Device for detecting the state of a built-in object, device for detecting an operation and method for detecting the state of a built-in object

Abstract

Device for detecting the state of a built-in object (2), which an information acquisition means (11) that acquires at least one physical piece of information that manifests itself outside an object (3) in which an item (2) is built, the state of which cannot be directly perceived from the outside, and a state-determining means (12) which determines the state of the object (2) on the basis of one or more acquired physical pieces of information, wherein the state-determining means (12) determines that the object is in an abnormal state or a state that will lead to a defect within a certain time period when a first physical information, acquired in advance as physical information at a time when the object (2) is in an abnormal state or a state that will lead to a defect within a certain time period, and a second physical information, acquired as needed by the information-acquisition means (11), are in agreement.

Description

Background of the inventionField of the invention

[0001] The present invention relates to a device for detecting the state of an installed object, a device for detecting an operation, and a method for detecting the state of an installed object. Background technology

[0002] Previously, technologies have been proposed in which physical information, such as strain or temperature, of machines and devices built into industrial machines, such as a reduction gear or a bearing, is detected by a sensor provided on these machines and devices themselves, and the state of the machines and devices is controlled based on the detected physical information. For example, Japanese Patent Laid-Open No.JP 2007-256 033 A (Detection of deterioration of lubricant of a bearing of a machine plant: Example of a machine plant: conveyor belt, for a carriage axle of a railway car) discloses a system in which the content of foreign matter in the lubricant of a bearing installed in the machine plant is optically detected by a sensor provided on a bearing and, on the basis of the detected content of foreign matter in the lubricant, deterioration of the lubricant is detected.

[0003] However, it often happens that there is not enough space inside or outside the machines and devices installed in an industrial machine to install a sensor for monitoring the condition of the machine or device. Furthermore, since the sensor requires multiple cables for sending detection signals and supplying electrical power, there is sometimes no space for these cables in the industrial machines, or wiring related to the operation of the industrial machine is not possible. This problem is particularly evident when monitoring machines and devices installed in an industrial machine that is already in use. Therefore, to provide a sensor for condition monitoring in machines and devices, measures must be taken, such as:This includes reducing the size of the sensor, increasing the size of the industrial machine to ensure sufficient space inside the machine, and using batteries or wireless communication. However, there are limitations to reducing the size of sensors or increasing the size of industrial machines, as well as the problem of changing batteries or installing antennas for wireless communication. As a result, the status of machines or devices built into an industrial machine, which cannot be directly accessed from the outside, cannot be monitored or recorded.

[0004] This problem also exists for structural components such as steel frames or reinforced concrete provided inside buildings or constructions such as wind turbines of wind power generators or the heliostat of the tower of a solar generator, elevated roads, bridges and buildings, fastening components provided inside them such as screws and various pipes laid inside for the water and sewage system and for electrical wiring or for the condition inside the concrete of buildings or constructions made of concrete such as wind turbines of wind power generators or the heliostat of the tower of a solar generator, elevated roads, bridges and buildings.

[0005] In addition, the same problem also exists for multi-purpose tunnels, water pipes, gas pipes and their connecting sections buried under roads and footpaths.

[0006] The same problem also exists, for example, for structures such as cross members and fastening components such as screws and rivets that are used in the body or door components of motor vehicles, trucks, railway wagons or construction machinery such as hydraulic excavators, in the hull of ships or in the fuselage of aircraft. Summary of the invention

[0007] The present invention has been made in consideration of these points and intends to provide an apparatus for detecting the state of an installed article, an apparatus for detecting an operation and a method for detecting the state of an installed article, by which the state can be monitored and detected without directly providing a sensor on machines or devices which are installed in, for example, an industrial machine, and to which direct detection of the state is not possible from the outside, including the use of a means such as a shaft connecting the inside of the industrial machine to the outside.

[0008] The present invention is a device for detecting the state of an installed object, comprising an information acquisition means that acquires at least one physical information of an object that is apparent outside an object in which an object whose state cannot be directly detected from the outside is installed, and a state determination means that determines the state of the object on the basis of the acquired physical information, wherein the state determination means determines that the state of the object is an abnormal state of the object or a state that leads to a defect within a certain period of time when a first physical information that was previously acquired as physical information in a state that was an abnormal state of the object or a state that leads to a defect within a certain period of time,and a second physical information acquired by the information acquisition means.

[0009] In the state detecting device according to the present invention, it is also possible that the information acquiring means acquires a plurality of physical information, and the state determining means determines the state of the object based on the acquired plurality of physical information.

[0010] In the device for detecting the state of an installed article according to the present invention, it is also possible for the state determining means to include an abnormality determining means for determining whether the state of the article is an abnormal state or not.

[0011] In the device for detecting the state of an installed object according to the present invention, it is also possible for the state determining means to include abnormality determining means that determines the state of the object as a state of the object being an abnormal state based on either the result of comparing the acquired plurality of physical information with an abnormality determination threshold for the respective plurality of physical information, or the result of comparing a combination of the acquired plurality of physical information with an abnormality determination threshold set for the combination of the plurality of physical quantities.

[0012] In the device for detecting the state of an installed article according to the present invention, it is also possible for the state determining means to have a defect predicting means that determines whether the state of the article is a state that leads to a defect within a certain period of time.In the device for detecting the state of an installed article according to the present invention, it is also possible for the defect prediction means to determine the state of the article as a state that leads to a defect within a certain period of time based on either the result of a comparison of the acquired plurality of physical information with a defect prediction determination threshold for the respective plurality of physical information, or the result of a comparison of a combination of the acquired plurality of physical information with a combination of defect prediction determination thresholds for the combination of the plurality of physical quantities.

[0013] In the device for detecting the state of an installed object according to the present invention, it is also possible for the state determination means to have a defect prediction means that determines whether it is a state that leads to a defect within a certain period of time based on a state determination model generated from history information of the acquired physical information.In the device for detecting the state of an installed article according to the present invention, the state determination model may be composed of two models, namely a defective state model consisting of historical information of an installed article that had a defect, and a normal state model consisting of historical information of an installed article that had no defect, wherein when the state of the article is more similar to the defective state model than to the normal state model, it is determined that it is a state that leads to a defect within a certain period of time.

[0014] In the device for detecting the state of a built-in object according to the present invention, the physical information exhibited outside the object may also include physical information of the surface of the object.

[0015] In the device for detecting the state of an installed object according to the present invention, the physical information of the surface of the object may also include at least one of temperature, position, strain, displacement, vibration, hue, brightness, color saturation, moisture content, fat content, as well as sound waves, ultrasonic waves, reflection rate of infrared light or other light.

[0016] In the device for detecting the state of a built-in object according to the present invention, the physical information exhibited outside the object may also include at least one of sound, odor, ultrasonic waves, electromagnetic waves, radiation, or excretions detectable outside the object.

[0017] In the device for detecting the state of an installed object according to the present invention, the information acquiring means may also acquire the physical information based on image captures of at least a part of the surface of the object.

[0018] In the device for detecting the state of a built-in object according to the present invention, a thermochromic component that changes color according to temperature may also be provided on the surface of the object, and the information acquiring means may acquire the temperature of the surface of the object based on an image of the thermochromic component.

[0019] In the device for detecting the state of an installed object according to the present invention, the information acquiring means may also have a shot magnification function by which at least a part of the surface of the object is magnified.

[0020] In the device for detecting the state of a built-in object according to the present invention, it is also possible to provide a driving means that drives the information acquiring means and changes the imaging area.

[0021] In the device for detecting the state of an installed object according to the present invention, the information acquiring means may also comprise a plurality of cameras each capturing different locations on the surface of the object.

[0022] In the device for detecting the state of a built-in object according to the present invention, the plurality of cameras may also be arranged to surround the object. In the device for detecting the state of a built-in object according to the present invention, the information acquisition means may also include an optical system capable of wide-angle or omni-azimuth imaging.

[0023] In the device for detecting the state of an installed object according to the present invention, it is also possible for the information acquisition means to comprise an unmanned aircraft equipped with a camera and a control means for the unmanned aircraft which records the surface of the object with the camera.

[0024] In the device for detecting the state of a built-in object according to the present invention, the information acquisition means can acquire at least a part of the three-dimensional shape of the surface of the object based on the image recordings, and acquire the physical information based on the acquired three-dimensional shape. In the device for detecting the state of a built-in object according to the present invention, the information acquisition means can be arranged away from the surface of the object, at least above the object.

[0025] In the device for detecting the state of an installed article according to the present invention, it is also possible for the state detecting means of the installed article to determine, based on the state at the time of construction or installation, whether the article is in an abnormal state or a state that leads to a failure within a certain period of time has been reached.

[0026] In the device for detecting the state of an installed article according to the present invention, it is also possible that the operation of the article is accompanied, and on the basis of the state at the start of the operation of the article, it is determined whether the article is in an abnormal state or a state that leads to a defect within a certain period of time has been reached.

[0027] In the device for detecting the state of an installed object according to the present invention, it is also possible that the object is an industrial robot having at least one rotating shaft, and the object is a reduction gear installed in the rotating shaft, and the information acquiring means acquires physical information of the surface of the rotating shaft.

