Reciprocating pump and flow state monitoring method

Abstract

In order to monitor the flow of a handled liquid without using a detection device that comes into contact with the liquid in a reciprocating pump, a reciprocating pump (1) according to the present invention comprises: a diaphragm (6); a pump chamber (21a); a plunger (5); an electromagnetic actuator (41); an actuator control unit (311); an acquisition unit (312) that acquires, as a specific current value, at least a control current value when the plunger is located at a specific position among control current values of the electromagnetic actuator; a storage unit (35) that stores, as a reference current value, at least a normal control current value when the plunger is located at the specific position among normal control current values which are control current values during normal times; and a determination unit (313) that determines whether a suction state and / or a discharge state is normal or abnormal, on the basis of the specific current value and the reference current value. The specific position is a position corresponding to a point at which, when an abnormality exists, a characteristic change corresponding to the abnormality appears in the current value waveform of the control current value for one stroke of the plunger.

Description

Reciprocating pump and flow condition monitoring method

[0001] The present invention relates to a reciprocating pump and a method for monitoring flow conditions.

[0002] A reciprocating pump includes a diaphragm, a plunger that reciprocates the diaphragm, and a pump chamber that houses the diaphragm. The reciprocating pump reciprocates the diaphragm through the reciprocating motion of the plunger, thereby repeatedly suctioning and discharging the pumped liquid within the pump chamber, thereby delivering a constant amount of the pumped liquid. Known reciprocating pumps include a detection device (e.g., a pressure gauge, a flow meter, a flow sensor, etc.) that comes into contact with the pumped liquid and monitors the suction and discharge states (hereinafter referred to as "flow states") of the pumped liquid (see, for example, Patent Documents 1 and 2).

[0003] JP 2018-21503 A JP 2019-173634 A

[0004] In the reciprocating pump disclosed in Patent Document 1, a mounting hole communicating with the pump chamber is disposed in a housing defining the pump chamber, and a pressure gauge is attached to the mounting hole. This configuration requires the machining of the mounting hole and the installation of a seal between the mounting hole and the pressure gauge. In the reciprocating pump disclosed in Patent Document 2, a pressure sensor is attached to the discharge pipe. This configuration requires the design and installation of a discharge pipe for the pressure sensor, as well as the installation of wiring for the pressure sensor. Furthermore, when the operation of the reciprocating pump is stopped in response to an abnormality detection, wiring is required to transmit the pressure gauge's detection signal to the reciprocating pump's control system. As such, the manufacturing costs of these reciprocating pumps increase due to the processing costs, material costs, and installation costs.

[0005] Furthermore, when the handled liquid is corrosive, there is little freedom in selecting the detection equipment, and dedicated detection equipment tends to be expensive. Furthermore, flow meters and flow sensors can malfunction if air bubbles get mixed in the piping. As such, there are technical challenges in monitoring the flow state using detection equipment that comes into contact with the handled liquid.

[0006] An object of the present invention is to monitor the flow state of the pumped liquid in a reciprocating pump without using a detection device that comes into contact with the liquid.

[0007] A reciprocating pump according to one embodiment of the present invention is a reciprocating pump that sucks and discharges pumped fluid by reciprocating a diaphragm, and includes the diaphragm, a pump chamber in which the diaphragm is housed, a plunger that reciprocates the diaphragm, an electromagnetic actuator that reciprocates the plunger, an actuator control unit that performs closed-loop control of the operation of the electromagnetic actuator, and a control current value used to control the operation of the electromagnetic actuator that specifies at least the control current value when the plunger is located at a specific position within one stroke of the plunger. The device comprises an acquisition unit that acquires the normal control current value as a current value, a memory unit that stores, as a reference current value, the normal control current value when the plunger is positioned at the specific position, among the normal control current values ​​which are the control current values ​​when the suction state and discharge state of the handled fluid are normal, and a judgment unit that judges whether the suction state and / or the discharge state are normal or abnormal based on the specific current value and the reference current value, wherein the specific position is a position corresponding to a point at which, when the abnormality exists, a characteristic change corresponding to the abnormality appears in the current value waveform of the control current value for one stroke of the plunger.

