JP2022019725A - Diagnostic device and diagnostic method for powder processing system, and computer program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a diagnostic device and a diagnostic method for a powder processing system, and a computer program.

SOLUTION: A diagnostic device for a powder processing system includes: an acquisition part for acquiring measurement data containing a flow rate of a fluid flowing in a powder processing system; an estimation part for estimating supply speed of a powder raw material, a power value of a pulverization rotor obtained from the powder processing system, pressure inside the powder processing system, or a particle size distribution of powder, on the basis of the measurement data acquired by the acquisition part, by using a learning model configured so that a calculation result is output regarding the supply speed of the powder raw material, the power value of the pulverization rotor, the pressure inside the powder processing system, or the particle size distribution of the powder, in accordance with input of the measurement data; a diagnosis part for diagnosing a deterioration degree of the pulverization rotor or a classification rotor, on the basis of an estimation result of the estimation part; and an output part for outputting a diagnosis result according to the diagnosis part.

SELECTED DRAWING: Figure 15

COPYRIGHT: (C)2022,JPO&INPIT

Description

The present invention relates to a diagnostic device, a diagnostic method, and a computer program for a powder processing system.

The powder processing process is composed of a combination of various processes such as storage, supply, transportation, pulverization, classification, mixing, drying, granulation, compounding, and spheroidization (see, for example, Patent Document 1). In order to stably obtain a product or intermediate having the quality desired by the user, it is necessary to appropriately monitor whether or not the state of various processes is appropriate.

Japanese Unexamined Patent Publication No. 2008-194592

However, there are various parameters indicating the state of the powder processing process, and a large number of measured values are output from the powder processing system. Therefore, it is difficult for the user to grasp the soundness of the powder processing system based on the measured values measured during the processing.

An object of the present invention is to provide a diagnostic device, a diagnostic method, and a computer program for a powder processing system capable of diagnosing the soundness of the powder processing system.

The diagnostic apparatus of the powder processing system according to one aspect of the present invention relates to a powder processing system including a crushing rotor for crushing a powder raw material or a classification rotor for classifying a powder obtained by crushing the powder raw material. , Acquire measurement data including at least one of the flow rate of the fluid flowing in the powder processing system, the rotation speed of the crushing rotor, the rotation speed of the classification rotor, and the vibration value of the vibration generated in the powder processing system. Depending on the acquisition unit and the input of the measurement data, the supply speed of the powder raw material, the power value of the crushing rotor, the pressure in the powder processing system, or the powder obtained from the powder processing system. Based on the measurement data acquired by the acquisition unit using a learning model configured to output the calculation results for the particle size distribution, the supply speed of the powder raw material, the power value of the crushing rotor, and the powder processing. An estimation unit that estimates the pressure in the system or the particle size distribution of the powder, a diagnosis unit that diagnoses the degree of deterioration of the crushing rotor or the classification rotor based on the estimation result by the estimation unit, and the diagnosis unit. It is equipped with an output unit that outputs the diagnosis result.

According to the present application, the soundness of the powder processing system can be diagnosed.

It is explanatory drawing explaining the structure of the powder processing system which is the diagnosis target of the diagnostic apparatus which concerns on Embodiment 1. FIG. It is a schematic cross-sectional view which shows the structure of the powder processing apparatus. It is a schematic cross-sectional view which shows the structure of the dust collector. It is a block diagram which shows the internal structure of the diagnostic apparatus which concerns on Embodiment 1. FIG. It is a block diagram which shows the internal structure of a terminal apparatus. It is a schematic diagram which shows the structural example of the learning model in Embodiment 1. FIG. It is a conceptual diagram which shows an example of the data which a diagnostic apparatus collects. It is a flowchart explaining the generation procedure of the learning model by a diagnostic apparatus. It is a flowchart explaining the diagnostic procedure of the diagnostic apparatus which concerns on Embodiment 1. FIG. It is a schematic diagram which shows the structural example of the learning model in Embodiment 2. It is a schematic diagram which shows the structural example of the learning model in Embodiment 3. It is a schematic diagram which shows the structural example of the learning model in Embodiment 4. It is a flowchart explaining the diagnostic procedure of the diagnostic apparatus which concerns on Embodiment 4. It is a schematic diagram which shows the structural example of the learning model in Embodiment 5. It is a flowchart explaining the diagnostic procedure of the diagnostic apparatus which concerns on Embodiment 5. It is a schematic diagram which shows the structural example of the learning model in Embodiment 6. It is a flowchart explaining the diagnostic procedure of the diagnostic apparatus which concerns on Embodiment 6.

Hereinafter, the present invention will be specifically described with reference to the drawings showing the embodiments thereof.
(Embodiment 1)
FIG. 1 is an explanatory diagram illustrating a configuration of a powder processing system 1 which is a diagnosis target of the diagnostic apparatus 100 according to the first embodiment. The diagnostic target of the diagnostic apparatus 100 is the powder processing system 1 that produces the powder desired by the user from the powder raw material. The powder processing system 1 includes, for example, a raw material supply machine 2, a hot air generator 3, a powder processing device 4, a cyclone 5, a dust collector 6, a blower 7, a product tank 8, and a dust collecting tank 9.

The raw material supply machine 2 is a device for supplying the powder raw material to the powder processing device 4. The powder raw material supplied by the raw material supply machine 2 to the powder processing device 4 is a raw material for producing powder of an inorganic material, an organic material, or a metallic material, and is, for example, a powder coating material, a battery material, or a magnetic material. , Toner materials, dyes, resins, waxes, polymers, pharmaceuticals, catalysts, metal powders, silica, solders, cements, foods, etc.

The raw material supply machine 2 is connected to the powder processing device 4 via the raw material supply path TP1. A screw feeder 21 (see FIG. 2) for transporting the powder raw material is provided in the raw material supply path TP1. The screw feeder 21 is preferable when the powder raw material is a solid and the powder raw material is continuously charged at a constant speed. A double damper, a rotary valve, or the like may be used instead of the screw feeder 21. Further, the hot air generated by the hot air generator 3 may be introduced into the raw material supply path TP1 and the powder raw material may be supplied to the powder processing apparatus 4 together with the hot air. Further, the raw material supply machine 2 manages the weight using a weight sensor S9 (see FIG. 4) such as a load cell, and the retention amount in the device is constant even when the powder processing device 4 performs continuous processing. The supply amount of the powder raw material may be adjusted so as to be. Further, the raw material supply machine 2 supplies the powder raw material per unit time (that is, the supply speed) based on the output of the built-in timer (not shown) and the supply amount of the powder raw material measured by the weight sensor S9. May be measured. The weight sensor S9 may be provided not only in the raw material supply machine 2 but also in the powder processing device 4, the cyclone 5, and the dust collector 6.

The hot air generator 3 is a device for generating hot air to be introduced into the powder processing device 4, and includes a heat source such as a heater, a blower, and a control device for controlling the temperature and flow rate of the hot air. The hot air generator 3 is not limited to the above configuration, and a known configuration may be used. For example, a part of the hot air introduced into the powder processing apparatus 4 may be recovered and circulated between the hot air generator 3 and the powder processing apparatus 4.

The hot air generator 3 is connected to the powder processing apparatus 4 via the gas introduction path TP2. The hot air generated by the hot air generator 3 is introduced into the powder processing apparatus 4 as a heat medium via the gas introduction path TP2. The temperature of the hot air generated by the hot air generator 3 is appropriately set according to the powder processed by the powder processing apparatus 4. For example, in the powder processing apparatus 4, hot air of about 200 ° C. to 600 ° C. may be generated to dry the powder.

The powder processing apparatus 4 is, for example, an apparatus having a crushing function for crushing a powder raw material supplied into the apparatus and a classification function for classifying the powder obtained by crushing the powder raw material. An example of the powder processing apparatus 4 having a crushing function and a classification function is ACM Pulverizer (registered trademark) manufactured by Hosokawa Micron Co., Ltd. The internal configuration of the powder processing apparatus 4 will be specifically described with reference to FIG.

The powder processed by the powder processing apparatus 4 is transported to the cyclone 5 via the powder transport path TP3. In the present embodiment, the particle size sensor S5 is installed in the middle of the powder transport path TP3 from the powder processing device 4 to the cyclone 5, and the particle size of the powder passing through the powder processing device 4 by the particle size sensor S5. Is measured at regular or regular timing (for example, every 5 seconds). The powder collected by the cyclone 5 is taken out to the product tank 8 and collected as a product.

The particle size sensor S5 is a device that measures the particle size distribution by using, for example, a laser diffraction / scattering method, and outputs the values of D10, D50, and D90. Here, D10, D50, and D90 represent particle diameters corresponding to cumulative 10%, 50%, and 90% from the small diameter side of the cumulative volume distribution in the particle size distribution, respectively. The cumulative volume distribution is a distribution representing the relationship between the particle size (μm) of the powder and the integration frequency (volume%) from the small diameter side. D50 is also generally referred to as an average particle diameter (median diameter). Instead of D10, D50, and D90, the particle size sensor S5 has a mode diameter representing the particle size having the largest appearance ratio in the frequency distribution of the particle size, or various arithmetic average values (number average, length average, area average, volume). (Average, etc.) may be output, or distribution data such as cumulative volume distribution may be output.
In the present embodiment, the particle size sensor S5 is installed in the powder transport path TP3, but the raw material supply machine 2, the path from the cyclone 5 to the product tank 8, and the dust collector 6 to the dust collector tank 9. The particle size sensor S5 may be installed at one or a plurality of places such as a path.

A dust collector 6 is connected to the cyclone 5 via a dust collection path TP4. The dust collector 6 includes a bug filter for collecting fine dust and the like that have passed through the cyclone 5. The gas that has passed through the bug filter of the dust collector 6 flows to the blower 7 through the exhaust air passage TP5 and is discharged from the discharge port of the blower 7. On the other hand, the fine dust or the like collected by the bug filter of the dust collector 6 is taken out to the dust collection tank 9 and collected.

A blower 7 is connected to the dust collector 6 via an exhaust passage TP5. By driving the blower 7, the gas flow from the powder processing apparatus 4 to the cyclone 5 (that is, the gas flow for taking out the powder from the powder processing apparatus 4) and the gas flow from the cyclone 5 to the dust collector 6 Form a flow. In the present embodiment, a flow rate sensor S2 (see FIG. 4) is installed in the middle of the exhaust passage TP5 from the dust collector 6 to the blower 7, and the discharge / suction flow rate when the powder is taken out from the powder processing device 4 is measured. Measure at regular or regular timing (for example, every 5 seconds). Instead of installing the flow rate sensor S2 in the exhaust air passage TP5, the flow rate is sent to one or more places such as the raw material supply path TP1, the gas introduction path TP2, the powder transport path TP3, the dust collection path TP4, and the discharge port of the blower 7. Sensor S2 may be installed.

