Trouble prediction method, system and program of conveyor belt

Abstract

A trouble prediction method of a conveyor belt is provided. In an acquisition step, measurement data related to powder (PW) that is being conveyed on the conveyor belt (41) is acquired via an optical sensor (3) installed apart from the conveyor belt (41). In the prediction step, a trouble that may occur in the conveyor belt (41) is predicted based on shape parameters related to a shape of the powder (PW) and reference information, the shape parameters being calculated from the measurement data. The reference information is information on a relationship between the shape parameters and various types of troubles.

Description

TROUBLE PREDICTION METHOD, SYSTEM AND PROGRAM OF CONVEYOR BELT

[0001] The present invention relates to a trouble prediction method, a system, and a program of a conveyor belt.

[0002] Patent document 1 discloses an apparatus of detecting meanderings of a conveyor belt.

[0003] JP 2000-118663A

[0004] However, in the known art such as Patent Literature 1, meanderings that are an example of troubles that have occurred in a conveyor belt can be detected, but other troubles are not supposed to be addressed. Furthermore, it is also impossible to predict the possibility of a trouble before it occurs. There is a growing need to understand the possibility of various troubles before they occur.

[0005] In view of the above circumstances, the present invention provides a trouble prediction method of a conveyor belt, or the like than can predict in advance various troubles that may occur in the conveyor belt.

[0006] According to an aspect of the present invention, a trouble prediction method of a conveyor belt is provided. This trouble prediction method includes the following steps. In the acquisition step, measurement data related to powder that is being conveyed on the conveyor belt is acquired via an optical sensor installed apart from the conveyor belt. In the prediction step, a trouble that may occur in the conveyor belt is predicted based on shape parameters related to a shape of the powder and reference information, the shape parameters being calculated from the measurement data. The reference information is information on a relationship between the shape parameters and various types of troubles.

[0007] In addition, it may be provided in each of the following forms.

[0008] (1)A trouble prediction method of a conveyor belt, comprising: an acquisition step of acquiring measurement data related to powder that is being conveyed on the conveyor belt via an optical sensor installed apart from the conveyor belt; and a prediction step of predicting a trouble that may occur in the conveyor belt based on shape parameters related to a shape of the powder and reference information, the shape parameters being calculated from the measurement data, the reference information being information on a relationship between the shape parameters and various types of the trouble.

[0009] According to such an aspect, various troubles that may occur in the conveyor belt can be predicted in advance.

[0010] (2) The trouble prediction method according to (1), wherein the shape parameters include a first coordinate that is a coordinate in a width direction of a belt of the conveyor belt in a region (including a contour and an interior) that represents the shape of the powder in the measurement data.

[0011] According to such an aspect, various troubles that may occur in the conveyor belt caused by the position of the powder can be predicted in advance.

[0012] (3) The trouble prediction method according to (2), wherein the first coordinate is a coordinate of a point where a second coordinate that is a vertical coordinate of the belt is maximum or maximal in the region.

[0013] According to such an aspect, various troubles that may occur in the conveyor belt caused by the position of powder can be predicted in advance with even greater accuracy.

[0014] (4) The trouble prediction method according to (3), wherein the prediction step predicts the trouble based on the two or more first coordinates calculated at different times during a conveyance of the powder.

[0015] According to such an aspect, various troubles that may occur in the conveyor belt caused by the position of the powder can be predicted in advance with even greater accuracy, considering changes in the shape of the powder PW over time.

[0016] (5) The trouble prediction method according to any one of (1) to (4), wherein the reference information includes, as one of the shape parameters, at least the shape parameters in a state where the powder is not being conveyed by the conveyor belt.

[0017] According to such an aspect, it is possible to distinguish the shape of the powder PW that is not being conveyed and, more accurately predict in advance various troubles that may occur in the conveyor belt 4 caused by the position of the powder PW.

[0018] (6) The trouble prediction method according to any one of (1) to (5), wherein the shape parameters are calculated in a state where the powder is loaded, and the belt of the conveyor belt is sunk in a direction of gravity.

[0019] According to such an aspect, it is possible to focus on the shape of the powder being conveyed and more accurately predict in advance various troubles that may occur in the conveyor belt 4 caused by the position of the powder PW.

[0020] (7) The trouble prediction method according to any one of (1) to (6), wherein the measurement data includes information on the belt of the conveyor belt as well as the shape of the powder, and the prediction step predicts a fall of at least part of the powder from the belt as the trouble based further on mechanical parameters of ends defining a width of the belt, the mechanical parameters being calculated from the measurement data.

[0021] According to such an aspect, various troubles such as a fall of the powder can be predicted in advance by focusing also on the condition of the ends of the belt of the conveyor belt.

[0022] (8) The trouble prediction method according to any one of (1) to (7), wherein the reference information is a learned model that has machine-learned a relationship between the shape of the powder and the trouble in advance, and the prediction step predicts the trouble by inputting the shape parameters into the learned model.

[0023] According to such an aspect, various troubles that may occur in the conveyor belt caused by the position of the powder can be predicted in advance as an appropriate decision by analogy from actual past situations.

[0024] (9) The trouble prediction method according to any one of (1) to (8), wherein the optical sensor is a LiDAR sensor, and the measurement data is point cloud data obtained via the LiDAR sensor.

[0025] According to such an aspect, various troubles that may occur in the conveyor belt can be predicted in advance, while ensuring both accuracy and cost.

[0026] (10) The trouble prediction method according to (9), wherein the LiDAR sensor is installed above the conveyor belt so as to look down a loading surface of the conveyor belt on which the powder is loaded.

[0027] According to such an aspect, various troubles that may occur in the conveyor belt can be predicted in advance with a simple configuration.

