the learning data (step S1308), and proceeds to step S1309 described later. That is, in step S1307, when it is determined that they are not similar, the range determination unit 365 sets the operation data obtained after the operation data obtained in the first exclusion range and the second exclusion range as the learning data.

[0261] Figure 13 The processing from step S1309 to step S1312 of Figure 10 is the same as the processing from step S1006 to step S1009 of

[0262] Therefore, the description thereof is omitted. Figure 11B Next, with reference to Figure 13 the processing of

[0263] will be specifically described. In the management device 300 of the present embodiment, the range of the operation data to be used as learning data is set to the operation data obtained during the period k6 and the period k7.

[0264] Thus, in the present embodiment, the operation data in the state set as normal before maintenance is compared with the operation data in the state set as normal after maintenance, and only in similar cases are both used as learning data.

[0265] Therefore, according to the present embodiment, even when the operation data in the state set as normal due to a change in the operation environment caused by maintenance (repair) changes, it will not be affected by the change. Therefore, according to the present embodiment, even after the completion of maintenance (after the implementation of maintenance), it is possible to appropriately detect the signs of a failure.

[0266] As described above, the preferred embodiments of the present invention have been described in detail, but the present invention is not limited to the above embodiments, and various modifications and substitutions can be made to the above embodiments without departing from the scope of the present invention.

[0267] Furthermore, this international application claims the priority based on Japanese Patent Application No. 2020-057513 filed on March 27, 2020, and incorporates the entire contents of Japanese Patent Application No. 2020-057513 into this international application.

[0268] Reference Signs

[0269] 30 - Controller, 31 - Operation Valve, 40 - Display Device, 42 - Input Device, 80 - Imaging Device, 80B, 80F, 80L, 80R - Cameras, 100 - Excavator, 300 - Management Device, 301 - Information Sending Unit, 302 - Operation Mode Acquisition Unit, 303 - Equipment Guidance Unit, 310 - Main Information Storage Unit, 320 - Defect Information Storage Unit, 330 - Action Information Storage Unit, 340 - Vehicle Information Storage Unit, 350 - Information Collection Unit, 360 - Data Range Determination Unit, 370 - Learning Unit, 380 - Abnormality Determination Unit.

Claims

1. A management device for an excavator, wherein, it has: a learning unit that learns the relationship between the operation data of the excavator and the degree of abnormality of the excavator using a data set as learning data. The data set sets the operation data corresponding to the period for determining the presence or absence of an abnormality of the excavator among the operation data indicating the actions of the excavator as input, and sets the information indicating no abnormality as output; and a range determination unit that determines the range of the operation data to be excluded from the learning data based on vehicle information including the operation data indicating the actions of the excavator and the defect information related to the defective conditions of the excavator. The operation data excluded from the learning data is the operation data acquired during a first period from the occurrence time of the defective condition indicated by the defect information to the time up to a specified period of retrospective, and the operation data acquired during a second period from the completion time of the repair indicated by the defect information to the time after a specified period has elapsed. Compare the first operation data acquired before the first period and the second operation data acquired after the second period among the operation data included in the learning data. Based on the result of the comparison, determine whether to exclude the first operation data from the learning data.

2. The management device for an excavator according to claim 1, wherein, The range determination unit determines the range of the learning data based on the operation data excluding the range, with the time for performing the process of determining the presence or absence of the abnormality as a reference.

3. The management device for an excavator according to claim 1 or 2, wherein, When performing the process of determining the presence or absence of the abnormality, the period of the operation data input to the learning unit and the period of the operation data set as the learning data are non - continuous.

4. The management device for an excavator according to claim 1, wherein, When it is determined that the first operation data is similar to the second operation data, the first operation data and the second operation data are set as learning data.

5. The management device for an excavator according to claim 1 or 2, wherein, The defect information includes information indicating the date of the start of maintenance of the excavator and information indicating the date of the completion of maintenance of the excavator.

6. The management device for an excavator according to claim 1 or 2, wherein, The display device is caused to display the determination result of the presence or absence of the abnormality and the information indicating the period of the operation data set as the learning data.

7. The management device for an excavator according to claim 1 or 2, which has: an abnormality determination unit that determines the presence or absence of an abnormality based on the degree of abnormality of the excavator. The learning unit receives the input of the operation data of the excavator extracted based on the execution date of the process of determining the presence or absence of the abnormality, and outputs the degree of abnormality of the excavator to the abnormality determination unit.

8. The management device for an excavator according to claim 1 or 2, wherein, The management device of the excavator receives, from the support device of the excavator, notification of the start date of the maintenance of the excavator input in the support device and the completion date of the maintenance of the excavator.

9. The management device of the excavator according to claim 8, wherein, The management device of the excavator is configured to: When a start button for inputting the start date of the maintenance of the excavator is operated in the support device, receive notification of the start date from the support device; When a completion button for inputting the completion date of the maintenance of the excavator is operated in the support device, receive notification of the completion date from the support device.

10. An excavator that communicates with a management device, wherein, The excavator has a communication device that transmits operation data indicating the operation of the excavator and information indicating the start date and completion date of the maintenance of the excavator to the management device having a learning unit and a range determination unit. The learning unit uses a data set as learning data to learn the relationship between the operation data of the excavator and the abnormality degree of the excavator. The data set sets the operation data corresponding to the period for determining the presence or absence of an abnormality of the excavator as input and sets information indicating no abnormality as output. The range determination unit determines the range of the operation data to be excluded from the learning data based on vehicle information including the operation data indicating the operation of the excavator and defect information related to the defective condition of the excavator. The operation data excluded from the learning data is the operation data acquired during a first period from the occurrence time of the defective condition indicated by the defect information to the time up to a specified period of retrospective and the operation data acquired during a second period from the completion time of the repair indicated by the defect information to the time after a specified period has elapsed. Compare the first operation data acquired before the first period and the second operation data acquired after the second period among the operation data included in the learning data. Based on the result of the comparison, determine whether to exclude the first operation data from the learning data.

11. The excavator according to claim 10, having: A start switch for inputting the start date of the maintenance of the excavator; and A completion switch for inputting the completion date of the maintenance of the excavator.

12. The excavator according to claim 11, having: A controller that controls the operation of the excavator, The controller associates and stores, in a storage device, information indicating a record when a person approaching within a specified range is detected by an operation of the start switch and information indicating an approach for maintenance.

Images