WO2021001897A1

WO2021001897A1 - Learning model generation method, computer program, learning model, control device, and control method

Info

Publication number: WO2021001897A1
Application number: PCT/JP2019/026150
Authority: WO
Inventors: 智浩北村
Original assignee: ホソカワミクロン株式会社
Priority date: 2019-07-01
Filing date: 2019-07-01
Publication date: 2021-01-07
Also published as: JP7311598B2; JPWO2021001897A1

Abstract

Provided are a learning model generation method, a computer program, a learning model, a control device, and a control method.　By using a computer, with respect to a powder treatment device which performs powder treatment including any one among dry treatment, mixing treatment, composite treatment, surface treatment, and granulation treatment, a learning model is generated which is configured to output a calculation result for a control parameter with respect to the powder treatment device, when measurement data indicating an operation state of the powder treatment device and powder data pertaining to powder obtained from the powder treatment device are acquired, the acquired measurement data and powder data are used as teacher data, and a target value for the powder, which is desired by a user, is input.

Description

Learning model generation method, computer program, learning model, control device, and control method

The present invention relates to a learning model generation method, a computer program, a learning model, a control device, and a control method.

The powder treatment 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 set the operating conditions of the powder processing apparatus.

Japanese Unexamined Patent Publication No. 2008-194592

However, the powder handled in the powder processing process exhibits a wide variety of properties depending on the type, size, shape, etc. of the powder. Therefore, there is a problem that the operating conditions set in the powder processing apparatus often depend on the experience and intuition of field engineers.

The present invention provides a learning model generation method, a computer program, a learning model, a control device, and a control method capable of determining control parameters related to a powder processing apparatus without depending on the experience and intuition of a field engineer. The purpose.

The method for generating a learning model according to one aspect of the present invention is a powder that uses a computer to perform powder treatment including any one of drying treatment, mixing treatment, compounding treatment, surface treatment, and granulation treatment. Regarding the processing device, measurement data indicating the operating state of the powder processing device and powder data related to the powder obtained from the powder processing device are acquired, and the acquired measurement data and powder data are used as teacher data. When a target value for the powder desired by the user is input, a learning model configured to output the calculation result for the control parameter for the powder processing apparatus is generated.

According to the present application, control parameters related to the powder processing apparatus can be determined without depending on the experience and intuition of field engineers.

It is a schematic diagram which shows the whole structure of the powder processing system which concerns on Embodiment 1. FIG. It is a schematic cross-sectional view which shows the structure of the powder processing apparatus which concerns on Embodiment 1. FIG. It is a block diagram which shows the internal structure of the control device which concerns on Embodiment 1. FIG. It is a block diagram which shows the internal structure of a terminal device. 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 control device collects. It is a flowchart explaining the generation procedure of the learning model by a control device. It is a flowchart explaining the control procedure by a control device. It is a schematic diagram which shows the structure of the powder processing apparatus which concerns on Embodiment 2. 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 structure of the powder processing apparatus which concerns on Embodiment 3. 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 structure of the powder processing apparatus which concerns on Embodiment 4. It is a schematic diagram which shows the structural example of the learning model in Embodiment 4. It is a schematic diagram which shows the structure of the powder processing apparatus which concerns on Embodiment 5. It is a schematic diagram which shows the structural example of the learning model in Embodiment 5. It is a flowchart explaining the re-learning procedure of a learning model. It is a flowchart explaining the adjustment procedure of a control parameter. It is a flowchart explaining the procedure of the process executed by a terminal device and a control device. It is a schematic diagram which shows an example of the target value input screen. It is a schematic diagram which shows an example of a monitoring screen. It is a flowchart explaining the procedure of the process executed by a terminal device and a control device. It is a schematic diagram which shows an example of the processing type selection screen. It is a conceptual diagram which shows the example of collecting data in Embodiment 10. It is a flowchart explaining the generation procedure of the learning model which concerns on Embodiment 10. It is a flowchart explaining the control procedure by a control device.

Hereinafter, the present invention will be specifically described with reference to the drawings showing the embodiments thereof.
(Embodiment 1)
FIG. 1 is a schematic view showing the overall configuration of the powder processing system 1 according to the first embodiment. The powder processing system 1 according to the first embodiment includes, for example, a raw material supply machine 2, a hot air generator 3, a powder processing device 4A, a cyclone 5, a dust collector 6, a blower 7, a product tank 8, a dust collecting tank 9, and a control. A device 100 (see FIG. 3) is provided.

The raw material supply machine 2 is a device for supplying the powder raw material to the powder processing device 4A. The powder raw material supplied by the raw material feeder 2 to the powder processing apparatus 4A 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, solder, cement, foods, etc.

The raw material supply machine 2 is connected to the powder processing device 4A 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 4A together with the hot air. Further, the raw material feeder 2 performs weight control using a weight sensor S1 (see FIG. 3) such as a load cell, and the amount of retention in the device is constant even when continuous processing is performed in the powder processing device 4A. 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 S1. May be measured. The weight sensor S1 may be provided not only in the raw material supply machine 2, but also in the powder processing device 4A, 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 4A, and includes a heat source such as a heating 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 4A may be recovered and circulated between the hot air generator 3 and the powder processing apparatus 4A.

The hot air generator 3 is connected to the powder processing device 4A via the gas introduction path TP2. The hot air generated by the hot air generator 3 is introduced into the powder processing apparatus 4A 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 4A. For example, hot air of about 200 ° C. to 600 ° C. may be generated in order to dry the powder in the powder processing apparatus 4A.

The powder processing device 4A according to the first embodiment is a device that performs a drying process. The powder processing apparatus 4A is an air flow dryer, and for example, a dry meister (registered trademark) manufactured by Hosokawa Micron Co., Ltd. is used. The powder processing apparatus 4A has a pulverizing function for pulverizing the powder raw material supplied into the apparatus and a classification function for classifying the powder obtained by pulverizing the powder raw material. The internal configuration of the powder processing apparatus 4A will be specifically described with reference to FIG.

The powder processed by the powder processing apparatus 4A is transported to the cyclone 5 via the powder transport path TP3. In the present embodiment, the particle size sensor S2 is installed in the middle of the powder transport path TP3 from the powder processing device 4A to the cyclone 5, and the particle size of the powder passing through the powder processing device 4A by the particle size sensor S2. Is measured at regular or regular timings (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 S2 is a device that measures the particle size distribution using, for example, a laser diffraction / scattering method, and outputs 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, D90, a mode diameter representing the particle diameter having the largest appearance ratio in the frequency distribution of the particle diameter may be used, and various arithmetic average values (number average, length average, area average, volume average, etc.) may be used. ) May be used.
In the present embodiment, the particle size sensor S2 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 are reached. It may be installed on a route or the like.

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 powder or the like that has 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 powder 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 air passage TP5. By driving the blower 7, the gas flow from the powder processing apparatus 4A to the cyclone 5 (that is, the gas flow for extracting the powder from the powder processing apparatus 4A) and the gas flow from the cyclone 5 to the dust collector 6 Form a flow. In the present embodiment, a flow rate sensor S3 (see FIG. 3) is installed in the middle of the exhaust air passage TP5 from the dust collector 6 to the blower 7, and the discharge / suction flow rate when taking out the powder from the powder processing device 4A is measured. Measure at regular or regular timing (for example, every 5 seconds). Instead of installing the flow rate sensor S3 in the exhaust air passage TP5, the flow rate sensor S3 is installed in the raw material supply path TP1, the gas introduction path TP2, the powder transport path TP3, the dust collection path TP4, the discharge port of the blower 7, and the like. You may.

The control device 100 is a device that controls the operation of the powder processing device 4A, and is configured so that necessary data can be exchanged between various devices and various sensors constituting the powder processing system 1. The control device 100 directly controls the operation of the powder processing device 4A by transmitting a control command for the powder processing device 4A to the powder processing device 4A. Further, the control device 100 transmits a control command to at least one of the raw material supply machine 2, the hot air generator 3, the cyclone 5, the dust collector 6, and the blower 7 connected to the powder processing device 4A to obtain powder. The operation of the processing device 4A may be indirectly controlled.

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 4A, the cyclone 5, the dust collector 6, and the blower 7 has been described, but is connected to the powder processing device 4A. 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.

FIG. 2 is a schematic cross-sectional view showing the configuration of the powder processing apparatus 4A according to the first embodiment. The powder processing apparatus 4A includes a cylindrical casing 410 that performs a drying process inside the powder processing apparatus 4A. 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, SPHC (Steel Plate Hot Commercial), stainless steel materials such as SUS304 and SUS316, iron casting materials such as FC20 and FC40, and stainless casting materials such as SCS13 and 14 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, 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, electropolished, coated with PTFE (Polytetrafluoroethylene), plated with nickel or the like. Treatment 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. The raw material input port 411 is preferably 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 lower than 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. To 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.

Here, depending on the treatment conditions of the powder treatment, a softening phenomenon may occur in the powder, the powders may be fused to each other, the particle size may vary, or the yield may decrease. Therefore, a temperature sensor S4, a moisture sensor S5, and a pressure sensor S6 (see FIG. 3) may be provided at one or a plurality of locations in the casing 410 to sequentially observe the temperature, moisture, pressure, and the like. Further, the flow rate sensor S3, the temperature sensor S4, the moisture sensor S5, and the pressure sensor S6 may be provided in each device or a path between the devices constituting the powder processing system 1. Usually, the temperature sensor S4 is often provided in the gas introduction path TP2, and the moisture sensor S5 is often provided in the powder transport path TP3. Further, it may be separately provided in the casing 410.

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. The cold air (refrigerant) may be introduced. 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 heat medium or a refrigerant 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 of the rotary disk 413A, and the rotation speed is adjusted to a desired speed by the power of the crushing motor 413M (see FIG. 3). It is configured to rotate. Here, the crushing motor 413M is one of the driving units included in the powder processing system 1. The rotation speed of the crushing motor 413M is measured by the rotation speed sensor S7 (see FIG. 3) 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 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-like structure may be used.

The crushing rotor 413 is rotated by the power of the crushing motor 413M to generate a swirling airflow in the casing 410, and the powder raw material introduced into the casing 410 is impacted, compressed, and ground by the action of the hammer 413B. The powder raw material is crushed by giving mechanical energy such as shearing. The crushing rotor 413 is an example of a rotating body that agitates the powder raw material.

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 tip may be attached so that the hammer 413B can withstand an impact force, or a composite of a metal such as ceramics or cermet having wear resistance and toughness and ceramics may be used.

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 SUS630 and quench hardening treatment of SUS630 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 pulverize the powder raw materials 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 tubular 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 fixed inside the casing 410. The method for fixing 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 4A. This is to control the flow state of the powder to be processed to an appropriate state inside the casing 410. In FIG. 2, the guide ring 414 whose inner diameter does not change in the vertical direction is shown, but the inner diameter may be continuously increased (or decreased) from the upper side to the lower side.

The powder produced by crushing the powder raw material is efficiently dried by violently contacting the hot air flowing from the lower part of the casing 410. The dried and pulverized powder is guided by an air flow to the classification rotor 415 provided on the upper part of the casing 410.

The classification rotor 415 is a rotor provided with a plurality of classification blades 415A arranged radially, and is configured to rotate at a desired rotation speed by the power of the classification motor 415M (see FIG. 3). Here, the classification motor 415M is one of the drive units included in the powder processing system 1. The rotation speed of the classification motor 415M is measured by the rotation speed sensor S8 (see FIG. 3) at regular or periodic timings (for example, at 5-second intervals). The classification rotor 415 is provided above the crushing rotor 413, and only the powder having a particle size smaller than a predetermined particle size is passed through the powder processed in the casing 410 by centrifugal force due to high-speed rotation. Only the body is guided to the powder outlet 416.

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 curing treatment after spraying in order to improve the durability of the device. Abrasion resistant treatments 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 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.

Here, 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.

The powder that has passed through the classification rotor 415 and is guided to the powder outlet 416 is taken out to the outside by the flow of gas produced by the blower 7, and is guided to the cyclone 5 and the dust collector 6 in the subsequent stage.

Hereinafter, the configuration of the control system in the powder processing system 1 will be described.
FIG. 3 is a block diagram showing an internal configuration of the control device 100 according to the first embodiment. The control device 100 is composed of a general-purpose or dedicated computer, 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 control device 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, to control the operation of each hardware unit, thereby as a control 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 more 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 device, and the like.

The computer program stored in the storage unit 102 includes a control program PG1 for controlling the operation of the device to be controlled, a particle size of the powder processed by the powder processing device 4A, and control parameters related to the powder processing device 4A. Includes a learning program PG2 and the like for learning the relationship with.

The computer program including the control program PG1 and the learning program PG2 may be provided by the non-temporary recording medium M1 in which the computer program is readablely recorded. The recording medium M1 is, for example, 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.

In the present embodiment, when the control unit 101 executes the learning program PG2, the control device 100 shifts to the learning phase. The learning phase is carried out, for example, when the control device 100 is introduced. Further, the learning phase may be executed when a user's instruction is received through the operation unit 106 or at a regular timing.