[0028] The present invention is an operation detecting device that detects, in a virtual space on a computer, the operation of an object built into an object whose state cannot be directly detected from the outside, and is an object operation detecting device into which the state of the object acquired by the built-in object state detecting device is input as information regarding the state of the object.The present invention is a method for detecting the state of an incorporated object, by which at least one physical information of an object is acquired that is external to an object in which an object whose state cannot be directly detected from the outside is installed, and based on the acquired physical information, it is determined whether the object is in an abnormal state or a state that will cause a failure within a certain period of time. According to the present invention, the state of an object that is incorporated in an object and whose state cannot be directly detected from the outside is detected, can be detected without directly providing a sensor to the object. Short description of the characters [ Fig. 1] Block diagram showing the apparatus for monitoring the condition of machines and equipment according to the present embodiment. [ Fig. 2] Diagram showing the example of an industrial machine to which the machine and equipment condition monitoring apparatus according to the present embodiment is applicable. [ Fig. 3] Diagram showing another example of an industrial machine to which the machine and equipment condition monitoring apparatus according to the present embodiment is applicable. [ Fig. 4] Block diagram showing details of the means for detecting the state of machines and equipment in the machine and equipment state monitoring apparatus according to the present embodiment. [ Fig. 5] Flowchart showing an example of the operation of the machine and equipment condition monitoring apparatus according to the present embodiment. [ Fig. 6] A block diagram showing the apparatus for monitoring the state of machines and equipment according to a first modified example of the present embodiment. [ Fig. 7] A diagram showing an application example to an industrial robot as a machine and equipment condition monitoring device according to a second modified example of the present embodiment. [ Fig. 8] Flowchart showing an example of the operation of the machine and equipment condition monitoring apparatus according to the second modified example of the present embodiment. [ Fig. 9] Explanatory diagram explaining the operation of defect prediction with respect to the operation example of the machine and equipment condition monitoring apparatus according to the second modified example of the present embodiment. [ Fig. 10] Explanatory diagram explaining the operation of defect prediction with respect to the operation example of the machine and equipment condition monitoring apparatus according to a third modified example of the present embodiment. [ Fig. 11] Explanatory diagram explaining, with respect to the operation example of the apparatus for monitoring the state of machines and equipment according to a fourth modified example of the present embodiment, the process of inferring the state of the monitored machines and equipment. [ Fig. 12] Explanatory diagram explaining the operation of defect prediction with respect to the operation example of the machine and equipment condition monitoring apparatus according to a fifth modified example of the present embodiment. [ Fig. 13] A diagram showing an application example to a travel motor for a construction machine as a machine and equipment condition monitoring device according to a sixth modified example of the present embodiment. [ Fig. 14] A diagram showing an application example to an automatic door as a machine and equipment condition monitoring device according to the sixth modified example of the present embodiment. [ Fig. 15] A diagram showing an application example to a wind turbine of a wind power generator as a machine and equipment condition monitoring device according to the sixth modified example of the present embodiment. [ Fig. 16] A diagram showing an application example to a heliostat for a solar generator as a machine and equipment condition monitoring device according to the sixth modified example of the present embodiment. [ Fig. 17] Block diagram showing details of the machine and equipment state detecting means in the machine and equipment state monitoring apparatus according to the seventh modified example of the present embodiment. [ Fig. 18] Flowchart showing an example of the operation of the machine and equipment condition monitoring apparatus according to the seventh modified example of the present embodiment. [ Fig. 19] Flowchart showing the continuation of Fig. 18 shows the example of operation of the machine and equipment condition monitoring apparatus according to the seventh modified example of the present embodiment. [ Fig. 20] Explanatory diagram explaining, with respect to the operation example of the machine and equipment condition monitoring apparatus according to the seventh modified example of the present embodiment, the operation of fault prediction based on the combination of the data of a plurality of physical quantities. [ Fig. 21] An explanatory diagram showing, with respect to the operation example of the apparatus for monitoring the state of machinery and equipment according to an eighth modified example of the present embodiment, the process of determining the state of the machinery and equipment to be monitored. [ Fig. 22] A diagram showing an operation detecting device according to a ninth modified example of the present embodiment. [ Fig. 23] Block diagram showing the structure condition monitoring apparatus according to a tenth modified example of the present embodiment. Detailed description of the invention

[0029] Hereinafter, as an example of a device for detecting the state of an installed object according to an embodiment of the present invention, a device for monitoring the state of machinery and equipment will be explained in detail with reference to the figures. Incidentally, the embodiments shown below are examples of embodiments of the present invention, but the present invention is not limited to these embodiments. Furthermore, in the figures relating to the present embodiment, components having the same or similar functions are denoted by the same or similar reference numerals, and repeated explanations are omitted. Furthermore, the dimensional proportions in the figures may deviate from the actual proportions for the purpose of explanation, or a part of the structure may be omitted from the drawings.

[0030] Fig. Fig. 1 is a block diagram showing the machine and equipment status monitoring device according to the present embodiment. The machine and equipment status monitoring device 1 according to the example of Fig. 1 is characterized in that it can monitor, independently of an industrial machine 3, which is an example of an object according to an embodiment of the present invention, from outside the industrial machine 3, the operating state of machines and devices 2 to be monitored, which are installed in the industrial machine 3 and whose state cannot be directly detected from the outside. The machines and devices 2 to be monitored are an example of the object according to an embodiment of the present invention. As in Fig. 1, the device 1 for monitoring the condition of machines and devices comprises an information acquisition means 11 and a machine and device condition determination means 12. (Information Acquisition Means 11)

[0031] The information acquisition means 11 acquires physical information that is visible outside the industrial machine 3 in which the machines and devices 2 to be monitored are installed. The installation of the machines and devices 2 to be monitored includes both the case where the machines and devices 2 to be monitored are completely housed inside the industrial machine 3 and cannot be directly detected from the outside in this state, and the case where part of the machines and devices 2 to be monitored are exposed outside the industrial machine 3, but their state cannot be directly detected from the outside.

[0032] Fig. Figure 2 is a diagram showing an example of an industrial machine 3 to which the machine and equipment condition monitoring device 1 according to the present embodiment is applicable. The industrial machine 3 is not further limited as long as it is a machine used for activities related to industry, i.e., the provision of products and services. As shown in Fig. 2, the industrial machine 3 can be, for example, an industrial robot 30A used, for example, for the production of motor vehicles, a reduction gear 30B built into the industrial robot 30A, a compressor 30C used, for example, for railways, a construction machine 30D, or a construction machine travel motor 30E.

[0033] Fig. Fig. 3 is a diagram showing another example of an industrial machine 3 to which the machine and equipment condition monitoring device 1 according to the present embodiment is applicable. Apart from the examples in Fig. 2, the industrial machine 3 can also be, for example, Fig. 3, a filling and packaging machine 30F used, for example, for packaging food, an aircraft 30G, a flight control actuator 30H that controls movable wings of an aircraft, an automatic door 30I, a reduction gear or bearing installed in a wind turbine 30J for a wind power generator, or a reduction gear or bearing installed in a heliostat 30K for a solar generator.

[0034] The machines and devices 2 to be monitored are not further limited, as long as they are machines and devices that are installed in an industrial machine 3. The machines and devices 2 to be monitored that are installed in the industrial robot 30A can also be, for example, the reduction gear 30B, a bearing, a motor, etc. The machines and devices 2 to be monitored that are installed in the reduction gear 30B can also be, for example, a gear, a bearing, a sealing component for a lubricating oil seal, etc. The machines and devices 2 to be monitored that are installed in the compressor 30C can also be, for example, an air compressor structure called an air end, etc. The machines and devices 2 to be monitored that are installed in the construction machine can also be, for example, the construction machine travel motor 30D, etc.The machines and devices 2 to be monitored that are installed in a construction machine travel motor 30D can, for example, also be a planetary gear, etc. The machines and devices 2 to be monitored that are installed in the filling and packaging machine 30E can, for example, also be a coupling, etc. The machines and devices 2 to be monitored that are installed in an aircraft can, for example, also be the flight control actuator 30F, etc. The machines and devices 2 to be monitored that are installed in the flight control actuator 30F can, for example, also be a valve, etc. The machines and devices 2 to be monitored that are installed in the automatic door 30I can, for example, also be an electric motor, etc. The machines and devices 2 to be monitored that are installed in the wind turbine 30J of a wind power generator can, for example, It could also be a reduction gear, a bearing, a motor, etc.The machines and devices 2 to be monitored, which are installed in the 30K heliostat for a solar generator, may also be, for example, a reduction gear, a bearing, a motor, etc.

[0035] The physical information of the industrial machine 3 that is exposed outside the industrial machine 3 is, for example, physical information of the surface 3a of the industrial machine 3. The physical information of the industrial machine 3 is information that can be acquired as quantified values ​​from the industrial machine 3. The surface 3a of the industrial machine 3 refers to parts of the industrial machine 3 that can be accessed from outside the industrial machine 3 through contact or non-contact. If the surface 3a is transparent, the surface 3a includes the transparent part and the part behind it. Apart from the surface 3a of the industrial machine 3 itself, the surface 3a may also include the surface of something that has been processed or coated on the surface 3a of the industrial machine to acquire physical information, or that has been glued to it.

[0036] The physical information of the surface 3a of the industrial machine 3 can be, for example, the temperature, position, strain, displacement, vibration, sound waves or ultrasonic waves, the reflection rate of infrared light or other light, the absorption rate of electromagnetic waves, hue, brightness, color saturation, moisture content, fat content, etc. The physical information that appears outside the industrial machine 3 is not limited to physical information of the surface 3a of the industrial machine 3. The physical information that appears outside the industrial machine 3 can also be, for example, sound or an odor emitted by the industrial machine 3 and detected outside the industrial machine 3. Furthermore, it can be the strength of radiation, for example,X-rays, electromagnetic waves or ultrasonic waves that penetrate or are transmitted into the interior of the industrial machine 3 and are detected outside the industrial machine 3, but they can also be excretions, such as exhaust gases, that are excreted by the industrial machine 3 and detected outside the industrial machine.

[0037] The information acquisition device 11 can be a contact-type information acquisition device 11 that acquires physical information when in contact with the surface 3a, or it can be a non-contact-type information acquisition device 11 that acquires physical information when not in contact with the surface 3a. Since a non-contact-type information acquisition device 11 has no contact with the industrial machine 3, it can detect the status of the machines and devices 2 to be monitored without negatively affecting the operation of the industrial machine 3.

[0038] The information acquisition means 11 may be, for example, a temperature measuring device, a position or displacement measuring potentiometer, a strain gauge measuring strain, a vibration measuring device, a moisture sensor measuring moisture content, or an oil content measuring device.

[0039] The contactless information acquisition means 11 may include, for example, a pyrometer for measuring temperature, a laser or overcurrent distance / displacement sensor for measuring position or displacement, a non-contact laser Doppler vibrometer for measuring vibrations, a sound or ultrasonic detector (including a 2-dimensional array) for emitting sound waves or ultrasonic waves and receiving their reflected waves, a camera for measuring the reflection rate, hue, brightness, or color saturation of infrared or other light, a sensor using microwaves or images for measuring moisture content, a sensor using a laser for measuring oil content, a microphone for measuring sound, a radiation meter for measuring radiation, an electromagnetic wave meter for measuring electromagnetic waves, an ultrasonic wave meter for measuring gaseous substances,Such as exhaust gases or organic gases. The contactless information acquisition means 11 may comprise a camera for acquiring the physical information based on images of at least a portion of the surface 3a of the industrial machine 3. By acquiring physical information based on images from a camera, the physical information can be acquired through a simple structure without negatively impacting the operation of the industrial machine 3.

[0040] The camera may also have a magnification function, i.e., a zoom mechanism, by which at least a portion of the surface 3a of the industrial machine 3 is recorded in a magnified manner. By recording the surface 3a in a magnified manner, specific physical information of the surface 3a can be acquired with high resolution. The camera may have an optical system, such as a fisheye lens, that allows wide-angle or omni-azimuth recording. By having an optical system with a wide angle of view, the camera's blind spot can be reduced for the surface 3a of the industrial machine 3.

[0041] Multiple cameras can also be provided to image different locations on the surface 3a of the industrial machine 3. By providing multiple cameras, the camera's blind spot on the surface 3a of the industrial machine 3 can be further reduced. The camera can be a color CCD or CMOS camera that records the hue, brightness, or color saturation of a pressure-sensitive ink applied to the surface 3a of the industrial machine 3 and changes its coloration according to the pressure, or of a temperature-sensitive ink, such as a thermochromatic substance, for example, that changes its hue according to the temperature.By using a color CCD or CMOS camera and a pressure-sensitive ink or a temperature-sensitive ink, precise physical information of the surface 3a of the industrial machine 3 can be acquired while simultaneously suppressing a negative influence on the industrial machine 3.

[0042] The camera can be a CCD or CMOS camera (color or black and white with sensitivity in the visible light range, or with sensitivity in the infrared range) that acquires the hue, brightness, color saturation, or light reflection rate. By acquiring the hue, brightness, color saturation, or light reflection rate, the moisture content or oil content adhering to the surface 3a of the industrial machine 3 can be acquired, so that precise physical information of the surface 3a of the industrial machine 3 can be acquired while simultaneously suppressing any negative influence on the industrial machine 3.