[0008] A reciprocating pump according to one embodiment of the present invention is a reciprocating pump that sucks and discharges pumped fluid by reciprocating a diaphragm, and includes the diaphragm, a pump chamber in which the diaphragm is housed, a plunger that reciprocates the diaphragm, an electromagnetic actuator that reciprocates the plunger, an actuator control unit that performs closed-loop control of the operation of the electromagnetic actuator, a current value acquisition unit that acquires, as a specific current value, a control current value when the plunger is located at a specific position within one stroke of the plunger, among control current values ​​used to control the operation of the electromagnetic actuator, and a moving speed of the plunger. a speed acquisition unit that acquires the number of strokes of the plunger; a stroke number acquisition unit that acquires the number of strokes of the plunger; a memory unit that stores a trained learning model that has been machine-learned to output whether the suction state or the discharge state of the handled fluid is normal or abnormal when the specific current value, the movement speed, and the stroke number are input; and an estimation unit that inputs the specific current value, the movement speed, and the stroke number into the learning model and estimates whether the suction state or the discharge state is normal or abnormal, and the specific position is a position that corresponds to a point at which a characteristic change corresponding to the abnormality appears in the current value waveform of the control current value for one stroke of the plunger when the abnormality is present.

[0009] A reciprocating pump in one embodiment of the present invention is a reciprocating pump that sucks and discharges pumped fluid by reciprocating a diaphragm, and includes: the diaphragm; a pump chamber in which the diaphragm is accommodated; a plunger that reciprocates the diaphragm; an electromagnetic actuator that reciprocates the plunger; an actuator control unit that performs closed-loop control of the operation of the electromagnetic actuator; a current value acquisition unit that acquires, as a specific current value, a control current value when the plunger is positioned at a specific position within one stroke of the plunger, among control current values ​​used to control the operation of the electromagnetic actuator; and a calculation unit that calculates an amount of change between the specific current value that serves as a reference and the latest specific current value among a plurality of the specific current values ​​acquired at predetermined sampling intervals. a speed acquisition unit that acquires the movement speed of the plunger; a stroke number acquisition unit that acquires the number of strokes of the plunger; a memory unit that stores a trained learning model that has been machine-learned to output whether or not an abnormality will occur in the suction state or discharge state of the handled fluid after a predetermined future period has elapsed when the specific current value, the amount of change, the movement speed, and the number of strokes are input; and a prediction unit that inputs the specific current value, the amount of change, the movement speed, and the number of strokes into the learning model and predicts whether or not the abnormality will occur in the suction state or the discharge state within the period, wherein the specific position is a position that corresponds to a point at which a characteristic change corresponding to the abnormality appears in the current value waveform of the control current value for one stroke of the plunger when the abnormality occurs.

[0010] In one embodiment of the present invention, a flow state monitoring method is a method for monitoring a flow state of a pumped liquid, the method being performed by a reciprocating pump comprising: a diaphragm; a pump chamber in which the diaphragm is accommodated; a plunger that reciprocates the diaphragm; an electromagnetic actuator that reciprocates the plunger; an actuator control unit that performs closed-loop control of the operation of the electromagnetic actuator; and a memory unit that stores a reference current value among control current values ​​used to control the operation of the electromagnetic actuator, the method being for determining whether a suction state and / or a discharge state of the pumped liquid are normal or abnormal, wherein the reference current value is a normal control current value that is the control current value when the suction state and the discharge state are normal, The normal control current value is at least the normal control current value when the plunger is located at a specific position within one stroke of the plunger, and the specific position is a position corresponding to a point at which, when the abnormality is present, a characteristic change corresponding to the abnormality appears in the current value waveform of the control current value for one stroke of the plunger, and the flow state monitoring method includes a step in which the reciprocating pump acquires, as a specific current value, the control current value when the plunger is located at least at the specific position among the control current values, and a step in which the reciprocating pump determines whether the suction state and / or the discharge state is normal or abnormal based on the specific current value and the reference current value.