[0028] (11) The trouble prediction method according to any one of (1) to (10), wherein the trouble is meanderings or a deviation of the belt of the conveyor belt, occurrence of a clogging or a deposit in a chute unit connecting the two or more conveyor belts, or a fall of at least part of the powder from the conveyor belt.

[0029] According to such an aspect, specific troubles that may occur in the conveyor belt can be predicted in advance.

[0030] (12) The trouble prediction method according to any one of (1) to (11), further comprising: a warning step of presenting a warning to a user, the warning including contents of the trouble predicted in the prediction step.

[0031] According to such an aspect, the user can notice the possibility of various kinds of troubles.

[0032] (13) A system for predicting a trouble that may occur in a conveyor belt comprising: at least one processor, wherein the processor is configured to execute a program so that each step of the trouble prediction method according to any one of (1) to (12) is executed.

[0033] According to such an aspect, various troubles that may occur in the conveyor belt can be predicted in advance.

[0034] (14) The system according to (13), comprising: an information processing apparatus and an optical sensor, wherein the optical sensor is configured to measure powder conveyed on the conveyor belt, the information processing apparatus includes the processor configured to execute the program, and is connected to the optical sensor and configured to acquire measurement data related to the powder from the optical sensor.

[0035] According to such an aspect, various troubles that may occur in the conveyor belt can be predicted in advance.

[0036] (15) A program, configured to allow at least one computer to execute each step of the trouble prediction method according to any one of (1) to (12).

[0037] According to such an aspect, various troubles that may occur in the conveyor belt can be predicted in advance. Of course, the present invention is not limited thereto.

[0038] Fig. 1 is a block diagram showing a hardware configuration of a system 1 according to one embodiment. Fig. 2 is a schematic diagram illustrating a positional relationship between a LiDAR sensor 3 and a conveyor belt 4. Fig. 3 is a block diagram showing the functions realized by a controller 23 or the like in an information processing apparatus 2. Fig. 4 is an activity diagram showing the flow of the process performed by the system 1. Fig. 5 is a schematic diagram indicating the positions of ends 41l and 41r in different states of a belt 41. Fig. 6 is a schematic diagram showing XY coordinates, which are an example of shape parameters of powder PW. Fig. 7 is an example of a screen on which a display controller 237 allows a display unit 24 to display a warning output by a warning unit 236. In particular, Fig. 7A shows a screen 5a that warns of the degree of meanderings or a deviation of the belt 41, and Fig. 7B shows a screen 5b that informs of predicted results of a possible trouble that may occur in the conveyor belt 4.

[0039] Hereinafter, an embodiment of the present invention will be described with reference to drawings. Various features described in the embodiment below can be combined with each other.

[0040] A program for realizing a software described in the present embodiment may be provided as a non-transitory computer-readable memory medium, may be provided to be downloaded via an external server, or may be provided so that the program is activated on an external computer and the program's function is realized on a client terminal (that is, the function is provided by so-called cloud computing).

[0041] A term "unit" in the present embodiment may include, for example, a combination of a hardware resource implemented as circuits in a broad sense and information processing of software that can be concretely realized by the hardware resource. Furthermore, various kinds of information are described in the present embodiment, and such information may be represented by, for example, physical values of signal values representing voltage and current, high and low signal values as a set of binary bits consisting of 0 or 1, or quantum superposition (so-called qubits), and communication and computation may be executed on a circuit in a broad sense.

[0042] The circuit in a broad sense is a circuit realized by properly combining at least a circuit, circuitry, a processor, a memory, and the like. In other words, a circuit includes an application specific integrated circuit (ASIC), a programmable logic device (e.g., simple programmable logic device (SPLD), a complex programmable logic device (CPLD), field programmable gate array (FPGA), and the like.

[0043] Embodiment 1. Hardware configuration This section will describe a hardware configuration of a system 1 according to one embodiment. Fig. 1 is a block diagram showing a hardware configuration of a system 1 according to one embodiment. As shown in Fig. 1, the system 1 comprises an information processing apparatus 2, a LiDAR sensor 3, and a conveyor belt 4. The information processing apparatus 2 and the LiDAR sensor 3 are electrically connected via a communication unit 21 and a transmission path 31 described below. The LiDAR sensor 3 is positioned so as to be able to measure the shape of the conveyor belt 4. The system 1 predicts a trouble that may occur in the conveyor belt 4. Alternatively, the system 1 monitors the conveyor belt 4.

[0044] In one embodiment, a "system" comprises one or more devices or components. Therefore, it should be noted that, for example, the information processing apparatus 2 alone may also be an example of a system as a component, the information processing apparatus 2 and the LiDAR sensor 3 may also be examples of a system as components, and the information processing apparatus 2, the LiDAR sensor 3 and the conveyor belt 4 may also be examples of a system as components.

[0045] (Information processing apparatus 2) The information processing apparatus 2 has a communication unit 21, a storage unit 22, a controller 23, a display unit 24, and an input unit 25, and these components are electrically connected inside the information processing apparatus 2 via a communication bus 20. Each component will be further described below.

[0046] Although wired communication methods such as USB, IEEE1394, Thunderbolt (registered trademark), wired LAN network communication, etc. are preferred, the communication unit 21 may include wireless LAN network communication, mobile communication such as LTE / 5G, Bluetooth (registered trademark) communication, etc. as necessary. That is, it is further preferable to implement as a set of these communication means. In other words, the information processing apparatus 2 may communicate various information from the outside via the communication unit 21 and a network. Specifically, the information processing apparatus 2 is connected to the LiDAR sensor 3 via the communication unit 21 and the transmission path 31. The LiDAR sensor 3 will be described later.

[0047] The storage unit 22 is configured to store various information as defined by the above description. This may be implemented as a storage device such as a solid state drive (SSD) storing various programs related to the information processing apparatus 2 that are executed by the controller 23, or as a memory such as a random access memory (RAM) that stores temporarily necessary information (argument, sequence, etc.) for program operation. The storage unit 22 stores various programs, variables, etc. related to the information processing apparatus 2 that are executed by the controller 23.