[0043] The camera can also be a TOF (time-of-flight) camera. A TOF camera is a distance image sensor camera that radiates pulsed or near-infrared light onto the imaging body, receives the reflected light of the near-infrared light from the imaging body with a TOF sensor, and can measure the distance to the imaging body based on the required reflection time of the received reflected light. The TOF camera can acquire the distance to the surface 3a of the industrial machine 3, i.e., depth information, so that a displacement of the industrial machine 3 can be measured without using a device for projecting a special pattern. Furthermore, vibration can also be acquired as the temporal change of the displacement, so that a strain can be measured as the temporal change rate of the displacement (integral value of the displacement).By using a TOF camera, the three-dimensional shape of surface 3a can be acquired easily and with high precision.

[0044] The contactless information acquisition means 11 may consist of a projector that projects a random pattern, a grid pattern, or a dot pattern onto the surface 3a of the industrial machine 3, a camera that records this pattern, and a computer that calculates the strain of the surface 3a based on the recorded image. In this case, the computer can calculate the strain based on the degree of deformation over time of the recorded pattern. The computer can also acquire the three-dimensional shape of at least a portion of the surface 3a of the industrial machine based on the recorded images of the pattern and acquire the strain based on the degree of deformation over time of the acquired three-dimensional shape. Based on the three-dimensional shape, the measurement precision of the strain can be improved.

[0045] The contactless information acquiring means 11 may also include a radar that acquires the absorption rate of the electromagnetic waves.

[0046] By measuring the absorption rate of electromagnetic waves and their change, a displacement of the surface 3a of the industrial machine 3 can be measured. Furthermore, vibration can also be acquired as a temporal change of the displacement, so that a strain can be measured as the temporal change rate of the displacement (integral value of the displacement).

[0047] The contactless information acquisition device 11 can also include a microphone that acquires sound. Multiple microphones can also be arranged either laterally or above the industrial machine 3, facing each other in a flat or flatly extending state. By arranging multiple microphones, sound can be precisely acquired at different locations on the surface 3a of the industrial machine 3.

[0048] The contactless information acquisition means 11 may also include an odor sensor that acquires an odor. There are no particular restrictions on the specific design of the odor sensor. For example, the odor sensor may be a semiconductor odor sensor that detects the amount of odor molecules absorbed on the semiconductor surface as a change in the resistance value of the semiconductor. Furthermore, the odor sensor may be a quartz oscillation odor sensor that has a sensitive film bonded to the surface of an oscillator and detects the increase in mass of the sensitive film associated with the amount of odor molecules absorbed by the sensitive film as a decrease in the resonant frequency of the oscillator.

[0049] The contactless information acquisition means 11 may also include a radiation detector that detects radiation such as X-rays that have penetrated the industrial machine 3 by a sensor and calculates the X-ray absorption rate by the industrial machine 3 based on the detected radiation such as X-rays, and an ultrasonic detector that detects ultrasonic waves that have penetrated the industrial machine 3 by a sensor and calculates the ultrasonic wave absorption rate by the industrial machine 3 based on the detected amount of ultrasonic waves. (Machinery and Equipment Condition Determination Tool 12)

[0050] Fig. 4 is a block diagram showing details of the machine and equipment condition detecting means 12 in the machine and equipment condition monitoring device 1 according to the present embodiment, which is an example of the condition detecting means according to the embodiment of the present invention. As shown in Fig. 4, the machine and equipment condition detecting means 12 includes a condition inference means 121 and a defect prediction means 122.

[0051] The state inference means 121 infers the state of the machines and devices 2 to be monitored based on the physical information acquired by the information acquisition means 11. Furthermore, the state inference means 121 determines whether the inferred state of the machines and devices 2 to be monitored is a specific state, such as an abnormal state. That is, the state inference means 121 functions as an abnormality determination means and determines whether the specific state is an abnormal state based on the acquired physical information.If an abnormal condition is detected, the condition inference means 121 outputs abnormality generation information, which notifies the abnormal condition of the monitored machines and devices 2, via a cable or wirelessly to a server 4 external to the machine and device condition monitoring device 1. Furthermore, the condition inference means 121 transmits the condition of the monitored machines and devices 2 to the defect prediction means 122.

[0052] The defect prediction means 122 determines whether the inferred state of the machines and devices 2 to be monitored is a state that will lead to a defect within a certain period of time or not. If it is determined that it is a state that will lead to a defect within a certain period of time, the defect prediction means 122 outputs defect prediction information, which notifies the prediction of a defect, to the server 4 via a cable or wirelessly. If neither an abnormal state nor a state that will lead to a defect within a certain period of time is detected, the defect prediction means 122 outputs machine and device state information, which represents the inferred state of the machines and devices to be monitored, to the server 4. Next, the information acquired by the machine and device state determination means 12, such asthe abnormality generation information, the defect prediction information or the machine and equipment condition information, also referred to as acquired information of the machine and equipment condition determining means 12.

[0053] The abnormal condition or the condition leading to a failure within a certain period of time of the machinery and equipment 2 to be monitored is not particularly limited, as long as it is a condition of the machinery and equipment 2 to be monitored that can be inferred based on physical information of the industrial machine 3 that is apparent outside the industrial machine 3. For example, the abnormal condition or the condition leading to a failure within a certain period of time may be a condition in which the temperature of the machinery and equipment 2 to be monitored is at or above a predetermined temperature.Furthermore, the abnormal condition or the condition leading to a defect within a certain period of time may be a condition in which a strain, displacement, or deformation of a certain part of the machines and devices 2 to be monitored is at or above a predetermined strain, displacement, or deformation. Furthermore, the abnormal condition or the condition leading to a defect within a certain period of time may be a condition in which the amplitude or cycle of the vibrations of a certain part of the machines and devices 2 to be monitored is at or above a predetermined amplitude or cycle.Furthermore, the abnormal condition or the condition leading to a defect within a certain period of time may be a condition in which an acceleration or jerk (rate of change of acceleration per unit time; also called "jerk") of a specific part of the machines and devices 2 to be monitored is at or above a predetermined acceleration or jerk. Furthermore, in the case where the information acquisition means 11 acquires multiple pieces of physical information, the abnormal condition or the condition leading to a defect within a certain period of time may be a condition in which the relationship between at least two points among the temperature, strain, displacement, vibration, acceleration, and jerk of the machines and devices 2 to be monitored is a predetermined specific relationship.Furthermore, the abnormal condition or the condition leading to a defect within a certain period of time can also be a condition in which the physical condition of several predetermined parts of the machines and devices 2 to be monitored matches a predetermined condition. In the examples of abnormal conditions or conditions leading to a defect within a certain period of time, "predetermined" can have a fixed meaning, but it can also have a changing meaning due to a certain relationship, for example, a change in the operating time or the installation environment of the industrial machine 3. As already explained, the abnormal conditions are determined by the condition inference means 121 as a defective condition of the machines and devices 2 to be monitored.Furthermore, a condition leading to a defect within a certain period of time is determined by the defect prediction means 122 as a defect prediction, which predicts the occurrence of a certain defect within a certain period of time.

[0054] The abnormal state or the state leading to a defect within a certain period of time may also be affected by the relationship between the position of the surface 3a on which the physical information is displayed and the position of the machinery and equipment 2 to be monitored. For example, even if the same value of physical information is acquired, according to the distance between the position of the surface 3a from which the physical information is acquired and the position of the machinery and equipment 2 to be monitored, the determination result as to whether the state of the machinery and equipment 2 to be monitored is an abnormal state or a state leading to a defect within a certain period of time may be different.Likewise, due to external factors of the machines and devices 2 to be monitored on the surface 3a from which the physical information is acquired, the determination result as to whether the condition of the machines and devices 2 to be monitored is an abnormal condition or a condition leading to a failure within a certain period of time may vary. External factors such as optical interference from sunlight, radio wave conditions, sound conditions, or wind may become temporary or permanent interference factors for the acquired physical information.

[0055] The machine and device condition determination means 12 is hardware, such as a computing device or a storage device. The machine and device condition determination means 12 may also be, at least in part, software. The machine and device condition determination means 12 may be incorporated into a machine and device condition monitoring device 1, but it is also possible for part of the structure (e.g., the defect prediction means 122) to be located in a device (e.g., a server or a database in a cloud) that can communicate with the machine and device condition monitoring device 1 via a network.

[0056] The server 4 can, for example, notify the user of acquired information of the machine and equipment condition determining means 12 inputted by the machine and equipment condition determining means 12 by displaying it on a screen. (Operational example)

[0057] Next, an operation example of the machine and equipment condition monitoring device 1 will be explained. Fig. Fig. 5 is a flowchart showing an example of the operation of the machine and equipment status monitoring device 1 according to the present embodiment. Incidentally, this flowchart is executed repeatedly as necessary. First, as shown in Fig. Fig. 5, physical information that appears outside the industrial machine 3 (step S1).

[0058] After the physical information has been acquired, the state inference means 121 infers the state of the machines and devices 2 to be monitored based on the acquired physical information (step S2).

[0059] Then, the state inference means 121 determines whether the inferred state of the monitored machines and devices 2 is an abnormal state or not (step S3). If the inferred state of the monitored machines and devices 2 is an abnormal state (step S3: Yes), the state inference means 121 outputs abnormality generation information to the outside (step S4).

[0060] On the other hand, if the inferred state of the machines and devices 2 to be monitored is not an abnormal state (step S3: No), the defect prediction means 122 determines whether the inferred state of the machines and devices 2 to be monitored is a state that will lead to a defect within a certain period of time or not (step S5). If the inferred state of the machines and devices 2 to be monitored is a state that will lead to a defect within a certain period of time (step S5: Yes), the defect prediction means 122 outputs defect prediction information to the outside (step S6).On the other hand, if the inferred state of the machines and devices 2 to be monitored is not a state that leads to a defect within a certain period of time (step S5: No), the defect prediction means 122 outputs machine and device state information corresponding to the inferred state of the machines and devices 2 to be monitored to the outside (step S7).

[0061] Even if an attempt is made to provide a sensor or wiring for directly detecting the condition of the machines and devices 2 to be monitored in the limited space inside the industrial machine 3, the provision of the sensor or wiring may be difficult due to limitations on the size or shape of the sensor, the power supply method, the data transmission means, or the amount of wiring.

[0062] In contrast, the machine and equipment condition monitoring device 1 of the present embodiment can acquire physical information of the surface 3a of the industrial machine 3, and infer the condition of the machines and equipment 2 to be monitored installed in the industrial machine 3 based on the correspondence between the acquired physical information and the condition of the machines and equipment 2 to be monitored installed in the industrial machine 3, which condition is set according to the acquired physical information. Incidentally, this correspondence can be set based on actual measurements, but it can also be set based on a physical calculation model.This allows the status of the monitored machines and devices 2 to be detected without providing a dedicated sensor or wiring for directly detecting the status of the monitored machines and devices 2 inside the industrial machine 3. Consequently, the present embodiment allows the status of the monitored machines and devices 2 to be monitored without directly attaching a sensor to the monitored machines and devices 2 installed in the industrial machine 3. (First Modified Example)

[0063] Next, a first modified example will be explained in which the monitoring result of the machine and device state monitoring device 1 is used for controlling the industrial machine 3. Fig. Fig. 6 is a block diagram showing the machine and equipment condition monitoring apparatus 1 according to the first modified example of the present embodiment. In the example of Fig. 1, the device 1 for monitoring the condition of machines and devices is independent of the industrial machine 3. In contrast, the device 1 for monitoring the condition of machines and devices of the first modified example is characterized in that the monitoring result of the machines and devices 2 to be monitored is used for the control of the industrial machine 3. Specifically, the machine and device condition determining means 12 in the Fig. 6, the acquired information of the machine and device state detecting means 12 is output to the industrial machine control device 5, which controls the operation of the industrial machine 3.