In the learning phase, the control unit 101 acquires the data of the wet content of the powder obtained from the powder processing apparatus 4A as the powder data, and the powder processing apparatus as the measurement data indicating the operating state of the powder processing apparatus 4A. Data such as temperature and flow rate of heat medium supplied to 4A, supply amount or supply speed of powder raw material, rotation speed of crushing rotor 413, rotation speed of classification rotor 415, pressure in casing 410, processing time of drying process, etc. To get. Here, the moisture content of the powder is measured by the moisture content sensor S5. The temperature and flow rate of the heat medium are measured by the temperature sensor S4 and the flow rate sensor S3. The supply amount or supply speed of the powder raw material is calculated based on the weight measured by the weight sensor S1. The rotation speed of the crushing rotor 413 is measured by the rotation speed sensor S7. The rotation speed of the classification rotor 415 is measured by the rotation speed sensor S8. The processing time of the drying process is measured by the built-in timer of the control unit 101. The processing time of the drying process corresponds to the operating time of the powder processing apparatus 4A in the batch processing, and corresponds to the residence time of the powder in the casing 410 in the continuous processing. These data may be collected in advance or after the transition to the learning phase.

The control unit 101 uses the collected powder data and measurement data as teacher data to learn the relationship between the moisture content of the powder obtained from the powder processing device 4A and the control parameters related to the powder processing device 4A. , Generate a learning model 210. Here, the control parameters related to the powder processing apparatus 4A are the temperature and flow rate of the heat medium supplied to the powder processing apparatus 4A, the supply amount or supply speed of the powder raw material, the rotation speed of the crushing rotor 413, and the rotation of the classification rotor 415. Includes speed, pressure in casing 410, and processing time for drying.

The learning model 210 generated in the learning phase is stored in the storage unit 102. The learning model 210 is defined by its definition information. The definition information of the learning model 210 includes, for example, the structural information of the learning model 210, parameters such as weights and biases between nodes.

Further, in the present embodiment, the control device 100 shifts to the operation phase when the control unit 101 executes the control program PG1. The operation phase is carried out after the learning model 210 is executed.

In the operation phase, the control device 100 controls the operation of the device including the powder processing device 4A. At this time, the control unit 101 receives the target value for the powder desired by the user. In the first embodiment, the target value is the moisture content of the powder. The control unit 101 inputs the received target value into the learning model 210, executes an operation using the learning model 210, and thereby supplies the temperature and flow rate of the heat medium supplied to the powder processing apparatus 4A and the powder raw material. The calculation result regarding the control parameters including the quantity or the supply speed, the rotation speed of the crushing rotor 413, the rotation speed of the classification rotor 415, the pressure in the casing 410, and the processing time of the drying process is acquired. The control unit 101 controls the operation of the device to be controlled based on the calculation result obtained from the learning model 210.

The input unit 103 includes a connection interface for connecting various devices and sensors. The device connected to the input unit 103 includes a raw material supply machine 2, a hot air generator 3, a powder processing device 4A, a cyclone 5, a dust collector 6, and a blower 7. The connection interface included in the input unit 103 may be a wired interface or a wireless interface. Data output from the raw material supply machine 2, the hot air generator 3, the powder processing device 4A, the cyclone 5, the dust collector 6, and the blower 7 is input to the input unit 103.

The sensors connected to the input unit 103 include a weight sensor S1, a particle size sensor S2, a flow rate sensor S3, a temperature sensor S4, a moisture sensor S5, a pressure sensor S6, a rotation speed sensor S7 for the crushing rotor 413, and a classification. The rotation speed sensor S8 for the rotor 415 is included. The measurement data output from these sensors is input to the input unit 103.

The sensor connected to the input unit 103 is not limited to the above sensor, and may include a dust concentration sensor for measuring the dust concentration, a sound pressure / frequency sensor for measuring the sound pressure or the frequency, and the like. Good. Furthermore, in order to measure the composition, physical properties, shape, etc. of the powder, BET (Brunauer-Emmett-Teller) value, NIR (NearInfrared), XRD (X-ray Diffraction), TG-DTA (Thermogravimetry-Differential Thermal Analysis) ), MS (Mass Spectrometry), SEM (Scanning Electron Microscope), FE-SEM (Field Emission-SEM), TEM (Transmission Electron Microscope), and other measuring devices that output data may be connected to the input unit 103.

Further, the control device 100 is configured to acquire data from various devices and sensors through the input unit 103, but when these devices and sensors are provided with a communication interface, the data may be acquired through the communication unit 105 described later. ..

The output unit 104 includes a connection interface for connecting the device to be controlled. The device connected to the output unit 104 includes a raw material supply machine 2, a hot air generator 3, a powder processing device 4A, a cyclone 5, a dust collector 6, and a blower 7. The connection interface included in the output unit 104 may be a wired interface or a wireless interface. The control unit 101 controls the operation of the device to be controlled by outputting a control command through the output unit 104. For example, when controlling the rotation speed of the crushing rotor 413, the control unit 101 generates a control command for the crushing motor 413M, which is a driving unit of the crushing rotor 413, and outputs the control command to the powder processing apparatus 4A through the output unit 104. , Control the rotation speed of the crushing rotor 413. The same applies to the case of controlling the rotation speed of the classification rotor 415, and the control unit 101 generates a control command for the classification motor 415M which is a drive unit of the classification rotor 415 and outputs the control command to the powder processing device 4A through the output unit 104. Thereby, the rotation speed of the classification rotor 415 is controlled. Further, when controlling the discharge / suction flow rate from the casing 410 included in the powder processing apparatus 4A, the control unit 101 generates a control command for the blower 7 and outputs the control command to the blower 7 through the output unit 104, whereby the casing 410. Controls the discharge / suction flow rate from.

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 the communication standard of LAN (Local Area Network) used in WiFi (registered trademark) and 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. The communication unit 105 may receive the operation data or the setting data transmitted from the terminal device 500 in order to accept the remote operation of the powder processing system 1. When the control unit 101 receives the operation data or the setting data transmitted from the terminal device 500 through the communication unit 105, the control unit 101 executes a process according to the received operation data or the setting data. For example, when the user receives the desired moisture data, the control unit 101 may execute a calculation using the learning model 210 to acquire control parameters related to the powder processing apparatus 4A. 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 accepts 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 control device 100 is provided with the operation unit 106, but the operation unit 106 is not indispensable, and the operation is received through a computer (for example, the terminal device 500) connected to the outside. 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 S6 received through the communication unit 105, or may display information based on various operations and various settings received through the operation unit 106. Further, the display unit 107 may display the calculation result by the learning model 210. In the present embodiment, the control 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 control device 100 has been described as a single computer, but it does not have to be a single computer, and may be composed of a plurality of computers or a plurality of virtual computers.

FIG. 4 is a block diagram showing the 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 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. Data and the like used during execution of the calculation are temporarily stored in the RAM included in the control unit 501.

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

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 control device 100 of the powder processing system 1. The data received by the terminal device 500 through the communication unit 503 includes screen data for displaying the interface screen of the control device 100 on the display unit 505, data indicating the setting state and operating state of the device including the powder processing device 4A, and the like. Including. Here, the data indicating the set state includes the temperature and flow rate of the heat medium, the supply amount or supply speed of the powder raw material, the rotation speed of the crushing rotor 413, the rotation speed of the classification rotor 415, the pressure in the casing 410, and the drying process. Contains the setting values related to the processing time of. In addition, the data showing the operating state includes the temperature and flow rate of the heat medium, the supply amount or supply speed of the powder raw material, the rotation speed of the crushing rotor 413, the rotation speed of the classification rotor 415, the pressure in the casing 410, and the drying process. Measurement data such as processing time is included. The data received by the communication unit 503 is output to the control unit 501. The data transmitted by the terminal device 500 through the communication unit 503 includes operation data, setting data, and the like when the powder processing system 1 is remotely controlled.

The operation unit 504 is equipped 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 of the control device 100, for example, based on the screen data received by the communication unit 503. Further, the display unit 505 displays the set state and the operating state of the device including the powder processing device 4A based on the data indicating the setting state and the operating state of the device including the powder processing device 4A received by the communication unit 503. You may.

Hereinafter, the learning model 210 used in the control device 100 will be described.
FIG. 5 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 for machine learning including deep learning, and is composed of a neural network. The learning model 210 includes an input layer 211,

intermediate layers

212A and 212B, and an output layer 213. In the example of FIG. 5, 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 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 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 target value data for the powder desired by the user. In the first embodiment, the target value for the powder is the moisture content of the powder obtained from the powder processing apparatus 4A. The target value data to be input to the input layer 211 of the learning model 210 does not have to be limited to scalar data, and may be data having some structure such as vector data and image data.

The target value data input to the learning model 210 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 nodes 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 calculation using the activation function including the weight and the bias set between the nodes is executed, and the calculation is transmitted to the subsequent layers one after another until the calculation result by the output layer 213 is obtained. Parameters such as weights and biases that connect the 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, powder data (moisture data) related to the powder obtained from the powder processing apparatus 4A, temperature and flow rate of the heat medium, supply amount or supply speed of the powder raw material, and rotation speed of the crushing rotor 413. , The rotation speed of the classification rotor 415, the pressure in the casing 410, and the measurement data including the processing time of the drying process are used as training data, and various parameters including weights and biases between nodes are learned by a predetermined learning algorithm. Can be done.

The output layer 213 outputs the calculation result for the control parameter for the powder processing apparatus 4A. Here, the control parameters are the temperature and flow rate of the heat medium supplied to the powder processing apparatus 4A, the supply amount or supply speed of the powder raw material, the rotation speed of the crushing rotor 413, the rotation speed of the classification rotor 415, and the inside of the casing 410. Includes pressure as well as processing time for drying. As the calculation result, for example, the probability indicating the quality of the combination of the plurality of control parameters described above may be output. Specifically, the output layer 213 is composed of n nodes from the first node to the nth node, and from the first node, the temperature of the heat medium is HT1, the flow rate of the heat medium is HF1, and the supply amount of the powder raw material. SA1, the rotation speed of the crushing rotor 413 is CV1, the rotation speed of the classification rotor 415 is CC1, the pressure in the casing 410 is CP1, and the probability P1 that the processing time of the drying process is DT1 is output, and heat is output from the second node. The medium temperature is HT2, the heat medium flow rate is HF2, the powder raw material supply amount is SA2, the rotation speed of the crushing rotor 413 is CV2, the rotation speed of the classification rotor 415 is CC2, the pressure inside the casing 410 is CP2, and the drying process. Outputs the probability P2 that the processing time is DT2, and ..., From the nth node, the temperature of the heat medium is HTn, the flow rate of the heat medium is HFn, the supply amount of the powder raw material is SAN, and the rotation speed of the crushing rotor 413 is CVn. , The probability Pn that the rotation speed of the classification rotor 415 is CCn, the pressure in the casing 410 is CPn, and the processing time of the drying process is DTn may be output. The number of nodes constituting the output layer 213 and the calculation result assigned to each node are not limited to the above examples, and can be appropriately designed.

The control device 100 includes powder data related to the powder obtained from the powder processing device 4A (data on the wet content of the powder in the first embodiment), the temperature of the heat medium supplied to the powder processing device 4A, and Measurement data including flow rate, supply amount or supply speed of powder raw material, rotation speed of crushing rotor 413, rotation speed of classification rotor 415, pressure in casing 410, and processing time of drying process are collected, and these data are collected. Is used as the teacher data to generate the learning model 210 as described above.

FIG. 6 is a conceptual diagram showing an example of data collected by the control device 100. The control unit 101 of the control device 100 acquires measurement data indicating the operating state of the powder processing device 4A and powder data related to the powder obtained from the powder processing device 4A through the input unit 103. Specifically, the control unit 101 measures the temperature of the heat medium (° C.) measured by the temperature sensor S4, the flow rate of the heat medium (m ³ / min) measured by the flow sensor S3, and the weight sensor S1. Supply amount of powder raw material calculated based on weight (kg / h), rotation speed (rpm) of crushing rotor 413 measured by rotation speed sensor S7, rotation speed of classification rotor 415 measured by rotation speed sensor S8. (Rotation), the pressure (atm) in the casing 410 measured by the pressure sensor S6, the processing time (min) of the drying process measured by the built-in timer of the control unit 101, and the like are acquired as measurement data. Further, the control unit 101 acquires the moisture content (%) of the powder measured by the moisture content sensor S5 as powder data. The control unit 101 stores the acquired measurement data and powder data in the storage unit 102 together with the time stamp. Note that FIG. 6 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.

The control unit 101 uses the collected data as teacher data to learn the relationship between the moisture content of the powder obtained from the powder processing apparatus 4A and the control parameters related to the powder processing apparatus 4A, as described above. Generate the training model 210.

Hereinafter, the operation of the control device 100 in the learning phase will be described.
FIG. 7 is a flowchart illustrating a procedure for generating the learning model 210 by the control device 100. The control unit 101 of the control device 100 collects measurement data indicating the operating state of the powder processing device 4A and powder data related to the powder obtained from the powder processing device 4A (step S101). As described above, the measurement data collected in step S101 includes the temperature and flow rate of the heat medium, the supply amount or supply speed of the powder raw material, the rotation speed of the crushing rotor 413, the rotation speed of the classification rotor 415, and the pressure in the casing 410. , And the processing time of the drying process. Further, the powder data collected in step S101 includes data indicating the moisture content of the powder obtained from the powder processing apparatus 4A. The collected measurement data and powder data are stored in the storage unit 102 together with the time stamp.