[0048] The controller 23 (processor) processes and controls overall operation pertaining to the information processing apparatus 2. The controller 23 is, for example, an unshown central processing unit (CPU). The controller 23 reads various functions pertaining to the information processing apparatus 2 by reading a predetermined program stored in the storage unit 22. In other words, information processing by software stored in the storage unit 22 is specifically realized by the controller 23, an example of hardware, thereby may be executed as each functional unit included in the controller 23. Further details on these will be described in the next section. It should be noted that the controller 23 is not limited to being singular and may be implemented with two or more controllers 23 for each function. Additionally, a combination thereof may be applied.

[0049] The display unit 24 may be included in a housing of the information processing apparatus 2 or may be externally attached, for example. The display unit 24 is configured to display a screen of graphical user interface (GUI) that is operable by a user. For instance, this is preferable to be implemented by using different display devices such as a CRT display, a liquid crystal display, an organic EL display, and a plasma display. In the case of an external device, the display unit 24 and the information processing apparatus 2 may be wired-connected, or they may be connected without direct wiring via wireless or satellite communication, etc.

[0050] The input unit 25 may be included in the housing of the information processing apparatus 2, or may be externally attached. For example, the input unit 25 may be integrated as a touch panel integrated with the display unit 24. With the touch panel, a user may input through tapping, swiping, or other operation. Of course, a switch button, mouse, a QWERTY keyboard, etc. may be employed instead of the touch panel. In other words, the input unit 25 receives operation inputs performed by the user. This input, treated as a command signal, is transferred to the controller 23 via the communication bus 20, and the controller 23 may execute predetermined control or calculation as necessary.

[0051] (LiDAR sensor 3) The LiDAR sensor 3 (an example of an optical sensor) is a sensor that irradiates a laser beam onto an object, measures the distance to the object by detecting the reflected light, and generates measurement data of a measuring object. More specifically, the measurement data is point cloud data obtained via the LiDAR sensor 3. The LiDAR sensor 3 includes an unshown laser light source, a light receiving unit, and a communication unit. A laser beam is irradiated from the laser light source, the light receiving unit receives reflected light reflected from the after-mentioned conveyor belt 4 or powder PW that is an example of a load, which is an object. Then, a microcomputer generates point cloud data, and the point cloud data can be output externally via the communication unit.

[0052] As shown in Fig. 1, the LiDAR sensor 3 is connected to the communication unit 21 of the information processing apparatus 2 via the transmission path 31. This configuration makes it possible to detect meanderings or a deviation of a belt 41 of a conveyor belt 4, as described below, while ensuring both accuracy and cost, as well as to predict in advance various troubles that may occur in the conveyor belt 4. The type of the transmission path 31 is not limited, and may be a USB or serial connection, for example, or may be connected via a network such as the Internet or an intranet. Also, for simplicity, one LiDAR sensor 3 is illustrated; however, in practice, a plurality of LiDAR sensors 3 may be implemented in the system 1.

[0053] In other words, the system 1 preferably comprises the information processing apparatus 2 and the LiDAR sensor 3 (an example of an optical sensor). The LiDAR sensor 3 (an example of an optical sensor) is configured to measure powder PW conveyed on the conveyor belt 4. The information processing apparatus 2 includes the controller 23 (an example of a processor) configured to execute a program for implementing each function. The information processing apparatus 2 is connected to the LiDAR sensor 3 and configured to acquire point cloud data (an example of measurement data) related to the powder PW (an example of a load) from the LiDAR sensor 3.

[0054] (Conveyor belt 4) Fig. 2 is a schematic diagram illustrating a positional relationship between the LiDAR sensor 3 and the conveyor belt 4. As shown in Fig. 2, a direction along the width of the belt 41 of the conveyor belt 4 is defined as an X-axis, a direction from bottom to top is defined as a Y-axis, and a direction of travel of the conveyor belt 4 is defined as a Z-axis. The conveyor belt 4 includes a belt 41 and rollers 42. The belt 41 is a sheet-like member and is provided so as to wrap around a plurality of rollers 42. The belt 41 is configured to be able to convey a load in the Z-axis direction in a state where the load such as powder PW is loaded on a loading surface 41f. The rollers 42 are cylindrical members extending in the X-axis direction and are arranged at regular intervals along the Z-axis direction. Each of the rollers 42 supports the belt 41 and rotates on its axis to rotate the belt 41, thereby sliding the loading surface 41f in the direction of travel. Furthermore, the belt 41 has ends 41l and 41r in its width direction, respectively. In one embodiment, the positions of the ends 41l and 41r are an example of mechanical parameters and can be used to specify the degree of meanderings or a deviation of the conveyor belt 4, as will be described later. In other words, each of the mechanical parameters is preferably a position of each of the ends 41l and 41r in the point cloud data. Details will be described later.

[0055] The above-mentioned LiDAR sensor 3 is installed above the conveyor belt 4. The LiDAR sensor 3 measures the shape of the surface of the load such as the powder PW that is conveyed by the conveyor belt 4 as seen from above, and outputs a measurement result, that is, point cloud data, to the above-mentioned information processing apparatus 2. In other words, the LiDAR sensor 3 is installed above the conveyor belt 4 so as to look down the loading surface 41f of the conveyor belt 4 on which the powder PW is loaded. By employing such a configuration, meanderings or a deviation of the belt 41 can be detected with a simple configuration, and various troubles that may occur in the conveyor belt 4 can be predicted in advance. The point cloud data obtained by the LiDAR sensor 3 indicates the shape of the surface of the load or the belt 41 in the X-Y plane including the LiDAR sensor 3 (a plane parallel to the plane including the X-axis and Y-axis and including the LiDAR sensor 3).