[0064] The industrial machine control device 5 controls the operation of the industrial machine 3 based on the acquired information of the machine and equipment condition determining means 12 inputted from the machine and equipment condition determining means 12. For example, if the acquired information of the machine and equipment condition determining means 12 is defect information, the industrial machine control device 5 stops the operation of the industrial machine 3 according to the defect information. Furthermore, if the acquired information of the machine and equipment condition determining means 12 is defect prediction information, the industrial machine control device 5 can stop the operation of the industrial machine 3 according to the defect prediction information, or limit the operating force or the movement range according to the defect prediction information.

[0065] In the first modified example, the status of the monitored machines and devices 2 can be monitored without directly attaching a sensor to the monitored machines and devices 2 installed in the industrial machine 3. Furthermore, the operation of the industrial machine 3 can be controlled based on the information acquired by the machine and device status detection means 12, so that it is possible to prevent a malfunction of the industrial machine 3 and continue operation until the maintenance time (Second Modified Example).

[0066] Next, a second modified example will be explained in which the state of a reduction gear 30B incorporated in an industrial robot 30A is monitored. Fig. 7 is a diagram showing an application example to an industrial robot 30A as the machine and equipment status monitoring device 1 according to a second modified example of the present embodiment.

[0067] As in Fig. 7, in the second modified example, the industrial machine 3 is an industrial robot 30A having at least one rotating shaft 306 to 310. More specifically, the industrial robot 30A includes a mounting member 300, first to fifth arms 301 to 305, and first to fifth rotating shafts 306 to 310.

[0068] The fastening part 300 is provided, for example, on the floor, at a specific fastening point P to which the industrial robot 30A is fastened.

[0069] The first rotating shaft 306 is connected to the fixing part 300 and one end of the first arm 301. The first rotating shaft 306 is installed in a first reduction gear 30B-1, which slows down and outputs the rotation of a motor (not shown), which rotates by transmitting the rotation output from the first reduction gear 30B-1 to the Z-direction in Fig. 5 (ie to the vertical) parallel axial direction.

[0070] The first arm 301 extends from the end connected to the fastening part 300 to the other end. The first arm 301 rotates about the axial direction of the first rotating shaft 306 due to the rotation of the first rotating shaft 306, which is provided further to the side of the fastening point P than the first arm 301.

[0071] The second rotating shaft 307 is connected to the other end of the first arm 301 and one end of the second arm 302. The second rotating shaft 307 is incorporated in a second reduction gear 30B-2, which decelerates and outputs the rotation of a motor (not shown), which rotates around the axial direction orthogonal to the Z direction by transmitting the rotation output from the second reduction gear 30B-2. Incidentally, in the state of Fig. 5, the axial direction of the second rotating shaft 307 is parallel to the Y direction, but the axial direction of the second rotating shaft 307 changes due to the rotation of the first rotating shaft 306, which is located further on the side of the fixing position P than the second rotating shaft 307 and has a different axial direction than the second rotating shaft 307. The second arm 302 extends from the end connected to the first arm 301 to the other end. The second arm 302 rotates about the axial direction of the first and second rotating shafts 306, 307 due to the rotation of the first and second rotating shafts 306, 307, which are provided further on the side of the fixing position P than the second arm 302.

[0072] The third rotating shaft 308 is connected to the other end of the second arm 302 and one end of the third arm 303. The third rotating shaft 308 is installed in a third reduction gear 30B-3, which decelerates and outputs the rotation of a motor (not shown), which rotates around the axial direction parallel to the axial direction of the second rotating shaft 307 by transmitting the rotation output from the third reduction gear 30B-3. Incidentally, in the state of Fig. 5, the axial direction of the third rotating shaft 308 is parallel to the Y direction, but the axial direction of the third rotating shaft 308 changes due to the rotation of the first rotating shaft 306, which is located further on the fixing position P side than the third rotating shaft 308 and has a different axial direction than the third rotating shaft 308. The third arm 303 extends from the end connected to the second arm 302 to the other end. The third arm 303 rotates about the axial direction of the first to third rotating shafts 306 to 308 due to the rotation of the first to third rotating shafts 306 to 308, which are located further on the fixing position P side than the third arm 303.

[0073] The fourth rotating shaft 309 is connected to the other end of the third arm 303 and one end of the fourth arm 304. The fourth rotating shaft 309 is incorporated in a fourth reduction gear 30B-4, which decelerates and outputs the rotation of a motor (not shown), which rotates around the axial direction orthogonal to the axial direction of the first to third rotating shafts 306 to 308 by transmitting the rotation output from the fourth reduction gear 30B-4. Incidentally, in the state of Fig. 5, the axial direction of the fourth rotating shaft 309 is parallel to the Y direction, but the axial direction of the fourth rotating shaft 309 changes due to the rotation of the first to third rotating shafts 306 to 308, which have a different axial direction than the fourth rotating shaft 309 further on the side of the fixing position P than the fourth rotating shaft 309.

[0074] The fourth arm 304 extends from the end connected to the third arm 303 to the other end. The fourth arm 304 rotates around the axial direction of the first to fourth rotating shafts 306 to 309 due to the rotation of the first to fourth rotating shafts 306 to 309, which are provided further on the side of the attachment point P than the fourth arm 304. The fifth rotating shaft 310 is connected to the other end of the fourth arm 304 and one end of the fifth arm 305. The fifth rotating shaft 310 is installed in a fifth reduction gear 30B-5, which decelerates and outputs the rotation of a motor (not shown), which rotates around the axial direction parallel to the second rotating shaft 307 by transmitting the rotation output from the fifth reduction gear 30B-5. Incidentally, in the state of Fig. 5, the axial direction of the fifth rotating shaft 310 is parallel to the Y direction, but the axial direction of the fifth rotating shaft 310 changes due to the rotation of the first and fourth rotating shafts 306, 309, which have a different axial direction than the fifth rotating shaft 310, further on the side of the fixing position P than the fifth rotating shaft 310.

[0075] The fifth arm 305 extends from the end connected to the fourth arm 304 to the other end. The fifth arm 305 rotates about the axial direction of the first to fifth rotating shafts 306 to 310 due to the rotation of the first to fifth rotating shafts 306 to 310, which are provided further to the side of the attachment point P than the fifth arm 305.

[0076] As in Fig. As shown in Fig. 7, in the second modified example, a temperature-sensitive coating 6 as an example of a thermochromatic substance that changes color according to temperature is provided on the surface 3a of the industrial robot 30A, specifically on the surface of the first and fourth rotating shafts 306, 309. The temperature-sensitive coating 6 is a film formed by applying a temperature-sensitive substance to the surface of the industrial machine 3. Incidentally, the temperature-sensitive coating 6 may also be provided on the surface of the rotating shafts 307, 308, and 310 apart from the first rotating shaft 306 and the fourth rotating shaft 309, or even apart from the rotating shafts on the surface 3a of the industrial robot 30A.

[0077] As in Fig. As shown in Figure 5, the information acquisition means 11 of the second modified example includes a plurality of cameras 110 that capture the temperature-sensitive coating 6 provided on the surface 3a of the industrial robot 30A. Furthermore, the information acquisition means 11 of the second modified example includes a control device 120 that acquires the temperature of the surface of the rotating shafts 306, 309 based on the color development of the temperature-sensitive coating 6 corresponding to the images captured by the cameras 110, i.e., the temperature. By the information acquisition means 11 of the second modified example, the temperature of the surface 3a of the industrial robot 30A can be accurately detected based on the images of the temperature-sensitive coating 6 directly provided on the surface 3a of the industrial robot 30A.

[0078] The plurality of cameras 110 are positioned at a distance from the surface 3a of the industrial robot 30A, and are positioned to surround the industrial robot 30A from the sides and from above. By arranging the cameras 110 to surround the industrial robot 30A, the blind spot of the cameras 110 at the temperature-sensitive coating 6, i.e., the surface 3a of the industrial robot 30A, can be reduced.

[0079] As in Fig. As shown in Figure 7, the machine and equipment condition monitoring device 1 of the second modified example includes a drive means 7 that drives the cameras 7 and changes the imaging range of the cameras 110. The drive means 7 may also include an actuator, e.g., a motor, that rotates and shifts the direction of the light axis of the cameras 110. The operation of the drive means 7 may also be controlled by a control device 1. For example, if the temperature-sensitive coating 6 has shifted according to the movement of the industrial robot 30A, the drive means 7 may move the respective cameras 110 so that they can continuously image the temperature-sensitive coating 6 following the shift of the temperature-sensitive coating.

[0080] The control device 120 also functions as the machinery and equipment condition determination means 12 of the second modified example (i.e., condition inference means 121 and defect prediction means 122). Based on the temperature of the surface 3a of the industrial robot 30A acquired from the images of the temperature-sensitive coating 6, the control device 120 determines whether or not the reduction gear 30B installed in the industrial robot 30A is in a specific condition, such as an abnormal condition or a condition that will lead to a defect within a specific period of time.

[0081] In the example in Fig. 7, the control device 120 determines whether the first reduction gear 30B-1 installed in the first rotating shaft 306 is in a certain state based on the temperature of the first rotating shaft 306. Furthermore, the control device 120 determines whether the fourth reduction gear 30B-4 installed in the fourth rotating shaft 309 is in a certain state based on the temperature of the fourth rotating shaft 309. (Operational example)

[0082] Next, an operation example of the machine and equipment condition monitoring apparatus 1 of the second modified example will be explained. Fig. Fig. 8 is a flowchart showing an example of the operation of the machine and equipment condition monitoring device 1 according to the second modified example. The machine and equipment condition monitoring device 1 executes the processing of the flowchart of Fig. 8 is executed continuously during the operation of the industrial robot 30A. Furthermore, this flowchart is executed repeatedly as necessary.

[0083] As in Fig. 8, in the second modified example, as an acquisition step (step S1) of the physical information of the surface 3a of the industrial machine 3, which is shown in Fig. 5, steps S11 and S12 are executed. Specifically, first, the cameras 110 capture an image of the temperature-sensitive coating 6 provided on the surface 3a of the industrial robot 30A (step S11). After the image of the temperature-sensitive coating 6 has been captured, the control device 120 measures the temperature distribution of the surface 3a of the industrial robot 30A based on the captured image of the temperature-sensitive coating 6 (step S12).