After collecting the measurement data and the particle size 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 determines the temperature and flow rate of the heat medium, the supply amount or supply speed of the powder raw material, the rotation speed of the crushing rotor 413, the rotation speed of the classification rotor 415, the pressure in the casing 410, and the processing time of the drying process. Only one set of the measured value and the wet content value of the powder obtained at that time is selected from the storage unit 102.

The control unit 101 inputs the wet content value included in the selected teacher data into the learning model 210 (step S103), and executes the calculation by the learning model 210 (step S104). That is, the control unit 101 inputs the moisture value to the nodes constituting the input layer 211 of the learning model 210, performs calculations using the weights and biases between the nodes in the

intermediate layers

212A and 212B, and outputs the calculation results. The process of outputting from the node of layer 213 is performed. In the initial stage before the start of learning, 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. The control unit 101 is in the process of optimizing (minimizing or maximizing) the error function by a gradient descent method such as the steepest descent method, and when the error function is below the threshold value (or above the threshold value), the learning is completed. to decide. In order to avoid the problem of overfitting, techniques such as cross-validation and early stopping may be adopted to end learning 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 an error backpropagation method that sequentially updates the weights and biases between the nodes from the output layer 213 of the learning model 210 toward the input layer 211. it 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.

As described above, in the learning phase, the control device 100 according to the present embodiment obtains the measurement data indicating the operating state of the powder processing device 4A and the powder data related to the powder obtained from the powder processing device 4A. collect. By using the collected data as teacher data, the control device 100 determines the temperature and flow rate of the heat medium according to the input of the target value (moisture value in the first embodiment) for the powder desired by the user. Learning model 210 that outputs calculation results related to control parameters that control the supply amount or supply speed of the powder raw material, the rotation speed of the crushing rotor 413, the rotation speed of the classification rotor 415, the pressure in the casing 410, and the processing time of the drying process. Can be generated.

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

Next, the operation of the control device 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. 8 is a flowchart illustrating a control procedure by the control device 100. The control unit 101 of the control device 100 receives an input of a target value (moisture content value in the first embodiment) for the powder desired by the user through the operation unit 106 (step S121).

Next, the control unit 101 inputs the received target value data to the input layer 211 of the learning model 210, and executes the calculation by the learning model 210 (step S122). At this time, the control unit 101 gives the received target value data to the node of the input layer 211. 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 the calculation result regarding the control parameter of the powder processing apparatus 4A. Specifically, each node of the output layer 213 has the temperature and flow rate of the heat medium, the supply amount or supply speed of the powder raw material, the rotation speed of the crushing rotor 413, the rotation speed of the classification rotor 415, and the pressure in the casing 410. And output the calculation result regarding the control parameter that controls the processing time of the drying process.

Next, the control unit 101 acquires the calculation result from the learning model 210 (step S123) and determines the control parameters used for control (step S124). In the output layer 213 of the learning model 210, for example, as a result of calculation regarding control parameters, the temperature of the heat medium is HTi, the flow rate of the heat medium is HFi, the supply amount of the powder raw material is SAi, and the rotation speed of the crushing rotor 413 is CVi. Each node outputs the probability Pi (i = 1 to n) that the rotation speed of the classification rotor 415 is CCi, the pressure in the casing 410 is CPi, and the processing time of the drying process is DTi. The control unit 101 has the temperature and flow rate of the heat medium corresponding to the highest probability among these probabilities Pi (i = 1 to n), the supply amount or supply speed of the powder raw material, the rotation speed of the crushing rotor 413, and the classification. The control parameters used for control are determined by specifying the combination of the rotation speed of the rotor 415, the pressure in the casing 410, and the processing time of the drying process.

Next, the control unit 101 executes control based on the control parameters determined in step S124 (step S125). That is, the control unit 101 generates and generates a control command for controlling the operation of the hot air generator 3 so that the temperature and the flow rate of the heat medium introduced into the casing 410 become the values determined in step S124, respectively. The control command is output to the hot air generator 3 through the output unit 104. Similarly, the control unit 101 has values determined in step S124 for each of the supply amount of the powder raw material, the rotation speed of the crushing rotor 413, the rotation speed of the classification rotor 415, the pressure in the casing 410, and the processing time of the drying process. As described above, a control command for controlling the operation of the raw material supply machine 2, the crushing rotor 413, the classification rotor 415, the blower 7, and the powder processing device 4A is generated, and the generated control command is output to each device through the output unit 104.

As described above, the control device 100 according to the present embodiment receives the target value for the powder desired by the user, and thus receives the temperature and flow rate of the heat medium, the supply amount or the supply speed of the powder raw material, and the crushing rotor. The control parameters that control the rotation speed of the 413, the rotation speed of the classification rotor 415, the pressure in the casing 410, and the processing time of the drying process can be determined. That is, in the present embodiment, the control parameters related to the powder processing apparatus 4A can be determined without depending on the experience and intuition of the field engineer. The control device 100 executes control based on the determined control parameters so that the powder having the target value can be obtained.

In the present 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 a search tree such as a decision tree, a regression tree, a random forest, and a gradient boosting tree. , 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. With learning models learned by methods using ensemble learning including boosting, bagging, etc., hierarchical clustering, non-hierarchical clustering, clustering methods including topic models, association analysis, 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.

In the present embodiment, the powder powder data (moisture content data) of the powder obtained from the powder processing apparatus 4A, the temperature and flow rate of the heat medium to be introduced into the powder processing apparatus 4A, the supply amount of the powder raw material, or The learning model 210 is generated by using the measurement data including the supply speed, the rotation speed of the crushing rotor 413, the rotation speed of the classification rotor 415, the pressure in the casing 410, and the processing time of the drying process as the teacher data. .. Alternatively, the learning model 210 may be generated by using the powder data (moisture data) and a part of the selected measurement data as the teacher data.

Further, the control device 100 further includes, as measurement data, at least one of the discharge / suction flow rate or temperature when the powder is taken out from the casing 410, and the drive power or drive current of the powder processing device 4A. May be used to generate the learning model 210.

Further, the control device 100 may generate the learning model 210 by using the teacher data including the yield per unit time (that is, the processing capacity) as the powder data. In this case, the control device 100 uses the powder data including the moisture content of the powder obtained from the powder processing device 4A and the yield per unit time and the above-mentioned measurement data as the teacher data, and the powder desired by the user. When the moisture content of the body and the yield per unit time are input, the learning model 210 configured to output the calculation result for the control parameters for the powder processing apparatus 4A may be generated.

Further, the control device 100 may generate a learning model 210 by using it for teacher data including raw material data. The raw material data used for the teacher data includes at least one of the moisture content, temperature, bulk density (or true density), and particle size of the powder raw material. The control device 100 uses the acquired powder data, raw material data, and measurement data as teacher data, and outputs the calculation result of the control parameters for the powder processing device 4A in response to the input of the powder data and the raw material data. The learning model 210 configured in the above may be generated.

Further, the control device 100 may generate the learning model 210 by using the teacher data including at least one of the product data, the exhaust gas data, and the environmental data. The product data used for the teacher data includes at least one of the temperature, bulk density (or true density), and particle size of the product powder. The exhaust gas data includes at least one of the flow rate and moisture of the gas discharged through the blower 7. Environmental data includes at least one of the temperature and humidity of the environment in which the powder treatment is performed. The control device 100 uses the acquired powder data, product data, exhaust gas data, environmental data, and measurement data as teacher data, and receives input of the powder data and at least one of product data, exhaust gas data, and environmental data. The learning model 210 configured to output the calculation result of the control parameter for the powder processing apparatus 4A may be generated.

Further, in the present embodiment, the direct heating type powder processing apparatus 4A for introducing hot air into the casing 410 to dry the powder has been described, but the powder processing apparatus 4A includes an indirect heating medium dryer and a medium. It may be a dryer such as a stirring type air flow dryer or a stirring type freeze dryer. As the indirect heating medium dryer, for example, a solid air (registered trademark), a torus disk, and a micron thermoprocessor (registered trademark) manufactured by Hosokawa Micron Co., Ltd. can be used. Further, as the medium stirring type airflow dryer, for example, Zelvis (registered trademark) manufactured by Hosokawa Micron Co., Ltd. can be used. As the stirring type freeze dryer, Active Freeze Dryer (registered trademark) manufactured by Hosokawa Micron Co., Ltd. can be used. Since these dryers are provided with a paddle, a screw, a rotor, etc. as a rotating body for stirring the powder raw material to be processed, among the control parameters obtained by the calculation using the learning model 210, the crushing rotor 413 Instead of the rotation speed of the above and the rotation speed of the classification rotor 415, the rotation speed of the rotating body may be used.

(Embodiment 2)
In the second embodiment, an application example of the mixing process will be described. The powder processing system 1 according to the second embodiment includes a powder processing device 4B that performs a mixing process instead of the powder processing device 4A that performs a drying process.

FIG. 9 is a schematic view showing the configuration of the powder processing apparatus 4B according to the second embodiment. The powder processing apparatus 4B includes an inverted conical casing 420 that performs a mixing process inside the powder processing apparatus 4B. The casing 420 is provided with raw

material input ports

421A and 421B, a swing arm 422, a screw 423, a powder outlet 424, and the like.

The raw

material input ports

421A and 421B are provided on the upper part of the casing 420. Raw material supply machines 2 for supplying different types of powder raw materials are connected to the raw

material input ports

421A and 421B. The different types of powder raw materials supplied from the raw material supply machine 2 are conveyed by the screw feeder 21 in the raw material supply path TP1 and are charged into the casing 420 from the raw

material input ports

421A and 421B.

The screw 423 provided inside the casing 420 is configured to rotate (rotate) around a long axis along the inner peripheral surface of the casing 420 by driving the screw motor 423M. The upper end of the screw 423 is connected to one end of the swing arm 422. The swing arm 422 is configured to rotate in a horizontal plane about the other end of the swing arm 422 by driving the swing motor 422M. The screw 423 is configured to rotate (revolve) along the inner peripheral surface of the casing 420 as the swing arm 422 rotates.

As the screw 423 rotates, the powder raw material in the vicinity of the screw 423 moves upward while being agitated, and the portion away from the screw 423 moves downward due to gravity. At the same time, the swing arm 422 rotates and the screw 423 revolves along the inner peripheral surface of the casing 420, so that the entire powder raw material in the casing 420 moves significantly. In this way, the powder processing apparatus 4B revolves while rotating the screw 423 to agitate the powder raw material charged in the casing 420. At this time, a plurality of types of powder raw materials can be quickly mixed with almost no segregation phenomenon.

The powder mixed by the action of the screw 423 is taken out to the outside of the powder processing apparatus 4B through the powder outlet 424 provided in the lower part of the casing 420.

The powder processing apparatus 4B may be provided with a jacket portion for adjusting the internal temperature of the casing 420. The jacket portion adjusts the internal temperature of the casing 420 by a fluid (heat medium or refrigerant) supplied from a tank provided separately.

The control device 100 according to the second embodiment acquires measurement data indicating the operating state of the powder processing device 4B and powder data related to the powder obtained from the powder processing device 4B through the input unit 103. The measurement data acquired by the control device 100 is the supply amount (or supply speed) of the powder raw material to be supplied to the powder processing device 4B, the rotation speed of the screw 423 for stirring the powder raw material, and the powder processing device 4B. Includes the supply amount (or supply speed) of the fluid (heat medium or refrigerant) to be supplied, and the processing time of the mixing process. The powder data is data on the mixing degree of the powder obtained from the powder processing apparatus 4B. The mixing degree is an index value indicating the degree of mixing of two or more kinds of powders, and can be measured by using, for example, a color difference meter, an absorbance meter, an NIR, a PH meter, image analysis, chromatography, or the like.

The control device 100 uses the acquired measurement data and powder data as teacher data to generate a learning model 220 (see FIG. 10). Since the procedure for generating the learning model 220 is the same as that in the first embodiment, the description thereof will be omitted.

The learning model 220 is configured to output the calculation result of the control parameters for the powder processing apparatus 4B when the target value (mixing degree) for the powder desired by the user is input. Here, the control parameters for the powder processing apparatus 4B are the supply amount (or supply speed) of the powder raw material, the rotation speed of the screw 423, the supply amount (or supply speed) of the fluid supplied to the powder processing apparatus 4B, and. Includes processing time for mixed processing.

FIG. 10 is a schematic diagram showing a configuration example of the learning model 220 in the second embodiment. Similar to the learning model 210 described in the first embodiment, the learning model 220 includes an input layer 221 having one or more nodes, an

intermediate layer

222A, 222B, and an output layer 223, respectively. The number of intermediate layers is not limited to two, and may be three or more.

The learning model 220 is configured to output the calculation result of the control parameters for the powder processing apparatus 4B in response to the input of the target value (mixing degree in the second embodiment) for the powder desired by the user. ..

When the control device 100 performs an operation using the learning model 220, the control device 100 inputs a target value (mixture degree) desired by the user into the learning model 220. The target value data input to the learning model 220 is output to the node included in the first intermediate layer 222A through the nodes constituting the input layer 221. The data input to the first intermediate layer 222A is output to the node included in the next intermediate layer 222B through the nodes constituting the intermediate layer 222A. 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 calculation using the activation function including the weight and the bias set between the nodes is executed, and the calculation is transmitted to the subsequent layers one after another until the calculation result by the output layer 223 is obtained.

