CN114746684B

CN114746684B - Machine learning device, data processing system, inference device, and machine learning method

Info

Publication number: CN114746684B
Application number: CN202180006800.2A
Authority: CN
Inventors: 青山文明
Original assignee: Kaneko Sangyo Co Ltd
Current assignee: Kaneko Sangyo Co Ltd
Priority date: 2020-03-16
Filing date: 2021-03-09
Publication date: 2023-12-29
Anticipated expiration: 2041-03-09
Also published as: JP2021148142A; JP6779456B1; KR102629089B1; CN114746684A; KR20220028064A; WO2021187212A1

Abstract

The invention provides a machine learning device and the like capable of grasping abnormality and abnormality symptoms in a fluid pressure driven valve with high accuracy without depending on experience of an operator or the like. A machine learning device (200) is applied to a fluid pressure driven valve (10) including a main valve (11), a driving device (12) that drives the main valve (11), and a solenoid valve (1) that controls supply and discharge of a driving fluid (A) to and from the driving device (12), and the machine learning device (200) is provided with: a learning data set storage unit (202) that stores a plurality of sets of learning data sets that are configured from input data including a valve opening degree of a main valve (11), a pressure of a driving fluid (A), a control parameter of a solenoid portion (3), and a temperature of a fluid pressure driving valve (10), and output data that is configured from diagnostic information of the fluid pressure driving valve (10) that has a correspondence with the input data; a learning unit (203) that learns a learning model that infers a correlation between input data and output data by inputting a plurality of sets of learning data sets; and a learned model storage unit (204) that stores a learned model.

Description

Machine learning device, data processing system, inference device, and machine learning method

Technical Field

The present invention relates to a machine learning device, a data processing system, an inference device, and a machine learning method for performing abnormality diagnosis of a valve system.

Background

Conventionally, a fluid pressure-driven valve is known in which a main valve is opened and closed by controlling a driving fluid using a solenoid valve. For example, patent document 1 discloses an emergency shutoff valve device that closes a ball valve (main valve) by controlling a driving fluid with an electromagnetic valve at an emergency when an abnormality occurs in a plant as a fluid pressure driving valve used in a piping of the plant, thereby shutting off the fluid flowing in the piping.

Prior art literature

Patent literature

Patent document 1: japanese patent laid-open No. 2009-97539

Disclosure of Invention

Problems to be solved by the invention

In the fluid pressure driven valve such as the emergency stop valve for the plant as disclosed in patent document 1, it is preferable that unexpected abnormality does not occur in order to improve the operation rate and reliability of the whole plant. Therefore, in such a fluid pressure driven valve, it is desirable to realize predictive maintenance for grasping the sign of an abnormality, in addition to performing post-maintenance for grasping the abnormality when the abnormality occurs.

Here, the abnormal symptoms of the fluid pressure-driven valve may appear as various images, but the causal relationship between the images that the fluid pressure-driven valve may appear to the abnormal symptoms has not been clearly determined. As a result, predictive maintenance of the fluid pressure actuated valve is performed based on judgment depending on experience (including implicit knowledge) of the operator, and there is a problem in that accuracy varies depending on the responsible operator.

The present invention has been made in view of the above-described problems, and an object thereof is to provide a machine learning device, a data processing system, an inference device, and a machine learning method for grasping abnormality and signs of abnormality (hereinafter, these will be collectively referred to as "abnormality" in the present invention) in a fluid pressure driven valve with high accuracy.

Solution for solving the problem

In order to achieve the above object, a machine learning device according to a first aspect of the present invention is applied to a fluid pressure driven valve 10 including at least a main valve 11, a driving device 12 that drives the main valve 11, and a solenoid valve 1 that includes a solenoid portion 3 that controls supply and discharge of a driving fluid a with respect to the driving device 12, as shown in fig. 1 to 4, and includes: a learning data set storage unit 202 that stores a plurality of sets of learning data sets including input data including a valve opening degree of the main valve 11, a pressure of the driving fluid 12, a control parameter of the solenoid portion 3, and a temperature of the fluid pressure driving valve 10, and output data including diagnostic information of the fluid pressure driving valve 10 in which a correspondence relationship with the input data is established; a learning unit 203 that learns a learning model that infers a correlation between the input data and the output data by inputting a plurality of sets of the learning data sets; and a learned model storage unit 204 that stores the learning model learned by the learning unit 203.

Effects of the invention

The machine learning device according to the present invention can provide a learned model that can infer the presence or absence of an abnormality in the fluid pressure-driven valve based on various information and the like that can be obtained when the fluid pressure-driven valve is operating stably. Therefore, by using the learned model, it is possible to accurately estimate the abnormality occurring in the fluid pressure actuated valve without depending on the experience of the operator.

Drawings

Fig. 1 is a schematic diagram showing an example of a fluid pressure driven valve to which a machine learning device or the like according to an embodiment of the present invention is applied.

Fig. 2 is a schematic diagram showing an example of a solenoid valve to which the machine learning device and the like according to one embodiment of the present invention are applied.

Fig. 3 is a block diagram showing an example of a solenoid valve to which the machine learning device and the like according to one embodiment of the present invention are applied.

Fig. 4 is a schematic block diagram of a machine learning device according to an embodiment of the present invention.

Fig. 5 is a diagram showing a configuration example (supervised learning) of data used in a machine learning apparatus or the like according to an embodiment of the present invention.

Fig. 6 is a diagram showing a configuration example (unsupervised learning) of data used in a machine learning device or the like according to an embodiment of the present invention.

Fig. 7 is a diagram showing an example of a neural network model for supervised learning implemented in a machine learning apparatus of an embodiment of the present invention.

Fig. 8 is a flowchart showing an example of a machine learning method of an embodiment of the present invention.

FIG. 9 is a schematic block diagram illustrating a data processing system in accordance with one embodiment of the present invention.

Fig. 10 is a flowchart showing an example of a data processing procedure of a data processing system according to an embodiment of the present invention.

Detailed Description

Hereinafter, embodiments for carrying out the present invention will be described with reference to the accompanying drawings. The range necessary for the description of the present invention for achieving the object is schematically shown below, and the range necessary for the description of the corresponding part of the present invention is mainly described, and the part omitted is based on the known technology.

Before explaining a machine learning device, a data processing system, an inference device, and a machine learning method according to an embodiment of the present invention, a fluid pressure driven valve to which the machine learning device and the like are applied will be first described.

(fluid pressure actuated valve)

Fig. 1 is a schematic diagram showing an example of a fluid pressure driven valve 10 according to an embodiment of the present invention. The fluid pressure driven valve 10 according to the present embodiment can be used as, for example, an emergency shut-off valve provided in a plant in a piping 100 through which various gases, oil, etc. flow, and configured to shut off the flow of the piping 100 when an emergency stop such as an abnormality occurs in the plant. The installation position and use of the fluid pressure actuated valve 10 are not limited to the above examples.