[0056] 2. Functional structure This section describes the functional structure of one embodiment. As mentioned above, information processing by software stored in the storage unit 22 is specifically realized by the controller 23, which is an example of hardware, and thus each functional unit included in the controller 23 can be executed. In other words, the system 1 has at least one processor (e.g., the controller 23). The processor is configured to execute a program so that each step of a monitoring method of the conveyor belt 4 or a trouble prediction method is performed. According to another aspect, the program allows at least one computer to execute each step of the monitoring method or the trouble prediction method of the conveyor belt 4. Each step includes, for example, an acquisition step, a correction step, a specification step, a prediction step, a warning step, and a display control step.

[0057] Fig. 3 is a block diagram showing functions realized by the controller 23 or the like in the information processing apparatus 2. Specifically, the controller 23 can function as a calculation unit 231, an acquisition unit 232, a correction unit 233, a specification unit 234, a prediction unit 235, a warning unit 236, and a display controller 237. Each functional unit will be outlined below.

[0058] The calculation unit 231 is configured to execute various operations in the information processing apparatus 2.

[0059] The acquisition unit 232 is configured to acquire external information via the communication unit 21. Specifically, as the acquisition step, the acquisition unit 232 acquires point cloud data (an example of measurement data) related to the belt 41 of the conveyor belt 4 via the LiDAR sensor 3 (an example of an optical sensor) installed apart from the conveyor belt 4. Details are described below.

[0060] As the correction step, the correction unit 233 corrects the reference position based on the amount of load loaded on the belt 41 and / or the start / stop of the conveyor belt 4. Details are described below.

[0061] The specification unit 234 is configured to specify various types of information. Specifically, as the specification step, the specification unit 234 specifies the degree of meanderings or a deviation of the belt 41 according to the positions of the ends defining the width of the belt 41 (an example of mechanical parameters), which is calculated from the point cloud data. Details are described below.

[0062] The prediction unit 235 is configured to predict various types of information. Specifically, as the prediction step, the prediction unit 235 predicts a trouble that may occur in the conveyor belt 4 based on shape parameters that is calculated from the point cloud data and is related to the shape of the powder PW (e.g., the X coordinate where the Y coordinate value is maximum), and reference information IF. Details are described below.

[0063] The warning unit 236 is configured to output various warnings as the warning step.

[0064] The display controller 237 is configured to output display information. Display information may be visual information itself, such as screens, images, icons, text, etc., generated in an aspect that is visible to the user, or may be rendering information for displaying screens, images, icons, text, etc. on the display unit 24. For example, the contents of the warning output by the warning unit 236 is displayed on the display unit 24 by the display controller 237 as the display control step.

[0065] 3. Overview of information processing This section describes an overview of the information processing executed by the system 1.

[0066] In one embodiment, assumed is that the case where the powder PW, an example of a load, is conveyed by the conveyor belt 4 to a yard, which is a storage location. In practice, the powder PW is conveyed to the destination yard via a plurality of conveyor belts 4, but here, the flow of the process will be described by focusing on a state in which the powder PW is being conveyed to one of these conveyor belts 4. Such a process is an example of the monitoring method of the conveyor belt 4 or the trouble prediction method of the conveyor belt 4.

[0067] Fig. 4 is an activity diagram showing the flow of the process performed by the system 1. First, the LiDAR sensor 3 measures the conveyor belt 4 while the powder PW is being conveyed on the conveyor 4 (Activity A101). Specifically, the LiDAR sensor 3 installed above the conveyor belt 4 continuously irradiates a laser beam toward the conveyor belt 4. The reflected light from the belt 41 and / or the powder PW is then detected.

[0068] The acquisition unit 232 then acquires point cloud data from the LiDAR sensor 3 (Activity A102). Specifically, the information processing apparatus 2 receives the point cloud data output from the LiDAR sensor 3 via the transmission path 31 through the communication unit 21 and allows the storage unit 22 to store the data. The point cloud data at this time includes the belt 41 of the conveyor belt 4 and the powder PW loaded on the belt 41.

[0069] The calculation unit 231 then calculates the positions of the ends 41l and 41r of the belt 41 from the current frame of the point cloud data (activity A103). The positions may be calculated using a rule-based algorithm that extracts edges of the point cloud data or using a learned model. In addition, the calculation unit 231 calculates the change in position by comparing the positions of the ends 41l and 41r calculated not only in the current frame but also in past frames (activity A104). The past frame may be one frame back or may be 10 frames back. Even more preferably, appropriate adjustments may be made depending on the frame rate or the control rate.

[0070] On the other hand, for the same frame, the calculation unit 231 extracts a point P, that is a point in the region occupied by the powder PW, as a shape parameter indicating the shape of the powder PW (activity A105). Preferably, the point P is the point where the Y-coordinate value is maximum in the region. In addition, the specification unit 234 specifies the X coordinate of the point P (Activity A106).

[0071] The various information calculated or specified in Activities A103 to A106 is then matched with the predetermined correction conditions (Activity A107). In the case where the correction conditions are met, at least part of the reference information IF, which is the information referred to in specifying a degree and predicting a trouble, is corrected (Activity A108). In the case where the correction conditions are not met, the process in Activity A108 is skipped. For example, it is assumed that the reference information IF is corrected according to the amount of the load, because the standard positions of the ends 41L and 41R may differ depending on the amount of the load. In addition, the standard positions of the ends 41l and 41r may differ depending on the start / stop and travel speed of the conveyor belt 4 itself, and thus it is recommended that corrections be performed in such cases as well. In other words, as the correction step, the correction unit 233 corrects the reference information IF (e.g., the reference positions described below) based on the amount of load loaded on the belt 41 and / or the start / stop of the conveyor belt 4.