[0084] Next, the state of the reduction gear corresponding to the position of the measured temperature distribution of the temperature-sensitive coating 6 is inferred (step S21). After the state of the reduction gear is inferred, the controller 120 determines whether the inferred state of the reduction gear is an abnormal state or not (step S3). That is, the controller 120 determines whether an abnormal state of the reduction gear has occurred at the specific location based on the result of comparing the temperature of the surface 3a of the industrial robot 30A at a specific position with a threshold value for determining the presence of an abnormal state (abnormality determination threshold value).Specifically, the control device 120 determines whether the temperature of the surface of the first rotating shaft 306 and the fourth rotating shaft 309 on which the temperature-sensitive coating 6 has been provided exceeds a certain threshold value or not.

[0085] The abnormal condition is, for example, a defect of the first reduction gear 30B-1 installed in the first rotating shaft 306 or the fourth reduction gear 30B-4 installed in the fourth rotating shaft 309.

[0086] Furthermore, the abnormality detection threshold is not limited to temperature, but varies according to the physical information acquired by the information acquisition means 11. Here, the "state of exceeding the abnormality detection threshold" as an example of a detection criterion of an abnormal state includes both the aforementioned case where the value of the physical information itself, such as temperature, exceeds an abnormality detection threshold, and the case where a value calculated for abnormality detection based on physical information exceeds the abnormality detection threshold (hereinafter also).

[0087] If the inferred state of the reduction gear is an abnormal state (step S3: Yes), the control device 120 outputs abnormality generation information to the outside (step S4).

[0088] On the other hand, if the inferred state of the reduction gear is not an abnormal state (step S3: No), the control device 120 determines whether or not the inferred state of the reduction gear is a state that leads to a failure within a certain period of time (step S5).

[0089] Fig. Fig. 9 is an explanatory diagram explaining the defect prediction process with respect to the operation example of the machine and equipment condition monitoring device 1 according to the second modified example of the present embodiment. As shown in Fig. 9, the control device 120 can also determine, for example, based on the result of comparing the temperature of the surface 3a of the industrial robot 30A at a specific position with a defect prediction determination threshold, whether or not it is a condition that leads to a defect within a specific period of time (step S5). In this case, for example, even if the temperature of the surface 3a of the industrial robot 30A at a specific position has exceeded the defect prediction determination threshold, the control device 120 can also determine that it is a condition that leads to a defect within a specific period of time.

[0090] Specifically, the control device 120 in the example of Fig. 9 the temporal change in the temperature of the surface of an industrial robot at a specific position with a defect temperature related to the specific position, at which a defect in the reduction gear occurs (defect prediction detection threshold). There are no particular restrictions on the specific acquisition method of the temporal change in temperature; for example, a regression analysis based on the temperature acquired up to the current detection time t2 can also be used. In the example of Fig. 9, the temporal change in temperature is acquired as a linear function of time, but it may also be acquired as a function other than a linear function. When comparing the temporal change in temperature with the defect prediction temperature, the control device 120 calculates, based on the temporal change in temperature, the time between a time t3 at which the temperature of the surface of the industrial robot is predicted to become a defect temperature and the current determination time t2 as a predetermined period of time. That is, the control device 120 determines that if the temporal change in temperature exceeds the defect temperature at time t3 after the predetermined period of time from the current determination time t2, the reduction gear will be in a defect state within the predetermined period of time.

[0091] Furthermore, the defect prediction detection threshold is not limited to temperature, but varies according to the physical information acquired by the information acquisition means 11. Furthermore, the "state of exceeding the defect prediction detection threshold" as an example of a detection criterion for defect prediction includes both the case where the value of the physical information itself exceeds a defect prediction detection threshold, and the case where a value calculated for defect prediction detection based on physical information, e.g., a temperature calculated based on the aforementioned temperature of the surface of the industrial robot, exceeds the defect prediction detection threshold (hereinafter also).

[0092] If the inferred state of the reduction gear is a state that leads to a failure within a certain period of time (step S5: Yes), the control device 120 outputs failure prediction information to the outside (step S6).

[0093] On the other hand, if the inferred state of the reduction gear is not a state that will cause a failure within a certain period of time (step S5: No), the control device 120 outputs machine and equipment state information corresponding to the inferred state of the reduction gears 30B-1, 30B-4 to the outside (step S71). According to the second modified example, by providing a temperature-sensitive coating 6 on the surface of the rotating shafts 306, 309 closest to the reduction gears 30B-1, 30B-2 and acquiring the temperature of the surface of the rotating shafts 306, 309 based on images of the temperature-sensitive coating 6, the state of the reduction gears 30B-1, 30B-4 can be detected based on the temperature without directly providing a sensor on the reduction gears. (Third modified example)

[0094] Next, a third modified example is explained in which the time-varying information from previously acquired physical information about a defect is used for defect prediction. Fig. 10 is an explanatory diagram explaining the defect prediction process with respect to the operation example of the machine and equipment condition monitoring apparatus 1 according to the third modified example of the present embodiment.

[0095] At Fig. In Figure 9 of the second modified example, an example was explained in which a defect prediction determination is performed based on whether the temperature of the surface of the industrial robot exceeds the defect temperature or not. In contrast, in the third modified example, as shown in Fig. 10, the control device 120 compares the temporal change in the temperature of the surface of the industrial robot (hereinafter also referred to as "acquired temporal change information") with the previously determined temporal change in the temperature of the surface of the industrial robot when a defect occurs (hereinafter also referred to as "defect temporal change information"), and calculates the degree of agreement between the two pieces of temporal change information. Then, if the calculated degree of agreement exceeds a defect prediction determination threshold, the control device 120 determines that the inferred state of the reduction gears is a state that will lead to a defect within a certain period of time.

[0096] More specifically, the defect time history information is information acquired in advance through tests and simulations, which, as in Fig. 10, which takes the information of the failure time t2 of the reduction gear unit as the starting point, with the start time of the industrial machine 3 as the starting point. After the industrial machine 3 is started, the control device 120 calculates the degree of agreement between the acquired time history information and the defect time history information at the determination time t1 back from the failure time t2 by the specified period of time, and performs a determination by comparing the calculated degree of agreement with a defect prediction determination threshold. The degree of agreement between the acquired time history information and the defect time history information is not limited to any specific form, as long as it is a value that is larger the smaller the difference (i.e., the temperature difference) between the time history information acquired up to the determination time t1 and the defect time history information.The degree of agreement may, for example, also be a value proportional to the inverse of the average value of the difference between the time-history information acquired up to the detection time t1 and the defect time-history information. According to the third modified example, a defect can be predicted by comparing the acquired time-history information with the defect time-history information without providing a sensor on the reduction gears, even if a defect cannot be easily predicted using a linear function or another function. (Fourth modified example)

[0097] Next, a fourth modified example will be explained in which the state of the machines and devices 2 to be monitored is inferred by comparing it with previously acquired physical information at a time when the machines and devices 2 to be monitored are in a certain state (without limitation to a defective state). Fig. 11 is an explanatory diagram explaining the inference process for the status of the monitored machines and devices 2 with respect to the operation example of the machine and device status monitoring apparatus 1 according to a fourth modified example of the present embodiment. As shown in Fig. 11, the machine and device state detection means 12 in the fourth modified example acquires in advance first physical information at a time when the machines and devices 2 to be monitored are in a certain state (without limitation to a defect state). In the example of Fig. 11, the first physical information is time-history information, which represents a change in physical information over time. The time-history information can be acquired, for example, through pre-conducted experiments or simulations. Furthermore, the time-history information can vary according to conditions such as the operating environment of the industrial robot 30A or the operating period.

[0098] The machine and device condition determination means 12 compares second physical information acquired by the information acquisition means 11 with the first physical information. Then, if the second physical information matches the first physical information, the machine and device condition determination means 12 determines that the machines and devices 2 to be monitored are in a certain state (without limitation to a defective state). The matching is not limited to a strict match, but may also include the case where the deviation of the second physical information from the first physical information is at or below a threshold value (i.e., the degree of match exceeds a threshold value).On the other hand, if the second physical information does not match the first physical information, the machine and device state determining means 12 determines that the machines and devices 2 to be monitored are not in a certain state.

[0099] According to the fourth modified example, it can therefore be easily inferred whether the machines and devices 2 to be monitored are in a certain state (without limitation to a defective state) or not by comparing the second physical information acquired by the information acquiring means 11 with first information known at the certain state (without limitation to a defective state) of the machines and devices 2 to be monitored. (Fifth modified example)

[0100] Next, a fifth modified example is explained in which a calculation model for defect prediction is created based on the maintenance history of the industrial machine 3, the condition history, the defect history and the maintenance history of the machines and devices 2 to be monitored in the past, and a defect prediction is carried out on the basis of this calculation model. Fig. 12 is an explanatory diagram explaining the defect prediction process with respect to the operation example of the machine and equipment condition monitoring apparatus 1 according to the fifth modified example of the present embodiment.

[0101] The defect prediction means 122 determines whether a condition is likely to cause a defect within a certain period of time based on a condition detection model generated from history information, which is the result of condition detection of installed items based on acquired physical information.

[0102] Specifically, the defect prediction means 122 includes a history database 1223 that stores a maintenance history of the industrial machine 3, which is historical information including at least a time at which maintenance of the industrial machine 3 was performed; a status history of the machines and devices 2 to be monitored, which stores at least the status of the machines and devices 2 to be monitored combined with the time at which this status occurred; a defect history of the machines and devices 2 to be monitored, which is historical information including at least a time at which the machines and devices 2 to be monitored had a defect; and a maintenance history of the machines and devices 2 to be monitored, which is historical information including at least a time at which maintenance of the machines and devices 2 to be monitored was performed.

[0103] If the history database 1223 also includes all of these histories, a defect prediction with greater precision is possible, although it does not necessarily have to include all of them; it is sufficient if it includes at least one. Furthermore, although the longer the retention period, the higher the precision with which a defect can be predicted, this can be appropriately determined based on the relationship between the product life of the machines and devices 2 to be monitored and the defect frequency of the machines and devices 2 to be monitored, as well as the volume of the storage medium (not shown) that stores the history database.

[0104] In the present modified example, the storage means and the state inference means 1221, which infers the state of the machines and devices 2 to be monitored, are physically provided as one piece, and the problem of the storage volume is effectively solved by using an external storage means, such as a server in a cloud via a communication line, such as the Internet, as the storage means.

[0105] The defect prediction means 122 further comprises a model creation means 1224, which creates a computational model for the defect prediction based on the various histories stored in the history database 1223. The model creation means 1224 creates, for example, a computational model for the defect prediction using a generally known model creation method in such a way that it is optimally suited to the condition history of the machines and devices 2 to be monitored and the temporal progression stored in the defect history of the machines and devices 2 to be monitored.

[0106] The model creation means 1224 may also provide two or more calculation models for defect prediction. For example, a normal state model creation means 12241 may be provided, which creates a calculation model of the normal state from the state history of the machines and devices 2 to be monitored using the historical information for which no defect occurred within the specified period of time, and a defect state model creation means 12242 may be provided, which creates a calculation model of the defective state using the historical information for which a defect occurred within the specified period of time. Based on the reliability of the assessment of the normal state and the defective state, three or more calculation models may also be provided. This allows for a higher-reliability assessment.