The output layer 223 outputs the calculation result of the control parameters for the powder processing apparatus 4B. As the calculation result, for example, the probability indicating the quality of the combination of the plurality of control parameters described above may be output. Specifically, the output layer 223 is composed of n nodes from the first node to the nth node, and the supply amount of the powder raw material is SA1 from the first node, the rotation speed of the screw 423 is SV1, and the fluid The probability P1 that the supply amount is FA1 and the processing time of the mixing process is MT1 is output, and ..., The supply amount of the powder raw material is SAN, the rotation speed of the screw 423 is SVn, and the fluid supply amount is FAn from the nth node. , The probability Pn that the processing time of the mixing process is MTn may be output. The number of nodes constituting the output layer 223 and the calculation result assigned to each node are not limited to the above examples, and can be appropriately designed.

The control unit 101 of the control device 100 has the highest probability of being output from the output layer 223 (in the present embodiment, the supply amount of the powder raw material, the rotation speed of the screw 423, and the powder processing device 4B). The control parameters used for control are determined by specifying the supply amount of the fluid to be supplied to the vehicle and the processing time of the mixing process). Based on the determined control parameters, the control unit 101 generates control commands for controlling the operations of the raw material supply machine 2, the screw motor 423M, the heat medium or refrigerant supply source, and the powder processing device 4B, and each of them is generated through the output unit 104. Output to the device.

As described above, in the second embodiment, with respect to the powder processing apparatus 4B that performs the mixing treatment, the supply amount (or the powder raw material) of the powder raw material is supplied according to the input of the target value (mixing degree) for the powder desired by the user. A learning model 220 that outputs calculation results related to control parameters for controlling the supply speed), the rotation speed of the screw 423, the supply amount (or supply speed) of the fluid supplied to the powder processing apparatus 4B, and the processing time of the mixing process. To generate. Further, by using the learning model 220, the supply amount (or supply speed) of the powder raw material, the rotation speed of the screw 423, the supply amount (or supply speed) of the fluid supplied to the powder processing apparatus 4B, and the mixing process The calculation result regarding the control parameter for controlling the processing time is obtained. The control device 100 can control the operation of the powder processing system 1 based on the calculation result of the learning model 220 so that the powder having a desired mixing degree can be obtained.

In the present embodiment, the powder data (mixing degree data) of the powder obtained from the powder processing apparatus 4B, the supply amount (or supply rate) of the powder raw material to be supplied to the powder processing apparatus 4B, and the powder A learning model using the rotation speed of the screw 423 for stirring the raw material, the supply amount (or supply speed) of the fluid supplied to the powder processing apparatus 4B, and the measurement data including the processing time of the mixing process as the teacher data. It was configured to generate 220. Alternatively, the powder data (moisture data) and a part of the selected measurement data may be used as the teacher data to generate the learning model 220.

Further, the control device 100 generates a learning model 220 by using the teacher data including at least one of the humidity in the casing 420 and the drive power or the drive current of the powder processing device 4B as the measurement data. May be good.

Further, the control device 100 may generate the learning model 220 by using the teacher data including the concentration of the powder obtained from the powder processing device 4B and the moisture content as the powder data. In this case, the control device 100 uses the powder data including the mixing degree, concentration and moisture of the powder obtained from the powder processing device 4B and the above-mentioned measurement data as the teacher data, and the powder desired by the user. When the mixing degree, concentration, and moisture content of the above are input, the learning model 220 configured to output the calculation result for the control parameters for the powder processing apparatus 4B may be generated.

Further, the control device 100 may generate the learning model 220 by using the teacher data including the raw material data. The raw material data used for the teacher data includes at least one of the mixing ratio, mixing degree, particle size, bulk density (or true density), and fluidity of the powder raw material. The control device 100 uses the acquired powder data, raw material data, and measurement data as teacher data, and outputs the calculation result of the control parameters for the powder processing device 4B in response to the input of the powder data and the raw material data. The learning model 220 configured in may be generated.

Further, the control device 100 may generate the learning model 220 by using the product data and the teacher data including at least one of the environmental data. The product data used for the teacher data includes at least one of the temperature of the product powder and the bulk density (or true density). Environmental data includes at least one of the temperature and humidity of the environment in which the powder treatment is performed. The control device 100 uses the acquired powder data, product data, environmental data, and measurement data as teacher data, and receives the input of the powder data and at least one of the product data and the environmental data to the powder processing device 4B. The learning model 220 configured to output the calculation result for the control parameter may be generated.

In the present embodiment, the powder processing apparatus 4B is a mixer that mixes powder by rotating a rotating body (screw 423). As such a mixer, for example, Nautamixer (registered trademark), Bitemix, and Cyclomix (registered trademark) manufactured by Hosokawa Micron Co., Ltd. can be used.

Further, instead of the mixer that mixes the powder by rotating the rotating body, a mixer that mixes by wind power mixing may be used. In this case, instead of the rotation speed of the rotating body, the flow rate (flow velocity) of the fluid introduced into the casing 420 may be a control parameter determined by the learning model 220.

(Embodiment 3)
In the third embodiment, an application example of the compounding process will be described. The powder treatment system 1 according to the third embodiment includes a powder treatment device 4C that performs a composite treatment instead of the powder treatment device 4A that performs the drying treatment.

FIG. 11 is a schematic view showing the configuration of the powder processing apparatus 4C according to the third embodiment. The powder processing apparatus 4C includes a horizontal cylindrical casing 430 that performs compounding processing inside the powder processing apparatus 4C. The casing 430 is provided with a raw material input port 431, a paddle 432, a powder outlet 433, and the like.

The raw material input port 431 is provided above the casing 430. A raw material supply machine 2 for supplying a powder raw material to be processed is connected to the raw material input port 431. The different types of powder raw materials supplied from the raw material supply machine 2 are conveyed by the screw feeder 21 in the raw material supply path TP1 and are charged into the casing 430 from the raw material input port 431. The powder raw material may be mixed in advance or may be separately supplied to the powder processing apparatus 4C.

The paddle 432 provided inside the casing 430 is configured to rotate at a peripheral speed of, for example, 40 m / s or more by driving the paddle motor 432M. The shape and arrangement of the paddle 432 is designed so that the impact force, compressive force, and shear force act uniformly on the individual powder particles. The powder processing apparatus 4C drives the paddle motor 432M and rotates the paddle 432 to apply impact force, compressive force, and shear force to the powder raw material put into the casing 430, and fine particles (child particles). Is dispersed and immobilized on particles (mother particles) of a larger size.

The powder complexed by the action of the paddle 432 is taken out to the outside of the powder processing apparatus 4C through the powder outlet 433 provided in the lower part of the casing 430.

The powder processing apparatus 4C may include a jacket portion for adjusting the internal temperature of the casing 430. The jacket portion adjusts the internal temperature of the casing 430 by a fluid (heat medium or refrigerant) supplied from a tank provided separately.

The control device 100 according to the third embodiment acquires measurement data indicating the operating state of the powder processing device 4C and powder data related to the powder obtained from the powder processing device 4C through the input unit 103. The measurement data acquired by the control device 100 includes the supply amount (or supply speed) of the powder raw material to be supplied to the powder processing device 4C, the rotation speed of the paddle 432, the load power of the powder processing device 4C, and the compounding process. Includes processing time. The powder data is data on the degree of compounding of the powder obtained from the powder processing apparatus 4C. The degree of compositing is an index showing the degree of integration of a plurality of types of powder, and can be measured using BET, NIR, XRD, TG-DTA, MS, SEM, FE-SEM, TEM, or the like. Is. The degree of compounding may be an index represented by a numerical value or an index represented by an image (image data).

The control device 100 uses the acquired measurement data and powder data as teacher data to generate a learning model 230 (see FIG. 12). Since the procedure for generating the learning model 230 is the same as that in the first embodiment, the description thereof will be omitted.

The learning model 230 is configured to output the calculation result of the control parameters for the powder processing apparatus 4C when the target value (composite degree) for the powder desired by the user is input. Here, the control parameters for the powder processing apparatus 4C include the supply amount (or supply rate) of the powder raw material, the rotation speed of the paddle 432, the load power of the powder processing apparatus 4C, and the processing time of the compounding process.

FIG. 12 is a schematic diagram showing a configuration example of the learning model 230 according to the third embodiment. Similar to the learning model 210 described in the first embodiment, the learning model 230 includes an input layer 231 having one or a plurality of nodes,

intermediate layers

232A and 232B, and an output layer 233, respectively. The number of intermediate layers is not limited to two, and may be three or more.

The learning model 230 is configured to output the calculation result of the control parameters for the powder processing apparatus 4C in response to the input of the target value (composite degree in the third embodiment) for the powder desired by the user. To.

When the control device 100 performs an operation using the learning model 230, the control device 100 inputs a target value (complexity degree) desired by the user into the learning model 230. The target value data input to the learning model 230 is output to the node included in the first intermediate layer 232A through the nodes constituting the input layer 231. The data input to the first intermediate layer 232A is output to the node included in the next intermediate layer 232B through the nodes constituting the intermediate layer 232A. 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 calculation using the activation function including the weight and the bias set between the nodes is executed, and the calculation is transmitted to the subsequent layers one after another until the calculation result by the output layer 233 is obtained.

The output layer 233 outputs the calculation result of the control parameters for the powder processing apparatus 4C. As the calculation result, for example, the probability indicating the quality of the combination of the plurality of control parameters described above may be output. Specifically, the output layer 233 is composed of n nodes from the first node to the nth node, and from the first node, the supply amount of the powder raw material is SA1, the rotation speed of the paddle 432 is PV1, and the load power. Outputs LP1 and the probability P1 that the processing time of the compounding process is CT1. From the nth node, the supply amount of the powder raw material is SAn, the rotation speed of the paddle 432 is PVn, the load power is LPn, and the compounding process is performed. The probability Pn that the processing time of the processing is CTn may be output. The number of nodes constituting the output layer 233 and the calculation result assigned to each node are not limited to the above examples, and can be appropriately designed.

The control unit 101 of the control device 100 has the highest probability of being output from the output layer 233 (in the present embodiment, the supply amount of the powder raw material, the rotation speed of the paddle 432, the load power, and the composite. The control parameters used for control are determined by specifying the combination of processing times of the chemical processing. The control unit 101 generates a control command for controlling the operation of the raw material supply machine 2, the paddle motor 432M, and the powder processing device 4C based on the determined control parameters, and outputs the control command to each device through the output unit 104.

As described above, in the third embodiment, with respect to the powder processing apparatus 4C that performs the compounding process, the supply amount of the powder raw material is obtained according to the input of the target value (composite degree) for the powder desired by the user. , Generates a learning model 230 that outputs calculation results for control parameters for controlling the rotational speed of the paddle 432, the load power, and the processing time of the mixing process. Further, by using the learning model 230, calculation results regarding the supply amount of the powder raw material, the rotation speed of the paddle 432, and the control parameters for controlling the processing time of the compounding process can be obtained. The control device 100 can control the operation of the powder processing system 1 based on the calculation result of the learning model 230 so that the powder having a desired degree of compounding can be obtained.

In the present embodiment, the powder data (data of the degree of compositing) of the powder obtained from the powder processing device 4C, the supply amount (or supply rate) of the powder raw material to be supplied to the powder processing device 4C, and the powder. The learning model 230 is generated by using the rotation speed of the paddle 432 for stirring the body material and the measurement data including the processing time of the compounding process as the teacher data. Alternatively, the learning model 230 may be generated by using the powder data (data of the degree of compounding) and a part of the selected measurement data as the teacher data.

Further, the control device 100 further includes at least one of the humidity and pressure in the casing 430, the temperature and flow rate of the refrigerant introduced in the casing 430, and the drive power or drive current of the powder processing device 4C as measurement data. The data may be used to generate the training model 230.

Further, the control device 100 may generate the learning model 230 by using the teacher data including the conductivity, thermal conductivity, and transmittance of the powder obtained from the powder processing device 4C as the powder data. Good. In this case, the control device 100 uses the powder data including the compositeness, conductivity, thermal conductivity, and permeability of the powder obtained from the powder processing device 4C and the above-mentioned measurement data as the teacher data. , A learning model configured to output the calculation results for the control parameters for the powder processing apparatus 4C when the compositeness, conductivity, thermal conductivity, and transmittance of the powder desired by the user are input. 230 may be generated.

Further, the control device 100 may generate the learning model 230 by using the teacher data including the raw material data. The raw material data used for the teacher data is at least the data obtained from the mixing ratio of the powder raw material, the bulk density (or true density), BET, NIR, XRD, TG-DTA, MS, SEM, FE-SEM, TEM, and the like. Includes one. The control device 100 uses the acquired powder data, raw material data, and measurement data as teacher data, and outputs the calculation result of the control parameters for the powder processing device 4C in response to the input of the powder data and the raw material data. The learning model 230 configured in may be generated.

Further, the control device 100 may generate the learning model 230 by using the product data and the teacher data including at least one of the environmental data. The product data used for the teacher data includes at least one of the bulk density (or true density) and the particle size of the product powder. Environmental data includes at least one of the temperature and humidity of the environment in which the powder treatment is performed. The control device 100 uses the acquired powder data, product data, environmental data, and measurement data as teacher data, and receives the input of the powder data and at least one of the product data and the environmental data to the powder processing device 4C. The learning model 230 configured to output the calculation result for the control parameter may be generated.