[0072] The various information calculated or specified in Activities A103 to A106 is then matched with the reference information IF (Activity A109). For example, as reference information IF, reference positions are set for the case where neither the meanderings nor the deviation occurs, and such reference positions are compared with the positions of the ends 41l and 41r in the point cloud data by the calculation unit 231. As a result, the degree of meanderings or a deviation of the belt 41 is specified. In other words, as the specification step, the specification unit 234 specifies the degree by comparing the positions of the ends 41l and 41r with the preset reference positions. The reference positions are the positions of the ends 41l and 41r obtained by measuring the belt 41 in a state where neither the meanderings nor the deviation occurs on the belt 41. According to such an aspect, the shift with respect to the preferred reference positions can be used to more accurately detect meanderings or a deviation of the belt 41. In the case where the reference information IF is corrected in activity A108, the specification unit 234 specifies the degree by comparing the positions of the ends 41l and 41r with the reference positions corrected by the correction unit 233 as the specification step. According to such an aspect, the reference positions can be varied according to the situation to more accurately detect meanderings or a deviation of the belt 41.

[0073] Alternatively, a look-up table summarizing the coordinates of the point P of the powder PW (an example of shape parameters) and the associated expected troubles may be employed as reference information IF. In such a case, the reference information IF preferably includes, as one of the shape parameters, at least the shape parameters in a state where the powder PW is not being conveyed by the conveyor belt 4. According to such an aspect, it is possible to distinguish the shape of the powder PW that is not being conveyed and, more accurately to predict in advance various troubles that may occur in the conveyor belt 4 caused by the position of the powder PW.

[0074] The display controller 237 then allows the display unit 24 to display at least part of the results of the series of information processing (activity A110). For example, the point cloud data itself may be permanently displayed on the display unit 24, or various information calculated or specified in activities A103 to A106 (e.g., the positions of the ends 41l and 41r, the point P in the powder PW, or the like) may be displayed on the display unit 24. In addition, in the case where the meanderings or the deviation of the belt 41 is specified, the degree of the meanderings or the deviation may be displayed.

[0075] The results of Activity A109 are then matched with the predetermined warning conditions (Activity A111). In the case where the warning condition is met, a warning is presented to the workers (an example of a user) who are waiting in the yard (Activity A112). In the case where the warning condition is not met, the processing of Activity A112 is skipped. For example, a warning is presented to the workers in the case where the degree of the meanderings or the deviation of the belt 41 exceeds a predetermined threshold value, or in the case where some kind of trouble is predicted from the coordinates of the P point of the powder PW.

[0076] Information processing related to such activities A101 to A112 is continuously executed for each frame rate of the LiDAR sensor 3 or for each control rate of the controller 23 thereby executing the monitoring method of the conveyor belt 4 or the trouble prediction method of the conveyor belt 4.

[0077] In summary, the monitoring method of the conveyor belt 4 according to one embodiment includes the following steps. In the acquisition step, the point cloud data (an example of measurement data) related to the belt 41 of the conveyor belt 4 is acquired via the LiDAR sensor 3 (an example of an optical sensor) installed apart from the conveyor belt 4. In the specification step, the degree of meanderings or a deviation of the belt 41 is specified according to the positions of the ends that defines the width of the belt 41 (an example of mechanical parameters), which is calculated from the point cloud data. According to such an aspect, meanderings or a deviation of the belt 41 can be more easily detected. In particular, in a preferred aspect, the positions of the ends 41l and 41r, which are easy to obtain in terms of information processing, can be used, thereby detecting the meanderings or the deviation of the belt 41.

[0078] The trouble prediction method of the conveyor belt 4 according to one embodiment also has the following steps. In the acquisition step, the point cloud data (an example of measurement data) related to the powder PW that is being conveyed on the conveyor belt 4 is acquired via the LiDAR sensor 3 (an example of an optical sensor) installed apart from the conveyor belt 4. In the prediction step, the trouble that may occur in the conveyor belt 4 is predicted based on the shape parameters that are calculated from the point cloud data and are related to the shape of the powder PW (e.g., the X coordinate where the Y coordinate value is maximum), and the reference information IF. The reference information IF is information on the relationship between the shape parameters and various troubles. According to such an aspect, various troubles that may occur in the conveyor belt 4 can be predicted in advance.

[0079] 4. Details This section further describes some of the flows of the process described in the previous section with reference to other figures. (Positions of ends 41l and 41r) Fig. 5 is a schematic diagram indicating the positions of the ends 41l and 41r in different states of the belt 41. This figure is a cross-sectional view of the belt 41 with one plane parallel to the XY plane as a cutting plane. A measurement direction MD indicates a direction of measurement by the LiDAR sensor 3. Different states ST0 to ST2 are also illustrated in this figure.

[0080] In State ST0, the power PW that is a load is not loaded on the belt 41. As illustrated, when no load is loaded, the belt 41 may float without being in contact with the roller 42 located in the center, and a gap 43 may be generated between the belt 41 and the roller 42. The positions of the ends 41l and 41r in State ST0 are shown as positions L0 and R0, respectively.

[0081] State ST1 is the state in which the powder PW that is a load is loaded on the belt 41. In such a case, the belt 41 sinks due to the gravity of the powder PW and is in contact with the roller 42 located in the center. The positions of the ends 41l and 41r in State ST1 are shown as positions L1 and R1, respectively. Note that since the belt 41 is sunk, the positions L1 and R1 are located inside in the width direction (X-axis direction) of the positions L0 and R0 in State ST0. In other words, in the state where the positions of the ends 41l and 41r both move inward in the width direction in the analysis results of point cloud data, the load can be certified as having been loaded on the belt 41. It is recommended that such rules be managed as one of types of reference information IF.

[0082] Since the sinking of the belt 41 does not change significantly when the conveyance amount is more than a certain amount, it is preferable to determine the meanderings or the deviation of the belt 41 on the condition that more than a certain amount of conveyance or sinking has occurred. The LiDAR sensor 3 may be used to understand the amount of conveyance, or new information may be obtained from other sensors, an external server, or the like.

