State determination system and state determination method

Abstract

To provide a state determination system and the like capable of appropriately determining a state of a crane winch.SOLUTION: The determination device includes an acquisition unit that acquires sound or vibration generated during operation of a crane winch, and a determination unit that determines a state of the crane winch based on the sound or vibration acquired by the acquisition unit using a learned model. The determination unit removes an unnecessary component from the sound or vibration acquired by the acquisition unit, and determines the state of the crane winch based on the sound or vibration from which the component has been removed.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates to a condition determination system and a condition determination method. [Background technology]

[0002] Patent Document 1 discloses a winding device for a crane that stops the winding operation of a crane wire by a crane winch in response to an overwinding detection signal output from a load sensor. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2013-256356 Summary of the Invention [Problem to be solved by the invention]

[0004] However, the technology disclosed in Patent Document 1 cannot properly determine the condition of the crane winch other than over-winding of the crane wire, including, for example, when an abnormality occurs in the crane winch itself.

[0005] Therefore, in one aspect, an object of the present invention is to provide a state determination system and the like that can appropriately determine the state of a crane winch. [Means for solving the problem]

[0006] In one embodiment, an acquisition unit that acquires sound or vibration generated during operation of the crane winch; a determination unit that determines a state of the crane winch based on the sound or vibration acquired by the acquisition unit using a trained model; A state determination system is provided, comprising: [Effects of the Invention]

[0007] According to one aspect of the present invention, the state of a crane winch can be appropriately determined. [Brief explanation of the drawings]

[0008] [Figure 1] 1 is a diagram illustrating a configuration of a state determination system according to an embodiment of the present invention. [Figure 2] FIG. 10 is a diagram showing a state in which a microphone and a camera are installed. [Figure 3] 3 is a flowchart showing a process in the state determination system of the present embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0009] FIG. 1 is a diagram showing the configuration of a state determination system according to this embodiment, and FIG. 2 is a diagram showing the state in which a microphone and a camera are installed.

[0010] As shown in Figure 1, the condition determination system 10 of this embodiment includes an acquisition unit 11 that acquires sounds or vibrations generated when the crane winch 50 (Figure 2) is operating, a determination unit 12 that uses a trained model to determine the condition of the crane winch 50 based on the sounds or vibrations acquired by the acquisition unit 11, and a notification unit 13 that notifies of the occurrence of an abnormality in the crane winch 50. In this embodiment, the determination unit 12 removes unnecessary components from the sound or vibration acquired by the acquisition unit 11, and determines the state of the crane winch 50 based on the sound or vibration from which the components have been removed.

[0011] Furthermore, the acquisition unit 11 can acquire images of the crane winch 50 during operation. In this case, the determination unit 12 can determine the state of the crane winch 50 based on the images acquired by the acquisition unit 11 as well as the sounds or vibrations acquired by the acquisition unit 11.

[0012] Furthermore, the acquisition unit 11 can acquire operation information indicating the operation state of the crane winch 50 from the control device 30 of the crane winch 50. In this case, the determination unit 12 can determine the state of the crane winch 50 based on the operation information acquired by the acquisition unit 11 as well as the sound or vibration acquired by the acquisition unit 11.

[0013] In the example of Figure 1, a terminal device 40 used by on-site staff or the like can be connected to the condition determination system 10, and the information handled by the condition determination system 10 can be accessed via the terminal device 40.

[0014] The condition determination system 10 can be configured using one or more computers (including a server and a terminal device). For example, some of the functions of the condition determination system 10 can be provided in the terminal device 40.

[0015] 2, a microphone 21 is installed near the crane winch 50 to capture the operating sound of the crane winch 50. The crane winch 50 is photographed by a camera 22 while in operation. A plurality of microphones 21 or a plurality of cameras 22 may be provided.

[0016] Next, the operation of the state determination system 10 will be described.

[0017] FIG. 3 is a flowchart showing the processing in the state determination system of this embodiment.