The fluid pressure driven valve 10 shown in fig. 1 includes: a main valve 11 disposed in the middle of the pipe 100; a fluid pressure type driving device 12 for driving a valve stem 13 connected to the main valve 11 according to the fluid pressure of the driving fluid, thereby opening and closing the main valve 11; and an electromagnetic valve 1 having a function of controlling supply and discharge of the driving fluid to and from the driving device 12.

The driving fluid used for the fluid pressure driving valve 10 is air (hereinafter simply referred to as "air") a compressed by a pneumatic gauge. The air a is supplied from the air supply source 14 to the solenoid valve 1 via the first air pipe 140, and then supplied to the driving device 12 via the second air pipe 141. Further, a communication cable 150 for transmitting and receiving various data between the external device 15 and the solenoid valve 1 and a power cable 160 for supplying power from the external power source 16 to the solenoid valve 1 are connected to the fluid pressure driven valve 10. The driving fluid is not limited to the air a, and may be other gas or liquid (e.g., oil).

The external device 15 is a device for transmitting and receiving various information to and from the fluid pressure driven valve 10, and is constituted by, for example, a complete equipment management computer (including a local server and a cloud server), a diagnostic computer used by an operator (maintenance and repair person), an external storage unit such as a USB memory, and an external HDD. The external device 15 can also be connected to a machine learning device 200 described later, and transmits various data constituting a learning data set. The external device 15 includes a report means including a GUI (Graphical User Interface: graphical user interface) or the like for reporting the occurrence of an abnormality and the contents thereof to an operator or the like when the fluid pressure driven valve 10 is abnormal. The communication between the external device 15 and the solenoid valve 1 may be wireless communication.

The fluid pressure actuated valve 10 of the present embodiment is driven by an airless closure method. Therefore, during the steady operation, the main valve 11 is opened by supplying air a (supply air) from the air supply source 14 to the driving device 12 via the solenoid valve 1, and during the emergency stop and the test operation, the main valve 11 is closed by discharging air a (exhaust air) from the driving device 12 via the solenoid valve 1. In this case, the fluid pressure driven valve 10 may be opened without air, and in this case, the air a is supplied to the driving device 12 to perform the closing operation, and the air a is discharged from the driving device 12 to perform the closing operation on the main valve 11.

The main valve 11 is a ball valve. The specific structure of the main valve 11 includes a valve body 110 disposed in the middle of the pipe 100 and a spherical valve element 111 rotatably provided in the valve body 110. A first end 130A of the valve stem 13 is connected to an upper portion of the valve body 111. The valve body 110 is rotated by the valve rod 13 corresponding to 0 to 90 degrees, and the valve element 111 is switched between a fully open state (state shown in fig. 1) and a fully closed state of the main valve 11. The valve used as the main valve 11 is not limited to a ball valve, and may be, for example, a butterfly valve or another two-position valve.

The driving device 12 employs a single-acting cylinder mechanism disposed between the main valve 11 and the solenoid valve 1. The specific configuration of the driving device 12 includes: a cylindrical cylinder 120; a pair of pistons 122A, 122B provided in the cylinder 120 so as to be capable of reciprocating linear movement, and coupled via a piston rod 121; a coil spring 123 provided on the first piston 122A side; an air supply and discharge port 124 formed on the second piston 122B side; a transmission mechanism 125 provided at a portion of the valve rod 13 disposed so as to extend through the cylinder 120 in the radial direction, the portion being orthogonal to the piston rod 121. The driving device 12 is not limited to the single-action type, and may be configured by other modes such as a multi-action type.

The first piston 122A is biased by the coil spring 123 to move the main valve 11 in the closing direction. The second piston 122B is pushed by the air a (air supply) supplied from the air supply/discharge port 124, so that the main valve 11 is operated in the opening direction (against the biasing force of the coil spring 123). The transmission mechanism 125 is configured by a rack-and-pinion mechanism, a link mechanism, a cam mechanism, or the like, and converts the reciprocating linear motion of the piston rod 121 into a rotational motion and transmits the rotational motion to the valve stem 13.

The valve stem 13 is formed in a shaft shape, and is rotatably disposed through the driving device 12. The first end 130A of the valve stem 13 is coupled to the main valve 11, and the second end 130B of the valve stem 13 is pivotally supported by the solenoid valve 1. The valve stem 13 may be configured such that a plurality of shafts are coupled via a coupling or the like.

The solenoid valve 1 has a function of controlling the supply and discharge of air a to and from the driving device 12, and is configured as a three-way solenoid valve that is normally closed (i.e., is "on" when energized and "off" when not energized). The solenoid valve 1 includes a spool valve portion 2 that switches a flow path of air a and a solenoid portion 3 that displaces the spool valve portion 2 in accordance with an energized state (when energized or when non-energized) in an interior of a housing portion 6 that functions as a casing of the indoor type or explosion-proof solenoid valve 1. The solenoid valve 1 is not limited to the three-way solenoid valve of the above type, and may be a three-position solenoid valve, a normally open solenoid valve, a four-way solenoid valve, or the like, and may be formed in various forms based on any combination thereof. In the present embodiment, the solenoid valve 1 is used as the pilot valve in the fluid pressure driven valve 10, but the use of the solenoid valve 1 is not limited thereto.

The spool 2 includes an input port 20 connected to the air supply source 14 via a first air pipe 140, an output port 21 connected to the driving device 12 via a second air pipe 141, and an exhaust port 22 for exhausting exhaust gas from the driving device 12.

The solenoid portion 3 displaces the spool portion 2 so as to communicate between the input port 20 and the output port 21 when energized, and displaces the spool portion 2 so as to communicate between the output port 21 and the exhaust port 22 when de-energized.

With the above-described series of configurations, when the solenoid valve 1 is in the energized state, air a (supplied air) from the air supply source 14 flows through the first air pipe 140, the input port 20, the output port 21, and the second air pipe 141 in this order, and is supplied to the air supply/discharge port 124, whereby the second piston 122B is pressed, and the coil spring 123 is compressed. When the valve rod 13 is driven to rotate via the piston rod 121 and the transmission mechanism 125 by an amount corresponding to the compression of the coil spring 123 by the piston rod 121, the valve body 111 rotates within the valve body 110, and the main valve 11 is operated in the fully open state.

On the other hand, when the solenoid valve 1 is in the non-energized state, the air a (exhaust gas) in the cylinder 120 flows through the second air pipe 141, the output port 21, and the exhaust port 22 in this order from the air supply/exhaust port 124, and is discharged to the outside air, whereby the pressing force of the second piston 122B is reduced, and the coil spring 123 returns from the compressed state. When the valve rod 13 is driven to rotate via the transmission mechanism 125 and the piston rod 121 moves by an amount corresponding to the restoration of the coil spring 123, the valve body 111 rotates in the valve body 110, and the main valve 11 is operated in the fully closed state.

Fig. 2 is a cross-sectional view showing an example of the solenoid valve 1 according to the embodiment of the present invention. As shown in fig. 2, the solenoid valve 1 of the present embodiment includes, in addition to the spool valve section 2 and the solenoid section 3, the following components: a plurality of sensors 4 that acquire states of respective portions of the solenoid valve 1; a substrate 5 on which at least one of the plurality of sensors 4 is mounted; and a housing portion 6 that houses the spool portion 2, the solenoid portion 3, the plurality of sensors 4, and the substrate 5.