[0107] For example, the defect prediction means 122 determines that it is a condition that will lead to a defect within a certain period of time if the condition of the machines and devices 2 to be monitored, which is input by the condition inference means 1221, is more similar to the defect condition model than to the normal condition model.

[0108] In the present modified example, the model creation means 12241 is physically integrated with, for example, a defect prediction calculation means 1222, although this is not limited to this, and a calculation model can also be created, for example, using a server in a cloud via a communication line, such as the Internet. This has the advantage, for example, that the calculation speed is comparatively faster than a combination of machines and devices, and the model creation algorithm is easier to subsequently modify. The calculation model created by the model creation means 1224 is sent to the defect prediction calculation means 1222. Then, the inferred information, which is the information inferred by the state inference means 1221, is input into the model, and a defect prediction is performed based on the calculation result.

[0109] In the event that there are several calculation models, the same information inferred by the state inference means 1221 is input into these models, and based on their calculation results, a defect prognosis is carried out on the basis of a predetermined evaluation basis.

[0110] In the present modified example, the defect prediction calculation means 1222 is physically integrated with, for example, the state inference means 1221, although this is not limited to this, and calculations can also be performed, for example, using a server in a cloud via a communication line, such as the Internet. This has the advantage, for example, that the calculation speed is comparatively faster than a combination of machines and devices, and the specified evaluation basis is easier to subsequently change. (Sixth modified example)

[0111] Next, application examples of the machine and device condition monitoring device 1 to industrial machines 3 other than the industrial robot 30A as a sixth modified example will be explained. Fig. 13 is a diagram showing, as the machine and equipment condition monitoring device 1 according to a modified example of the fifth modified example of the present embodiment, an application example to travel motors 30E for a construction machine. Fig. 14 is a diagram showing an application example to an automatic door 30I as the machine and equipment status monitoring device 1 according to a modified example of the sixth modified example of the present embodiment.

[0112] In the example of Fig. 13, the information acquisition means 11 (not shown) of the machine and equipment condition monitoring device 1 acquires physical information (e.g., strain or vibration of the surface of the construction machine travel motor 30D1) exhibited outside the construction machine travel motor 30D1 based on images of the travel motor 30D1 for a construction machine taken by a camera 110 installed on the construction machine 30D. Then, the machine and equipment condition determination means 12 of the machine and equipment condition monitoring device 1 performs condition determination (e.g., abnormality determination or defect prediction determination) of the monitored machines and equipment (e.g., a planetary gear) installed in the travel motor 30D1 for a construction machine based on the acquired physical information.

[0113] In the example of Fig. 14, the information acquiring means 11 (not shown) of the machinery and equipment condition monitoring device 1 acquires physical information (e.g., the temperature, which is reflected as color development of the temperature-sensitive coating 6) of the surface of the crossbeam part of the automatic door 30I based on images of a crossbeam part of the automatic door 30I taken by a camera 110 installed near the automatic door 30I (e.g., on the ceiling). Then, the machinery and equipment condition determining means 12 of the machinery and equipment condition monitoring device 1 performs a condition determination (e.g., an abnormality determination or a defect prediction determination) of an electric motor 30I1 installed in the crossbeam part based on the acquired physical information.

[0114] In the example of Fig. 15, the information acquisition means 11 (not shown) of the machinery and equipment condition monitoring device 1 acquires physical information (e.g., the temperature, which is reflected in the color development of the temperature-sensitive coating 6) of the surface of the nacelle of the wind turbine 30J based on images of the nacelle of the wind turbine 30J taken by a camera 110 mounted on a remotely controlled aircraft DR (drone) flying near the wind turbine 30J for a wind power generator. Then, the machinery and equipment condition determination means 12 of the machinery and equipment condition monitoring device 1 performs a determination (e.g., an abnormality determination or a defect prediction determination) of the accelerator 30J1 and the drive motor 30J2 installed in the nacelle based on the acquired physical information.

[0115] In the example of Fig. 16, the information acquisition means 11 (not shown) of the machine and equipment condition monitoring device 1 acquires physical information (e.g., the temperature, which is reflected as color development of the temperature-sensitive coating 6) of the surface of the drive device of the heliostat 30K based on images of the drive device of a heliostat 30K taken by a camera 110 mounted on a remote-controlled aircraft DR (drone) flying near the heliostat 30K for a solar generator. Then, the machine and equipment condition determination means 12 of the machine and equipment condition monitoring device 1 performs a condition determination (e.g.,an abnormality determination or a defect prediction determination) of a reduction gear 30K1 or an electric motor 30K2 driving the reduction gear.

[0116] According to the sixth modified example, by applying the machine and equipment condition monitoring device 1 to various industrial machines 3, the universal applicability of the machine and equipment monitoring device 1 can be improved. (Seventh Modified Example) Next, a sixth modified example will be explained in which the condition of the machines and equipment 2 to be monitored is determined based on multiple physical information.

[0117] Fig. 17 is a block diagram showing details of the machine and equipment state detecting means 12 in the machine and equipment state monitoring apparatus 1 according to a seventh modified example of the present embodiment. As shown in Fig. 17, in the seventh modified example, the information acquisition means 11 can acquire a plurality of physical information #1 to #n (n is a natural number ≧ 2; hereinafter, the same) of the industrial machine 3 that appears outside the industrial machine 3. There is no particular limitation on the specific implementation of the plurality of physical information, and it may be, for example, a combination of two or more types selected from temperature, position, strain, displacement, vibration, sound waves or ultrasonic waves, the reflection rate of infrared light or other light, the absorption rate of electromagnetic waves, hue, brightness, color saturation, moisture content, fat content, sound, odor, intensity of X-rays or other rays, electromagnetic waves or ultrasonic waves, or excretions such as exhaust gases.The concrete design of the information acquisition means 11 for acquiring this physical information corresponds to the example already presented. (State-Conclusion Means 121)

[0118] The state inference means 121 infers the state of the machines and devices 2 to be monitored based on the acquired plurality of physical information #1 to #n in the case where a plurality of physical information #1 to #n has been acquired. In addition, the state inference means 121 determines whether or not the inferred state of the machines and devices 2 to be monitored is a specific state, such as an abnormal state or a state that leads to a failure within a specific period of time.

[0119] The state inference means 121 determines whether the state of the monitored machinery and equipment 2 is an abnormal state or not based on the result of a comparison with the abnormality determination threshold value corresponding to each acquired plurality of physical information #1 to #n (hereinafter also referred to as "individual abnormality determination threshold value"). Furthermore, the state inference means 121 determines whether the state of the monitored machinery and equipment 2 is an abnormal state or not based on the result of a comparison with a set of abnormality determination threshold values ​​corresponding to the combination of the acquired plurality of physical information #1 to #n.

[0120] The individual abnormality determination threshold is a threshold value that can be used to determine whether the state of the monitored machinery and equipment 2 is abnormal based on the individual values ​​of the plurality of physical information items #1 to #n. The individual abnormality determination threshold is, for example, a value greater than a certain threshold value of the physical information. The state inference means 121 can also determine that the state of the monitored machinery and equipment 2 is abnormal if at least one piece of physical information exceeds the corresponding individual abnormality determination threshold.The set of abnormality detection thresholds is a threshold that can be used to determine whether the state of the monitored machinery and equipment 2 is abnormal based on a combination of the multiple pieces of physical information #1 to #n. The set of abnormality detection thresholds may, for example, be a combination of thresholds that are smaller than the individual abnormality detection thresholds for the respective multiple pieces of physical information #1 to #n. Furthermore, the number of thresholds constituting the set of abnormality detection thresholds may be smaller than the number of measured physical information items, or it may be a single threshold.

[0121] The state inference means 121 can also perform an abnormality determination based on the set of abnormality detection thresholds if no abnormal condition was detected during the abnormality determination based on the individual abnormality detection thresholds. Depending on the state of the machines and devices 2 to be monitored, an abnormality determination based on the set of abnormality detection thresholds can also be performed instead of the abnormality determination based on the individual abnormality detection thresholds. In the event that the state of the machines and devices 2 to be monitored is not suitable for an abnormality determination using an individual abnormality detection threshold, e.g.In the case where there is an interrelationship between the plurality of physical information #1 to #n as described above, abnormality detection can also be performed only on the basis of the set of abnormality detection thresholds.

[0122] Even if it cannot be determined that an abnormal condition exists because physical information #1 to #n does not exceed the individual abnormality detection threshold during an individual assessment, an abnormal condition must be determined during an overall assessment in which multiple pieces of physical information #1 to #n are combined. For example, if all of the multiple pieces of physical information #1 to #n show a value that does not exceed the individual abnormality detection threshold but is close to the individual abnormality detection threshold, it may be necessary to judge an abnormal condition when viewed as a whole.Furthermore, depending on the condition of the monitored machinery and equipment 2, as a determination method for the condition of the monitored machinery and equipment 2, performing abnormality determination by an overall judgment of a combination of physical information #1 to #n may be more suitable than performing abnormality determination based on an individual judgment of physical information #1 to #n. According to the fifth modified example, by performing abnormality determination based on both an individual abnormality determination threshold and a set of abnormality determination thresholds, an abnormal condition can be appropriately detected.

[0123] If it is determined that the state of the machines and devices 2 to be monitored is an abnormal state, the state inference means 121 outputs abnormality generation information to the outside. (Defect Forecasting Tool 122)

[0124] The defect prediction means 122 determines whether the condition of the monitored machinery and equipment 2 is a condition that leads to a defect within a certain period of time based on the result of a comparison with the defect prediction determination thresholds (hereinafter also referred to as "individual prediction determination thresholds") corresponding to the respective acquired plurality of physical information #1 to #n. Furthermore, the defect prediction means 122 determines whether the condition of the monitored machinery and equipment 2 is a condition that leads to a defect within a certain period of time based on the result of a comparison with a set of defect prediction determination thresholds corresponding to the combination of the acquired plurality of physical information #1 to #n.The individual prediction determination thresholds are thresholds that can be used to determine, based on the individual values ​​of the multiple physical information items #1 to #n, whether the condition of the monitored machinery and equipment 2 is a condition that will lead to a failure within a certain period of time. The individual prediction determination thresholds can also be, for example, the failure temperature, which is illustrated by the example of [the figure]. Fig. 9, or the threshold of the degree of agreement of the time history information, which is shown in the example of Fig. 10. The defect prediction means 122 can also determine that the condition of the machines and devices 2 to be monitored is a condition that will lead to a defect within a certain period of time if, by comparison with the individual prediction determination threshold, it is judged that at least one piece of physical information exceeds the individual prediction determination threshold.

[0125] The set of defect prediction detection thresholds are thresholds that can be used to determine, based on a combination of the multiple pieces of physical information #1 to #n, whether the condition of the machines and devices 2 to be monitored is a condition that will lead to a defect within a certain period of time. The set of defect prediction detection thresholds can, for example, also be a combination of thresholds that are smaller than the individual prediction detection thresholds corresponding to the respective multiple pieces of physical information #1 to #n. Furthermore, depending on the condition of the machines and devices 2 to be monitored, the number of thresholds constituting the set of defect prediction detection thresholds can also be fewer than the number of physical information items, and it can also be a single threshold.