In the example of FIG. 1, the powder processing system 1 including the raw material supply machine 2, the hot air generator 3, the powder processing device 4, the cyclone 5, the dust collector 6, and the blower 7 has been described, but is connected to the powder processing device 4. The equipment to be used is not limited to the above, and it is possible to construct the powder processing system 1 by combining various equipment. Further, in the present embodiment, the blower 7 is used to form a flow of gas for taking out the powder from the powder processing apparatus 4, but a pump may be used instead of the blower 7. Further, in the present embodiment, the configuration is such that a gas (hot air or cold air) is introduced into the powder processing apparatus 4, but a configuration in which a liquid is introduced may be used instead of the configuration in which the gas is introduced.

The diagnostic device 100 is configured to exchange necessary data with and from various devices and various sensors constituting the powder processing system 1. The diagnostic apparatus 100 acquires various measurement data measured by the powder processing system 1 and estimates the state of the powder processing system 1 using a learning model 210 (see FIG. 4) described later. Then, the diagnostic apparatus 100 estimates the soundness of the components constituting the powder processing system 1 based on the estimation result of the state of the powder processing system 1. Health is diagnosed based on the degree of deterioration of the component or the presence or absence of abnormalities in the component. The diagnostic device 100 notifies the terminal device 500 used by, for example, the user or the administrator of the powder processing system 1 of the diagnosis result. In the following description, the user and the administrator of the powder processing system 1 are not distinguished and are simply described as a user.

FIG. 2 is a schematic cross-sectional view showing the configuration of the powder processing apparatus 4. The powder processing apparatus 4 includes a cylindrical casing 410 that performs powder processing inside the powder processing apparatus 4. The casing 410 is provided with a raw material input port 411, a gas introduction port 412, a crushing rotor 413, a guide ring 414, a classification rotor 415, a powder outlet 416, and the like.

As the material of the casing 410, a known material conventionally used for the casing of the powder processing apparatus may be used. Specifically, iron-based steel materials such as SS400, S25C, S45C and SPHC (Steel Plate Hot Commercial), stainless steel materials such as SUS304 and SUS316, iron casting materials such as FC20 and FC40, and stainless steel casting materials such as SCS13 and 14 are used. Metal, ceramics, glass, etc. may be used. Further, aluminum, other wood, or synthetic resin may be used as long as an abrasion resistant material is attached to the inner wall surface.

In order to improve the durability of the equipment, the inner surface of the casing 410 is subjected to plating treatment such as hard chrome plating treatment, abrasion resistant thermal spraying material treatment such as tungsten carbide spraying, metal vapor deposition under vacuum, carbon vapor deposition of diamond structure, etc. Abrasion resistant treatment may be applied.

Further, when powder treatment of a low melting point resin component such as toner is performed in the casing 410, if the fixing component is mixed in the product, the quality becomes poor. Therefore, in order to prevent turbulence of the air flow due to adhesion or fixation of the treated powder or blockage in the casing 410, the inner surface of the casing 410 is buffed, electrolytically polished, coated with PTFE (Polytetrafluoroethylene), or plated with nickel or the like. Processing may be applied.

The casing 410 is provided with a raw material input port 411 for charging the powder raw material supplied from the raw material supply machine 2 into the casing 410. It is preferable that the raw material input port 411 is provided at a position above the rotary disk 413A of the crushing rotor 413. The powder raw material supplied from the raw material supply machine 2 is conveyed by the screw feeder 21 in the raw material supply path TP1 and is charged into the casing 410 from the raw material input port 411.

The casing 410 is provided with a gas introduction port 412 for introducing hot air (gas) from the hot air generator 3 into the casing 410. The gas introduction port 412 is connected to the hot air generator 3 via the gas introduction path TP2. The position of the gas introduction port 412 is not particularly limited, but it is preferably provided at a position below the crushing rotor 413 so that the gas is introduced into the casing 410 via the rotating crushing rotor 413. In the present embodiment, the gas is introduced from the direction intersecting the rotation direction of the crushing rotor 413, but the gas may be introduced along the rotation direction of the crushing rotor 413.

The gas introduced from the gas introduction port 412 forms an air flow that circulates while swirling inside the casing 410, and reaches the cyclone 5 and the dust collector 6 from the powder outlet 416 via the classification rotor 415 from the inside of the casing 410. do. The air flow in the casing 410 may be formed by suction by the blower 7 connected via the cyclone 5 and the dust collector 6, or may be formed by blowing (pressurizing) from the gas introduction port 412 side. The type of gas introduced into the casing 410 may be appropriately determined according to the target processed product. For example, air may be used, or an inert gas such as nitrogen or argon may be used to prevent oxidation.

Depending on the powder to be treated, the temperature rise in the casing 410 may cause a softening phenomenon in the powder, and the powders may be fused to each other to cause variation in particle size or decrease in yield. .. Therefore, temperature sensors (not shown) may be provided at one or a plurality of locations in the casing 410 to control the temperature inside the casing 410. For example, when the powder to be treated is toner having a low melting point, for example, the temperature of the gas introduced into the casing 410 may be adjusted so that the exhaust temperature at the powder outlet 416 is 35 to 55 ° C. ..
Further, temperature sensors may be provided at one or a plurality of locations such as the cyclone 5, the dust collector 6, the product tank 8, and the dust collecting tank 9.

Further, in the present embodiment, the hot air from the hot air generator 3 is introduced into the casing 410, but using a cold air generator (not shown in the figure), the temperature is about −20 ° C. to 5 ° C. in the casing 410. It may be configured to introduce the cold air of. In this case, the gas introduced into the casing 410 is preferably a dehumidified gas in order to prevent dew condensation. In addition, when processing foods that lose their flavor due to heat or are easily deteriorated, cold air adjusted to 0 to 15 ° C. may be used.

Further, in order to adjust the internal temperature of the casing 410, a jacket portion may be provided around the casing 410. The jacket portion adjusts the internal temperature of the casing 410 by circulating and supplying a heating fluid or a cooling fluid from a tank provided separately.

The crushing rotor 413 is a rotor including a rotary disk 413A and a plurality of hammers 413B protruding upward from the upper peripheral edge portion of the rotary disk 413A. The powder processing apparatus 4 includes a drive mechanism (not shown) including a crushing motor and bearings for rotationally driving the crushing rotor 413. The crushing rotor 413 is configured to rotate at a desired rotation speed by the power of the crushing motor. The rotation speed of the crushing rotor 413 is measured by the rotation speed sensor S7 (see FIG. 4) at regular or periodic timings (for example, at 5-second intervals). A plurality of hammers 413B of the crushing rotor 413 are arranged at equal intervals in the circumferential direction on the upper peripheral edge portion of the rotary disk 413A. The shape, size, number, and material of the hammer 413B are appropriately designed according to the required particle size, circularity, and the like of the product powder. For example, although FIG. 2 shows a rod-shaped hammer 413B, it may be a rectangular parallelepiped hammer or a trapezoidal hammer in a plan view. Further, instead of the hammer 413B, a blade-shaped structure may be used.

The crushing rotor 413 is rotated by the power of the crushing motor to generate a swirling airflow in the casing 410, and by the action of the hammer 413B, the powder raw material introduced in the casing 410 is impacted, compressed, crushed, and sheared. The powder raw material is crushed by giving mechanical energy such as.

As the material of the crushing rotor 413, a known material conventionally used for the crushing rotor of the powder processing apparatus may be used. For example, SS400, S25C, S45C, SUS304, SUS316, SUS630 and the like can be used. Further, for the hammer 413B, a cemented carbide chip may be attached so as to withstand the impact force, or a composite of a metal such as ceramics or cermet having abrasion resistance and toughness and ceramics may be used. good.

Further, the surface of the crushing rotor 413 is subjected to plating treatment such as hard chrome plating, abrasion resistant thermal spraying material treatment such as tungsten carbide spraying, metal vapor deposition under vacuum, and carbon vapor deposition of diamond structure in order to improve the durability of the device. Abrasion resistant treatment such as, quenching hardening treatment of SUS630, and the like may be performed. Further, the surface of the crushing rotor 413 may be buffed, electrolytically polished, coated with PTFE, or plated with nickel or the like.

In the present embodiment, the configuration of the hammer 413B protruding upward from the upper peripheral edge portion of the rotating disk 413A has been described, but a hammer protruding downward from the lower surface peripheral edge portion of the rotating disk 413A may be used. The hammer protruding downward in this way does not directly crush the powder raw material in the casing 410, but can form a strong swirling airflow in the casing 410, so that the powder raw materials collide with each other. The powder raw material can be indirectly crushed.

Further, a crushing liner may be provided on the inner peripheral surface of the casing 410 at a position facing the hammer 413B. The crushing liner is a cylindrical member having a central axis along the rotation axis direction of the crushing rotor 413, and even if the inner peripheral surface of the tubular member is provided with a triangular, corrugated, or wedge-shaped groove. good.

The guide ring 414 is a cylindrical member for generating a swirling air flow in the casing 410 and guiding the powder processed in the casing 410 to the classification rotor 415. The guide ring 414 is arranged coaxially with the crushing rotor 413 above the crushing rotor 413 and is fixed inside the casing 410. The fixing method of the guide ring 414 is not particularly limited, but it is necessary to fix the guide ring 414 without rotating inside the casing 410 during the operation of the powder processing apparatus 4. This is to control the flow state of the powder to be processed to an appropriate state inside the casing 410. In the example of FIG. 2, the guide ring 414 whose inner diameter continuously increases from the lower side to the upper side in the casing 410 is shown, but the guide ring whose inner diameter continuously decreases toward the upper side. It may be a guide ring whose inner diameter does not change in the vertical direction.

The classification rotor 415 is a rotor provided with a plurality of classification blades 415A arranged radially. As the material of the classification rotor 415, a known material conventionally used for the classification rotor of the powder processing apparatus may be used. For example, SS400, S25C, S45C, SUS304, SUS316, titanium, titanium alloy, aluminum alloy and the like can be used. In addition, the surface of the classification rotor 415 is specially subjected to heat hardening treatment such as carburizing and quenching, tungsten carbide spraying material treatment, plating treatment such as hard chrome plating, and heat hardening treatment after spraying in order to improve the durability of the equipment. Abrasion resistant treatment such as thermal spray material treatment, metal vapor deposition performed under vacuum, and carbon vapor deposition of a diamond structure may be performed.

Further, in order to prevent the powder from adhering to or sticking to the classification rotor 415, the surface of the classification rotor 415 may be buffed, electrolytically polished, coated with PTFE, or plated with nickel or the like.

The powder processing apparatus 4 includes a drive mechanism (not shown) including a classification motor and bearings for rotationally driving the classification rotor 415. The classification rotor 415 is configured to rotate at a desired rotation speed by the power of the classification motor. The rotation speed of the classification rotor 415 is measured by the rotation speed sensor S8 (see FIG. 4) at regular or periodic timings (for example, at 5-second intervals). The classification rotor 415 is provided above the crushing rotor 413, and the powder processed in the casing 410 is passed through only the powder having a particle size smaller than the predetermined particle size by the centrifugal force due to the high-speed rotation. Only the body is guided to the powder outlet 416.