The housing section 6 includes: a first housing portion 60 that houses the spool 2; a second housing portion 61 adjacent to the first housing portion 60 and housing the solenoid portion 3, the plurality of sensors 4, and the substrate 5; junction box 62 connects communication cable 150 and power cable 160. The first and second housing portions 60 and 61 are made of a metal material such as aluminum, for example.

The first housing portion 60 has openings (not shown) that function as the input port 20, the output port 21, and the exhaust port 22, respectively.

The second housing portion 61 includes: a cylindrical housing 610 having both ends (a first housing end 610a and a second housing end 610 b) open; a main body 611 disposed inside the housing 610; a solenoid cover 612 that covers the solenoid portion 3 fixed to the first housing end 610a to isolate outside air; and a junction box cover 613 that covers the junction box 62 fixed to the second housing end 610b to isolate outside air.

The housing 610 has: a shaft insertion port 610c formed at a lower portion of the housing 610, into which the second end 130B of the valve stem 13 is inserted; a body insertion port 610d formed at an upper portion of the housing 610, into which the body 611 is inserted; a cable insertion port 610e formed on the second housing end 610b side into which the communication cable 150 and the power cable 160 are inserted.

The first housing portion 60 and the second housing portion 61 are formed so as to penetrate the main body 611: a first flow path 63 that branches from the input side flow path 26 and communicates between the input side flow path 26 and the first pressure sensor 40; a second flow path 64 that branches from the output side flow path 27 and communicates between the output side flow path 27 and the second pressure sensor 41; and a spool flow path 65 through which air a that has caused the spool 2 to interlock with the solenoid 3 flows.

The spool valve section 2 includes: a spool hole 23 formed in a second housing portion 61 functioning as a spool housing; a spool 24 movably disposed in the spool hole 23; a spool spring 25 that biases the spool 24; an input-side flow path 26 that communicates between the input port 20 and the spool hole 23; an output-side flow path 27 that communicates the output port 21 and the spool hole 23; and an exhaust flow path 28 that communicates between the exhaust port 22 and the spool hole 23.

The solenoid section 3 includes: a solenoid housing 30; a solenoid coil 31 housed in the solenoid case 30; a movable iron core 32 movably disposed within the solenoid coil 31; a fixed iron core 33 disposed in a fixed state in the solenoid coil 31; and a solenoid spring 34 that biases the movable iron core 32.

When the solenoid valve 1 is switched from the non-energized state to the energized state, a coil current flows through the solenoid coil 31 in the solenoid portion 3, and the solenoid coil 31 generates an electromagnetic force, and the movable iron core 32 is attracted by the fixed iron core 33 against the urging force of the solenoid spring 34 by the electromagnetic force, thereby switching the flow state of the air a flowing through the spool flow path 65. In the spool portion 2, the state of the air a flowing through the spool passage 65 is switched, whereby the spool 24 moves against the urging force of the spool spring 25, and the state between the communication input port 20 and the exhaust port 22 is switched to the state between the communication input port 20 and the output port 21.

The substrate 5 includes: a first substrate 50 having substrate surfaces 500A and 500B along the stem 13 inserted from the shaft insertion port 610 c; a second substrate 51 disposed close to the junction box 62; and a third substrate 52 disposed near the solenoid portion 3.

The main body 611, the solenoid portion 3, and the third substrate 52 are disposed on the first substrate surface 500A side of the substrate surfaces 500A, 500B of the first substrate 50. The second substrate 51 and the junction box 62 are disposed on the second substrate surface 500B side opposite to the first substrate surface 500A side.

The sensors 4 are disposed at appropriate positions of the first substrate 50, the second substrate 51, and the third substrate 52. The sensor 4 includes, for example: a first pressure sensor 40 that measures the fluid pressure of the air a flowing in the input-side flow path 26 and the first flow path 63; a second pressure sensor 41 that measures the fluid pressure of the air a flowing in the output-side flow path 27 and the second flow path 64; and a main valve opening sensor 42 that measures a rotation angle at which the valve stem 13 is driven to rotate, and acquires valve opening information of the main valve 11 based on the rotation angle.

The main valve opening sensor 42 is constituted by, for example, a magnetic sensor, measures the magnetic field intensity generated by the permanent magnet 131 attached to the second end 130B of the valve stem 13, and obtains valve opening information of the main valve 11 based on the magnetic field intensity. The main valve opening sensor 42 is preferably placed at a position facing the outer periphery of the valve stem 13 around the axis in the first substrate surface 500A of the first substrate 5 arranged along the valve stem 13 inserted from the shaft insertion port 610 c. Accordingly, the main valve opening sensor 42 mounted on the first substrate 50 and the second end 130B of the valve stem 13 can be disposed close to each other in the housing portion 6 without wasting the disposing space, and the valve opening information can be accurately acquired.

Fig. 3 is a block diagram showing an example of the solenoid valve 1 according to the embodiment of the present invention. As shown in fig. 3, the electromagnetic valve 1 has an electrical configuration example including a control unit 7 for controlling the electromagnetic valve 1, a communication unit 8 for communicating with an external device 15, and a power supply circuit unit 9 connected to an external power supply 16, in addition to the substrate 3 and the sensor 4.

The plurality of sensors 4 are provided as a sensor group for measuring physical quantities of each portion, and include, in addition to the first pressure sensor 40, the second pressure sensor 41, and the main valve opening sensor 42: a voltage sensor 43 that measures a supply voltage to the solenoid section 3; a current/resistance sensor 44 that measures a current value at the time of energization and a resistance value at the time of non-energization in the solenoid portion 3; a temperature sensor 45 that measures the internal temperature of the housing 6; and a magnetic sensor 46 that measures the intensity of the magnetic field generated by the solenoid section 3.

The plurality of sensors 4 are provided as a sensor group for acquiring information on the operation histories of the respective sections, and include: an operation timer (timer) 47 that measures at least one of a total of energization times to the solenoid portion and a current energization linked time as an operation time of the solenoid portion 3; and an operation counter (counter) 48 for counting the number of operations of the solenoid valve 1, the driving device 12, and the main valve 11.

The sensors 40 to 48 are not limited to the individual sensors described above, and may be provided with the function of other sensors in combination with a specific sensor, and the other sensors may not be provided individually. For example, the magnetic sensor 46 may measure the magnetic field intensity generated by the solenoid unit 3, and the current value at the time of energization in the solenoid unit 3 may be obtained from the magnetic field intensity, whereby the current/resistance sensor 44 may not be separately provided. Further, the microcontroller 70 may incorporate the function of the sensor or may implement a part of the function of the sensor, for example, the microcontroller 70 may incorporate the operation timer 47 and the operation counter 48, and the operation timer 47 and the operation counter 48 may not be separately provided.