[0018] In step S102 of FIG. 3, the acquisition unit 11 acquires (updates) the operation sound (sound during operation) captured by the microphone 21.

[0019] In step S104, the acquisition unit 11 acquires (updates) an image captured by the camera 22.

[0020] In step S105, the determination unit 12 removes noise from the operation sound acquired in step S102. Here, for example, it is possible to filter out sounds in a frequency band that has a low correlation with the state of the crane winch 50. In this case, for example, it is possible to remove sounds in a frequency band different from sounds generated during normal or abnormal conditions as noise.

[0021] Furthermore, unnecessary sound components can be removed as noise based on whether or not the operation sound is periodic. Any method for removing noise can be used. For example, by utilizing the fact that the operating sound of the crane winch 50 is periodic, non-periodic sounds can be removed by regarding them as noise at the work site related to other work that is unrelated to the crane winch 50. Alternatively, if a sudden non-periodic sound is likely to indicate an abnormality in the crane winch 50, it is possible to extract the sudden sound in a specific frequency band and remove the other specific sound as noise unrelated to the abnormality.

[0022] In order to be able to determine a wide range of conditions of the crane winch 50, a plurality of noise removal methods may be combined. In this case, the determination unit 12 can make inferences for a plurality of sounds after noise removal by each method. This makes it possible, for example, to distinguish between and accurately detect a plurality of types of abnormality.

[0023] In step S106, the determination unit 12 infers and determines the state of the crane winch 50 using a trained model based on the sound after noise has been removed (step S105). Here, the determination unit 12 can determine whether the state of the crane winch 50 is abnormal or not using a trained model prepared in advance. Furthermore, if the state of the crane winch 50 is abnormal, the type (attribute) of the abnormality can be determined.

[0024] Furthermore, the determination unit 12 can make inferences based not only on the sound at that time but also on changes in the sound repeatedly acquired in step S102. In this case, for example, it is possible to accurately detect the occurrence of an abnormality that brings about a change in the frequency band or volume of the sound generated from the crane winch 50.

[0025] In step S108, it is determined whether the result of the determination in step S106 indicates an abnormality in the crane winch 50. If the determination is affirmative, the process proceeds to step S110, and if the determination is negative, the process proceeds to step S102.

[0026] In step S110, the notification unit 13 issues an alert to notify that an abnormality has occurred in the crane winch 50, and notifies the terminal device 40 of the occurrence of the abnormality, and then the process proceeds to step S102.

[0027] 3, the determination unit 12 makes a determination based on the sound captured by the microphone 21, but the determination unit 12 may also make a determination based on the vibration of the crane winch 50 in addition to or instead of the sound captured by the microphone 21. In this case, an acceleration sensor attached to the crane winch 50 or attached near the crane winch 50, for example, can be used as a sensor for detecting the vibration.

[0028] In this case, in step S105, the determination unit 12 removes noise from the vibration acquired in step S102. Here, for example, it is possible to filter out vibrations in a band that has a low correlation with the state of the crane winch 50. Furthermore, it is possible to remove unnecessary vibration components as noise based on whether or not the vibration is periodic. Furthermore, in order to be able to determine a wide range of the state of the crane winch 50, it is possible to combine multiple methods for removing noise. Furthermore, the determination unit 12 can make inferences based not only on the vibration at that time, but also on changes in the sound repeatedly acquired in step S102. In this case, it is possible to accurately detect the occurrence of an abnormality that causes a change in the frequency band or magnitude of the vibration generated by the crane winch 50, for example.

[0029] Furthermore, in step S106, the determination unit 12 may make a determination by taking into account the image acquired in step S104. For example, the determination unit 12 can use a trained model prepared in advance to infer the state of the crane winch 50 based on the operation sound and the image.

[0030] Furthermore, operation information indicating the operating state of the crane winch 50 can be acquired from the control device 30 by the acquisition unit 11, and the acquired operation information can be used during inference. For example, by acquiring the rotation speed of the crane winch 50, it is possible to grasp the period of change in operation sound associated with rotation, the frequency band of operation sound under normal conditions, etc. This can improve the accuracy of inference by the determination unit 12. Furthermore, as will be described later, the operation information from the control device 30 may be reflected in the trained model.