In the present embodiment, the powder processing apparatus 4C is a composite processing apparatus that composites powder by rotating the paddle 432. As such a composite processing apparatus, for example, Nobilta (registered trademark), Nanocura (registered trademark), and Mechanofusion (registered trademark) manufactured by Hosokawa Micron Co., Ltd. can be used.
In the present embodiment, the compounding is performed using the powder processing apparatus 4C, but the mixing treatment may be performed with the target value as the mixing degree, and the target value may be spherical with the target value as the circularity for a single raw material. Surface treatment such as chemical conversion may be performed.

(Embodiment 4)
In the fourth embodiment, an application example of the surface treatment will be described. The powder processing system 1 according to the fourth embodiment includes a powder processing device 4D that performs surface treatment instead of the powder processing device 4A that performs drying treatment.

FIG. 13 is a schematic view showing the configuration of the powder processing apparatus 4D according to the fourth embodiment. The powder processing apparatus 4D includes a cylindrical casing 440 that performs surface treatment inside the powder processing apparatus 4D. The casing 440 is provided with a raw material input port 441, a gas introduction port 442, a crushing rotor 443, a classification rotor 444, a powder outlet 445, a fine powder outlet 446, and the like.

The raw material input port 441 is provided above the crushing rotor 443. A raw material supply machine 2 for supplying powder raw materials is connected to the raw material input port 441. 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 440 from the raw material input port 441.

The casing 440 is provided with a gas introduction port 442. The position of the gas introduction port 442 is not particularly limited, but it is preferably provided at a position below the crushing rotor 443 so that the gas is introduced into the casing 440 via the rotating crushing rotor 443.

The crushing rotor 443 is a rotor including a rotary disk 443A and a plurality of hammers 443B protruding upward from the upper peripheral edge of the rotary disk 443A, and the rotation speed is adjusted to a desired speed by the power of a crushing motor 413M (see FIG. 3). It is configured to rotate. Here, the crushing motor 413M is one of the driving units included in the powder processing system 1. The rotation speed of the crushing motor 413M is measured by the rotation speed sensor S7 (see FIG. 3) at regular or periodic timings (for example, at 5-second intervals).

The crushing rotor 443 is rotated by the power of the crushing motor 413M to generate a swirling airflow in the casing 440, and the powder raw material introduced into the casing 440 is impacted, compressed, and ground by the action of the hammer 443B. Surface treatment is applied to the powder raw material by applying mechanical energy such as shearing. The crushing rotor 443 is an example of a rotating body that agitates the powder raw material. The powder produced by surface-treating the powder raw material is guided to the classification rotor 444 provided in the upper part of the casing 440 by the air flow flowing in from the lower part of the casing 440.

The classification rotor 444 is a rotor provided with a plurality of classification blades 444A arranged radially, and is configured to rotate at a desired rotation speed by the power of the classification motor 415M (see FIG. 3). Here, the classification motor 415M is one of the drive units included in the powder processing system 1. The rotation speed of the classification motor 415M is measured by the rotation speed sensor S8 (see FIG. 3) at regular or periodic timings (for example, at 5-second intervals). The classification rotor 444 is provided above the crushing rotor 443, and the powder processed in the casing 440 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 fine powder outlet 446. On the other hand, the powder to which the classification rotor 444 is not added is guided to the powder outlet 445 and taken out as a product powder to the outside of the powder processing apparatus 4D.

The powder processing apparatus 4D may be provided with a jacket portion for adjusting the internal temperature of the casing 440. The jacket portion adjusts the internal temperature of the casing 440 by a fluid (heat medium or refrigerant) supplied from a tank provided separately.

The control device 100 according to the fourth embodiment acquires measurement data indicating the operating state of the powder processing device 4D and powder data related to the powder obtained from the powder processing device 4D through the input unit 103. The measurement data acquired by the control device 100 includes the supply amount (or supply speed) of the powder raw material supplied to the powder processing device 4D, the rotation speed of the crushing rotor 443, the rotation speed of the classification rotor 444, and the surface treatment processing time. And the load power of the powder processing apparatus 4D. The powder data is data on the circularity of the powder obtained from the powder processing apparatus 4D. The circularity is an index representing the shape of powder (particles) obtained from the powder processing apparatus 4D, and is measured by, for example, a particle shape analyzer or image analysis.

The control device 100 uses the acquired measurement data and powder data as teacher data to generate a learning model 240 (see FIG. 14). Since the procedure for generating the learning model 240 is the same as that in the first embodiment, the description thereof will be omitted.

The learning model 240 is configured to output the calculation result of the control parameters for the powder processing apparatus 4D when the target value (mixing degree) for the powder desired by the user is input. Here, the control parameters for the powder processing apparatus 4D are the supply amount (or supply speed) of the powder raw material, the rotation speed of the crushing rotor 443, the rotation speed of the classification rotor 444, the surface treatment processing time, and the powder processing apparatus. Includes 4D load power.

FIG. 14 is a schematic diagram showing a configuration example of the learning model 240 in the fourth embodiment. Similar to the learning model 210 described in the first embodiment, the learning model 240 includes an input layer 241 having one or a plurality of nodes,

intermediate layers

242A and 242B, and an output layer 243, respectively. The number of intermediate layers is not limited to two, and may be three or more.

The learning model 240 is configured to output the calculation result of the control parameter for the powder processing apparatus 4D in response to the input of the target value (circularity in the fourth embodiment) for the powder desired by the user. ..

When the control device 100 performs an calculation using the learning model 240, the control device 100 inputs a target value (circularity) desired by the user into the learning model 240. The target value data input to the learning model 240 is output to the node included in the first intermediate layer 242A through the nodes constituting the input layer 241. The data input to the first intermediate layer 242A is output to the node included in the next intermediate layer 242B through the nodes constituting the intermediate layer 242A. 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 calculation using the activation function including the weight and the bias set between the nodes is executed, and the calculation is transmitted to the subsequent layers one after another until the calculation result by the output layer 243 is obtained.

The output layer 243 outputs the calculation result for the control parameters for the powder processing apparatus 4D. As the calculation result, for example, the probability indicating the quality of the combination of the plurality of control parameters described above may be output. Specifically, the output layer 243 is composed of n nodes from the first node to the nth node, and from the first node, the supply amount of the powder raw material is SA1, the rotation speed of the crushing rotor 443 is CV1, and the classification is performed. The rotation speed of the rotor 444 is CC1, the load power is LP1, the probability P1 of the surface treatment processing time ST1 is output, ..., The supply amount of the powder raw material is SAN, and the rotation speed of the crushing rotor 443 is from the nth node. CVn, the rotation speed of the classification rotor 444 is CCn, the load power is LPn, and the probability Pn of the surface treatment processing time STn may be output. The number of nodes constituting the output layer 243 and the calculation result assigned to each node are not limited to the above examples, and can be appropriately designed.

The control unit 101 of the control device 100 has the highest probability of being output from the output layer 243 (in the present embodiment, the supply amount of the powder raw material, the rotation speed of the crushing rotor 443, and the classification rotor 444. The control parameters used for control are determined by specifying the combination of rotation speed, load power, and surface treatment time). The control unit 101 generates a control command for controlling the operation of the raw material supply machine 2, the crushing motor 413M, and the powder processing device 4D based on the determined control parameters, and outputs the control command to each device through the output unit 104.

As described above, in the fourth embodiment, with respect to the powder processing apparatus 4D that performs the surface treatment, the supply amount (or the powder raw material) of the powder raw material is supplied according to the input of the target value (circularity) for the powder desired by the user. (Supply speed), rotation speed of crushing rotor 443, rotation speed of classification rotor 444, load power of powder processing apparatus 4D, and learning model 240 that outputs calculation results regarding control parameters for controlling the processing time of surface treatment. Generate. Further, by using the learning model 240, the supply amount (or supply speed) of the powder raw material, the rotation speed of the crushing rotor 443, the rotation speed of the classification rotor 444, the load power of the powder processing apparatus 4D, and the surface treatment treatment. The calculation result regarding the control parameter for controlling the time is obtained. The control device 100 can control the operation of the powder processing system 1 based on the calculation result of the learning model 240 so that the powder having a desired circularity can be obtained.

In the present embodiment, the powder data (circularity data) of the powder obtained from the powder processing apparatus 4D, the supply amount (or supply speed) of the powder raw material to be supplied to the powder processing apparatus 4D, and the crushing rotor. The learning model 240 is generated by using the measurement data including the rotation speed of 443, the rotation speed of the classification rotor 444, the load power of the powder processing apparatus 4D, and the processing time of the surface treatment as the teacher data. Alternatively, the powder data (circularity data) and a part of the selected measurement data may be used as the teacher data to generate the learning model 240.

Further, the control device 100 uses the measurement data as the drive power or drive current of the powder processing device 4D, the flow rate of the fluid (gas or liquid) to be introduced into the casing 440, the flow velocity or temperature, the temperature inside the casing 440, and the powder. The learning model 240 may be generated using teacher data further including at least one of the flow rate or temperature of the refrigerant (or heat medium) supplied to the processing apparatus 4D.

Further, the control device 100 generates a learning model 240 by using the teacher data including the fluidity of the powder obtained from the powder processing device 4D and the bulk density (or true density) as the powder data. May be good. Fluidity and bulk density are measured using a known powder tester. The control device 100 uses the powder data including the fluidity and bulk density (or true density) of the powder obtained from the powder processing device 4D and the above-mentioned measurement data as the teacher data, and the powder desired by the user. When the roundness, fluidity, and bulk density (or true density) of the body are input, the learning model 240 configured to output the calculation result for the control parameters for the powder processing apparatus 4D may be generated. .. Further, in addition to the above-mentioned fluidity and bulk density, measurement data including any one of the BET value, the powder temperature, and the particle size may be used.

Further, the control device 100 may generate the learning model 240 by using the teacher data including the raw material data. The raw material data used for the teacher data includes at least one of the BET value, particle size, circularity, bulk density, and fluidity of the powder raw material. The control device 100 uses the acquired powder data, raw material data, and measurement data as teacher data, and outputs the calculation result of the control parameters for the powder processing device 4D in response to the input of the powder data and the raw material data. The learning model 240 configured in may be generated.

Further, the control device 100 may generate the learning model 240 by using the teacher data including at least one of the product data and the environmental data. The product data used for the teacher data may include the moisture content of the product powder. Environmental data includes at least one of the temperature and humidity of the environment in which the powder treatment is performed. The control device 100 uses the acquired powder data, product data, environmental data, and measurement data as teacher data, and receives the input of the powder data and at least one of the product data and the environmental data to the powder processing device 4D. The learning model 240 configured to output the calculation result for the control parameter may be generated.

In the present embodiment, the powder treatment device 4D is a surface treatment device that performs surface treatment by rotating the crushing rotor 443 and the classification rotor 444. As such a surface treatment apparatus, for example, Faculty (registered trademark) manufactured by Hosokawa Micron Co., Ltd. can be used.
Further, in the present embodiment, the blower 7 is used to form a flow of gas for extracting the powder from the powder processing apparatus 4D, but a pump may be used instead of the blower 7.
Further, in the present embodiment, the configuration is such that the gas is introduced into the powder processing apparatus 4D, but the configuration in which the liquid is introduced may be used instead of the configuration in which the gas is introduced.

(Embodiment 5)
In the fifth embodiment, an application example for the granulation treatment will be described. The powder treatment system 1 according to the fifth embodiment includes a powder treatment device 4E that performs a granulation treatment instead of the powder treatment device 4A that performs the drying treatment.

FIG. 15 is a schematic view showing the configuration of the powder processing apparatus 4E according to the fifth embodiment. The powder processing apparatus 4E includes a cylindrical casing 450 that performs granulation processing inside the powder processing apparatus 4E. The casing 450 is provided with a raw material input port 451, an additive input port 452, a rotor 453, a flexible wall 454, a powder outlet 455, and the like.

The raw material input port 451 is provided above the casing 450. A raw material supply machine 2 for supplying a powder raw material to be processed is connected to the raw material input port 451. 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 450 from the raw material input port 451.

The additive input port 452 is provided on the upper part of the casing 450 so as to face the raw material input port 451. The additive charged into the casing 450 through the additive charging port 452 is a liquid such as water or oil. The additive inlet 452 includes a nozzle (not shown) for spraying the additive. The powder processing apparatus 4E uniformly humidifies the inside of the casing 450 by spraying droplets from the additive input port 452. Humidification progresses, and the powder that has become more than the water content at which granulation begins repeats cohesive growth so as to wrap the sprayed droplets, and grows into porous granulated particles.

The rotor 453 provided inside the casing 450 is configured to rotate around an axis in the vertical direction by driving the rotor motor 453M. The rotor 453 is provided with a knife blade 453A, and by giving the powder raw material three types of movements of swirling, rotating, and consolidation by gravity, homogeneous mixing, humidification, and granulation are performed.

Further, a flexible wall 454 is provided on the inner peripheral surface of the casing 450. The outer peripheral surface of the flexible wall 454 is surrounded by a roller gauge 454B to which a large number of rollers 454A are attached. The flexible wall 454 has a function of being deformed by the vertical movement of the roller gauge 454B and removing agglomerates adhering to the inner wall by humidification.