The control unit 7 includes: a microcontroller 70 that processes information indicating the states of the respective parts of the solenoid valve 1 acquired by the plurality of sensors 4 and controls the respective parts of the solenoid valve 1; and a valve test switch 71 for controlling the state of energization of the solenoid portion 3 and performing an opening/closing operation of the main valve 11 during test operation.

The microcontroller 70 includes a processor (not shown) such as a CPU (Central Processing Unit: central processing unit) and a Memory including a ROM (Read Only Memory), a RAM (Random Access Memory: random access Memory), and the like. The microcontroller 70 can include functionality to implement the data processing system 300 described below in this embodiment.

The valve test switch 71 is configured to receive a command from the microcontroller 70 when a predetermined test operation condition is satisfied, and to perform a stroke test of the fluid pressure driven valve 10 as a test operation.

The stroke test is performed by, for example, any one of a full stroke test and a partial stroke test. The full-stroke test is performed as follows: the main valve 11 is switched from the energized state to the non-energized state in the fully open state to be operated in the fully closed state, and is switched from the non-energized state to the energized state to be returned to the fully open state in the fully closed state. The partial stroke test is performed as follows: the main valve 11 is not operated in the fully closed state (i.e., the plant is not stopped), and the main valve 11 is partially closed to a predetermined opening degree by being switched from the energized state to the non-energized state in the fully open state, and is returned to the fully open state by being switched from the non-energized state to the energized state in the partially closed state.

For example, when an execution time based on an execution frequency (for example, 1 year) specified by the manager as a set value, a specific specified date and time arrives, an execution command from the external device 15 is received, or a test button (not shown) provided in the solenoid valve 1 is operated by the manager, the test operation (stroke test) may be executed as a test operation condition is satisfied.

(machine learning device)

In the fluid pressure driven valve 10 having the above-described series of configurations, by having the above-described plurality of sensors 4, various information of the fluid pressure driven valve 10 can be obtained at the time of steady operation and at the time of unsteady operation (for example, at the time of test operation including opening and closing operation, at the time of emergency stop). Therefore, the machine learning device 200 that learns an inference model (learned model) capable of inferring diagnostic information of the fluid pressure driven valve 10 based on information (state variables) that can be acquired from the fluid pressure driven valve 10 will be described below. The machine learning device 200 described herein includes not only a machine learning device provided as a device for performing an individual action, but also a machine learning device provided in the form of a non-transitory computer-readable medium storing a program for causing an arbitrary processor to perform an action described below or one or more instructions for causing an arbitrary processor to perform the action.

Fig. 4 is a schematic block diagram of a machine learning device 200 according to an embodiment of the present invention. As shown in fig. 4, the machine learning device 200 of the present embodiment includes a learning data set acquisition unit 201, a learning data set storage unit 202, a learning unit 203, and a learning model storage unit 204.

The learning data set acquisition unit 201 is an interface unit for acquiring a plurality of data constituting a learning (training) data set from various devices connected via, for example, a wired or wireless communication line. Here, examples of the various devices connected to the learning data set acquisition means 201 include an external device 15, an operator computer PC1 used by an operator of the fluid pressure driven valve 10, and the like. In fig. 4, the learning data set acquisition unit 201 is shown as being connected to the external device 15 and the computer PC1, but the external device 15 and the worker computer PC1 may be configured by the same computer. In the learning data set acquisition unit 201, it is possible to acquire, for example, detection data of the plurality of sensors 4 of the fluid pressure-driven valve 10 as input data from the external device 15, and to acquire, for example, diagnostic information of the fluid pressure-driven valve 10 associated with the input data as output data from the operator computer PC 1. The input data and the output data correlated with each other constitute one learning data set described later.

Fig. 5 is a diagram showing a configuration example (supervised learning) of data used in the machine learning device 200 according to the embodiment of the present invention. Fig. 6 is a diagram showing a configuration example (unsupervised learning) of data used in the machine learning device 200 according to an embodiment of the present invention. Fig. 5 and 6 are also appropriately referred to in the description of the data processing system and the inference means.

The learning data set is a data set for use in machine learning described later, and as shown in fig. 5 and 6, the input data includes at least the valve opening of the main valve 11, the pressure of the air a, the control parameter of the solenoid portion 3, and the temperature of the fluid pressure driven valve 10, and the output data includes diagnostic information of the fluid pressure driven valve 10. The details of these various data are described below as examples, but the present invention is not limited thereto.

The valve opening of the main valve 11 is a value of the open/close state of the main valve 11, and can be obtained from the main valve opening sensor 42. The temperature of the fluid pressure-driven valve 10 is a value of the fluid pressure-driven valve 10, particularly the internal temperature, and can be obtained by the temperature sensor 45.

The pressure of the air a is preferably the pressure of the air a flowing through each portion inside the fluid pressure driven valve 10, and specifically, preferably includes at least one of the solenoid valve input side pressure of the air a supplied from the air supply source 14 to the solenoid valve 1, the solenoid valve output side pressure of the air a supplied from the solenoid valve 1 to the driving device 12, and the pressure difference between the solenoid valve input side pressure of the air a and the solenoid valve output side pressure of the air a. The solenoid valve output side pressure of the air a is the pressure of the air a supplied from the solenoid valve 1 to the driving device 12, and includes the pressure of the air a (supplied air) when the air a is supplied from the solenoid valve 1 to the driving device 12 and the pressure of the air a (discharged air) when the air a is discharged from the driving device 12 to the outside air via the solenoid valve 1. The solenoid valve input side pressure of the air a can be obtained by the first pressure sensor 40, the solenoid valve output side pressure of the air a can be obtained by the second pressure sensor 41, and the pressure difference can be calculated from the outputs of the first pressure sensor 40 and the second pressure sensor 41.

The control parameter of the solenoid unit 3 is various parameter information for controlling the solenoid unit 3, and specifically preferably includes at least one of a supply voltage to be supplied to the solenoid coil 31, a current value when the solenoid coil 31 is energized, a resistance value when the solenoid coil 31 is not energized, an operation time of the solenoid unit 3, and a magnetic field strength generated in the solenoid unit 3. The supply voltage to the solenoid coil 31 can be obtained by the voltage sensor 43, the current value when the solenoid coil 31 is energized and the resistance value when the solenoid coil 31 is not energized can be obtained by the current/resistance sensor 44, the operation time of the solenoid unit 3 can be obtained by the operation timer 47, and the magnetic field strength generated in the solenoid unit 3 can be obtained by the magnetic sensor 46.

The valve opening of the main valve 11, the pressure of the air a, the control parameter of the solenoid portion 3, and the temperature of the fluid pressure-driven valve 10, which constitute input data, may be constituted by one piece of data (time data) at a specific time, or may be constituted by a plurality of pieces of data (time data) acquired at a plurality of different times within a predetermined period, as shown in brackets in fig. 5 and 6. When each data is composed of time series data, the time series data of the valve opening degree of the main valve 11, the time series data of the pressure of the air a, the time series data of the control parameter of the solenoid portion 3, and the time series data of the temperature of the fluid pressure driven valve 10 may be data acquired at a plurality of times with the same sampling period and the same phase (state without a phase difference), or may be data different in at least one of the sampling period and the phase. In order to effectively improve the accuracy of the reasoning, the latter is preferably in the form of time series data.