[0083] State ST2 is a state in which the belt 41 deviates to the left with the powder PW that is a load loaded on the belt 41. The positions of the ends 41l and 41r in State ST2 are shown as positions L2 and R2, respectively. As is clear from comparing both States ST1 and ST2, the position L2 is to the left of the position L1, and the position R2 is to the left of the position R1. In other words, in the case where both ends 41l and 41r move in the same direction (here to the left) or are located relatively close to one side, it can be certified that a deviation has occurred. It is recommended that such rules be managed as one of types of reference information IF.

[0084] To summarize the above, the specification unit 234 may further specify that the ends 41l and 41r have moved in a direction of narrowing or widening the width based on the positions L1, L2, R1, R2, or the like (an example of the mechanical parameters) as a specification step. According to such an aspect, meanderings or a deviation of the belt 41 can be more accurately detected by distinguishing the elements that are not meanderings or a deviation of the belt 41. Even more preferably, the reference positions of the ends 41l and 41r used as reference information IF may be the positions of the ends 41l and 41r obtained in a state where the load is loaded and the belt 41 is sunk in a direction of gravity, as shown in State ST1. According to such an aspect, it is possible to more accurately detect meanderings or a deviation of the belt 41 such as in State ST2 by distinguishing the state in which the belt 41 is floating (State ST0).

[0085] By the way, regarding the shape parameters of the powder PW, it should be noted that the shape of the powder PW when naturally loaded (energetically stable) will differ depending on the physical properties of the powder PW, but not only this, the shape of the powder PW will also change depending on the state of the belt 41 (e.g., States ST0 to ST2). Therefore, the shape parameters such as the coordinates of the P point may be calculated in a state where the powder PW is loaded and the belt 41 is sunk in the direction of gravity (e.g., State ST1). According to such an aspect, it is possible to focus on the shape of the powder PW being conveyed and more accurately predict in advance various troubles that may occur in the conveyor belt 4 caused by the positions of the powder PW.

[0086] (Coordinates of the powder PW) Fig. 6 is a schematic diagram showing the XY coordinates, which are an example of the shape parameters of the powder PW. This figure is a cross-sectional view of the belt 41 with one plane parallel to the XY plane as a cutting plane. The measurement direction MD indicates the direction of measurement by the LiDAR sensor 3.

[0087] The point P in which the Y-coordinate is maximum in the region occupied by the powder PW is illustrated in Fig. 6. In one embodiment, the prediction of trouble is performed by focusing on the relationship between the coordinates of the point P within the specified X-coordinate range and the trouble. That is, the shape parameters of the powder PW include the first coordinate (e.g., X coordinate), which is the coordinate in the width direction of the belt 41 of the conveyor belt 4, in the region (including the contour and the interior) that represents the shape of the powder PW in the measurement data. According to such an aspect, it is possible to predict in advance various troubles that may occur in the conveyor belt 4 caused by the position of the powder PW. And preferably, the first coordinate (e.g., the X coordinate) is, in the region, a coordinate of the point where the second coordinate (e.g., the Y coordinate) that is the vertical coordinate of the belt 41 is maximum or maximal. According to such an aspect, it is possible to predict in advance various troubles that may occur in the conveyor belt 4 caused by the position of the powder PW with even greater accuracy.

[0088] In the case where the amount of the powder PW conveyed is small, analysis based on shape parameters such as the coordinate of the P point may be difficult. In such a case, analysis regarding prediction may be performed only when the conveyance amount is more than a certain amount. In the case where the belt 41 and the roller 42 are not in contact with each other, as in State ST0 in Fig. 5, the belt 41 may sink during the conveyance of the powder PW, and the point cloud data of the conveyor belt 4 to be reference may differ. Therefore, the analysis may be performed only when the conveyance amount is more than a certain amount.

[0089] (Example of warning screen) Fig. 7 is an example of a screen on which the display controller 237 allows the display unit 24 to display a warning output by the warning unit 236. In particular, Fig. 7A shows a screen 5a that warns of the degree of meanderings or a deviation of the belt 41, and Fig. 7B shows a screen 5b that informs of the predicted results of a possible trouble that may occur in the conveyor belt 4.

[0090] In an area 51 of the screen 5a, the text information is drawn as "a deviation at level 3 has occurred on the conveyor belt No. 2. Please stop the operation and work on the correction. The deviation at level 3 is an example of degree, and multiple levels (e.g., five levels) may be defined depending on the meanderings or the deviation of the belt 41. In addition, the warning may include information indicating the degree as well as information on what action the user should take. Specifically, the degree of meanderings or a deviation of the belt 41 is specified depending on which range of the degree (e.g., levels 1 to 5) set as the reference information IF the positions of the ends 41l and 41r of the belt 41 calculated by the calculation unit 231 fall within. Then, in the case where the degree is at a level where a warning should be issued (e.g., level 3 or higher), the warning is output by the warning unit 236.

[0091] In an area 52 of the screen 5b, the text information is drawn as "the position of the load being conveyed on the conveyor No. 3 is deviated. The load may fall!" Specifically, it is determined whether the coordinates of the powder PW calculated by the calculation unit 231 correspond to a pattern that causes the trouble set as reference information IF, and if so, a warning is output by the warning unit 236 as a predicted trouble. The screen 5b warns of "a fall of the loads " as an example of troubles, but it is not limited thereto. Preferably, the trouble is meanderings or a deviation of the belt 41 of the conveyor belt 4, occurrence of a clogging or a deposit in a chute unit (not shown) connecting the two or more conveyor belts 4, or a fall of at least part of the powder PW from the conveyor belt 4. According to such an aspect, specific troubles that may occur in the conveyor belt 4 can be predicted in advance. The unshown chute unit will be supplemented. The unshown chute unit is designed to receive the powder PW that has been conveyed from one conveyor belt 4 through an inlet of the chute unit and further guide the powder PW downward to transfer the powder PW to another downstream conveyor belt 4. For example, in the case where there is a change in the shape parameter of the powder PW (e.g., the X coordinate of the P point) immediately after passing through the chute unit, troubles such as occurrence of a clogging or a deposit in the chute unit may be predicted.