The particle size of the powder passing through the classification rotor 415 can be set by controlling the rotation speed of the classification rotor 415 and the like. That is, by controlling the rotation speed of the classification rotor 415, powder having a particle size smaller than a predetermined particle size can be taken out from the casing 410. On the other hand, the powder that cannot pass through the classification rotor 415 circulates in the casing 410 and is repeatedly processed.

FIG. 3 is a schematic cross-sectional view showing the configuration of the dust collector 6. The dust collector 6 includes a housing unit 600, a filter unit 610, a wiping unit 620, and a discharging unit 630.

The housing portion 600 is a collection of a top brenham 601 for accommodating the wiping portion 620, a tube sheet 602 for fixing the filter portion 610 and separating the clean side and the dirty side, and a straight body portion 603 for accommodating the filter portion 610. It is composed of a hopper unit 604 that sends powder to the discharge unit 630. The material of the housing portion 600 is selected according to the type of powder to be collected, the gas temperature, the gas composition, the purpose of use, and the like, and for example, a rolled steel material for general structure (SS400), a stainless steel material, and the like are used. Further, an introduction port 603A for introducing a fluid (for example, gas) containing powder is provided in the lower portion of the straight body portion 603, and an discharge port for discharging the fluid after dust collection is provided on the upper surface of the top Brenham 601. 601A is provided.

The filter unit 610 includes a filter cloth 611. The filter cloth 611 is formed into a tubular shape (for example, a cylindrical shape) by a felt cloth. The material of the filter cloth 611 is selected according to the properties of the dust to be collected, the gas temperature, the gas composition, and the like, and for example, fibers such as polyester, polypropylene, acrylic, nylon-6, and glass fiber are used. A retainer (not shown) is inserted inside the filter cloth 611 to maintain its shape. When the fluid containing the powder flows from the outside (downstream side) to the inside (upstream side) of the filter cloth 611, the powder contained in the fluid is collected by the outer surface of the filter cloth 611. The fluid that has passed through the filter cloth 611 (that is, the fluid from which the powder has been removed) is discharged to the outside from the discharge port 601A.

The wiping unit 620 includes a blow tube 621 that injects compressed air onto the filter cloth 611. The pressure of the compressed air injected from the blow tube 621 is appropriately set according to the filter area and the like. In the example of FIG. 3, the powder is collected by the filter cloth 611 on the left side, and the powder is removed by the filter cloth 611 on the right side. May be carried out appropriately on both filter cloths 611.

The discharge unit 630 includes, for example, a rotary valve 631. Since there is a pressure difference between the inside and the outside of the dust collector 6 during operation, the discharge unit 630 discharges the powder while performing an air lock.

Next, the configurations of the diagnostic device 100 and the terminal device 500 will be described.
FIG. 4 is a block diagram showing an internal configuration of the diagnostic apparatus 100 according to the first embodiment. The diagnostic device 100 is a computer such as a server device, and includes a control unit 101, a storage unit 102, an input unit 103, an output unit 104, a communication unit 105, an operation unit 106, and a display unit 107.

The control unit 101 includes, for example, a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), and the like. The ROM included in the control unit 101 stores a control program or the like that controls the operation of each hardware unit included in the diagnostic apparatus 100. The CPU in the control unit 101 executes a control program stored in the ROM and various computer programs stored in the storage unit 102, which will be described later, and controls the operation of each hardware unit to control the operation of each unit of the hardware as a diagnostic device according to the present invention. To realize the function of. The RAM included in the control unit 101 temporarily stores data and the like used during execution of the calculation.

The control unit 101 is configured to include a CPU, ROM, and RAM, but is a GPU (Graphics Processing Unit), FPGA (Field Programmable Gate Array), DSP (Digital Signal Processor), quantum processor, volatile or non-volatile memory. It may be one or a plurality of arithmetic circuits or control circuits including the above. Further, the control unit 101 may have functions such as a clock for outputting date and time information, a timer for measuring the elapsed time from giving the measurement start instruction to giving the measurement end instruction, and a counter for counting the number.

The storage unit 102 includes a storage device that uses a hard disk, a flash memory, or the like. The storage unit 102 stores a computer program executed by the control unit 101, various data acquired from the outside, various data generated inside the apparatus, and the like.

The computer program stored in the storage unit 102 includes a diagnostic program PG1 for diagnosing the soundness of the powder processing system 1. The control unit 101 of the diagnostic apparatus 100 diagnoses the soundness of the powder processing system 1 based on the measurement data measured by the powder processing system 1 by executing the diagnostic program PG1, and outputs the diagnosis result. Execute the process.

The computer program including the diagnostic program PG1 may be provided by a non-temporary recording medium M1 in which the computer program is readablely recorded. The recording medium M1 is a portable memory such as a CD-ROM, a USB memory, a compact flash (registered trademark), an SD (Secure Digital) card, or a micro SD card. The control unit 101 reads various programs from the recording medium M1 using a reading device (not shown in the figure), and stores the read various programs in the storage unit 102.

The input unit 103 includes a connection interface for connecting various devices and sensors. The connection interface included in the input unit 103 may be a wired interface or a wireless interface. The device connected to the input unit 103 includes a raw material supply machine 2, a hot air generator 3, a powder processing device 4, a cyclone 5, a dust collector 6, and a blower 7. Data transmitted from the raw material supply machine 2, the hot air generator 3, the powder processing device 4, the cyclone 5, the dust collector 6, and the blower 7 are input to the input unit 103. Further, the sensor connected to the input unit 103 includes a pressure sensor S1, a flow rate sensor S2, a power sensor S3, a concentration sensor S4, a particle size sensor S5 and the like. The measurement data measured by these sensors S1 to S5 is input to the input unit 103.

Here, the pressure sensor S1 is installed in the casing 410 included in the powder processing apparatus 4, upstream and downstream of the dust collector 6, upstream of the blower 7, and the like, and measures the pressure (static pressure) at these installation locations. do. The flow rate sensors S2 are installed on the upstream and downstream sides of the casing 410, the upstream and downstream sides of the cyclone 5, the upstream and downstream sides of the dust collector 6, and measure the flow rate of the fluid at these installation locations. The power sensor S3 is installed in the blower 7 and measures the power of the blower 7. The concentration sensor S4 is installed on the upstream side and the downstream side of the dust collector 6 and measures the dust content concentration at these installation locations. Instead of installing the concentration sensor S4 on the upstream side and the downstream side of the dust collector 6, the concentration sensor S4 may be installed on the upstream side and the downstream side of the filter cloth 611 in the dust collector 6. As described above, the particle size sensor S5 is installed in the middle of the powder transport path TP3 from the powder processing device 4 to the cyclone 5, and measures the particle size of the powder passing through the powder processing device 4.

Sensors such as an image pickup sensor S6, a rotation speed sensor S7, S8, a weight sensor S9, a temperature sensor S10, a vibration sensor S11, and a sound sensor S12 may be connected to the input unit 103. The image pickup sensor S6 is, for example, a sensor such as a CCD (Charge-Coupled Device) or CMOS (Complementary Metal Oxide Semiconductor) for taking an image of the filter cloth 611 from the downstream side of the filter cloth 611, and outputs image data. Further, the image pickup sensor S6 may be installed inside or outside the powder processing device 4 in order to observe the internal state of the powder processing device 4. The rotation speed sensors S7 and S8 are installed in the powder processing apparatus 4 and measure the rotation speeds of the crushing rotor 413 and the classification rotor 415, respectively. The weight sensor S9 measures the weight of the powder raw material supplied by the raw material supply machine 2 per unit time (that is, the supply speed of the powder raw material). The temperature sensor S10 is installed at one or a plurality of locations of the powder processing system 1 and measures the temperature at the installation location. The temperature sensor S10 may measure the environmental temperature. The vibration sensor S11 is installed at one or a plurality of locations of the powder processing system 1 and measures the vibration at the installed locations. The sound sensor S12 is installed at one or a plurality of locations of the powder processing system 1 and measures sound (AE: Acoustic Emission) at the installed locations.

The output unit 104 includes a connection interface for connecting various devices and sensors. The connection interface included in the output unit 104 may be a wired interface or a wireless interface. The device connected to the output unit 104 includes a raw material supply machine 2, a hot air generator 3, a powder processing device 4, a cyclone 5, a dust collector 6, and a blower 7.

The communication unit 105 includes a communication interface for transmitting and receiving various communication data. The communication interface included in the communication unit 105 is, for example, a communication interface conforming to a LAN (Local Area Network) communication standard used in WiFi (registered trademark) or Ethernet (registered trademark). Alternatively, a communication interface conforming to a communication standard such as Bluetooth (registered trademark), ZigBee (registered trademark), 3G, 4G, 5G, LTE (Long Term Evolution) may be used.

The communication unit 105 communicates with, for example, the terminal device 500 used by the user of the powder processing system 1. When the diagnosis result of the powder processing system 1 is obtained, the control unit 101 of the diagnosis device 100 transmits information indicating the diagnosis result from the communication unit 105 to the terminal device 500. Further, the control unit 101 may generate screen data of the user interface screen to be displayed on the terminal device 500, and may transmit the generated screen data to the terminal device 500 through the communication unit 105.

The operation unit 106 is provided with an input interface such as a keyboard and a mouse, and receives various operations and various settings. The control unit 101 performs appropriate processing based on various operations and various settings received through the operation unit 106, and stores the setting information in the storage unit 102 as necessary. In the present embodiment, the diagnostic device 100 is configured to include the operation unit 106, but the operation unit 106 is not indispensable, and the operation is received through a computer connected to the outside (for example, the terminal device 500). You may.

The display unit 107 includes a display panel such as a liquid crystal panel or an organic EL (Electro-Luminescence) panel, and displays information to be notified to the user. The display unit 107 may display, for example, the measurement data of the various sensors S1 to S5 received through the communication unit 105, or may display information based on various operations and various settings received through the operation unit 106. In the present embodiment, the diagnostic device 100 is configured to include the display unit 107, but the display unit 107 is not essential and outputs information to be notified to the user to an external computer (for example, the terminal device 500). , The information may be displayed on the output destination computer.

In the present embodiment, the diagnostic device 100 has been described as a single computer, but it does not have to be a single computer, and may be configured by a plurality of computers or may be configured by a plurality of virtual computers.

FIG. 5 is a block diagram showing an internal configuration of the terminal device 500. The terminal device 500 is a computer such as a personal computer, a smartphone, or a tablet terminal, and includes a control unit 501, a storage unit 502, a communication unit 503, an operation unit 504, and a display unit 505.