The diagnostic information of the fluid pressure driven valve 10 is information showing whether or not the fluid pressure driven valve 10 has some kind of abnormality at the time of abnormality diagnosis of the fluid pressure driven valve 10, and various forms may be adopted as the data form thereof. The abnormality includes not only a post-abnormality in which occurrence of the abnormality is determined at the time of abnormality diagnosis, but also an abnormality symptom in which occurrence of the future abnormality is predicted while the abnormality is within an allowable range determined to be normal at the time of abnormality diagnosis.

For example, as shown in fig. 5, the diagnostic information as one embodiment may be composed of information indicating either a normal state (no abnormality) or an abnormal state (presence of abnormality) of the fluid pressure driven valve 10. In this case, the diagnostic information is classified into 2 values, for example, a value indicating that the fluid pressure-driven valve 10 is in a normal state is "0", a value indicating that the fluid pressure-driven valve 10 is in an abnormal state is "1", and the operator may input the corresponding value in a form associated with the input data using the work computer PC 1. In this case, information related to specific abnormal content is not necessarily required.

As indicated by the broken line in fig. 5, the information indicating the abnormality in the diagnostic information may include details of the abnormality, and the details of the abnormality may include, for example, malfunction of the main valve 11, malfunction of the air a circuit, malfunction of the air a supply pressure from the air supply source 14, malfunction of the supply voltage in the solenoid portion 3, deterioration and breakage of the solenoid coil 31, short circuit, lifetime of the solenoid portion 3, abnormal heat generation of the fluid pressure driven valve 10, malfunction of the iron core, and the like. In this case, the diagnostic information is classified into a plurality of values (3 or more), for example, a value indicating that the fluid pressure driven valve 10 is in a normal state is "0", a value indicating that an abnormality of the fluid pressure driven valve 10 belongs to a malfunction of the main valve 11 is "1", a value indicating that an abnormality of the fluid pressure driven valve 10 belongs to an air a circuit malfunction is "2", and the values are uniquely specified in advance based on the content of each abnormality in the same manner as described below, and then the corresponding values may be inputted in a form associated with the input data by the operator using the work computer PC 1. By setting such diagnostic information, it is possible to prepare a learning data set including not only the occurrence of an abnormality but also information including the specific content of the abnormality when the abnormality has occurred (corresponding to abnormality 1/abnormality 2/…/abnormality n shown in fig. 6). As the diagnostic information of the above-described embodiment, in machine learning described below, the present invention is used in the case of performing supervised learning (see fig. 5).

As the diagnostic information of the fluid pressure driven valve 10, information other than the above may be used. For example, as shown in fig. 6, other diagnostic information can be used that indicates only that the fluid pressure actuated valve 10 is not abnormal but normal. In this case, since the diagnostic information includes only information indicating that the fluid pressure-driven valve 10 is normal, the learning data set including the diagnostic information as the output data is necessarily only a data set composed of the input data and the output data when the fluid pressure-driven valve 10 is normal. Therefore, the output data of the learning data set in this case is always the same, and it is needless to say that the learning data set does not necessarily have the output data as data, as will be understood by those skilled in the art. As diagnostic information of this other mode, in machine learning described below, it is used in the case of performing unsupervised learning (see fig. 6).

Alternatively, the input data in the learning data set may include the total operation time of the fluid pressure driven valve 10, the operation time after the last power on of the fluid pressure driven valve 10, the number of operations of the main valve 11, the number of operations of the driving device 12, the number of operations of the solenoid portion 3, and the opening/closing time of the main valve 11. The total operation time of the fluid pressure driven valve 10 and the operation time after the last power on of the fluid pressure driven valve 10 can be obtained by the operation timer 47, the number of operations of the main valve 11, the number of operations of the driving device 12, and the number of operations of the solenoid portion 3 can be obtained by the operation counter 48, and the opening and closing time of the main valve 11 can be obtained by a timer or the like, not shown. Increasing the type of input data is generally useful for improving the accuracy of estimating a learned model obtained after machine learning, but using input data having a low degree of correlation with diagnostic information may rather prevent improvement of the accuracy of estimating a learned model. Therefore, the number and type of data used as input data should be appropriately selected in consideration of the state of the fluid pressure driven valve 10 to which the learned model is applied, and the like.

The learning data set storage unit 202 is a database for storing a plurality of data for constituting the learning data set acquired by the learning data set acquisition unit 201 as one learning data set in association with the relevant input data and output data. The specific structure of the database constituting the learning data set storage means can be appropriately adjusted. For example, in fig. 4, the learning data set storage unit 202 and the learning model storage unit 204 described later are shown as independent storage units for convenience of explanation, but they may be constituted by a single storage medium (database).

The learning unit 203 performs machine learning by using the plurality of learning data sets stored in the learning data set storage unit 202, thereby generating a learning model in which the correlation between the input data and the output data included in the plurality of learning data sets is learned. In the present embodiment, as will be described in detail later, a specific method of machine learning employs supervised learning using a neural network. However, the specific method of machine learning is not limited to this, and other learning methods may be employed as long as the correlation between the input and the output can be learned from the learning data set. For example, ensemble learning (random forest, lifting algorithm, etc.) can also be used.

The learned model storage unit 204 is a database for storing the learned model generated by the learning unit 203. The learned model stored in the learned model storage unit 24 is applied to an actual system via a communication line including the internet and a storage medium according to a request. The specific application of the learned model to the actual system (data processing system 300) will be described in detail later.

Next, a learning method in the learning unit 203 using the plurality of learning data sets obtained as described above will be described centering on supervised learning. Fig. 7 is a diagram showing an example of a neural network model for supervised learning implemented in a machine learning apparatus of an embodiment of the present invention. The neural network in the neural network model shown in fig. 7 is composed of l neurons (x 1 to x 1) located at the input layer, m neurons (y 11 to y1 m) located at the first intermediate layer, n neurons (y 21 to y2 n) located at the second intermediate layer, and o neurons (z 1 to zo) located at the output layer. The first intermediate layer and the second intermediate layer are also called hidden layers, and may have a plurality of hidden layers in addition to the first intermediate layer and the second intermediate layer as a neural network, or may use only the first intermediate layer as a hidden layer. In fig. 7, a neural network model in which a plurality of (o) output layers are set is illustrated, but for example, in the case where the above-described diagnostic information is determined by one value, that is, in the case where the number of training data to be described later is 1 (only t 1), the number of neurons of the output layers may be 1 (only z 1).

In addition, nodes connecting neurons between layers are covered between the input layer and the first intermediate layer, between the first intermediate layer and the second intermediate layer, and between the second intermediate layer and the output layer, and the nodes and weights wi (i is a natural number) are associated with each other.