[0126] If the defect prediction means 122 does not detect a condition that leads to a defect within a certain time period, the defect prediction means 122 can also perform a defect prediction determination based on the set of defect prediction determination thresholds. Furthermore, if the condition inference means 121 judges that the condition is not abnormal, the defect prediction means 122 can also perform a defect prediction determination based on the set of defect prediction determination thresholds. Depending on the condition of the machines and devices 2 to be monitored, a defect prediction determination can also be performed based on the set of defect prediction determination thresholds instead of the defect prediction determination based on the individual forecast determination thresholds.

[0127] It may happen that even if a defect is not predicted because the multiple pieces of physical information #1 to #n do not exceed the individual prediction determination thresholds in an individual assessment, it can be judged that signs of a defect are showing in an overall assessment in which the multiple pieces of physical information #1 to #n are combined. Furthermore, it may happen that, depending on the condition of the machines and equipment 2 to be monitored, performing a defect prediction determination using an overall assessment of a combination of the physical information #1 to #n is more suitable as a determination method for the condition of the machines and equipment 2 to be monitored than performing a defect prediction determination based on an individual assessment of the physical information #1 to #n.According to the seventh modified example, based on both the individual prediction detection thresholds and the set of defect prediction detection thresholds, a future defect can be appropriately predicted. (Operational example)

[0128] Next, an operation example of the machine and equipment condition monitoring apparatus 1 of the seventh modified example will be explained. Fig. 18 is a flowchart showing an example of the operation of the machine and equipment condition monitoring apparatus 1 according to the seventh modified example of the present embodiment. As shown in Fig. As shown in Figure 18, the information acquisition means 11 first acquires physical information exhibited outside the industrial machine 3 (step S1). After acquiring the physical information, the information acquisition means 11 determines whether the acquired physical information is only one (ie, one type) or not (step S8).

[0129] If the physical information is only one (step 8: Yes), the information acquisition means 11 infers the state of the machines and devices 2 to be monitored based on the acquired physical information (step S201). Next, the state inference means 121 determines whether the physical information exceeds the abnormality detection threshold (step S31).

[0130] If the abnormality detection threshold is exceeded (step S31: Yes), the state inference means 121 determines that the state of the monitored machinery and equipment 2 is an abnormal state and outputs abnormality generation information to the outside (step S4). On the other hand, if the abnormality detection threshold is not exceeded (step S31: No), the state inference means 121 determines that the state of the monitored machinery and equipment 2 is not an abnormal state.

[0131] If it is determined that the state of the machines and devices 2 to be monitored is not an abnormal state, the defect prediction means 122 determines whether the physical information exceeds the defect prediction determination threshold or not (step S51).

[0132] If the defect prediction determination threshold is exceeded (step S51: Yes), the defect prediction means 122 judges that the state of the machines and devices 2 to be monitored is a state that leads to a defect within a certain period of time (ie, a defect is predicted), and outputs defect prediction information to the outside (step S6).

[0133] On the other hand, if the defect prediction detection threshold is not exceeded (step S51: No), the defect prediction means 122 outputs machine and device state information corresponding to the inferred state of the machines and devices 2 to be monitored to the outside (step S7).

[0134] Fig. 19 is a flowchart showing the continuation of Fig. 18 shows the example of operation of the machine and equipment condition monitoring apparatus 1 according to the seventh modified example of the present embodiment.

[0135] If the acquired physical information is not only a (step S8 of Fig. 18: No), the information acquisition means 11 infers the state of the machines and devices 2 to be monitored based on the acquired plurality of physical information (step S203).

[0136] Next, the state inference means 121 determines whether or not at least one of the plurality of physical information #1 to #n exceeds the individual abnormality determination threshold (step S32).

[0137] If an individual abnormality detection threshold is exceeded (step S32: Yes), the state inference means 121 judges that the state of the monitored machinery and equipment 2 is abnormal and outputs abnormality generation information to the outside (step S41). On the other hand, if the individual abnormality detection threshold is not exceeded (step S32: No), the state inference means 121 determines whether the plurality of physical information items #1 to #n exceed the set of abnormality detection thresholds (step S33).

[0138] If the set of abnormality detection thresholds is exceeded (step S33: Yes), the state inference means 121 judges that the state of the monitored machinery and equipment 2 is an abnormal state and outputs abnormality generation information to the outside (step S41). On the other hand, if the set of abnormality detection thresholds is not exceeded (step S33: No), the state inference means 121 judges that the state of the monitored machinery and equipment 2 is not an abnormal state.

[0139] If it has been judged that the state of the machines and devices 2 to be monitored is not an abnormal state, the defect prediction means 122 determines whether or not at least one of the plurality of physical information #1 to #n exceeds the individual prediction determination threshold (step S52).

[0140] If the individual prediction determination threshold is exceeded (step S52: Yes), the defect prediction means 122 judges that the state of the machines and devices 2 to be monitored is a state that leads to a defect within a certain period of time (ie, a defect is predicted), and outputs defect prediction information to the outside (step S61).

[0141] On the other hand, if the individual prediction determination threshold is not exceeded (step S53: No), the defect prediction means 122 determines whether or not the plurality of physical information #1 to #n exceeds the set of defect prediction determination thresholds (step S53).

[0142] If the set of defect prediction determination thresholds is exceeded (step S53: Yes), the defect prediction means 122 judges that the state of the machines and devices 2 to be monitored is a state that leads to a defect within a certain period of time and outputs defect prediction information to the outside (step S61). On the other hand, if the set of defect prediction determination thresholds is not exceeded (step S53: No), the defect prediction means 122 judges that the state of the machines and devices 2 to be monitored is not a state that leads to a defect within a certain period of time and outputs machine and device state information corresponding to the state of the machines and devices 2 to be monitored (step S72).

[0143] Fig. 20 is an explanatory diagram explaining the defect prediction process based on the combination of data from a plurality of physical quantities, with respect to the operation example of the machine and equipment condition monitoring device 1 according to the seventh modified example of the present embodiment. Fig. Figure 20 shows the degree of agreement of the time history information of the physical information #1 to #3 and the magnitude relationship to the individual prognostic determination thresholds as a result of a comparison with the individual prognostic determination thresholds corresponding to the physical information #1 to #3, respectively. Fig. 20 further shows, as a result of comparison with the set of defect prediction detection thresholds corresponding to the physical information #1 to #3, the respective matching degree of the time history information of the physical information #1 to #3 and the size relationship to the set of defect prediction detection thresholds.

[0144] In the example of Fig. 20, the respective degree of agreement of the time course information of the physical information #1 to #3 is smaller than the corresponding individual prediction determination threshold. Therefore, in the example of Fig. 20 in the defect prediction determination (step S52) by the Fig. 19, a negative result is obtained. On the other hand, in the example of Fig. 19 the combination of the respective agreement degrees of the time history information of the physical information #1 to #3 is greater than the set of defect prediction detection thresholds. Therefore, in the example of Fig. 19 in the defect prediction determination (step S53) by the Fig. 18, a positive detection result is achieved. That is, in the example of Fig. 19, it is determined that this is a condition that will lead to a defect within a certain period of time. According to the seventh modified example, an abnormal condition can be appropriately detected based on both the individual prediction detection thresholds and the set of defect prediction detection thresholds. Furthermore, a future defect can be appropriately predicted based on both the individual prediction detection thresholds and the set of defect prediction detection thresholds. (Eighth modified example)

[0145] Next, an eighth modified example is explained in which the state of the monitored machines and equipment 2 is determined using the state at the start time as a reference, which is the time at which the monitored machines and equipment 2 start operating. The start time of operation is the time when the monitored machines and equipment 2 stop operating, for example, during the night, and resume operating the next morning. Fig. 21 is an explanatory diagram explaining, with respect to the operation example of the machine and equipment condition monitoring apparatus 1 according to an eighth modified example of the present embodiment, the determination process for the condition of the monitored machine and equipment 2.

[0146] In the modified eighth example, the information acquisition means 11 acquires, as physical information of the surface 3a of the industrial robot 30A, the displacement of the surface of the first to third rotating shafts 306 to 309 (see FIG. Fig. 21). The specific method for acquiring the displacement of the surface of the first to third rotating shafts 306 to 309 is not particularly limited. For example, the information acquisition means 11 may also acquire the traces of the markings as displacements of the surface of the rotating shafts 306 to 308 based on images of markers provided at a specific location on the surface of the first to third rotating shafts 306 to 308, which were captured by cameras 110.

[0147] The information acquisition means 11 acquires the displacement of the surfaces of the first to third rotating shafts 306 to 309 both at the start-up time of the industrial robot 30A and at the time of detecting the state of the machinery and equipment. At both the start-up time and the detection time, the control device 120 inputs the same operation command to the industrial robot 30A, and the information acquisition means 11 acquires the displacement of the surfaces of the first to third rotating shafts 306 to 308 of the industrial robot 30A operating under the same operation command.

[0148] The control device 120 calculates, for each of the first to third rotating shafts 306 to 308, the deviation between the surface displacement at the start time and the surface displacement at the detection time (hereinafter also referred to as displacement deviation). The specific type of displacement deviation is not particularly limited, and may be, for example, the difference in the total displacement amount during the aforementioned operation under the same operation command (e.g., the moving distance of the markers). The control device 120 determines whether or not the first to third reduction gears 30B-1 to 30B-3 installed in the first to third rotating shafts 306 to 308 are in an abnormal state based on a comparison of the calculated displacement deviation with a threshold value.The control device 120 determines that the reduction gears 30B-1 to 3B-3 are in an abnormal state when the displacement deviation exceeds the threshold value.

[0149] When determining the presence of an abnormal condition, the control device 120 may also perform a determination taking into account the position of the surface whose displacement was acquired. In the example of Fig. 21, for example, the third rotating shaft 308 is positioned at a position farther away from the attachment point P than the first rotating shaft 306 and the second rotating shaft 307. Therefore, by adding the displacement deviation of the first and second rotating shafts 306, 307, the displacement deviation of the third rotating shaft 308 can become larger than the displacement deviation of the first and second rotating shafts 306, 307. As a result, the threshold value of the displacement deviation of the third rotating shaft in determining the presence of an abnormal state can be larger than the threshold value of the displacement deviation of the first rotating shaft 306 and the second rotating shaft 307. Thereby, the state of the third reduction gear 30B-3 installed in the third rotating shaft 308 can be accurately determined.

[0150] According to the eighth modified example, by determining whether the reduction gear 30B is in an abnormal state or not based on the state at the start time, external influences such as variations in physical information caused by different operating environments such as seasons and temperatures can be reduced, and the accuracy of determining the state of the reduction gear 30B can be improved. Incidentally, the above explanation has been given for determining an abnormal state, but the same applies to determining a state leading to failure within a certain period of time.

[0151] Furthermore, the control device 120 can also determine whether the reduction gear 30B is in a certain state or not by referring to the state of the reduction gear 30B at the time of installation of the industrial robot 30A. By referring to the state at the time of installation, the time for replacing the reduction gear 30B due to aging wear can be detected. (Ninth modified example)

[0152] Next, as a ninth modified example, an example of an operation detecting device equipped with the machine and equipment state monitoring device 1 will be explained. Fig. 22 is a diagram showing an operation detecting apparatus 10 according to the ninth modified example of the present embodiment.