The control unit 501 includes, for example, a CPU, a ROM, a RAM, and the like. The ROM included in the control unit 501 stores a control program or the like that controls the operation of each hardware unit included in the terminal device 500. The CPU in the control unit 501 executes a control program stored in the ROM and various computer programs stored in the storage unit 502 described later, and controls the operation of each hardware unit. Further, the control unit 501 may be provided with functions such as a clock for outputting date and time information, a timer for measuring the elapsed time from the time when the measurement start instruction is given to the time when the measurement end instruction is given, and a counter for counting the number. The RAM included in the control unit 501 temporarily stores data and the like used during the execution of the calculation.

The storage unit 502 includes a storage device using a hard disk, a flash memory, or the like. The storage unit 502 stores computer programs executed by the control unit 501, various data acquired from the outside, various data generated inside the apparatus, and the like. The computer program stored in the storage unit 502 may include an application program for accessing the diagnostic device 100 from the terminal device 500.

The computer program stored in the storage unit 102 may include a display program that accesses the diagnostic device 100 from the terminal device 500 and displays information provided by the diagnostic device 100. The computer program including the display program may be provided by a non-temporary recording medium in which the computer program is readablely recorded. The recording medium is, for example, a portable memory such as a CD-ROM, a USB memory, a compact flash (registered trademark), an SD card, or a micro SD card. The control unit 501 reads various programs from the recording medium using a reading device (not shown in the figure), and stores the read various programs in the storage unit 502.

The communication unit 503 includes a communication interface for transmitting and receiving various data. The communication interface included in the communication unit 503 is, for example, a communication interface conforming to the LAN communication standard used in WiFi (registered trademark) and Ethernet (registered trademark). Alternatively, a communication interface conforming to communication standards such as Bluetooth (registered trademark), ZigBee (registered trademark), 3G, 4G, 5G, and LTE may be used.

The communication unit 503 communicates with, for example, the diagnostic device 100. The data received by the terminal device 500 from the diagnostic device 100 through the communication unit 503 is the diagnosis result of the powder processing system 1 by the diagnostic device 100, screen data for displaying the interface screen provided by the diagnostic device 100 on the display unit 505, and the like. including. The data transmitted by the terminal device 500 to the diagnostic device 100 through the communication unit 503 includes a request for transmission of the diagnostic result and the like.

The operation unit 504 is provided with an input interface such as a keyboard and a mouse, and accepts various operations and various settings. The control unit 501 performs appropriate processing based on various operations and various settings received through the operation unit 504, and stores the setting information in the storage unit 502 as necessary.

The display unit 505 includes a display panel such as a liquid crystal panel or an organic EL panel, and displays information to be notified to the user. The display unit 505 displays the interface screen provided by the diagnostic apparatus 100, for example, based on the screen data received by the communication unit 503. Further, the display unit 505 may display the diagnosis result of the powder processing system 1 received by the communication unit 503.

Hereinafter, the learning model 210 used in the diagnostic apparatus 100 will be described.
FIG. 6 is a schematic diagram showing a configuration example of the learning model 210 according to the first embodiment. The learning model 210 is, for example, a learning model of machine learning including deep learning, and is configured by a neural network. The learning model 210 includes an input layer 211, intermediate layers 212A, 212B, and an output layer 213. In the example of FIG. 6, two intermediate layers 212A and 212B are described, but the number of intermediate layers is not limited to two and may be three or more.

The input layer 211, the intermediate layers 212A, 212B, and the output layer 213 have one or more nodes, and the nodes of each layer are coupled to the nodes existing in the front and rear layers in one direction with a desired weight and bias. Has been done. The same number of data as the number of nodes included in the input layer 211 is input to the input layer 211 of the learning model 210. In the present embodiment, the data input to the node of the input layer 211 is the measurement data measured by the powder processing system 1. Specifically, the pressure in the casing 410 measured by the pressure sensor S1, the flow rate measured by the flow rate sensor S2, the power value of the blower 7 measured by the power sensor S3, and the dust content concentration measured by the concentration sensor S4. , Measurement data including the particle size distribution measured by the particle size sensor S5 is input to the node of the input layer 211. The measurement data input to the input layer 211 of the learning model 210 does not have to be limited to scalars, and may be data having some structure such as vector data and image data.

The input measurement data is output to the node included in the first intermediate layer 212A through the nodes constituting the input layer 211. The data input to the first intermediate layer 212A is output to the node included in the next intermediate layer 212B through the nodes constituting the intermediate layer 212A. At this time, the output is calculated using the activation function including the weights and biases set between the nodes. Hereinafter, in the same manner, the operation using the activation function including the weight and the bias set between the nodes is executed, and the operation is transmitted to the subsequent layers one after another until the operation result by the output layer 213 is obtained. Parameters such as weights and biases that connect nodes are learned by a predetermined learning algorithm. As a learning algorithm for learning various parameters, for example, a learning algorithm for deep learning is used. In the present embodiment, the above-mentioned measurement data and data indicating the state of the powder processing system 1 are used as teacher data, and when the measurement data is input, the calculation result for the powder processing system 1 is output. As described above, various parameters including weights and biases between nodes are learned by a predetermined learning algorithm.

The output layer 213 outputs the calculation result for the state of the powder processing system 1. As the state of the powder processing system 1, the learning model 210 according to the first embodiment outputs a calculation result for the pressure loss of the filter cloth 611. Specifically, the output layer 213 is composed of n nodes from the first node to the nth node, and the probability P1 that the pressure loss of the filter cloth 611 is PL1 is output from the first node, and the probability P1 is output from the second node. The probability P2 that the pressure loss of the filter cloth 611 is PL2 is output, and ..., the probability Pn that the pressure loss of the filter cloth 611 is PLn is output from the nth node. The number of nodes constituting the output layer 213 and the calculation result assigned to each node are not limited to the above example, and can be appropriately designed.

The diagnostic apparatus 100 collects data measured for the powder processing system 1 (pressure in the casing 410, flow rate, power value of the blower 7, dust content concentration, particle size distribution, and pressure loss in the filter cloth 611). , These measurement data are used as the teacher data to generate the training model 210 as described above.

FIG. 7 is a conceptual diagram showing an example of data collected by the diagnostic apparatus 100. The control unit 101 of the diagnostic device 100 acquires the measurement data measured by the various sensors S1 to S5 through the input unit 103. Specifically, the pressure in the casing 410 measured by the pressure sensor S1, the flow rate of the fluid measured by the flow sensor S2 (m ³ / min), and the power value of the blower 7 measured by the power sensor S3. (KW), dust content concentration (g / m ³ ) at the inlet (upstream side) and outlet (downstream side) of the dust collector 6 measured by the concentration sensor S4, particle size distribution (μm) measured by the particle size sensor S5. , The pressure at the inlet (upstream side) and the outlet (downstream side) of the dust collector 6 measured by the pressure sensor S1 is acquired. The control unit 101 stores the acquired measurement data in the storage unit 102 together with the time stamp. At this time, the pressure loss (atm) of the filter cloth 611 measured as the pressure difference between the inlet (upstream side) and the outlet (downstream side) of the dust collector 6 may be stored in the storage unit 102. Note that FIG. 7 shows an example in which the data collected at 5-second intervals is stored in the storage unit 102, but the time interval for collecting data is not limited to 5 seconds and may be set arbitrarily.

Using the collected measurement data as teacher data, the control unit 101 uses the measurement data including the pressure in the casing 410, the flow rate, the power value of the blower 7, the dust content concentration, and the particle size distribution, and the pressure loss of the filter cloth 611. The relationship with is learned, and the learning model 210 as described above is generated.

Hereinafter, the operation of the diagnostic apparatus 100 in the learning phase will be described.
FIG. 8 is a flowchart illustrating a procedure for generating a learning model 210 by the diagnostic apparatus 100. The control unit 101 of the diagnostic apparatus 100 collects the measurement data measured in the powder processing system 1 (step S101). As described above, the measurement data collected in step S101 includes the pressure in the casing 410, the flow rate of the fluid, the power value of the blower 7, the pressure loss of the filter cloth 611, the dust content concentration, and the particle size distribution. The collected measurement data is stored in the storage unit 102 together with the time stamp.

After collecting the measurement data, the control unit 101 selects a set of teacher data from the data stored in the storage unit 102 (step S102). That is, the control unit 101 stores the pressure in the casing 410, the flow rate of the fluid, the power value of the blower 7, the dust content concentration, the measured value of the particle size distribution, and the pressure loss of the filter cloth 611 obtained at that time. Select only one set from 102.

The control unit 101 inputs the measured values of the pressure in the casing 410, the flow rate of the fluid, the power value of the blower 7, the dust content concentration, and the particle size distribution included in the selected teacher data into the learning model 210 (step S103). The calculation by the learning model 210 is executed (step S104). That is, the control unit 101 inputs the measured values to the nodes constituting the input layer 211 of the learning model 210, performs operations using the weights and biases between the nodes in the intermediate layers 212A and 212B, and outputs the operation results to the output layer. The process of outputting from the node of 213 is performed. In the initial stage before the learning is started, it is assumed that the definition information describing the learning model 210 is given an initial value.

Next, the control unit 101 evaluates the calculation result obtained in step S104 (step S105), and determines whether or not the learning is completed (step S106). Specifically, the control unit 101 can evaluate the calculation result by using an error function (also referred to as an objective function, a loss function, or a cost function) based on the calculation result obtained in step S104 and the teacher data. When the error function becomes less than or equal to the threshold value (or more than the threshold value) in the process of optimizing (minimizing or maximizing) the error function by the gradient descent method such as the steepest descent method, the control unit 101 determines that the learning is completed. to decide. In addition, in order to avoid the problem of overfitting, techniques such as cross-validation and early stopping may be adopted and learning may be terminated at an appropriate timing.

When it is determined that the learning is not completed (S106: NO), the control unit 101 updates the weights and biases between the nodes of the learning model 210 (step S107), returns the process to step S102, and another teacher. Continue learning with the data. The control unit 101 may update the weights and biases between the nodes by using the error back propagation method in which the weights and biases between the nodes are sequentially updated from the output layer 213 of the learning model 210 toward the input layer 211. can.

When it is determined that the learning is completed (S106: YES), the control unit 101 stores the learned learning model 210 in the storage unit 102 (step S108), and ends the process according to this flowchart.

In the present embodiment, the diagnostic apparatus 100 is configured to generate the learning model 210, but an external server (not shown) for generating the learning model 210 may be provided and the learning model 210 may be generated by the external server. In this case, the diagnostic device 100 may acquire the learning model 210 from an external server by communication or the like, and store the acquired learning model 210 in the storage unit 102.

Next, the operation of the diagnostic apparatus 100 in the operation phase will be described. In the operation phase, it is assumed that the learning model 210 has already been learned.

FIG. 9 is a flowchart illustrating a diagnostic procedure of the diagnostic apparatus 100 according to the first embodiment. In the first embodiment, a procedure for diagnosing the soundness of the filter cloth 611 of the dust collector 6 provided as a component of the powder processing system 1 will be described.