[0092] In other words, as the warning step, the warning unit 236 presents a warning to the user when the degree specified in the specification unit 234 exceeds the threshold value, and presents a warning to the user including the contents of the trouble predicted by the prediction unit 235. According to such an aspect, the user can notice any trouble such as meanderings or a deviation of the belt 41. In addition, according to such an aspect, the user can notice the possibility of various kinds of troubles.

[0093] Each aspect displayed on the display unit 24 is only an example, and warning sounds may be output, or a patrol lamp with a contact output, which is placed in a prominent position, may be lit, or the like. In other words, the aspect is not limited to visual information, but may be other five-sense information such as auditory information, tactile information, etc., or a combination of two or more of these. In such a case, it is possible to save more time and effort for workers who regularly monitor the display unit 24.

[0094] Other The system 1 according to one embodiment can also be implemented in the following aspect.

[0095] According to one embodiment, it is possible to specify the degree of meanderings or a deviation of the conveyor belt 4 and to predict possible troubles that may occur in the conveyor belt 4. The point cloud data obtained by the LiDAR sensor 3 may be displayed on the display unit 24 of the information processing apparatus 2, so that on-site staff (an example of a user) can visually specify the degree of meanderings or a deviation. In other words, the acquisition unit 232 acquires point cloud data related to the belt 41 of the conveyor belt 4 via the pre-installed LiDAR sensor 3 as an acquisition step, and the display controller 237 allows the display device to display the point cloud data so as to be visible to the user as a display control step. This assists the user in determining the degree of meanderings or a deviation of the belt 41. According to such an aspect, meanderings or a deviation of the belt 41 can be detected with a simpler configuration.

[0096] In one embodiment, rules with predetermined reference positions, threshold values, or the like were used as the reference information IF, but the present invention is not limited thereto. In other words, the reference information IF is not limited to databases, look-up tables, etc., but can also be a mathematical model that mathematically relates multiple pieces of information, or a learned model in which the correlation of multiple pieces of information has been machine-learned in advance. Preferably, the reference information IF is a learned model. For example, the reference information IF is a learned model that has machine-learned a relationship between the shape of the powder PW and the trouble in advance. The prediction unit 235 predicts a trouble by inputting shape parameters into the learned model as the prediction step. According to such an aspect, it is possible to predict in advance various troubles that may occur in the conveyor belt 4 caused by the position of the powder PW, as an appropriate decision by analogy from actual past situations. In addition, the reference information IF may be a learned model that is a machine learning model in which a relationship between the positions of the ends and the meanderings or the deviation of the belt 41 has been learned in advance. In such a case, the specification unit 234 specifies the degree by inputting the positions of the ends into the learned model as the specification step. According to such an aspect, meanderings or a deviation of the belt 41 can be more accurately detected as an appropriate decision by analogy from actual past conditions.

[0097] In one embodiment, the X-coordinate (the first coordinate) was employed as the shape parameter of the powder PW. However, since the measurement by the LiDAR sensor 3 and the calculation by the information processing apparatus 2 are executed continuously, multiple X-coordinates may be employed in a time series. For example, in the case where the value of the X coordinate changes significantly over time, it is assumed that part of the powder PW may have fallen as a trouble. Such a rule may be one of types of reference information IF. In other words, as the prediction step, the prediction unit 235 predicts a trouble based on the two or more X-coordinates (first coordinate) calculated at different times during the conveyance of the powder PW. According to such an aspect, it is possible to predict in advance various troubles that may occur in the conveyor belt 4 caused by the position of the powder PW with even greater accuracy, considering changes in the shape of the powder PW over time.

[0098] In one embodiment, the position information of the ends 41l and 41r is employed as an example of mechanical parameters of the ends 41l and 41r. However, the degree of meanderings or a deviation of the belt 41 may be specified using velocity information, which is information on difference (differential) for each frame of the position information. Also, the degree of meanderings or a deviation of the belt 41 may be specified using acceleration information, which is information on the difference (differential) for each frame of the velocity information. Preferably, the relationship between a plurality of mechanical parameters and the degree of meanderings or a deviation of the belt 41 may be set as the reference information IF. In such a case, the reference information IF may be a physical model or a learned model in which a machine learning has been performed in advance.

[0099] By combining the positions of the ends 41l and 41r (an example of mechanical parameters) and the coordinates of the point P of the powder PW (an example of shape parameters), which are obtained from the point cloud data, troubles that may occur in the conveyor belt 4 may be predicted. For example, the positions of the ends 41l and 41r and the coordinate of the point P of the powder PW can be used to predict the possibility of the fall of the powder PW. In such a case, as reference information IF, a look-up table storing the relationship between the positions of the ends 41l and 41r and the coordinate of the point P of the powder PW may be employed, or a learned model in which the relationship between the two has been machine-learned may be employed. In other words, the point cloud data (an example of measurement data) includes information on the belt 41 of the conveyor belt 4 as well as the shape of the powder PW. As the prediction step, the prediction unit 235 predicts the fall of at least part of the powder PW from the belt 41 as a trouble, based further on the positions of the ends 41l and 41r defining the width of the belt 41 (an example of mechanical parameters), calculated from the point cloud data. According to such an aspect, it is possible to predict in advance various troubles such as a fall of the powder PW by focusing also on the condition of the ends 41l and 41r of the belt 41 of the conveyor belt 4.