[0031] Next, we will explain how to create a trained model.

[0032] The trained model can be created, for example, by machine learning the normal operating sounds of the crane winch 50. In this case, by learning the operating sounds under various conditions, the accuracy of inference can be improved.

[0033] Furthermore, a trained model corresponding to the sound generated when an abnormality occurs can be created by machine learning the operating sound of the crane winch 50 when an abnormality occurs. In this case, by creating multiple trained models that differ from each other according to the type (attribute) of the abnormality, it becomes possible to accurately determine each type of abnormality.

[0034] During machine learning, images of the crane winch 50, or data obtained by combining the image data after processing the images with the operating sound (data during normal or abnormal conditions), can also be used as the learning subject. In this case, by performing inference based on the sound captured by the microphone 21 and the image captured by the camera 22 during judgment in the judgment unit 12, it becomes possible to judge the state of the crane winch 50 with higher accuracy.

[0035] Furthermore, a trained model can be created by machine learning data (normal or abnormal data) that combines operation information obtained from the control device 30 with operation sounds. In this case, by performing inference based on the sound captured by the microphone 21 when making a judgment in the judgment unit 12 and the operation information obtained from the control device 30, it becomes possible to judge the state of the crane winch 50 with higher accuracy.

[0036] Abnormal sounds from the crane winch 50 are one indicator of wear or damage to internal components of the crane winch 50, such as gears, bearings, and brakes. Conventionally, abnormalities have been identified and addressed through, for example, reports from the crane operator or inspections. Therefore, there is a risk that abnormalities may be overlooked if the operator does not notice the abnormal sound or if an inspection is not performed. However, according to this embodiment, for example, if the artificial intelligence detects an abnormal sound from among sounds captured by a microphone 21 installed near the crane winch 50, an alert can be issued and on-site personnel can be notified. Since a notification is issued as soon as an abnormal sound from the crane winch 50 is detected, on-site response can be promptly performed. Furthermore, detection of abnormal sounds can also predict failure of the crane winch 50, preventing shortages and improving safety.

[0037] As described above, according to this embodiment, the trained model is used to determine the state of the crane winch based on the sound or vibration acquired by the acquisition unit 11. Therefore, it becomes possible to appropriately determine the state of the crane winch, including when an abnormality occurs in the crane winch itself.

[0038] Although the embodiments have been described in detail above, the present invention is not limited to the specific embodiments, and various modifications and changes are possible within the scope of the claims. In addition, it is also possible to combine all or a plurality of components of the above-described embodiments. [Explanation of symbols]

[0039] 10. Status Judgment System 11 Acquisition Department 12 Judgment section 13 Notification Department

Claims

1. an acquisition unit that acquires sound or vibration generated during operation of the crane winch; a determination unit that determines a state of the crane winch based on the sound or vibration acquired by the acquisition unit using a trained model; A state determination system comprising:

2. The state determination system according to claim 1 , wherein the determination unit removes unnecessary components from the sound or vibration acquired by the acquisition unit, and determines the state of the crane winch based on the sound or vibration from which the components have been removed.

3. The acquisition unit further acquires an image of the crane winch during operation, The state determination system according to claim 1 , wherein the determination unit determines the state of the crane winch based on the image acquired by the acquisition unit as well as the sound or vibration acquired by the acquisition unit.

4. The acquisition unit further acquires operation information indicating an operation state of the crane winch, The state determination system according to claim 1 , wherein the determination unit determines the state of the crane winch based on the sound or vibration acquired by the acquisition unit and the operation information acquired by the acquisition unit.

5. an acquisition step of acquiring sound or vibration generated during operation of the crane winch; a determination step of determining a state of the crane winch based on the sound or vibration acquired by the acquisition step using a trained model; A state determination method comprising:

Images