The neural network in the neural network model of the present embodiment learns the correlation between the valve opening of the main valve 11, the pressure of the air a, the control parameters of the solenoid unit 3, and the temperature of the fluid pressure driven valve 10 and the diagnostic information of the fluid pressure driven valve 10 using the learning data set. Specifically, the valve opening of the main valve 11, the pressure of the air a, the control parameter of the solenoid portion 3, and the temperature of the fluid pressure-driven valve 10, which are state variables, are respectively associated with neurons of the input layer, and the value of a neuron located on the output layer is calculated using a general method of calculating the output value of a neural network, that is, using a method of calculating the value of a neuron on the output side in the form of the sum of the value of a neuron on the input side connected to a neuron on the output side and the number of the multiplication value of a weight wi associated with a node connecting the neuron on the output side and the neuron on the input side, for all neurons other than the neuron on the input layer. The information acquired as the state variable is input in what form when the state variable is input to the neurons of the input layer, and may be appropriately set in consideration of the accuracy of the generated learning model, and the like. Specifically, preprocessing can be performed on specific input data in order to adjust the number of neurons corresponding to each input data, or in order to adjust the number to a value that can correspond to a neuron.

Then, the calculated values of the o neurons z1 to zo located in the output layer, that is, 1 or more pieces of diagnostic information are compared with training data t1 to which 1 or more pieces of diagnostic information are also included as part of the learning data set in the present embodiment, respectively, an error is obtained, and the weights wi (back propagation) corresponding to the nodes are repeatedly adjusted so that the obtained error becomes smaller.

When the series of steps repeated a predetermined number of times or when the error is smaller than a predetermined condition such as an allowable value is satisfied, learning is ended, and (all weights wi corresponding to the nodes) of the neural network model is stored in the learned model storage unit 204 as a learned model.

(machine learning method)

In connection with the foregoing, the present invention provides a machine learning method. The machine learning method according to the present invention will be described below with reference to fig. 5 (learning stage), fig. 6 (learning stage), fig. 7, and fig. 8. Fig. 8 is a flowchart showing an example of a machine learning method of an embodiment of the present invention. In the machine learning method shown below, the description is based on the machine learning device 200, but the configuration to be used is not limited to the machine learning device 200. The machine learning method is implemented by using a computer, but various computers can be applied to the computer, and examples thereof include a computer device constituting the external device 15, the work computer PC1, or the microcontroller 70, a server device disposed on a network, and the like. The specific configuration of the computer may include, for example, at least an arithmetic device including a CPU, a GPU, and the like, a storage device including a volatile or nonvolatile memory, a communication device for communicating with a network or other devices, and a bus connecting these devices.

As the supervised learning of the machine learning method of the present embodiment, as a preparation in advance for starting the machine learning, first, a desired number of learning data sets (see fig. 5) are prepared, and the prepared plurality of learning data sets are stored in the learning data set storage unit 202 (step S11). The number of learning data sets prepared here can be set in consideration of the inference accuracy required for the finally obtained learning model.

The method of preparing the learning data set used in the supervised learning can employ several methods. For example, when an abnormality occurs in the specific fluid pressure-driven valve 10 or when an operator recognizes an abnormality symptom, various pieces of information at the time of steady operation of the fluid pressure-driven valve 10 at this time are acquired using the plurality of sensors 4 or the like, and the operator uses the work computer PC1 or the like to determine and input diagnostic information in a form correlated with these pieces of information, thereby preparing input data and output data (for example, the value of the output data in this case is "1") constituting the learning data set. Further, a method of preparing a desired number of learning data sets by repeating such a job can be employed. As a method for preparing the learning data set, various methods such as actively producing an abnormal state in the fluid pressure driven valve 10 to obtain the learning data set can be used in addition to such a method. However, since various information of the fluid pressure driven valves 10 tends to be peculiar to each fluid pressure driven valve 10, it is preferable that data constituting the learning data set is collected from only a predetermined one of the fluid pressure driven valves 10, and a learning model obtained by machine learning described later is applied to the fluid pressure driven valves 10. The learning data set includes not only a data set including input/output data when an abnormality occurs, but also a predetermined amount of learning data set including input data and output data (for example, the value of the output data at this time is "0") in a normal state of the fluid pressure actuated valve 10 when no abnormality occurs.

When step S11 is completed, a neural network model before learning is next prepared for starting learning by the learning unit 203 (S12). The neural network model before learning prepared here has, for example, a structure shown in fig. 7 as its structure, and the weights of the respective nodes are set to initial values. Then, from among the plurality of learning data sets stored in the learning data set storage unit 202, for example, one learning data set is randomly selected (step S13), and input data in the one learning data set is input to the input layer of the neural network model before learning (see fig. 7) that is prepared (step S14).

Here, the value of the output layer (see fig. 7) generated as a result of the above-described step S14 is a value generated by the neural network model before learning, and thus is a value different from the desired result, that is, a value showing information different from the correct diagnostic information in most cases. Therefore, next, machine learning is performed using the diagnostic information as training data in one of the learning data sets acquired in step S13 and the value of the output layer generated in step S13 (step S15). The machine learning performed here may be, for example, a process (back propagation) of comparing diagnostic information constituting training data with values of an output layer and adjusting weights that have been associated with respective nodes in a neural network model before learning to obtain a preferable output layer. The number and form of the values output to the output layer of the neural network model before learning are the same as those of the training data in the learning data set to be learned.

If the machine learning described here is specifically exemplified, it is assumed that the diagnostic information constituting the training data is composed of any value (binary classification) of "0" for the normal case and "1" for the abnormal case, and that the value of the output layer is a predetermined value of 0 to 1, specifically, for example, a value of "0.63" when the value of the output data in the one learning data set selected in step S13 is "1". Therefore, in step S15, if the same input data is input to the input layer, the weights corresponding to the nodes of the neural network model under learning are adjusted so that the value obtained by the neural network model under learning approaches "1".

When machine learning is performed in step S15, it is determined whether further machine learning is required to be continued, for example, based on the remaining number of learning data sets that are not learned and stored in the learning data set storage unit 202 (step S16). Then, when the machine learning is continued (no in step S16), the routine returns to step S13, and when the machine learning is ended (yes in step S16), the routine proceeds to step S17. In the case of continuing the machine learning, the steps S13 to S15 are performed a plurality of times on the neural network model under learning using the learning data set that has not been learned. The accuracy of the finally generated learned model generally increases in proportion to this number.

When the machine learning is completed (yes in step S16), the neural network generated by adjusting the weights associated with the nodes in a series of steps is stored in the learning model storage unit 204 as a learning model (step S17), and a series of learning processes is completed. The learned model stored here can be applied to the data processing system 300 described later.

In the learning process and the machine learning method of the machine learning apparatus described above, it is described that the accuracy is improved by repeatedly performing the machine learning process on one (before-learning) neural network model a plurality of times in order to generate one learned model, and a learned model sufficient to be applied to the data processing system 300 is obtained. However, the present invention is not limited to this acquisition method. For example, a learned model in which a predetermined number of machine learning operations have been performed may be stored in advance as one candidate in the plurality of learned model storage sections 204, a data set for validity determination may be input to the plurality of learned model groups, the (neuron values of the) output layer may be generated, and the accuracy of the values specified in the output layer may be compared and studied, thereby selecting an optimal learned model suitable for the data processing system 300. The validity determination data set may be constituted by the same data set as the learning data set used for learning, and may not be used for learning.