[0100] In the case of specifying the degree of meanderings or a deviation of the belt 41, the load conveyed by the conveyor belt 4 is not limited to the powder PW. For example, the load may be gravel, iron ore, coal, or other raw materials whose grain size is larger than the powder PW.

[0101] Although the system 1 according to one embodiment can specify the degree of meanderings or a deviation of the belt 41 and predict possible troubles that may occur in the belt 41, some process may be performed to take countermeasures depending on this result. For example, in the case where there is a possibility of falling the powder PW, control of chemical injection to prevent this by making the powder PW contain moisture may be performed. For example, a mechanism control may be used to automatically correct the meanderings or the deviation of the belt 41.

[0102] The LiDAR sensor 3 is a preferred example of an optical sensor, and the point cloud data is a preferred example of measurement data obtained by the optical sensor. However, the present invention is not limited thereto, and other types of optical sensors such as an infrared camera may be employed. In such a case, depth information may be obtained as stereo vision. Alternatively, the optical sensor may be placed at an oblique angle (e.g., 30-60 degrees) for measurement, instead of looking down from above. Even in such a case, the positions of the ends 41l and 41r and the shape parameters of the powder PW can be calculated.

[0103] On the other hand, in the case where the LiDAR sensor 3 is used in principle, it is possible to reliably obtain information in the distance direction of the target. By combining such information with the physical property information of the powder PW that is known in advance, the shape of the powder PW and its changeability can also be estimated. Such information is managed as reference information IF, and thus it is possible to predict troubles that may occur in the conveyor belt 4 with greater accuracy.

[0104] Finally, various embodiments of the present invention have been described, but these are presented as examples and are not intended to limit the scope of the invention. The novel embodiment can be implemented in various other forms, and various omissions, replacements, and modifications can be made without departing from the abstract of the invention. The embodiments and its modifications are included in the scope and abstract of the invention and are included in the scope of the invention described in the claims and the equivalent scope thereof.

[0105] 1 System 2 Information processing apparatus 20 Communication bus 21 Communication unit 22 Storage unit 23 Controller 231 Calculation unit 232 Acquisition unit 233 Correction unit 234 Specification unit 235 Prediction unit 236 Warning unit 237 Display controller 24 Display unit 25 Input unit 3 LiDAR sensor 31 Transmission path 4 Conveyor belt 41 Belt 41f loading surface 41l End 41r End 42 Roller 43 Gap 5a Screen 5b Screen 51 Area 52 Area IF Reference information MD Measurement direction PW Powder ST0 State ST1 State ST2 State

Claims

1. A trouble prediction method of a conveyor belt, comprising: an acquisition step of acquiring measurement data related to powder that is being conveyed on the conveyor belt via an optical sensor installed apart from the conveyor belt; and a prediction step of predicting a trouble that may occur in the conveyor belt based on shape parameters related to a shape of the powder and reference information, the shape parameters being calculated from the measurement data, the reference information being information on a relationship between the shape parameters and various types of the trouble.

2. The trouble prediction method according to claim 1, wherein the shape parameters include a first coordinate that is a coordinate in a width direction of a belt of the conveyor belt in a region (including a contour and an interior) that represents the shape of the powder in the measurement data.

3. The trouble prediction method according to claim 2, wherein the first coordinate is a coordinate of a point where a second coordinate that is a vertical coordinate of the belt is maximum or maximal in the region.

4. The trouble prediction method according to claim 3, wherein the prediction step predicts the trouble based on the two or more first coordinates calculated at different times during a conveyance of the powder.

5. The trouble prediction method according to any one of claims 1 to 4, wherein the reference information includes, as one of the shape parameters, at least the shape parameters in a state where the powder is not being conveyed by the conveyor belt.

6. The trouble prediction method according to any one of claims 1 to 5, wherein the shape parameters are calculated in a state where the powder is loaded, and the belt of the conveyor belt is sunk in a direction of gravity.

7. The trouble prediction method according to any one of claims 1 to 6, wherein the measurement data includes information on the belt of the conveyor belt as well as the shape of the powder, and the prediction step predicts a fall of at least part of the powder from the belt as the trouble based further on mechanical parameters of ends defining a width of the belt, the mechanical parameters being calculated from the measurement data.

8. The trouble prediction method according to any one of claims 1 to 7, wherein the reference information is a learned model that has machine-learned a relationship between the shape of the powder and the trouble in advance, and the prediction step predicts the trouble by inputting the shape parameters into the learned model.

9. The trouble prediction method according to any one of claims 1 to 8, wherein the optical sensor is a LiDAR sensor, and the measurement data is point cloud data obtained via the LiDAR sensor.

10. The trouble prediction method according to claim 9, wherein the LiDAR sensor is installed above the conveyor belt so as to look down a loading surface of the conveyor belt on which the powder is loaded.

11. The trouble prediction method according to any one of claims 1 to 10, wherein the trouble is meanderings or a deviation of the belt of the conveyor belt, occurrence of a clogging or a deposit in a chute unit connecting the two or more conveyor belts, or a fall of at least part of the powder from the conveyor belt.

12. The trouble prediction method according to any one of claims 1 to 11, further comprising: a warning step of presenting a warning to a user, the warning including contents of the trouble predicted in the prediction step.

13. A system for predicting a trouble that may occur in a conveyor belt, comprising: at least one processor, wherein the processor is configured to execute a program so that each step of the trouble prediction method according to any one of claims 1 to 12 is executed.

14. The system according to claim 13, comprising: an information processing apparatus and an optical sensor, wherein the optical sensor is configured to measure powder conveyed on the conveyor belt, the information processing apparatus includes the processor configured to execute the program, and is connected to the optical sensor and configured to acquire measurement data related to the powder from the optical sensor.

15. A program, configured to allow at least one computer to execute each step of the trouble prediction method according to any one of claims 1 to 12.

Images