[0153] The device 10 for detecting an operation comprises the device 1 for monitoring the state of machines and devices and a computer 8. The device 10 for detecting an operation detects the operation of the industrial robot 30A in a virtual space of a memory area 81 on the computer 8. In the example of Fig. 22, the device 10 for detecting an operation in the virtual space detects the operation of a plurality of industrial robots 30A arranged in a real space to form a production line. In the example of Fig. 22, a device 1 for monitoring the status of machines and devices is provided for each industrial robot 30A, whereby the number of devices 1 for monitoring machines and devices does not necessarily have to correspond to the number of industrial robots 0. For example, one device 1 for monitoring machines and devices can also monitor the status of several industrial robots 30A.

[0154] The operation detection device 10 includes an input means 82. The input means 82 inputs the information acquired by the machine and equipment monitoring device 1 as information regarding the state of the industrial robot 30A into the storage area 81. The information acquired by the machine and equipment monitoring device 1 includes, for example, the aforementioned information acquired by the machine and equipment state detection means 12 (i.e., the abnormality generation information, the defect prediction information, and the machine and equipment state information). The information acquired by the machine and equipment monitoring device 1 may also include physical information acquired by the information acquisition means 11. The input means 82 may also be, for example, a CPU.The input means 82 can also regularly acquire and input information from the device 1 for monitoring machines and devices.

[0155] According to the ninth modified example, the current operating state of the industrial robot 30A in a real space can be captured in a virtual space. This allows the operation of the industrial robot 30A in the real space to be simulated in the virtual space, and furthermore, a simulation result of a process change performed in the virtual space can be reflected in the real space. (Tenth modified example)

[0156] Next, a tenth modified example will be explained in which the state of an object to be monitored installed in a civil engineering structure 30L is monitored by means of the machine and equipment monitoring device 1. Fig. Fig. 23 is a block diagram showing the machine and equipment condition monitoring apparatus 1 according to the tenth modified example of the present embodiment. In the example of Fig. 1, the device 1 for monitoring the condition of machines and equipment is independent of the civil engineering structure 30L.

[0157] A civil engineering structure 30L can be, for example, the wind turbine of a wind power generator, the heliostat of a solar generator tower, an elevated road, a bridge, or a building. A monitoring object 20 can be, for example, a structural component provided inside the civil engineering structure 30L, such as a steel framework or reinforced concrete, a fastening component provided inside, such as a screw, or various types of pipes laid inside for the water and sewage system or for electrical wiring. Furthermore, a section inside the concrete of a civil engineering structure or a concrete building structure, such as a wind turbine of a wind power generator, the tower for a solar generator, an elevated road, a bridge, or a building, can be cited as a monitoring object 20.Furthermore, roads and footpaths may also be listed as civil engineering structures 30L, and multi-purpose tunnels buried beneath roads and paths, water pipes, gas pipes and their connecting sections may be listed as objects to be monitored.

[0158] The rest is identical to the present embodiment and the other modified examples of the present embodiment, so this explanation is omitted.

[0159] In the above embodiment and the respective modified examples, condition monitoring was mainly explained, but this is not limited to this, and it is also possible to acquire the condition of the machines and devices 2 to be monitored at any time during any period, if necessary. In this case, it is also possible for physical information to be acquired by the information acquisition means 11 only at any time during any period, or for the information acquisition means 11 to continuously acquire physical information, and the machine and device condition determination means 12 to perform the determination only at any time during any period.

[0160] The above embodiment and the respective modified examples can also be combined as appropriate, and individual structures of the above embodiment or the respective modified examples can also be arbitrarily combined or partially omitted unless there is a special indication to this effect.

[0161] The implementation of the present invention is not limited to the above embodiment, and various modifications conceivable by a person skilled in the art are also included, and the effects of the present invention are not limited to the above content. That is, various additions, modifications, and partial omissions are possible within a scope that does not deviate from the conceptual idea and purpose of the present invention, which can be derived from the content defined in the claims and their equivalents.

Claims

[1] Device for detecting the state of a built-in object (2), which an information acquisition means (11) which acquires at least one physical information item which is visible outside an object (3) in which an object (2) is installed, the state of which cannot be directly detected from the outside, and a state detection means (12) which detects the state of the object (2) on the basis of one or more acquired physical information, wherein the condition determining means (12) determines that it is an abnormal condition of the object or a condition that leads to a defect within a certain period of time when a first physical information acquired in advance as physical information at a time point that it is an abnormal condition of the object (2) or a condition that leads to a defect within a certain period of time and a second physical information acquired by the information acquiring means (11) as needed match. [2] An apparatus (1) for detecting the state of an installed object according to claim 1, wherein said information acquiring means (11) acquires a plurality of physical information, and said state determining means (12) determines the state of the object based on the acquired plurality of physical information. [3] The device (1) for detecting the state of an installed article according to claim 1 or 2, wherein the state determining means (12) comprises an abnormality determining means (121) for determining whether the state of the article (2) is an abnormal state or not. [4] The device (1) for detecting the state of an installed article according to claim 3, wherein the abnormality determination means (121) determines whether the state of the article (2) is an abnormal state or not with respect to the state of the article (2) on the basis of at least either the result of a comparison of the acquired plurality of physical information with an abnormality determination threshold corresponding to each plurality of physical information, or the result of a comparison of a combination of the acquired plurality of physical information with a combined abnormality determination threshold corresponding to the combination of the plurality of physical quantities. [5] Device (1) for detecting the state of an installed article according to one of claims 1 to 4, wherein the state detecting means (12) comprises a defect predicting means (122) which determines whether the state of the article (2) is a state which leads to a defect within a certain period of time or not. [6] The device (1) for detecting the state of an installed article according to claim 5, wherein the defect prediction means (122) determines whether the state of the article (2) is a state that leads to a defect within a certain period of time or not based on at least one of the result of a comparison of the acquired plurality of physical information with a defect prediction determination threshold corresponding to the respective plurality of physical information, and the result of a comparison of a combination of the acquired plurality of physical information with a combination of defect prediction determination thresholds corresponding to the combination of the plurality of physical quantities. [7] The device (1) for detecting the state of an installed object according to any one of claims 1 to 4, wherein the state detecting means (12) comprises a defect predicting means (122) which determines whether or not it is a state that leads to a defect within a certain time based on a state detecting model created from history information which is the result of detecting the state of the installed object based on acquired physical information. [8] The device (1) for detecting the state of an installed article according to claim 7, wherein the state determination model consists of two models, namely a defective state model created from historical information of an installed article that had a defect, and a normal state model created from historical information of an installed article that did not have a defect, wherein the defect prediction means determines that it is a condition that leads to a defect within a certain period of time when the state of the article is more similar to the defective state model than to the normal state model. [9] Device (1) for detecting the state of an installed object according to one of claims 1 to 8, wherein the physical information appearing outside the object (3) also includes physical information of the surface (3a) of the object (3). [10] The device (1) for detecting the state of an installed object according to claim 9, wherein the physical information of the surface (3a) of the object (3) includes at least one of the temperature of the surface (3a) of the object (3), the position, the strain, the displacement, the vibration, the hue, the brightness, the color saturation, the moisture content, the fat content, as well as the sound waves, the ultrasonic waves and the reflection rate of infrared light or other light. [11] Device (1) for detecting the state of an installed object according to one of claims 1 to 9, wherein the physical information exhibited outside the object (3) comprises at least one of sound, smell, ultrasonic waves, electromagnetic waves, radiation or excretions detectable outside the object (3). [12] The device (1) for detecting the state of an installed object according to one of claims 9 or 10, wherein the information acquiring means (11) acquires the physical information on the basis of image recordings of at least a part of the surface (3a) of the object (3). [13] The device (1) for detecting the state of an installed object according to claim 12, wherein a thermochromatic substance (6) which changes color according to temperature is provided on the surface (3a) of the object (3), and the information acquiring means (11) acquires the temperature of the surface (3a) of the object (3) based on images of the thermochromatic substance (6). [14] The device (1) for detecting the state of an installed object according to claim 12 or 13, wherein the information acquiring means (11) has a recording magnification function by which at least a part of the surface (3a) of the object (3) is recorded in an enlarged manner. [15] The device (1) for detecting the state of a built-in object according to one of claims 13 or 14, further comprising a driving means (7) which drives the information acquiring means (11) and changes the imaging area. [16] Device (1) for detecting the state of an installed object according to one of claims 13 to 15, wherein the information acquisition means (11) comprises a plurality of cameras (110) each recording different locations of the surface (3a) of the object (3). [17] The device (1) for detecting the state of an installed object according to claim 16, wherein the plurality of cameras (110) are arranged to surround the object (3). [18] Device (1) for detecting the state of an installed object according to one of claims 13 to 15, wherein the information acquisition means (11) comprises an unmanned aircraft equipped with a camera (110) and a control means for the unmanned aircraft which records the surface of the object with the camera. [19] An apparatus (1) for detecting the state of an installed object according to any one of claims 13 to 18, wherein the information acquiring means (11) comprises an optical system capable of wide-angle or omni-azimuth imaging. [20] The device (1) for detecting the state of an installed object according to any one of claims 13 to 19, wherein the information acquiring means (11) acquires at least a part of the three-dimensional shape of the surface (3a) of the object (3) on the basis of the image recordings, and acquires physical information on the basis of the acquired three-dimensional shape. [21] Device (1) for detecting the state of an installed object according to one of claims 13 to 20, wherein the information acquisition means (11) is arranged away from the surface (3a) of the object (3), at least above the object (3). [22] Device (1) for detecting the state of an installed object according to one of claims 1 to 21, wherein the state detecting means (12) determines with reference to the state during erection or installation of the object (3) whether it is an abnormal state of the object (2) or a state has been reached which leads to a defect within a certain period of time. [23] Device (1) for detecting the state of an installed object according to one of claims 1 to 22, wherein the object (3) is involved in an operation, and the state determination means (12) determines with reference to the state at the start of operation of the object (3) whether it is an abnormal state of the object (2) or a state has been reached which leads to a defect within a certain period of time. [24] The device (1) for detecting the state of an installed object according to any one of claims 1 to 23, wherein the object (3) is an industrial robot (30A) having at least one rotating shaft (306 to 310), and the object (2) is a reduction gear (30B) installed in the rotating shaft (306 to 310), and the information acquiring means (11) acquires physical information of the surface of the rotating shaft (306 to 310). [25] An operation detection device (10) which detects, in a virtual space on a computer (8), the operation of an object in which an article is installed, the state of which cannot be directly detected from the outside, the device (10) for detecting the operation of an object into which the state of the article acquired by the device (1) for detecting the state of a built-in article according to any one of claims 1 to 24 is input as information relating to the state of the object. [26] A method for detecting the state of an object, by which at least one item of physical information of an object is acquired which is apparent outside an object in which an object is installed, the state of which cannot be directly detected from the outside, wherein it is determined on the basis of the acquired physical information whether the state of the object is abnormal or a state which leads to a defect within a certain period of time, when a first item of physical information acquired in advance as physical information at a time which is an abnormal state of the object or a state which leads to a defect within a certain period of time and a second item of physical information acquired as needed agree.

Images