As described above, by applying the machine learning apparatus and the machine learning method of the present embodiment, the following learning-completed model can be obtained: from various data acquired by the plurality of sensors 4 provided at appropriate positions of the fluid pressure driven valve 10, diagnostic information showing whether an abnormality (including a post abnormality and an abnormality sign) has occurred can be accurately derived.

In the learning method and the machine learning method of the machine learning apparatus 200 described above, "supervised learning" is described. However, as a method of generating the learned model, other known "supervised learning" methods such as Convolutional Neural Network (CNN) and the like may be used, and "unsupervised learning" using a learning data set including the diagnostic information of the other modes described above, that is, information indicating that only the fluid pressure driven valve 10 is not abnormal but normal as diagnostic information constituting the output data may also be used (see fig. 6). By using "unsupervised learning", even in the case where only the information on the normal state of the fluid pressure-driven valve 10 can be obtained with respect to the diagnostic information in the output data in which the correspondence relation with the input data is established, the learned model can be obtained by learning the correlation relation indicating the characteristics of the normal state of the input data and the output data as shown in "learning stage" of fig. 6. In this case, in the case of reasoning in the data processing system 300 described later, the input data determined to be not in accordance with the feature of the normal state by the predetermined amount is regarded as not being in the normal state, that is, as being in the abnormal state, and thus, the reasoning of the diagnostic information can be performed. As a specific method of the "unsupervised learning", for example, a known method using an automatic encoder or the like, which is schematically shown in fig. 6, can be used, and a detailed description thereof will be omitted.

(data processing System)

Next, an application example of the learned model generated by the machine learning device 200 and the machine learning method will be described with reference to fig. 9. FIG. 9 is a schematic block diagram illustrating a data processing system in accordance with one embodiment of the present invention.

As the data processing system 300 of the present embodiment, a system mounted in the microcontroller 70 of the fluid pressure driven valve 10 is exemplified. It should be noted that, with respect to the data processing system 300, at least a part thereof may be applied to other devices, such as other devices connected to the external device 15 and the fluid pressure driven valve 10.

The data processing system 300 includes at least an input data acquisition unit 301, an inference unit 302, a learned model storage unit 303, and a reporting unit 304.

The input data acquisition unit 301 is an interface unit connected to the plurality of sensors 4 included in the fluid pressure driven valve 10, and configured to acquire data output from each sensor 4. The input data acquisition unit 301 acquires at least the valve opening of the main valve 11, the pressure of the air a, the control parameter of the solenoid portion 3, and the temperature of the fluid pressure driven valve 10. In the example shown in fig. 9, all the sensors 4 included in the fluid pressure driven valve 10 are connected so that all the input data usable in the reasoning described later can be acquired, but the sensor 4 to be connected to the input data acquisition means 301 can be appropriately selected according to the learning model or the like used in the reasoning means 302 described later. The inference result of the inference unit 302 is preferably stored in a storage unit, not shown, and the stored past inference result can be used as a learning data set for online learning, for example, for further improving the inference accuracy of the learned model in the learned model storage unit 303.

The inference unit 302 is configured to infer whether or not an abnormality has occurred in the fluid pressure-driven valve 10 based on various data of the fluid pressure-driven valve 10 acquired by the input data acquisition unit 301. For example, a learned model learned by the machine learning device 200 and the machine learning method is used for this reasoning, and the learned model is stored in the learned model storage unit 303 made up of an arbitrary storage medium. The inference unit 302 has not only a function of performing inference processing using a learned model, but also a preprocessing function of adjusting input data acquired by the input data acquisition unit 301 to a desired form as preprocessing of the inference processing and inputting the input data to the learned model, and a post-processing function of finally judging the presence or absence (anomaly (normal) or presence of anomaly (anomaly) by applying, for example, a predetermined threshold to an output value outputted from the learned model as post-processing of the inference processing.

As described above, the learned model storage unit 303 is a storage medium for storing the learned model used in the inference unit 302. The number of learned models stored in the learned model storage unit 303 is not limited to one. For example, a plurality of learning models having different amounts of input data or different learning methods (for example, supervised learning and unsupervised learning performed by the machine learning apparatus 200 or the like) can be stored, and these learning models can be selectively used.

The reporting unit 304 is configured to report the reasoning result of the reasoning unit 302 to an operator or the like. The specific reporting method may be various methods, for example, by transmitting the inference result to the external device 15 through the communication unit 8, displaying on a GUI of the external device 15, or the like, or by providing a light emitting member, a speaker, or the like in advance in the fluid pressure actuated valve 10 and operating them, it is possible to report the presence or absence of an abnormality to an operator or the like.

Next, a data processing procedure of the data processing system having the above configuration will be described with reference to fig. 5 (inference stage), fig. 6 (inference stage), and fig. 10. Fig. 10 is a flowchart showing an example of a data processing procedure of the data processing system 300 of one embodiment of the present invention.

When the power supply from the external power source 16 to the solenoid valve 1 of the fluid pressure driven valve 10 is started, and the abnormality diagnosis of the fluid pressure driven valve 10 is started, the input data acquisition unit 301 acquires various data indicating the states of the respective portions of the fluid pressure driven valve 10 acquired by the plurality of sensors 4 (step S21). At the time when the input data acquisition means 301 obtains desired input data (the valve opening of the main valve 11, the pressure of the air a, the control parameter of the solenoid portion 3, and the temperature of the fluid pressure driven valve 10 (see fig. 5 and 6)), the inference means 302 infers based on the input data (step S22). In this case, the learning model for reasoning is preferably determined in advance. In addition, in the case where the learning-completed model determined in advance requires prescribed time series data as its input data, for example, the inference in step S22 is performed after the necessary data amount is acquired in the input data acquisition unit 301.

Specifically, the inference unit 302 performs preprocessing on input data and inputs the input data to the learning model completion model, and performs post-processing on an output value from the learning model completion model, thereby judging whether or not an abnormality (including a post-abnormality and an abnormality precursor) has occurred as an inference result. In the post-processing of the supervised learning (see "inference phase" of fig. 5), the inference unit 302 compares the output value of the learning model completion model (a numerical value between 0 and 1 if it is a binary classification) with a predetermined threshold value, determines that there is an "abnormality (abnormality)" if the output value of the learning model completion model is equal to or greater than the predetermined threshold value, and determines that there is no abnormality (normal) "if it is less than the predetermined threshold value, and outputs the determination result as an inference result. In the post-processing of the unsupervised learning (see "inference stage" in fig. 6), the inference unit 302 obtains a difference (distance) between the output value (feature amount) of the learning model completion model and the feature amount based on the input data, determines that "abnormality (abnormality)" is present if the difference (distance) is equal to or greater than a predetermined threshold value, and determines that "abnormality (normal)" is absent if the difference (distance) is smaller than the predetermined threshold value, and outputs the determination result as an inference result.