The control unit 101 of the diagnostic apparatus 100 receives the input of measurement data measured by various sensors (step S121). The measurement data that receives the input in step S121 includes the pressure in the casing 410, the flow rate of the fluid, the power value of the blower 7, the dust content concentration, and the particle size distribution. Further, the control unit 101 accepts input of measurement data including the pressure loss of the filter cloth 611 for comparison with the estimated value described later.

Next, the control unit 101 inputs data including the pressure in the casing 410, the flow rate of the fluid, the power value of the blower 7, the dust content concentration, and the value of the particle size distribution from the received measurement data to the input layer 211 of the learning model 210. Is input to, and the calculation by the learning model 210 is executed (step S122). The data given to the node of the input layer 211 is output to the node of the adjacent intermediate layer 212A. In the intermediate layer 212A, an operation using an activation function including weights and biases between nodes is performed, and the operation result is output to the intermediate layer 212B in the subsequent stage. In the intermediate layer 212B, an operation using an activation function including weights and biases between nodes is further performed, and the operation result is output to each node of the output layer 213. Each node of the output layer 213 outputs a calculation result indicating the state of the powder processing system 1. Specifically, each node of the output layer 213 outputs the calculation result regarding the pressure loss of the filter cloth 611.

Next, the control unit 101 acquires the calculation result from the learning model 210 (step S123), and estimates the pressure loss of the filter cloth 611 (step S124). The output layer 213 of the learning model 210 outputs, for example, the probability Pi (i = 1 to n) that the pressure loss of the filter cloth 611 is PLi as a calculation result regarding the pressure loss of the filter cloth 611 from each node. By identifying the pressure loss with the highest probability among these probabilities Pi (i = 1 to n), the control unit 101 determines the pressure in the casing 410, the flow rate of the fluid, the power value of the blower 7, and the dust content concentration. , And the pressure loss of the filter cloth 611 estimated from the value of the particle size distribution is obtained.

Next, the control unit 101 diagnoses the degree of deterioration of the filter cloth 611 based on the pressure loss of the filter cloth 611 estimated in step S124 (step S125). At this time, the control unit 101 compares the pressure loss of the filter cloth 611 obtained as the measurement result with the pressure loss of the filter cloth 611 estimated from the learning model 210, and the deterioration of the filter cloth 611 is based on the comparison result. The degree can be diagnosed. The MT method (Mahalanobis-Taguchi Method) may be used when comparing the measured value and the estimated value of the pressure loss. For example, when the pressure loss of the filter cloth 611 obtained as the measurement result is higher than the pressure loss of the filter cloth 611 obtained as the estimation result, the control unit 101 determines that the filter cloth 611 is deteriorated and is in a closed state. Can be diagnosed. Further, when the pressure loss of the filter cloth 611 obtained as the measurement result is lower than the pressure loss of the filter cloth 611 obtained as the estimation result, the control unit 101 diagnoses that the filter cloth 611 is deteriorated and damaged. be able to.

Next, the control unit 101 determines whether or not an alarm is necessary based on the diagnosis result in step S125 (step S126). The control unit 101 determines that an alarm is required when the pressure loss of the filter cloth 611 obtained as the measurement result is higher (or lower) by a certain percentage or more than the pressure loss of the filter cloth 611 obtained as the estimation result. When it is determined that an alarm is necessary (S126: YES), the control unit 101 outputs an alarm (step S128). Specifically, the control unit 101 outputs an alarm by transmitting alarm information indicating that the deterioration of the filter cloth 611 is more than expected from the communication unit 105 to the terminal device 500. Further, the control unit 101 may output an alarm by displaying the alarm information indicating that the deterioration of the filter cloth 611 is more than expected on the display unit 107.

When it is determined in step S126 that the alarm is not required (S126: NO), the control unit 101 outputs the diagnosis result in step S125 (step S127). At this time, the control unit 101 may display the diagnosis result on the display unit 107. Further, the control unit 101 may transmit the diagnosis result from the communication unit 105 to the terminal device 500.

Further, the control unit 101 may select a filter cloth recommended to be attached to the dust collector 6 according to the degree of deterioration of the filter cloth 611, and transmit information on the selected filter cloth to the terminal device 500. For example, when the degree of deterioration of the filter cloth 611 estimated by using the learning model 210 is larger than the preset assumed value, the control unit 101 selects a filter cloth having a longer product life and selects the filter cloth. Information can be transmitted to the terminal device 500.

As described above, the diagnostic apparatus 100 according to the present embodiment is configured to output the calculation result regarding the state of the powder processing system 1 when the measurement data measured for the powder processing system 1 is input. The state of the powder processing system 1 can be estimated using the learning model, and the soundness (degree of deterioration and presence / absence of abnormality) of the filter cloth 611 can be diagnosed based on the estimation result of the learning model 210.

In this embodiment, a configuration for acquiring calculation results related to control parameters using a machine learning learning model 210 configured by a neural network has been described, but the learning model 210 is limited to a model obtained by using a specific method. Not done. For example, instead of the neural network by deep learning, a learning model by a perceptron, a convolutional neural network, a recurrent neural network, a residual network, a self-organizing map, or the like may be used.

In addition, instead of the above neural network learning model, a regression analysis method including linear regression, logistic regression, support vector machine, etc., and a method using search trees such as decision trees, regression trees, random forests, and gradient boosting trees. , Bayesian estimation method including simple bays, AR (Auto Regressive), MA (Moving Average), ARIMA (Auto Regressive Integrated Moving Average), time series prediction method including state space model, clustering method including K neighborhood method, etc. , Boosting, methods using ensemble learning including bagging, hierarchical clustering, non-hierarchical clustering, clustering methods including topic models, association analysis, learning models learned by other methods including emphasis filtering, etc. There may be.
Further, a learning model may be constructed using multivariate analysis including PLS (Partial Least Squares) regression, multiple regression analysis, principal component analysis, factor analysis, cluster analysis and the like.

(Embodiment 2)
In the second embodiment, a configuration for performing a diagnosis based on image data obtained by imaging the inside of the filter cloth 611 will be described.

FIG. 10 is a schematic diagram showing a configuration example of the learning model 220 in the second embodiment. The learning model 220 is, for example, a learning model based on a CNN (Convolutional Neural Networks), and includes an input layer 221, an intermediate layer 222, and an output layer 223. The learning model 220 is trained in advance so as to output a calculation result regarding the pressure loss of the filter cloth 611 in response to the input of image data obtained by imaging the inside of the filter cloth 611 using the image pickup sensor S6. Since the learning method is the same as that of the first embodiment, the description thereof will be omitted.

Image data obtained by imaging the inside of the filter cloth 611 is input to the input layer 221. The image data input to the input layer 221 is transmitted to the intermediate layer 222.

The intermediate layer 222 is composed of, for example, a convolutional layer 222A, a pooling layer 222B, and a fully connected layer 222C. A plurality of convolution layers 222A and pooling layers 222B may be provided alternately. The convolution layer 222A and the pooling layer 222B extract the features of the image data input through the input layer 221 by the calculation using the nodes of each layer. The fully connected layer 222C combines the data whose feature portions are extracted by the convolution layer 222A and the pooling layer 222B into one node, and outputs the feature variable converted by the activation function. The feature variable is output to the output layer 223 through the fully connected layer 222C.

The output layer 223 includes one or more nodes. The output layer 223 converts the probability of the pressure loss of the filter cloth 611 into a probability based on the feature variable input from the fully connected layer 222C of the intermediate layer 222 by using the softmax function, and the probability that the pressure loss of the filter cloth 611 is a specific value is calculated for each node. Output from. That is, the output layer 223 is composed of n nodes from the first node to the nth node, the probability P1 that the pressure loss of the filter cloth 611 is PL1 is output from the first node, and the filter cloth 611 is output from the second node. Outputs the probability P2 that the pressure loss of is PL2, and outputs the probability Pn that the pressure loss of the filter cloth 611 is PLn from the nth node.

Similar to the first embodiment, the control unit 101 of the diagnostic apparatus 100 compares the pressure loss of the filter cloth 611 estimated by using the learning model 220 with the pressure loss of the filter cloth 611 obtained as measurement data. The degree of deterioration of the filter cloth 611 and the presence or absence of abnormalities can be diagnosed.

As described above, the diagnostic apparatus 100 according to the second embodiment inputs the image data obtained by imaging the inside of the filter cloth 611 into the learning model 220 to check the degree of deterioration of the filter cloth 611 and the presence or absence of abnormalities. Can be diagnosed.

Although the learning model 220 by CNN is shown in the example of FIG. 10, the machine learning model for constructing the learning model 220 can be arbitrarily set. For example, instead of CNN, a learning model based on R-CNN (Region-based CNN), YOLO (You Only Look Once), SSD (Single Shot Detector) and the like may be set.

(Embodiment 3)
In the third embodiment, a configuration for estimating the replacement time of the filter cloth 611 based on the time series data will be described.

FIG. 11 is a schematic diagram showing a configuration example of the learning model 230 according to the third embodiment. The learning model 230 is a Seq2Seq (Sequence to Sequence) model, which is a kind of recurrent neural network.

The learning model 230 includes m encoders E1 to Em for inputting time-series data and n decoders D1 to Dn for outputting time-series data. The m and n of the index are integers of 2 or more. In FIG. 11, the encoders E1 to Em and the decoders D1 to Dn are described as a single block, but have a plurality of layers of about 2 to 8 including an input layer and a hidden layer. Since the internal structures of the encoders E1 to Em and the decoders D1 to Dn and the method of learning the internal parameters of the encoders E1 to Em and the decoders D1 to Dn are known, detailed description thereof will be omitted. In the present embodiment, the pressure in the casing 410 measured by the pressure sensor S1, the flow rate measured by the flow rate sensor S2, the power value of the blower 7 measured by the power sensor S3, and the dust content concentration measured by the concentration sensor S4. , And the internal parameters of the learning model 230 are learned so that the time-series data of the particle size distribution measured by the particle size sensor S5 is input to the learning model 230 and the time-series data of the pressure loss in the filter cloth 611 is output. Will be done.

In FIG. 11, the horizontal direction represents a time step, and indicates that the procedure is proceeding from the left direction to the right direction in the figure. Time-series data of any one of the pressure, the flow rate, the power value of the blower 7, the dust content concentration, and the particle size distribution in the casing 410 is input to each of the encoders E1 to Em. For example, when the output of the pressure sensor S1 is input to the learning model 230, the control unit 101 of the diagnostic apparatus 100 sequentially inputs the pressure values K1, K2, ... Km obtained as time series data to the encoders E1 to Em. do.

In the hidden layer of the encoders E1 to Em, the time series data of the input pressure is recorded as the internal vector ct as the internal state. Here, t represents a time step and takes a value of 1 to m in the encoders E1 to Em. The internal vector ct is passed to the value step for each input, and when all the inputs are completed, the internal vector cm to be passed to the decoder D1 is obtained.