The powder (granulated particles) produced in the casing 450 is taken out to the outside of the powder processing apparatus 4E through the powder outlet 455 provided in the lower part of the casing 450.

The control device 100 according to the fifth embodiment acquires measurement data indicating the operating state of the powder processing device 4E and powder data related to the powder obtained from the powder processing device 4E through the input unit 103. The measurement data acquired by the control device 100 includes the supply amount (or supply speed) of the powder raw material to be supplied to the powder processing device 4E, the rotation speed of the rotor 453, the pressure in the casing 450, and the addition to be charged into the casing 450. Includes the amount of agent input. The powder data is data on the particle size and shape of the powder obtained from the powder processing apparatus 4E. The shape of the powder is data obtained by a particle shape analyzer or image analysis.

The control device 100 uses the acquired measurement data and powder data as teacher data to generate a learning model 250 (see FIG. 16). Since the procedure for generating the learning model 250 is the same as that in the first embodiment, the description thereof will be omitted.

The learning model 250 is configured to output the calculation result of the control parameters for the powder processing apparatus 4E when the target values (particle diameter and shape) for the powder desired by the user are input. Here, the control parameters for the powder processing apparatus 4E include the supply amount (or supply speed) of the powder raw material, the rotation speed of the rotor 453, the pressure in the processing chamber, and the input amount of the additive.

FIG. 16 is a schematic diagram showing a configuration example of the learning model 250 in the fifth embodiment. Similar to the learning model 210 described in the first embodiment, the learning model 250 includes an input layer 251 having one or a plurality of nodes,

intermediate layers

252A and 252B, and an output layer 253, respectively. The number of intermediate layers is not limited to two, and may be three or more.

The learning model 250 is configured to output the calculation result of the control parameter for the powder processing apparatus 4E in response to the input of the target value (particle diameter and shape in the fifth embodiment) for the powder desired by the user. Will be done.

When the control device 100 performs the calculation using the learning model 250, the control device 100 inputs the target values (particle diameter and shape) desired by the user into the learning model 250. The target value data input to the learning model 250 is output to the node included in the first intermediate layer 252A through the nodes constituting the input layer 251. The data input to the first intermediate layer 252A is output to the node included in the next intermediate layer 252B through the nodes constituting the intermediate layer 252A. 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 calculation using the activation function including the weight and the bias set between the nodes is executed, and the calculation is transmitted to the subsequent layers one after another until the calculation result by the output layer 253 is obtained.

The output layer 253 outputs the calculation result for the control parameters for the powder processing apparatus 4E. As the calculation result, for example, the probability indicating the quality of the combination of the plurality of control parameters described above may be output. Specifically, the output layer 253 is composed of n nodes from the first node to the nth node, from the first node, the supply amount of the powder raw material is SA1, the rotation speed of the rotor 453 is RV1, and the casing 450. The probability P1 that the pressure inside is CP1 and the input amount of the additive is AA1 is output, ..., The supply amount of the powder raw material is SAN, the rotation speed of the rotor 453 is RVn, and the pressure inside the casing 450 from the nth node. May output CPn and the probability Pn such that the amount of the additive added is AAn. The number of nodes constituting the output layer 253 and the calculation result assigned to each node are not limited to the above examples, and can be appropriately designed.

The control unit 101 of the control device 100 has the highest probability of being output from the output layer 253 (in the present embodiment, the supply amount of the powder raw material, the rotation speed of the rotor 453, and the pressure in the casing 450). , And the combination of the input amount of the additive), the control parameter used for the control is determined. The control unit 101 generates a control command for controlling the operation of the raw material supply machine 2, the rotor motor 453M, the nozzle, and the like based on the determined control parameters, and outputs the control command to each device through the output unit 104.

As described above, in the fifth embodiment, with respect to the powder processing apparatus 4E that performs the granulation treatment, the powder raw material is supplied according to the input of the target values (particle size and shape) for the powder desired by the user. A learning model 250 is generated that outputs calculation results regarding control parameters for controlling the amount, the rotation speed of the rotor 453, the pressure in the casing 450, and the input amount of the additive. Further, by using the learning model 250, calculation results regarding control parameters for controlling the supply amount of the powder raw material, the rotation speed of the rotor 453, the pressure in the casing 450, and the input amount of the additive can be obtained. The control device 100 can control the operation of the powder processing system 1 based on the calculation result of the learning model 250 so that the powder having a desired particle size and shape can be obtained.

In the present embodiment, the powder powder data (particle size and shape data) of the powder obtained from the powder processing apparatus 4E, the supply amount (or supply rate) of the powder raw material to be supplied to the powder processing apparatus 4E, The learning model 250 is generated by using the measurement data including the rotation speed of the rotor 453 for stirring the powder raw material, the pressure in the casing 450, and the amount of the additive to be charged into the casing 450 as the teacher data. It was configured. Alternatively, the learning model 250 may be generated by using the powder data (particle size and shape data) and a part of the selected measurement data as the teacher data.

Further, the control device 100 may generate the learning model 250 by using the teacher data including at least one of the humidity in the casing 450, the drive power of the powder processing device 4E, or the drive current as the measurement data. ..

Further, the control device 100 generates a learning model 250 as powder data by using the teacher data further including the fluidity, bulk density, hardness, water absorption amount or oil absorption amount of the powder obtained from the powder processing device 4E. You may. In this case, the control device 100 uses the powder data including the fluidity, bulk density, hardness, water absorption amount or oil absorption amount of the powder obtained from the powder processing device 4E and the above-mentioned measurement data as the teacher data. When the fluidity, bulk density, hardness, water absorption amount or oil absorption amount of the powder desired by the user is input, the learning model 250 configured to output the calculation result for the control parameters for the powder processing apparatus 4E is provided. It should be generated.

Further, the control device 100 may generate the learning model 250 by using the teacher data including the raw material data. The raw material data used for the teacher data includes at least one of the BET value, particle size, fluidity, and bulk density of the powder raw material. The control device 100 uses the acquired powder data, raw material data, and measurement data as teacher data, and outputs the calculation result of the control parameters for the powder processing device 4E in response to the input of the powder data and the raw material data. The learning model 250 configured in the above may be generated.

Further, the control device 100 may generate the learning model 250 by using the product data and the teacher data including at least one of the environmental data. The product data used for the teacher data includes the moisture content of the product powder. Environmental data includes at least one of the temperature and humidity of the environment in which the powder treatment is performed. The control device 100 uses the acquired powder data, product data, environmental data, and measurement data as teacher data, and receives the input of the powder data and at least one of the product data and the environmental data to the powder processing device 4E. A learning model 250 configured to output the calculation results for the control parameters may be generated.

In the present embodiment, the powder processing device 4E is a granulation processing device that granulates by adding an additive and rotating the rotor 453. As such a granulation processing apparatus, for example, a flexomix manufactured by Hosokawa Micron Co., Ltd. can be used. It can also be applied to a fluidized bed type granulation processing apparatus that performs the same processing. As such a granulation processing apparatus, Agromaster (registered trademark) manufactured by Hosokawa Micron Co., Ltd. can be used.

(Embodiment 6)
In the sixth embodiment, the re-learning procedure of the learning model 210 described in the first embodiment will be described.
Since the overall configuration of the powder processing system 1 and the configuration of each device in the powder processing system 1 are the same as those in the first embodiment, the description thereof will be omitted.

FIG. 17 is a flowchart illustrating a re-learning procedure of the learning model 210. As described in the first embodiment, in the operation phase, the control unit 101 of the control device 100 receives the input of the target value (moisture content) for the powder desired by the user, and learns the received moisture content data. By inputting to the model 210, the calculation result related to the control parameter is acquired. The control unit 101 controls the operation of each device constituting the powder processing system 1 based on the calculation result acquired from the learning model 210. In the operation phase, the control unit 101 determines the temperature and flow rate of the heat medium, the supply amount or supply speed of the powder raw material, the rotation speed of the crushing rotor 413, the rotation speed of the classification rotor 415, the pressure in the casing 410, and the drying process. The measurement data including the processing time and the powder data (moisture data) of the powder obtained from the powder processing apparatus 4A may be collected. The collected measurement data and powder data are stored in the storage unit 102 together with the time stamp.

The control unit 101 compares the actually measured value indicated by the powder data with the target value input by the user at an appropriate timing after the start of the operation phase (step S601). The control unit 101 determines whether or not to execute re-learning based on the comparison result (step S602).

When the measured value is close to the target value (for example, when the difference between the two is less than 10%), the control unit 101 determines that re-learning is not executed (S602: NO), and ends the process according to this flowchart.

On the other hand, when the measured value is not close to the target value (for example, when the difference between the two is 10% or more), the control unit 101 determines that re-learning is executed (S602: YES).

When it is determined to execute re-learning, the control unit 101 reads the data collected after the start of operation from the storage unit 102 (step S603) and selects the teacher data (step S604). The re-learning procedure after step S603 may be executed at a timing when the powder processing system 1 is not operating.

The control unit 101 inputs the wet content data included in the selected teacher data into the learning model 210 (step S605), and executes the calculation by the learning model 210 (step S606). That is, the control unit 101 inputs moisture data to the nodes constituting the input layer 211 of the learning model 210, performs calculations using weights and biases between the nodes in the

intermediate layers

212A and 212B, and outputs the calculation results. The process of outputting from the node of layer 213 is performed.

Next, the control unit 101 evaluates the calculation result obtained by the calculation in step S606 (step S607), and determines whether or not the learning is completed (step S608). 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 S606 and the teacher data.

When it is determined that the learning is not completed (S608: NO), the control unit 101 updates the weights and biases between the nodes of the learning model 210 (step S609), returns the process to step S604, and another teacher. Continue learning with the data. The control unit 101 may update the weights and biases between the nodes by using an error backpropagation method that sequentially updates the weights and biases between the nodes from the output layer 213 of the learning model 210 toward the input layer 211. it can.

When it is determined that the learning is completed (S608: YES), the control unit 101 stores the learned learning model 210 in the storage unit 102 (step S610), and ends the process according to this flowchart.

As described above, since the control device 100 according to the present embodiment relearns the learning model 210 as necessary, the accuracy of powder processing by the powder processing system 1 even after the start of the operation phase. Can be enhanced.

Although the re-learning procedure of the learning model 210 has been described in the sixth embodiment, the learning models 220 to 250 described in the second to fifth embodiments can be re-learned by the same procedure. ..

(Embodiment 7)
In the seventh embodiment, the procedure for adjusting the control parameters in the powder processing apparatus 4A will be described.
Since the overall configuration of the powder processing system 1 and the configuration of each device in the powder processing system 1 are the same as those in the first embodiment, the description thereof will be omitted.

FIG. 18 is a flowchart illustrating a control parameter adjustment procedure. As described in the first embodiment, in the operation phase, the control unit 101 of the control device 100 receives the input of the target value (moisture content) for the powder desired by the user, and learns the data of the received target value. By inputting to the model 210, the calculation result related to the control parameter is acquired. The control unit 101 controls the operation of each device constituting the powder processing system 1 based on the calculation result acquired from the learning model 210. In the operation phase, the control unit 101 determines the temperature and flow rate of the heat medium, the supply amount or supply speed of the powder raw material, the rotation speed of the crushing rotor 413, the rotation speed of the classification rotor 415, the pressure in the casing 410, and the drying process. The measurement data including the processing time and the powder data (moisture data) of the powder obtained from the powder processing apparatus 4A may be collected. The collected measurement data and powder data are stored in the storage unit 102 together with the time stamp.

The control unit 101 compares the actually measured value indicated by the powder data with the target value input by the user at an appropriate timing after the start of the operation phase (step S701). The control unit 101 determines whether or not to adjust the control parameter based on the comparison result (step S702).

When the measured value is close to the target value (for example, when the difference between the two is less than 10%), the control unit 101 determines that the control parameter is not adjusted (S702: NO), and ends the process according to this flowchart.

On the other hand, when the measured value is not close to the target value (for example, when the difference between the two is 10% or more), the control unit 101 determines that the control parameter is adjusted (S702: YES).

When it is determined that the control parameter is to be adjusted, the control unit 101 adjusts the control parameter according to the comparison result in step S702 (step S703), and operates the device including the powder processing device 4A based on the adjusted control parameter. Control (step S704). For example, when the measured value of the moisture content is lower (higher) than the target value, the control unit 101 may control the operation of the hot air generator 3 so that the temperature of the heat medium becomes higher (lower). Further, the control unit 101 may control the operation of the crushing motor 413M so that the rotation speed of the crushing rotor 413 becomes high (low). Further, the control unit 101 may control the operation of the classification motor 415M so that the rotation speed of the classification rotor 415 becomes high (low). Further, the control unit 101 may control the operation of the blower 7 so that the pressure in the casing 410 becomes low (high). Further, the control unit 101 may control the operation of the raw material supply machine 2 so that the supply amount (or supply speed) of the powder raw material is low (high). Further, the control unit 101 may control the operation of the powder processing apparatus 4A so that the processing time of the drying process becomes longer (shorter).

As described above, in the present embodiment, when a deviation of a certain value or more occurs between the measured value and the target value during the operation of the powder processing system 1, the control parameter is adjusted so that the deviation becomes small. be able to.