Then, in step S22, the inference by the inference unit 302 is performed, and if the inference result indicates "no abnormality (normal)" (no in step S23), the routine returns to step S21, and a series of inferences is continued. On the other hand, as shown in fig. 5 and 6, when the result of the inference indicates "abnormality (abnormality)" as the result of the inference is "abnormality (abnormality)", that is, abnormality (including a post abnormality and an abnormality sign) occurs in the fluid pressure actuated valve 10, the result of the inference is reported to the operator or the like by the reporting unit 304 (step S24). After the abnormality is reported in step S24, the routine returns to step S21 to continue a series of reasoning. The fluid pressure driven valve 10 may be stopped at the stage of detecting the abnormality, depending on the use of the fluid pressure driven valve 10 and the content of the detected abnormality.

(inference means)

The present invention can be provided not only by the data processing system 300 described above, but also by an inference means for performing inference. In this case, the inference means comprises a memory and at least one processor, wherein the processor is capable of executing a series of processes. The series of processes includes: a process of acquiring input data including a valve opening of the main valve, a pressure of the driving fluid, a control parameter of the solenoid, and a temperature of the fluid pressure driving valve; and a process of inferring diagnostic information in the fluid pressure driven valve 10 when the input data is input. By providing the present invention in the manner of the above-described inference means, it is possible to easily adapt to various fluid pressure driven valves 10, as compared with the case of installing the data processing system 300. Those skilled in the art will of course understand that: in this case, when the inference means performs the process of inferring the diagnostic information, the inference means may be implemented by the inference means 302 of the data processing system using the learning model obtained by the machine learning means and the machine learning method in the present invention, as described in the present specification.

The present invention is not limited to the above-described embodiments, and can be variously modified and implemented within a scope not departing from the gist of the present invention. And these are included in the technical idea of the present invention.

Description of the reference numerals

1: an electromagnetic valve; 3: a solenoid section; 4: a sensor; 10: a fluid pressure actuated valve; 11: a main valve; 12: a (fluid pressure) drive; 14: an air supply source; 15: an external device; 26: an input-side flow path; 27: an output-side flow path; 28: an exhaust flow path; 30: a solenoid housing; 31: a solenoid coil; 32: a movable iron core; 40: a first pressure sensor; 41: a second pressure sensor; 42: a main valve opening sensor; 43: a voltage sensor; 44: a current/resistance sensor; 45: a temperature sensor; 46: a magnetic sensor; 47: running a timer (timer); 48: an action counter (counter); 70: a microcontroller; 100: piping; 200: a machine learning device; 201: a learning data set acquisition unit; 202: a learning data set storage unit; 203: a learning unit; 204: a model storage unit after learning; 300: a data processing system; 301: an input data acquisition unit; 302: an inference unit; 303: a model storage unit after learning; 304: a reporting unit; a: air (driving fluid); PC1: a work computer.

Claims

1. A machine learning device is applied to a fluid pressure driven valve including at least a main valve, a driving device that drives the main valve, and a solenoid valve that includes a solenoid portion that controls supply and discharge of driving fluid to and from the driving device, and includes:

a learning data set storage unit that stores a plurality of sets of learning data sets including input data including a valve opening of the main valve, a pressure of the driving fluid, a control parameter of the solenoid portion, and a temperature of the fluid pressure driving valve, and output data including diagnostic information of the fluid pressure driving valve in which a correspondence relation with the input data is established;

a learning unit that learns a learning model that infers a correlation between the input data and the output data by inputting a plurality of sets of the learning data sets; and

and a learning-completed model storage unit that stores the learning model learned by the learning unit.

2. The machine learning device of claim 1 wherein,

the diagnostic information is information indicating any one of a normal and an abnormal of the fluid pressure driven valve.

3. The machine learning device of claim 1 wherein,

the diagnostic information is information indicating only that the fluid pressure driven valve is not abnormal but normal.

4. The machine learning apparatus of any one of claims 1 to 3, wherein,

the fluid pressure driven valve is connected to a driving fluid supply source for supplying the driving fluid to the solenoid valve,

the pressure of the driving fluid constituting the input data is at least one of a solenoid valve input side pressure of the driving fluid supplied from the driving fluid supply source to the solenoid valve, a solenoid valve output side pressure of the driving fluid supplied from the solenoid valve to the driving device, and a pressure difference between the solenoid valve input side pressure and the solenoid valve output side pressure.

5. The machine learning apparatus of claim 1 wherein,

the control parameter of the solenoid unit constituting the input data is at least one of a supply voltage of the solenoid unit, a current value when the solenoid unit is energized, a resistance value of the solenoid unit, an operation time of the solenoid unit, and a magnetic field strength of the solenoid unit.

6. The machine learning apparatus of claim 1 wherein,

the fluid pressure driven valve further includes a timer for detecting an operation time of the fluid pressure driven valve,

the input data also includes a total operating time of the fluid pressure actuated valve obtained by the timer and an operating time after last power-on of the fluid pressure actuated valve.

7. The machine learning apparatus of claim 1 wherein,

the fluid pressure driven valve further includes a counter for detecting the number of operations of the main valve, the driving device, and the solenoid portion,

the input data further includes the number of operations of the main valve, the number of operations of the driving device, and the number of operations of the solenoid portion.

8. The machine learning apparatus of claim 1 wherein,

the input data also includes the opening and closing times of the main valve.

9. A data processing system for a fluid pressure driven valve provided with at least a main valve, a driving device that drives the main valve, and a solenoid valve that includes a solenoid portion that controls supply and discharge of driving fluid to and from the driving device, the data processing system comprising:

An input data acquisition unit that acquires input data including a valve opening of the main valve, a pressure of the driving fluid, a control parameter of the solenoid portion, and a temperature of the fluid pressure driving valve; and

an inference unit that inputs the input data acquired by the input data acquisition unit to a learned model generated by the machine learning device according to any one of claims 1 to 8, and infers diagnostic information of the fluid pressure driven valve.

10. An inference device for a fluid pressure driven valve comprising at least a main valve, a driving device for driving the main valve, and a solenoid valve including a solenoid portion for controlling supply and discharge of driving fluid to and from the driving device,

the inference means is provided with a memory and at least one processor,

the at least one processor is configured to perform the following:

acquiring input data including a valve opening of the main valve, a pressure of the driving fluid, a control parameter of the solenoid portion, and a temperature of the fluid pressure driving valve; and

when the input data is entered, diagnostic information of the fluid pressure actuated valve is inferred.

11. A machine learning method using a computer applied to a fluid pressure driven valve including at least a main valve, a driving device that drives the main valve, and an electromagnetic valve including a solenoid portion that controls supply and discharge of driving fluid to and from the driving device, the machine learning method comprising:

storing a plurality of sets of learning data including input data including a valve opening of the main valve, a pressure of the driving fluid, a control parameter of the solenoid portion, and a temperature of the fluid pressure driving valve, and output data including diagnostic information of the fluid pressure driving valve in which a correspondence relation with the input data is established;

learning a learning model for reasoning a correlation between the input data and the output data by inputting a plurality of sets of the learning data sets;

and storing the learned learning model.