The internal vector cm in the final encoder Em is passed to the decoder D1. A reserved word instructing the start of output is input to the decoder D1. In the example of FIG. 11, <go> is described as a reserved word, but it may be a fixed value set in advance. When the internal vector cm is passed from the encoder Em and a reserved word instructing the start of output is input, the decoder D1 outputs the output h1 and the internal vector changes to cm + 1. The output h1 of the decoder D1 is used for input to the decoder D2 in the next step. The internal vector cm + 1 of the decoder D1 is used as the internal state of the decoder D2 in the next step. In this way, the output ht and the internal vector ct of D1, D2, ... Dn-1 are sequentially input to the next decoders D2, D3, ..., Dn, and the final decoder Dn represents the end of output. The operation is executed sequentially until> is output.

As a result of the above, the n-1 outputs h1 to hn-1 obtained by the decoders D1 to Dn-1 are the final outputs of the learning model 230. These outputs h1 to hn-1 represent time-series data indicating an estimated value of the pressure loss in the filter cloth 611.

The control unit 101 acquires time-series data indicating an estimated value of the pressure loss in the filter cloth 611 from the decoders D1 to Dn-1 of the learning model 230. The control unit 101 sequentially repeats the input of the time-series data to the encoders E1 to Em and the acquisition of the time-series data output from the decoders D1 to Dn-1, so that the time-series change of the pressure loss in the filter cloth 611 is performed. Can be derived.

Further, the control unit 101 may estimate the replacement time of the filter cloth 611 based on the time-series change of the pressure loss estimated by the learning model 230. For example, the control unit 101 may set a threshold value for the pressure loss in advance, and estimate the time when the pressure loss estimated by the learning model 230 is equal to or greater than or equal to the threshold value as the replacement time of the filter cloth 611.

As described above, the diagnostic apparatus 100 according to the third embodiment can estimate the time-series change with respect to the pressure loss of the filter cloth 611 by using the learning model 230. Further, the diagnostic apparatus 100 can estimate the replacement time of the filter cloth 611 based on the time-series change of the pressure loss estimated by the learning model 230.

(Embodiment 4)
In the fourth embodiment, a configuration for diagnosing the degree of deterioration of the pumping system of the fluid containing the powder in the powder processing system 1 will be described. Here, the pumping system for the fluid containing the powder includes a pipe such as the powder transport path TP3 through which the fluid flows, and a blower 7 for generating the fluid in the pipe.

FIG. 12 is a schematic diagram showing a configuration example of the learning model 240 in the fourth embodiment. The learning model 240 is, for example, a learning model of machine learning including deep learning, and is configured by a neural network. The learning model 240 includes an input layer 241, intermediate layers 242A, 242B, and an output layer 243. In the example of FIG. 12, two intermediate layers 242A and 242B are described, but the number of intermediate layers is not limited to two and may be three or more.

The input layer 241, the intermediate layer 242A, 242B, and the output layer 243 have one or more nodes, and the nodes of each layer are coupled to the nodes existing in the previous and next layers in one direction with a desired weight and bias. Has been done. The same number of data as the number of nodes included in the input layer 241 is input to the input layer 241 of the learning model 240. In the present embodiment, the data input to the node of the input layer 241 is the measurement data measured by the powder processing system 1. Specifically, the rotation speed of the crushing rotor 413 measured by the rotation speed sensor S7, the rotation speed of the classification rotor 415 measured by the rotation speed sensor S8, the pressure loss of the filter cloth 611 measured by the pressure sensor S1, and the temperature. The measurement data including the temperature in the system measured by the sensor S10 is input to the node of the input layer 241. The measurement data input to the input layer 241 of the learning model 240 does not have to be limited to scalars, and may be data having some structure such as vector data and image data.

The learning model 240 according to the fourth embodiment outputs the calculation results regarding the power value of the blower 7, the flow rate of the fluid flowing through the powder processing system 1, and the pressure in the system in response to the input of these measurement data. In addition, it is learned by a predetermined learning algorithm. Since the learning method is the same as that of the first embodiment, the description thereof will be omitted.

The output layer 243 of the learning model 240 outputs the calculation results of the power value of the blower 7, the flow rate of the fluid flowing through the powder processing system 1, and the pressure in the system as the state of the powder processing system 1. Specifically, the output layer 243 is composed of n nodes from the first node to the nth node, the power value of the blower from the first node is BP1, the flow rate in the system is FV1, and the pressure SP1 in the system. A certain probability P1 is output, the probability P2 that the power value of the blower is BP2, the flow rate in the system is FV2, and the pressure SP2 in the system is output from the second node, ..., The power value of the blower is BPn from the nth node. , The probability Pn that the flow rate in the system is FVn and the pressure SPn in the system is output. The number of nodes constituting the output layer 243 and the calculation result assigned to each node are not limited to the above example, and can be appropriately designed.

FIG. 13 is a flowchart illustrating a diagnostic procedure of the diagnostic apparatus 100 according to the fourth embodiment. In the fourth embodiment, a procedure for diagnosing the soundness of a fluid containing powder in a pumping system will be described.

The control unit 101 of the diagnostic apparatus 100 receives the input of measurement data measured by various sensors (step S401). The measurement data that receives the input in step S401 includes the rotation speed of the crushing rotor 413, the rotation speed of the classification rotor 415, the pressure loss of the filter cloth 611, and the temperature in the system. Further, the control unit 101 accepts input of measurement data including the power value of the blower 7, the flow rate in the system, and the pressure in the system for comparison with the estimated value described later.

Next, the control unit 101 inputs data including the rotation speed of the crushing rotor 413, the rotation speed of the classification rotor 415, the pressure loss of the filter cloth 611, and the temperature in the system from the received measurement data as the input layer of the learning model 240. Input to 241 and execute the calculation by the learning model 240 (step S402). The learning model 240 outputs, as calculation results, calculation results regarding the power value of the blower 7, the flow rate of the fluid flowing through the powder processing system 1, and the pressure in the system.

Next, the control unit 101 acquires the calculation result from the learning model 240 (step S403), estimates the power value of the blower 7, the flow rate of the fluid flowing through the powder processing system 1, and the pressure in the system (step S404). In the output layer 243 of the learning model 240, the probability Pi (i = 1 to n) that the power value of the blower 7 is BPi, the flow rate of the fluid flowing through the powder processing system 1 is FVi, and the pressure in the system is SPi as the calculation result. ) Is output from each node. By identifying the state having the highest probability among these probability Pis (i = 1 to n), the control unit 101 determines the rotation speed of the crushing rotor 413, the rotation speed of the classification rotor 415, and the pressure loss of the filter cloth 611. , And the power value of the blower 7 estimated from the temperature in the system, the flow rate in the system, and the pressure in the system are obtained.

Next, the control unit 101 diagnoses the degree of deterioration in the pumping system based on the estimation result of step S404 (step S405). At this time, the control unit 101 has the power value of the blower 7 obtained as a measurement result, the flow rate in the system, the pressure in the system, the power value of the blower 7 estimated from the learning model 240, the flow rate in the system, and the flow rate in the system. The pressure in the casing 410 is compared, and the degree of deterioration in the pumping system is diagnosed based on the comparison result. The control unit 101 may use the MT method when comparing the measured value and the estimated value. For example, when the flow rate per power is larger than the estimated value, the control unit 101 can diagnose that deterioration or abnormality has occurred in the pumping system because there is a possibility of leakage from outside the system. Further, when the flow rate per power is smaller than the estimated value, the control unit 101 may diagnose that the pumping system is deteriorated or abnormal because there is a possibility of blockage in the system or deterioration of parts of the blower 7. Can be done. Further, the control unit 101 compares the measured value and the estimated value with respect to the pressure in the system, and identifies the part where the difference between the measured value and the estimated value is more than the threshold value, so that the blocked part or the leak occurs. The location may be specified.

Next, the control unit 101 determines whether or not an alarm is necessary based on the diagnosis result in step S405 (step S406). When the difference between the measured value and the estimated value is greater than or equal to the threshold value, the control unit 101 determines that an alarm is required. When it is determined that an alarm is necessary (S406: YES), the control unit 101 outputs an alarm (step S408). Specifically, the control unit 101 outputs an alarm by transmitting alarm information indicating that the deterioration of the pumping system is more than expected from the communication unit 105 to the terminal device 500. Further, the control unit 101 may output an alarm by displaying the alarm information indicating that the deterioration of the pumping system is more than expected on the display unit 107.

When it is determined in step S406 that an alarm is not required (S406: NO), the control unit 101 outputs the diagnosis result in step S405 (step S407). At this time, the control unit 101 may display the diagnosis result on the display unit 107. Further, the control unit 101 may transmit the diagnosis result from the communication unit 105 to the terminal device 500.

As described above, the diagnostic apparatus 100 according to the present embodiment can diagnose the soundness (degree of deterioration and presence / absence of abnormality) of the pumping system based on the estimation result of the learning model 240.

In the present embodiment, the measurement data measured at a certain timing is input to the learning model 240, but the time-series change may be diagnosed by using the same learning model as in the third embodiment. ..

(Embodiment 5)
In the fifth embodiment, a configuration for diagnosing the degree of deterioration of the crushing rotor 413 and the classification rotor 415 included in the powder processing apparatus 4 and the drive mechanism thereof will be described. In the following description, when it is not necessary to distinguish between the crushing rotor 413 and the classification rotor 415, it is also simply referred to as a rotor.

FIG. 14 is a schematic diagram showing a configuration example of the learning model 250 in the fifth embodiment. The learning model 250 is, for example, a learning model of machine learning including deep learning, and is configured by a neural network. The learning model 250 includes an input layer 251, intermediate layers 252A, 252B, and an output layer 253. In the example of FIG. 14, two intermediate layers 252A and 252B are described, but the number of intermediate layers is not limited to two and may be three or more.

The input layer 251 and the intermediate layers 252A, 252B, and the output layer 253 have one or more nodes, and the nodes of each layer are coupled to the nodes existing in the previous and next layers in one direction with a desired weight and bias. Has been done. The same number of data as the number of nodes included in the input layer 251 is input to the input layer 251 of the learning model 250. In the present embodiment, the data input to the node of the input layer 251 is the measurement data measured by the powder processing system 1. Specifically, the measurement data including the rotation speed of the crushing rotor 413 measured by the rotation speed sensor S7, the rotation speed of the classification rotor 415 measured by the rotation speed sensor S8, and the vibration value measured by the vibration sensor S11. It is input to the node of the input layer 251. The measurement data input to the input layer 251 of the learning model 250 does not have to be limited to scalars, and may be data having some structure such as vector data and image data.

The learning model 250 according to the fifth embodiment is a calculation result regarding the power value of the rotor, the pressure in the casing 410, the processing capacity (supply rate of the powder raw material), and the particle size distribution with respect to the input of these measurement data. Is trained by a predetermined learning algorithm so as to output. Since the learning method is the same as that of the first embodiment, the description thereof will be omitted.