In the seventh embodiment, an application example to the powder processing apparatus 4A has been described, but it goes without saying that the same method can be applied to the powder processing apparatus 4B to 4E.

(Embodiment 8)
In the eighth embodiment, a configuration for controlling the operation of the powder processing system 1 from the terminal device 500 will be described.
Since the overall configuration of the powder processing system 1 and the configuration of each device in the powder processing system 1 are the same as those in the first embodiment, the description thereof will be omitted.

FIG. 19 is a flowchart illustrating a procedure of processing executed by the terminal device 500 and the control device 100. The control unit 501 of the terminal device 500 accesses the control device 100 through the communication unit 503 (step S801).

When the control unit 101 of the control device 100 receives access from the terminal device 500 when the powder processing device 4A is not operating, for example, the screen of the target value input screen for receiving the target value of the powder desired by the user. Data is generated (step S802), and the generated screen data is transmitted from the communication unit 105 to the terminal device 500 (step S803). When an access is received from the terminal device 500 while the powder processing device 4A is operating, the control unit 101 may execute the processes after step S803 described below.

The control unit 501 of the terminal device 500 receives the screen data of the target value input screen transmitted from the control device 100 from the communication unit 503 (step S804). The control unit 501 displays the target value input screen on the display unit 505 based on the received screen data (step S805), and accepts the input of the target value for the powder desired by the user (step S806). The target value received in step S806 is, for example, the moisture content of the powder.

FIG. 20 is a schematic diagram showing an example of a target value input screen. The target value input screen shown as an example in FIG. 20 shows a state in which "moisture content" is set in the item of "target value 1" and a target value for moisture content is set in the item below it. In this target value input screen, for example, data on the yield of powder desired by the user per unit time may be accepted. Further, raw material data, product data, exhaust gas data, environmental data, etc. to be given to the learning model 210 may be accepted. The control unit 501 transmits the data of the target value received in step S806 to the control device 100 (step S807).

The control unit 101 of the control device 100 receives the target value data transmitted from the terminal device 500 from the communication unit 105 (step S808). The control unit 101 inputs the received target value data into the learning model 210, and executes the calculation by the learning model 210 (step S809). Next, the control unit 101 acquires the calculation result from the learning model 210 (step S810), and determines the control parameter based on the calculation result (step S811). The control unit 101 controls the operation of the powder processing system 1 including the powder processing apparatus 4A based on the determined control parameters (step S812).

During the operation of the powder processing system 1, the control unit 101 determines the moisture content of the powder obtained from the powder processing device 4A, the temperature and flow rate of the heat medium, the supply amount (or supply speed) of the powder raw material, and the crushing rotor 413. Measurement data on the rotation speed of the above, the rotation speed of the classification rotor 415, the pressure in the casing 143, the processing time in the powder processing apparatus 4A, and the like are acquired from the input unit 103 at any time. The control unit 101 generates screen data of the monitoring screen including at least one of the above measurement data and an overall view of the powder processing system 1 at an appropriate timing (step S813). The control unit 101 transmits the generated screen data from the communication unit 105 to the terminal device 500 (step S814).

The control unit 501 of the terminal device 500 receives the screen data transmitted from the control device 100 from the communication unit 503 (step S815). The control unit 501 displays the monitoring screen on the display unit 505 based on the received screen data (step S816).

FIG. 21 is a schematic diagram showing an example of the monitoring screen. The example shown in FIG. 21 shows a state in which a monitoring screen including measurement data obtained from the powder processing apparatus 4A and an overall view of the powder processing system 1 is displayed on the display unit 505. Instead of the measurement data, or together with the measurement data of the particle size, a target value input by the user, a control value set based on the calculation result of the learning model 210, and the like may be displayed on the monitoring screen.

As described above, in the present embodiment, the powder processing system 1 can be remotely controlled by using the terminal device 500, and the operating status of the powder processing system 1 can be monitored by the terminal device 500.

In the eighth embodiment, an application example to the powder processing apparatus 4A has been described, but it goes without saying that the same method can be applied to the powder processing apparatus 4B to 4E.

(Embodiment 9)
In the ninth embodiment, a configuration for selecting the type of powder treatment in the terminal device 500 will be described.
The overall configuration of the powder processing system 1 and the configuration of each device in the powder processing system 1 are the same as those in the first embodiment, but the storage unit 102 of the control device 100 has the first to fifth embodiments. The learning models 210 to 250 described in the above are stored, and the control device 100 controls the operations of the powder processing devices 4A to 4E based on the calculation results using the learning models 210 to 250.

FIG. 22 is a flowchart illustrating a procedure of processing executed by the terminal device 500 and the control device 100. The control unit 501 of the terminal device 500 accesses the control device 100 through the communication unit 503 (step S901).

When the control unit 101 of the control device 100 receives access from the terminal device 500 when the powder processing device 4A is not operating, the control unit 101 generates screen data of a processing type selection screen for selecting the powder processing type. (Step S902), the generated screen data is transmitted from the communication unit 105 to the terminal device 500 (step S903). When the access is received from the terminal device 500 while the powder processing device 4A is operating, the control unit 101 may execute the processes after step S803 of the flowchart shown in FIG.

The control unit 501 of the terminal device 500 receives the screen data of the processing type selection screen transmitted from the control device 100 from the communication unit 503 (step S904). The control unit 501 displays the processing type selection screen on the display unit 505 based on the received screen data (step S905), and accepts the type selection (step S906). FIG. 23 is a schematic diagram showing an example of the processing type selection screen. On this processing type selection screen, a radio button arranged as a component of the user interface is used to select the type of powder processing desired by the user from among drying processing, mixing processing, compounding processing, surface treatment, and granulation processing. Is configured to accept. When the control unit 501 accepts the type selection, the control unit 501 transmits the selection result from the communication unit 503 to the control device 100 (step S907).

The control unit 101 of the control device 100 receives the selection result transmitted from the terminal device 500 from the communication unit 503 (step S908). After receiving the selection result, the control unit 101 executes the processes after step S802 of the flowchart shown in FIG. That is, the control unit 101 inputs the input of the target value to the corresponding learning model 210 (or learning models 220 to 250) according to the type of the powder processing selected by the user. , The process of acquiring the calculation result, the process of controlling the operation of the powder processing system 1 based on the acquired calculation result, and the like may be executed.

As described above, in the ninth embodiment, any of the learning models 210 to 250 is selected according to the type of powder processing selected by the user, and the selected learning model 210 (or learning models 220 to 250) is selected. The operation of the powder processing system 1 can be controlled by using the calculation result of.

(Embodiment 10)
In the tenth embodiment, a configuration for generating the learning model 210 according to the type of the powder raw material will be described.
Since the overall configuration of the powder processing system 1 and the configuration of each device in the powder processing system 1 are the same as those in the first embodiment, the description thereof will be omitted.

The control device 100 according to the tenth embodiment collects measurement data indicating the operating state of the powder processing device 4A and powder data of the powder obtained from the powder processing device 4A for each type of powder raw material. To do. FIG. 24 is a conceptual diagram showing an example of collecting data according to the tenth embodiment. The control unit 101 of the control device 100 acquires measurement data indicating the operating state of the powder processing device 4A and powder data related to the powder obtained from the powder processing device 4A through the input unit 103. The measurement data and powder data acquired by the control unit 101 are the same as those in the first embodiment. The control unit 101 stores the acquired measurement data and powder data in the storage unit 102 together with the time stamp and information on the type of the powder raw material. Here, the information regarding the type of the powder raw material may be character information indicating the type name of the powder raw material, or may be an arbitrary identifier that can specify the type of the powder raw material. Information on the type of powder raw material may be received through the operation unit 106 before data collection or after data collection.

The control unit 101 uses the collected data as teacher data to learn the relationship between the moisture content of the powder obtained from the powder processing apparatus 4A and the control parameters related to the powder processing apparatus 4A, as described above. Generate the training model 210. The example of FIG. 24 shows a state in which data is collected for each type of powder raw material such as magnesium hydroxide, lithium cobalt oxide, and calcium phosphate and stored in the storage unit 102.

The control unit 101 uses the data collected for each type of powder raw material as the teacher data to generate the learning model 210.

FIG. 25 is a flowchart illustrating a procedure for generating the learning model 210 according to the tenth embodiment. The control unit 101 of the control device 100 receives information regarding the type of the powder raw material prior to the generation of the learning model 210 (step S1001). For example, the control unit 101 displays a screen for inquiring the user about the type of the powder raw material on the display unit 107, and can receive information on the type of the powder raw material through the displayed screen. Further, the control unit 101 may receive information on the type of the powder raw material by transmitting an inquiry about the type of the powder raw material from the communication unit 105 to the terminal device 500 and receiving a reply from the terminal device 500. ..

Next, the control unit 101 reads the data stored in association with the type received in step S1001 from the storage unit 102 (step S1002). The control unit 101 selects the data to be used for the teacher data from the read data (step S1003). That is, from the read data, the control unit 101 determines the temperature and flow rate of the heat medium, the supply amount or supply speed of the powder raw material, the rotation speed of the crushing rotor 413, the rotation speed of the classification rotor 415, and the pressure in the casing 410. Select only one set of the measured value of the processing time of the drying process and the value of the wet content of the powder obtained at that time.

Next, the control unit 101 inputs the wet content data included in the selected teacher data into the learning model 210 (step S1004), and executes the calculation by the learning model 210 (step S1005). That is, the control unit 101 inputs the moisture value to the nodes constituting the input layer 211 of the learning model 210, performs calculations using the weights and biases between the nodes in the

intermediate layers

Next, the control unit 101 evaluates the calculation result obtained in step S1005 (step S1006), and determines whether or not the learning is completed (step S1007). 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 S1005 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 order to avoid the problem of overfitting, techniques such as cross-validation and early stopping may be adopted to end learning at an appropriate timing.

When it is determined that the learning is not completed (S1007: NO), the control unit 101 updates the weights and biases between the nodes of the learning model 210 (step S1008), returns the process to step S1003, and another teacher. Continue learning with the data. The control unit 101 may update the weights and biases between the nodes by using an error backpropagation method that sequentially updates the weights and biases between the nodes from the output layer 213 of the learning model 210 toward the input layer 211. it can.

When it is determined that the learning is completed (S1007: YES), the control unit 101 stores the learned learning model 210 in the storage unit 102 in association with the information regarding the type of the powder raw material (step S1009), and this flowchart. Ends the processing by.

As described above, the control device 100 according to the tenth embodiment can generate the learning model 210 according to the type of the powder raw material. The learning model 210 in the tenth embodiment is the same as that in the first embodiment, but instead of the learning model 210, the learning models 220 to 250 described in the second to fifth embodiments are generated for each type of powder raw material. You may.

FIG. 26 is a flowchart illustrating a control procedure by the control device 100. The control unit 101 of the control device 100 receives information on the type of powder raw material and a target value (moisture content) of the powder desired by the user through the operation unit 106 (step S1221).

Next, the control unit 101 reads the learning model 210 according to the type of the received powder raw material from the storage unit 102 (step S1222). The control unit 101 inputs the received moisture data to the input layer 211 of the learning model 210 read in step S1222, and executes the calculation by the learning model 210 (step S1223). At this time, the control unit 101 gives the received moisture data to the node of the input layer 211, and executes the calculation by the

intermediate layers

212A and 212B. The output layer 213 of the learning model 210 is used for the temperature and flow rate of the heat medium, the supply amount or supply speed of the powder raw material, the rotation speed of the crushing rotor 413, the rotation speed of the classification rotor 415, the pressure in the casing 410, and the drying process. Outputs the calculation result related to the control parameters that control the processing time.

Next, the control unit 101 acquires the calculation result from the learning model 210 (step S1224) and determines the control parameters used for control (step S1225). Based on the probability of being output as a calculation result, the control unit 101 includes the temperature and flow rate of the heat medium, the supply amount or supply speed of the powder raw material, the rotation speed of the crushing rotor 413, the rotation speed of the classification rotor 415, and the inside of the casing 410. The combination of pressure and processing time for drying processing may be determined.

Next, the control unit 101 executes control based on the control parameters determined in step S1225 (step S1226). That is, the control unit 101 generates and generates a control command for controlling the operation of the hot air generator 3 so that the temperature and the flow rate of the heat medium introduced into the casing 410 become the values determined in step S1225, respectively. The control command is output to the hot air generator 3 through the output unit 104. Similarly, the control unit 101 has values determined in step S1225 for each of the supply amount of the powder raw material, the rotation speed of the crushing rotor 413, the rotation speed of the classification rotor 415, the pressure in the casing 410, and the processing time of the drying process. As described above, a control command for controlling the operation of the raw material supply machine 2, the crushing rotor 413, the classification rotor 415, the blower 7, and the powder processing device 4A is generated, and the generated control command is output to each device through the output unit 104.

As described above, the control device 100 according to the present embodiment can determine the control parameters by using the learning model 210 according to the type of the powder raw material. The control device 100 can perform control based on the determined control parameters so that a powder having a desired particle size can be obtained.

In the present embodiment, the control parameters are determined using the learning model 210 generated for each type of powder raw material. However, the learning models 220 to 250 described in the second to fifth embodiments are used as powder. The control parameters may be determined by using the generated learning models 220 to 250, which are generated for each type of body material.