The output layer 253 of the learning model 250 outputs the calculation results regarding the power value of the rotor, the pressure in the casing 410, the processing capacity (supply rate of the powder raw material), and the particle size distribution as the state of the powder processing system 1. .. Specifically, the output layer 253 is composed of n nodes from the first node to the nth node, the power value of the rotor from the first node is RP1, the pressure in the casing 410 is CP1, and the processing capacity is PV1. The probability P1 that the particle size distribution is DD1 is output, and the probability P2 that the power value of the rotor is RP2, the pressure in the casing 410 is CP2, the processing capacity is PV2, and the particle size distribution is DD2 is output from the second node. ..., The probability Pn that the power value of the rotor is RPn, the pressure in the casing 410 is CPn, the processing capacity is PVn, and the particle size distribution is DDn is output from the nth node. The number of nodes constituting the output layer 253 and the calculation result assigned to each node are not limited to the above example, and can be appropriately designed.

FIG. 15 is a flowchart illustrating a diagnostic procedure of the diagnostic apparatus 100 according to the fifth embodiment. In the fifth embodiment, a procedure for diagnosing the soundness of the rotor and its drive mechanism will be described.

The control unit 101 of the diagnostic apparatus 100 receives the input of measurement data measured by various sensors (step S501). The measurement data that receives the input in step S501 includes the rotation speed of the crushing rotor 413, the rotation speed of the classification rotor 415, and the vibration value in the system. Further, the control unit 101 accepts input of measurement data including the power value of the rotor, the pressure in the casing 410, the processing capacity, and the particle size distribution for comparison with the estimated value described later.

Next, the control unit 101 inputs data including the rotation speed of the crushing rotor 413, the rotation speed of the classification rotor 415, and the vibration value in the system from the received measurement data to the input layer 251 of the learning model 250 for learning. The calculation by the model 250 is executed (step S502). The learning model 250 outputs the calculation result regarding the power value of the rotor, the pressure in the casing 410, the processing capacity, and the particle size distribution as the calculation result.

Next, the control unit 101 acquires the calculation result from the learning model 250 (step S503), and estimates the power value of the rotor, the pressure in the casing 410, the processing capacity, and the particle size distribution (step S504). The output layer 253 of the learning model 250 has a probability Pi (i = 1 to n) that the power value of the rotor is RPi, the pressure in the casing 410 is CPi, the processing capacity is PVi, and the particle size distribution DDi as the calculation result. Output from each node. By identifying the state having the highest probability among these probabilities Pi (i = 1 to n), the control unit 101 determines the rotation speed of the crushing rotor 413, the rotation speed of the classification rotor 415, and the vibration value in the system. The power value of the rotor estimated from, the pressure in the casing 410, the processing capacity, and the particle size distribution are obtained.

Next, the control unit 101 diagnoses the degree of deterioration of the rotor and its drive mechanism based on the estimation result of step S504 (step S505). At this time, the control unit 101 has the rotor power value, the pressure in the casing 410, the processing capacity, and the particle size distribution obtained as the measurement result, the rotor power value estimated from the learning model 250, and the pressure in the casing 410. , Processing capacity, and particle size distribution are compared, and the degree of deterioration of the rotor and the drive mechanism is diagnosed based on the comparison result. The control unit 101 may use the MT method when comparing the measured value and the estimated value. For example, when the measured processing capacity is lower than the estimated value, the control unit 101 may wear the crushing rotor 413, deteriorate the drive mechanism including the bearing, and increase the load due to the adhesion of foreign matter. And it can be diagnosed that the drive mechanism has deteriorated. Further, when the measured vibration value is higher than the estimated value, the control unit 101 may deteriorate the drive mechanism including the bearing, increase the vibration due to the imbalance due to uneven wear, and so on. Therefore, the rotor and the drive mechanism Can be diagnosed as having deteriorated. Further, when the particle size distribution is larger than the estimated value, the control unit 101 may have a decrease in the aperture ratio due to the adhesion of foreign matter to the classification rotor 415, a decrease in the rotation speed due to deterioration of the drive mechanism, and the like. Therefore, it can be diagnosed that the rotor and the drive mechanism have deteriorated.

Next, the control unit 101 determines whether or not an alarm is necessary based on the diagnosis result in step S505 (step S506). When the difference between the measured value and the estimated value is greater than or equal to the threshold value, the control unit 101 determines that an alarm is required. When it is determined that an alarm is necessary (S506: YES), the control unit 101 outputs an alarm (step S508). Specifically, the control unit 101 outputs an alarm by transmitting alarm information from the communication unit 105 to the terminal device 500 to the effect that the deterioration of the rotor or its drive mechanism is greater than expected. Further, the control unit 101 may output an alarm by displaying the alarm information indicating that the deterioration of the rotor or its drive mechanism is more than expected on the display unit 107.

When it is determined in step S506 that the alarm is not required (S506: NO), the control unit 101 outputs the diagnosis result in step S505 (step S507). At this time, the control unit 101 may display the diagnosis result on the display unit 107. Further, the control unit 101 may transmit the diagnosis result from the communication unit 105 to the terminal device 500.

As described above, the diagnostic apparatus 100 according to the present embodiment can diagnose the soundness (degree of deterioration and presence / absence of abnormality) of the rotor and its drive mechanism based on the estimation result of the learning model 250.

In the present embodiment, the measurement data measured at a certain timing is input to the learning model 250, but the time-series change may be diagnosed by using the same learning model as in the third embodiment. ..

(Embodiment 6)
In the sixth embodiment, a configuration for diagnosing the degree of deterioration of the drive mechanism provided in the powder processing system 1 will be described. Here, the drive mechanism is a drive mechanism for rotationally driving a rotating body including a crushing rotor 413 and a classification rotor 415 provided in the powder processing system 1. Not limited to the crushing rotor 413 and the classification rotor 415, any rotating body such as a paddle or a screw may be used. In the following example, a configuration for diagnosing the degree of deterioration of the drive mechanism for driving the rotors (crushing rotor 413 and classification rotor 415) will be described.

FIG. 16 is a schematic diagram showing a configuration example of the learning model 260 in the sixth embodiment. The learning model 260 is, for example, a learning model of machine learning including deep learning, and is configured by a neural network. The learning model 260 includes an input layer 261, intermediate layers 262A, 262B, and an output layer 263. In the example of FIG. 16, two intermediate layers 262A and 262B are described, but the number of intermediate layers is not limited to two and may be three or more.

The input layer 261 and the intermediate layers 262A, 262B, and the output layer 263 have one or more nodes, and the nodes of each layer are coupled to the nodes existing in the previous and next layers in one direction with a desired weight and bias. Has been done. The same number of data as the number of nodes included in the input layer 261 is input to the input layer 261 of the learning model 260. In the present embodiment, the data input to the node of the input layer 261 is the measurement data measured by the powder processing system 1. Specifically, the supply speed of the powder raw material measured by the weight sensor S9, the rotation speed of the rotor measured by the rotation speed sensors S7 and S8, and the power value of the rotor measured by the power sensor (not shown in the figure). Measurement data including the particle size distribution measured by the particle size sensor S5, the environmental temperature measured by the temperature sensor S10, and the sound measured by the sound sensor S12 is input to the node of the input layer 261. The measurement data input to the input layer 261 of the learning model 260 does not have to be limited to scalars, and may be data having some structure such as vector data and image data.

The learning model 260 according to the sixth embodiment is trained by a predetermined learning algorithm so as to output the calculation result about the vibration value and the temperature of the drive mechanism in response to the input of these measurement data. Since the learning method is the same as that of the first embodiment, the description thereof will be omitted.

The output layer 263 of the learning model 260 outputs the calculation result about the vibration value and the temperature of the drive mechanism as the state of the powder processing system 1. Specifically, the output layer 263 is composed of n nodes from the first node to the nth node, and the probability P1 that the vibration value of the drive mechanism is DV1 and the temperature is DT1 is output from the first node. The probability P2 that the vibration value of the drive mechanism is DV2 and the temperature is DT2 is output from the two nodes, and the probability Pn that the vibration value of the drive mechanism is DVn and the temperature is DTn is output from the nth node. The number of nodes constituting the output layer 263 and the calculation result assigned to each node are not limited to the above example, and can be appropriately designed.

FIG. 17 is a flowchart illustrating a diagnostic procedure of the diagnostic apparatus 100 according to the sixth embodiment. In the sixth embodiment, a procedure for diagnosing the soundness of the drive mechanism will be described.

The control unit 101 of the diagnostic apparatus 100 receives the input of measurement data measured by various sensors (step S601). The measurement data that receives the input in step S601 includes the supply speed of the powder raw material, the rotation speed and power value of the rotor, the particle size distribution, the environmental temperature, and the sound emitted from the system. Further, the control unit 101 accepts input of measurement data including the vibration value and temperature of the drive mechanism for comparison with the estimated value described later.

Next, the control unit 101 obtains data including the supply speed of the powder raw material, the rotation speed and power value of the rotor, the particle size distribution, the environmental temperature, and the sound emitted from the system among the received measurement data of the learning model 260. Input to the input layer 261 and execute the calculation by the learning model 260 (step S602). The learning model 260 outputs the calculation result regarding the vibration value and the temperature of the drive mechanism as the calculation result.

Next, the control unit 101 acquires the calculation result from the learning model 260 (step S603), and estimates the vibration value and the temperature of the drive mechanism (step S604). The output layer 263 of the learning model 260 outputs the probability Pi (i = 1 to n) that the vibration value of the drive mechanism is DVi and the temperature of the drive mechanism is DTi as the calculation result from each node. By identifying the state having the highest probability among these probabilities Pi (i = 1 to n), the control unit 101 determines the supply speed of the powder raw material, the rotation speed and power value of the rotor, the particle size distribution, and the environment. Obtain the vibration value and temperature of the drive mechanism estimated from the temperature and the sound emitted from the system.

Next, the control unit 101 diagnoses the degree of deterioration of the drive mechanism based on the estimation result of step S604 (step S605). At this time, the control unit 101 compares the vibration value and temperature of the drive mechanism as a measurement result with the vibration value and temperature of the drive mechanism estimated from the learning model 260, and based on the comparison result, the degree of deterioration of the drive mechanism. To diagnose. The control unit 101 may use the MT method when comparing the measured value and the estimated value. For example, when the vibration value and temperature of the drive mechanism to be measured are higher than the estimated values, the control unit 101 may deteriorate the drive mechanism because there is a possibility that the vibration may increase due to imbalance due to fitting wear or uneven wear. It can be diagnosed as occurring.

Next, the control unit 101 determines whether or not an alarm is necessary based on the diagnosis result in step S605 (step S606). When the difference between the measured value and the estimated value is greater than or equal to the threshold value, the control unit 101 determines that an alarm is required. When it is determined that an alarm is necessary (S606: YES), the control unit 101 outputs an alarm (step S608). Specifically, the control unit 101 outputs an alarm by transmitting alarm information indicating that the deterioration of the drive mechanism is more than expected from the communication unit 105 to the terminal device 500. Further, the control unit 101 may output an alarm by displaying the alarm information indicating that the deterioration of the drive mechanism is more than expected on the display unit 107.

When it is determined in step S606 that the alarm is not required (S606: NO), the control unit 101 outputs the diagnosis result in step S605 (step S607). At this time, the control unit 101 may display the diagnosis result on the display unit 107. Further, the control unit 101 may transmit the diagnosis result from the communication unit 105 to the terminal device 500.

As described above, the diagnostic device 100 according to the present embodiment can diagnose the soundness (degree of deterioration and presence / absence of abnormality) of the drive mechanism based on the estimation result of the learning model 260.

In the present embodiment, the measurement data measured at a certain timing is input to the learning model 260, but the time-series change may be diagnosed by using the same learning model as in the third embodiment. ..

The embodiments disclosed this time should be considered to be exemplary in all respects and not restrictive. The scope of the present invention is shown by the scope of claims, not the above-mentioned meaning, and is intended to include all modifications within the meaning and scope equivalent to the scope of claims.

For example, the learning models 210 to 260 include not only the various parameters described above, but also image data obtained by imaging an arbitrary component of the powder processing system 1, oxygen concentration in the powder processing system 1, and powder processing system 1. Particle size, moisture, temperature, density, particle size, circularity, mixing degree, fluidity, BET (Brunauer --Emmett --Teller) value, NIR (Near Infrared), XRD (X-ray Diffraction) of the powder obtained from ), TG-DTA (Thermogravimetry-Differential Thermal Analysis), MS (Mass Spectrometry), SEM (Scanning Electron Microscope), FE-SEM (Field Emission-SEM), and TEM (Transmission Electron Microscope). It may be a learning model learned by using it.

Further, the powder processing apparatus 4 is not limited to the apparatus that performs the pulverization treatment and the classification treatment. For example, instead of the crushing treatment and the classification treatment, a drying treatment of the powder raw material, a mixing treatment of mixing two or more kinds of powder raw materials (or powder raw materials and liquid), and binding of two or more kinds of powder particles to each other. Perform surface treatment such as compounding treatment and smoothing of powder surface, or at least one granulation treatment for processing a powder raw material composed of one or a plurality of components into granules larger than the raw material using a binder or the like. It may be a processing device.

Further, the powder processing system 1 may include a plurality of powder processing devices 4, 4, ..., 4 for performing different types of processing. In this case, the learning model is prepared for each processing type of the powder processing apparatus 4. Further, the powder processing system 1 accepts the selection for the powder processing type through the operation unit 106 or the communication unit 105, and measures the learning model prepared corresponding to the selected powder processing type. The state of the powder processing system may be estimated by inputting data.

1 ... Powder processing system, 2 ... Raw material supply machine, 3 ... Hot air generator, 4 ... Powder processing equipment, 5 ... Cyclone, 6 ... Dust collector, 7 ... Blower, 611 ... Filter cloth, S1 ... Pressure sensor, S2 ... Flow sensor, S3 ... Power sensor, S4 ... Concentration sensor, S5 ... Particle size sensor, 100 ... Diagnostic device, 101 ... Control unit, 102 ... Storage unit, 103 ... Input unit, 104 ... Output unit, 105 ... Communication unit, 106 ... Operation unit, 107 ... Display unit, 500 ... Terminal device, 501 ... Control unit, 502 ... Storage unit, 503 ... Communication unit, 504 ... Operation unit, 505 ... Display unit

Claims (11)

With respect to a powder processing system including a crushing rotor for crushing a powder raw material or a classification rotor for classifying a powder obtained by crushing the powder raw material, the flow rate of a fluid flowing in the powder processing system, the crushing rotor. And an acquisition unit that acquires measurement data including at least one of the rotation speed of the classifying rotor, the rotation speed of the classification rotor, and the vibration value of the vibration generated in the powder processing system.
Depending on the input of the measurement data, the supply speed of the powder raw material, the power value of the crushing rotor, the pressure in the powder processing system, or the particle size distribution of the powder obtained from the powder processing system. Based on the measurement data acquired by the acquisition unit using a learning model configured to output the calculation results, the supply speed of the powder raw material, the power value of the crushing rotor, the pressure in the powder processing system, and the like. Or an estimation unit that estimates the particle size distribution of the powder, and
A diagnostic unit for diagnosing the degree of deterioration of the crushing rotor or the classification rotor based on the estimation result by the estimation unit.
A diagnostic device for a powder processing system including an output unit that outputs diagnostic results by the diagnostic unit. The acquisition unit acquires measurement data including the supply speed of the powder raw material, the power value of the crushing rotor, the pressure in the powder processing system, or the particle size distribution of the powder.
The diagnostic unit uses the supply speed of the powder raw material, the power value of the crushing rotor, the pressure in the powder processing system, or the particle size distribution of the powder, which are included in the measurement data, and the learning model. By comparing the supply speed of the powder raw material, the power value of the crushing rotor, the pressure in the powder processing system, or the particle size distribution of the powder, the crushing rotor or the classification rotor The diagnostic apparatus according to claim 1, wherein the degree of deterioration of the particles is diagnosed. Regarding a powder processing system including a drive mechanism for rotationally driving a rotating body provided for powder processing, the supply speed of the powder raw material, the power value and the rotation speed of the crushing rotor for crushing the powder raw material, and the powder. The power value and rotation speed of the classification rotor that classifies the powder obtained by crushing the body raw material, the particle size distribution of the powder obtained from the powder processing system, the environmental temperature, and the sound emitted from the powder processing system. An acquisition unit that acquires measurement data including at least one,
Vibration in the drive mechanism based on the measurement data acquired by the acquisition unit using a learning model configured to output the calculation result of the vibration value or temperature in the drive mechanism in response to the input of the measurement data. An estimation unit that estimates the value or temperature, and
A diagnostic unit that diagnoses the degree of deterioration of the drive mechanism based on the vibration value or temperature estimated by the estimation unit.
A diagnostic device for a powder processing system including an output unit that outputs diagnostic results by the diagnostic unit. The measurement data acquired by the acquisition unit includes the vibration value or temperature in the drive mechanism.
The diagnostic unit compares the vibration value or temperature in the drive mechanism included in the measurement data with the vibration value or temperature in the drive mechanism estimated using the learning model, thereby deteriorating the drive mechanism. The diagnostic device according to claim 3, wherein the degree is diagnosed. The diagnostic unit diagnoses the presence or absence of an abnormality in the powder processing system and determines the presence or absence of an abnormality.
The diagnostic device according to any one of claims 1 to 4, wherein the output unit outputs an alarm when the diagnostic unit diagnoses that there is an abnormality in the powder processing system. The acquisition unit is an image data obtained by imaging a component of the powder processing system, an oxygen concentration in the powder processing system, a particle size, a moisture content, a temperature, and a density of the powder obtained from the powder processing system. , Particle size, roundness, mixing degree, fluidity, BET (Brunauer --Emmett --Teller) value, NIR (Near Infrared), XRD (X-ray Diffraction), TG-DTA (Thermogravimetry --Differential Thermal Analysis), MS ( Any of claims 1 to 5 for acquiring measurement data including at least one of Mass Spectrometry), SEM (Scanning Electron Microscope), FE-SEM (Field Emission --SEM), and TEM (Transmission Electron Microscope) data. The diagnostic device according to one. It is equipped with a reception unit that accepts selections for the type of powder processing performed by the powder processing system.
Claim 1 estimates the state of the powder processing system by inputting the measurement data into a learning model provided corresponding to the type of powder processing selected by the reception unit. The diagnostic device according to any one of claims 6. The computer
With respect to a powder processing system including a crushing rotor for crushing a powder raw material or a classification rotor for classifying a powder obtained by crushing the powder raw material, the flow rate of a fluid flowing in the powder processing system, the crushing rotor. The measurement data including at least one of the rotation speed of the classifying rotor, the rotation speed of the classification rotor, and the vibration value of the vibration generated in the powder processing system is acquired.
Depending on the input of the measurement data, the supply speed of the powder raw material, the power value of the crushing rotor, the pressure in the powder processing system, or the particle size distribution of the powder obtained from the powder processing system. Based on the acquired measurement data using a learning model configured to output the calculation result, the supply speed of the powder raw material, the power value of the crushing rotor, the pressure in the powder processing system, or the powder. Estimate the particle size distribution of
Based on the estimation result, the degree of deterioration of the crushing rotor or the classification rotor is diagnosed.
A diagnostic method for a powder processing system that outputs diagnostic results. The computer
Regarding a powder processing system including a drive mechanism for rotationally driving a rotating body provided for powder processing, the supply speed of the powder raw material, the power value and the rotation speed of the crushing rotor for crushing the powder raw material, and the powder. The power value and rotation speed of the classification rotor that classifies the powder obtained by crushing the body raw material, the particle size distribution of the powder obtained from the powder processing system, the environmental temperature, and the sound emitted from the powder processing system. Acquire measurement data including at least one and
Using a learning model configured to output the calculation result of the vibration value or temperature in the drive mechanism in response to the input of the measurement data, the vibration value or temperature in the drive mechanism is calculated based on the acquired measurement data. Estimate and
Based on the estimated vibration value or temperature, the degree of deterioration of the drive mechanism is diagnosed, and
A diagnostic method for a powder processing system that outputs diagnostic results. On the computer
With respect to a powder processing system including a crushing rotor for crushing a powder raw material or a classification rotor for classifying a powder obtained by crushing the powder raw material, the flow rate of a fluid flowing in the powder processing system, the crushing rotor. The measurement data including at least one of the rotation speed of the classifying rotor, the rotation speed of the classification rotor, and the vibration value of the vibration generated in the powder processing system is acquired.
Depending on the input of the measurement data, the supply speed of the powder raw material, the power value of the crushing rotor, the pressure in the powder processing system, or the particle size distribution of the powder obtained from the powder processing system. Based on the acquired measurement data using a learning model configured to output the calculation result, the supply speed of the powder raw material, the power value of the crushing rotor, the pressure in the powder processing system, or the powder. Estimate the particle size distribution of
Based on the estimation result, the degree of deterioration of the crushing rotor or the classification rotor is diagnosed.
A computer program for executing processes that output diagnostic results. On the computer
Regarding a powder processing system including a drive mechanism for rotationally driving a rotating body provided for powder processing, the supply speed of the powder raw material, the power value and the rotation speed of the crushing rotor for crushing the powder raw material, and the powder. The power value and rotation speed of the classification rotor that classifies the powder obtained by crushing the body raw material, the particle size distribution of the powder obtained from the powder processing system, the environmental temperature, and the sound emitted from the powder processing system. Acquire measurement data including at least one and
Using a learning model configured to output the calculation result of the vibration value or temperature in the drive mechanism in response to the input of the measurement data, the vibration value or temperature in the drive mechanism is calculated based on the acquired measurement data. Estimate and
Based on the estimated vibration value or temperature, the degree of deterioration of the drive mechanism is diagnosed, and
A computer program for executing processes that output diagnostic results.

Images