As described above, the control device 100 according to the present embodiment can determine the control parameters by using the learning model 210 according to the type of the powder raw material. The control device 100 can perform control based on the determined control parameters so that a powder having a desired moisture content is obtained.

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

For example, in the first to fifth embodiments, the powder processing system 1 includes one powder processing device 4A to 4E, respectively, but the powder processing system 1 includes a plurality of powder processing devices 4A to 4E. It may be constructed by combining.

1 ... Powder processing system, 2 ... Raw material supply machine, 3 ... Hot air generator, 4A-4E ... Powder processing device, 5 ... Cyclone, 6 ... Dust collector, 7 ... Blower, 8 ... Product tank, 9 ... Dust collection tank , 410 ... Casing, 411 ... Raw material input port, 412 ... Gas introduction port, 413 ... Crushing rotor, 414 ... Guide ring, 415 ... Classification rotor, 416 ... Powder outlet, S1 ... Weight sensor, S2 ... Particle size sensor, S3 ... Flow sensor, S4 ... Temperature sensor, S5 ... Moisture sensor, S6 ... Pressure sensor, S7 ... Rotation speed sensor, S8 ... Rotation speed sensor, 100 ... Control device, 101 ... Control unit, 102 ... Storage unit, 103 ... Input unit, 104 ... Output unit, 105 ... Communication unit, 106 ... Operation unit, 107 ... Display unit, 413M ... Grinding motor, 415M ... Classification motor, 500 ... Terminal device, 501 ... Control unit, 502 ... Storage unit, 503 ... Communication unit, 504 ... Operation unit, 505 ... Display unit

Claims

Using a computer
With respect to a powder processing apparatus that performs powder treatment including any one of drying treatment, mixing treatment, compounding treatment, surface treatment, and granulation treatment, measurement data showing an operating state of the powder treatment device and the above-mentioned Obtain powder data related to powder obtained from powder processing equipment,
Using the acquired measurement data and powder data as teacher data, when a target value for the powder desired by the user is input, it is configured to output the calculation result for the control parameters for the powder processing apparatus. How to generate a learning model.
The target value is the moisture content of the powder desired by the user.
The method for generating a learning model according to claim 1, wherein the learning model is generated by using the powder data containing the wet content measured for the powder obtained by the drying treatment as the teacher data.
The target value further includes the yield of the powder desired by the user per unit time.
The method for generating a learning model according to claim 2, wherein the learning model is generated by using the powder data including the yield per unit time measured for the powder obtained by the drying treatment as the teacher data.
The measurement data includes the temperature and flow rate of the heat medium, the supply amount or supply speed of the powder raw material, the rotation speed of the rotating body for stirring the powder raw material, the pressure in the processing chamber for processing the powder raw material, and the pressure in the processing chamber. , Including the processing time of the drying process
When the target value is input, the temperature and flow rate of the heat medium, the supply amount or supply speed of the powder raw material, the rotation speed of the rotating body, the pressure in the processing chamber, and the processing time of the drying treatment are displayed. The method for generating a learning model according to claim 2 or 3, wherein a learning model configured to output a calculation result for the including control parameters is generated.
The measurement data further includes the rotation speed of the rotating body for classifying the powder.
The learning according to claim 4, wherein when the target value is input, a learning model configured to output a calculation result for a control parameter including a rotation speed of a rotating body for classifying the powder is generated. How to generate a model.
The target value is the mixing degree of the powder desired by the user.
The method for generating a learning model according to claim 1, wherein the learning model is generated by using the powder data including the mixing degree of the powder obtained by the mixing process as the teacher data.
The target value further includes the concentration or moisture content of the powder desired by the user.
The method for generating a learning model according to claim 6, wherein the learning model is generated by using the powder data including the concentration or the moisture measured for the powder obtained by the mixing treatment as the teacher data.
The measurement data includes the supply amount or supply speed of the powder raw material, the rotation speed of the rotating body for stirring the powder raw material, the supply amount or supply speed of the fluid supplied to the powder processing apparatus, the flow rate or the pressure. In addition, the processing time of the mixing process is included.
When the target value is input, the control parameters including the supply amount or supply speed of the powder raw material, the rotation speed of the rotating body, the supply amount or supply speed of the fluid, and the processing time of the mixing process. The method for generating a learning model according to claim 6 or 7, wherein a learning model configured to output a calculation result is generated.
The target value is the degree of compounding of the powder desired by the user.
The method for generating a learning model according to claim 1, wherein the learning model is generated by using the powder data including the degree of compounding measured for the powder obtained by the compounding process as the teacher data.
The target value further includes the conductivity, thermal conductivity or transmittance of the powder desired by the user.
The generation of the learning model according to claim 9, wherein the learning model is generated by using the powder data including the conductivity, thermal conductivity or transmittance measured for the powder obtained by the compounding treatment as the teacher data. Method.
The measurement data includes the supply amount or supply speed of the powder raw material, the rotation speed of the rotating body for stirring the powder raw material, the processing time of the compounding process, and the load power of the powder processing device. ,
When the target value is input, control parameters including the supply amount or supply speed of the powder raw material, the rotation speed of the rotating body, the processing time of the compounding process, and the load power of the powder processing apparatus. The method for generating a learning model according to claim 9 or 10, wherein a learning model configured to output the calculation result of the above is generated.
The target value is the circularity of the powder desired by the user.
The method for generating a learning model according to claim 1, wherein the learning model is generated by using the powder data including the circularity measured for the powder obtained by the surface treatment as the teacher data.
The target value further includes the density or fluidity of the powder desired by the user.
The method for generating a learning model according to claim 12, wherein the learning model is generated by using the powder data including the density or fluidity measured for the powder obtained by the surface treatment as the teacher data.
The measurement data includes the supply amount or supply speed of the powder raw material, the rotation speed of the rotating body for stirring the powder raw material, the rotation speed of the rotating body for classifying the powder, and the processing time of the surface treatment. , And the load power of the powder processing apparatus.
When the target value is input, the supply amount or supply speed of the powder raw material, the rotation speed of the rotating body for stirring and classification, the treatment time of the surface treatment, and the load power of the powder treatment apparatus. The method for generating a learning model according to claim 12 or 13, wherein a learning model configured to output a calculation result for a control parameter including the above is generated.
The method for generating a learning model according to any one of claims 12 to 14, wherein the fluid flowing in the system including the powder processing apparatus is a gas or a liquid.
The target value includes the particle size and shape of the powder desired by the user.
The method for generating a learning model according to claim 1, wherein the learning model is generated by using powder data including the particle diameter and shape measured for the powder obtained by the granulation treatment as teacher data.
The target value further includes the density, fluidity, hardness, water absorption amount, or oil absorption amount of the powder desired by the user.
The 16th claim, wherein the learning model is generated by using the powder data including the density, fluidity, hardness, water absorption amount, or oil absorption amount measured for the powder obtained by the granulation treatment as the teacher data. How to generate a training model.
The measurement data is input to the supply amount or supply speed of the powder raw material, the rotation speed of the rotating body for stirring the powder raw material, the pressure in the processing chamber for processing the powder raw material, and the processing chamber. Including the input amount of additives,
When the target value is input, control parameters including the supply amount or supply speed of the powder raw material, the rotation speed of the rotating body, the pressure in the processing chamber, and the input amount of the additive to be charged into the processing chamber. The method for generating a learning model according to claim 16 or 17, wherein a learning model configured to output the calculation result of the above is generated.
The measurement data includes the driving power or driving current of the powder processing apparatus, the flow rate or temperature of the refrigerant supplied to the powder processing apparatus, and the discharge / suction flow rate when the powder is taken out from the processing chamber of the powder processing apparatus. Alternatively, the temperature and at least one of the temperature, humidity or pressure in the processing chamber are further included.
When the target value is input, at least one of the driving power or driving current of the powder processing apparatus, the flow rate or temperature of the refrigerant, the discharge / suction flow rate or temperature, or the temperature, humidity or pressure in the processing chamber. The method for generating a learning model according to any one of claims 1 to 18, wherein a learning model configured to output a calculation result for a control parameter including one is generated.
The teacher data further includes raw material data of the powder raw material used in the powder processing.
When the target value and the raw material data of the powder raw material to be supplied to the powder processing apparatus are input by using the teacher data including the raw material data, the calculation result for the control parameter is output. The method for generating a learning model according to any one of claims 1 to 19, wherein the learning model is generated.
The raw material data includes moisture, temperature, density, particle size, circularity, mixing degree, fluidity, BET (Brunauer --Emmett --Teller) value, NIR (Near Infrared), and XRD (X-ray) of the powder raw material. At least one of Diffraction), TG-DTA (Thermogravimetry-Differential Thermal Analysis), MS (Mass Spectrometry), SEM (Scanning Electron Microscope), FE-SEM (Field Emission-SEM), and TEM (Transmission Electron Microscope) data. 20. The method of generating a learning model according to claim 20.
The teacher data further includes environmental data when performing the powder processing.
Learning configured to output the calculation result for the control parameter when the target value and the environment data for executing the powder processing are input by using the teacher data including the environment data. The method for generating a learning model according to any one of claims 1 to 21.
Re-acquire the measurement data and powder data used for the teacher data,
The method for generating a learning model according to any one of claims 1 to 22, wherein the re-acquired measurement data and the powder data are used as teacher data to relearn the learning model.
On the computer
With respect to a powder processing apparatus that performs powder treatment including any one of drying treatment, mixing treatment, compounding treatment, surface treatment, and granulation treatment, measurement data showing an operating state of the powder treatment device and the above-mentioned Obtain powder data related to powder obtained from powder processing equipment,
Using the acquired measurement data and powder data as training data, when a target value for the powder desired by the user is input, it is configured to output the calculation result for the control parameters for the powder processing apparatus. A computer program for executing the process of generating a learning model.
Input layer, where the target value for the powder desired by the user is input
An output layer that outputs calculation results for control parameters for a powder processing apparatus that performs powder processing including any one of drying treatment, mixing treatment, compounding treatment, surface treatment, and granulation treatment, and the powder. Using the powder data related to the powder obtained from the processing apparatus and the measurement data indicating the operating state of the powder processing apparatus as the teacher data, the target value input to the input layer and the output layer output. It has an intermediate layer that has learned the relationship with the calculation result.
A learning model in which, when a target value for a powder desired by a user is input to the input layer, a calculation is performed in the intermediate layer and the computer functions to output a calculation result for the control parameter.
A reception unit that accepts target values for the powder desired by the user,
Powder data related to powder obtained from a powder treatment apparatus that performs powder treatment including any one of drying treatment, mixing treatment, compounding treatment, surface treatment, and granulation treatment, and the powder treatment apparatus. It is configured to output the calculation result of the control parameters for the powder processing device when the target value received by the reception unit is input after learning using the measurement data indicating the operating state as the teacher data. With the learning model
An arithmetic processing unit that inputs the target value received by the reception unit into the learning model and executes an operation by the learning model.
A control device including a control unit that controls operations related to the powder processing device based on the calculation result of the learning model.
The learning model is provided for each type of powder processing executed by the powder processing apparatus.
The reception section accepts selections for the type of powder processing,
The calculation processing unit selects a learning model corresponding to the type of powder processing received by the reception unit, and executes the calculation by the learning model by inputting the target value into the selected learning model. The control device described in.
The learning model is provided for each type of powder raw material.
The reception section accepts selections for the type of powder raw material,
The calculation processing unit selects a learning model corresponding to the type of powder raw material received by the reception unit, and inputs the target value to the selected learning model to execute the calculation by the learning model. The control device described in.
It is provided with an adjustment unit that adjusts the control parameters according to the target value, the powder data, and the comparison result.
The control device according to any one of claims 26 to 28, wherein the adjusting unit controls the operation of the powder processing device based on the adjusted control parameters.
A generator that generates screen data for displaying at least one of the target value, the powder data, the measurement data, the operation history of the powder processing apparatus, and the apparatus configuration diagram including the powder processing apparatus. ,
The control device according to any one of claims 26 to 29, further comprising an output unit for outputting the generated screen data.
From claim 26, the control unit controls at least one operation of the powder processing apparatus and a raw material feeder, a hot air generator, a cyclone, a dust collector, a blower, and a pump connected to the powder processing apparatus. The control device according to any one of claims 30.
Accepting the target value for the powder desired by the user,
Powder data related to powder obtained from a powder processing apparatus that performs powder processing including any one of drying treatment, mixing treatment, compounding treatment, surface treatment, and granulation treatment, and the received target value. The measurement data indicating the operating state of the powder processing apparatus is learned as the teacher data, and the calculation result of the control parameters for the powder processing apparatus is output in response to the input of the target value. Fill in the configured learning model and
Execute the calculation by the learning model
A control method for controlling the operation of the powder processing apparatus based on the calculation result of the learning model.
On the computer
Accepting the target value for the powder desired by the user,
Powder data related to powder obtained from a powder processing apparatus that performs powder processing including any one of drying treatment, mixing treatment, compounding treatment, surface treatment, and granulation treatment, and the received target value. The measurement data indicating the operating state of the powder processing apparatus is learned as the teacher data, and the calculation result of the control parameters for the powder processing apparatus is output in response to the input of the target value. Fill in the configured learning model and
Execute the calculation by the learning model
A computer program for executing a process for controlling the operation of the powder processing apparatus based on the calculation result of the learning model.