[0011] According to the present invention, the flow state of the pumped liquid can be monitored without using a detection device that comes into contact with the liquid in the reciprocating pump.

[0012] 1 is a schematic cross-sectional view showing an embodiment of a reciprocating pump according to the present invention; FIG. 2 is a functional block diagram of the reciprocating pump of FIG. 1; (a) is a schematic diagram showing an example of a current value waveform in a normal state, (b) is a schematic diagram showing an example of a current value waveform in an abnormal state, and (c) is a schematic diagram showing an example of a change in the current value waveform; FIG. 3 is a schematic diagram showing an example of information stored in a memory unit provided in the reciprocating pump of FIG. 1; (a) is a schematic diagram explaining operation modes of the reciprocating pump of FIG. 1, where (a) shows a standard mode, (b) shows a high viscosity mode, and (c) shows a low pulsation mode; FIG. 4 is a flowchart showing an example of operation of the reciprocating pump of FIG. 1; (b) is a schematic diagram showing an example of a control current value in an abnormal state; FIG. 5 is a functional block diagram showing a second embodiment of a reciprocating pump according to the present invention; (c) is a schematic diagram showing an example of information stored in a memory unit provided in the reciprocating pump of FIG. 9; and (d) is a flowchart showing an example of operation of the reciprocating pump of FIG. 9. Fig. 13 is a schematic diagram showing an example of information stored in a storage unit included in the reciprocating pump of Fig. 12. Fig. 14 is a flowchart showing an example of the operation of the reciprocating pump of Fig. 12.

[0013] Embodiments of a reciprocating pump (hereinafter referred to as "the pump") and a flow condition monitoring method (hereinafter referred to as "the method") according to the present invention will be described below. In the following description, reference will be made to the drawings as appropriate. In the drawings, identical members and elements are designated by the same reference numerals, and duplicate explanations will be omitted. Furthermore, the dimensional proportions of the elements may be exaggerated for the sake of convenience, and are not limited to the proportions shown in the drawings.

[0014] Reciprocating Pump (1) Configuration of Reciprocating Pump (1) Fig. 1 is a schematic cross-sectional view showing an embodiment of the pump. Fig. 2 is a functional block diagram of the pump.

[0015] The pump 1 is an electromagnetically driven diaphragm pump that sucks and discharges pumped liquid. The pump 1 includes a housing 2, a control device 3, a drive unit 4, a plunger 5, a diaphragm 6, an operation unit 7, a connection unit 8, a suction pipe L1, a discharge pipe L2, a suction valve V1, and a discharge valve V2. The basic configuration of the pump 1 is the same as that of a known reciprocating pump. Therefore, in the following description, a description of the basic configuration of the pump 1 will be omitted.

[0016] The housing 2 accommodates the control device 3, the drive unit 4, the plunger 5, and the diaphragm 6. The housing 2 includes a pump head unit 21 and a main body unit 22.

[0017] The pump head 21 houses the diaphragm 6 and sucks pumped fluid from the suction pipe L1 and discharges the pumped fluid to the discharge pipe L2. The pump head 21 includes a pump chamber 21a, a suction flow path 21b, and a discharge flow path 21c. The pump chamber 21a houses the diaphragm 6. The suction flow path 21b is a flow path that communicates with the pump chamber 21a and the suction pipe L1. The discharge flow path 21c is a flow path that communicates with the pump chamber 21a and the discharge pipe L2.

[0018] The main body 22 houses the control device 3, the drive unit 4, and the plunger 5. The main body 22 includes an electric chamber 22a and a machine chamber 22b. The electric chamber 22a houses the control device 3. The electric chamber 22a is disposed above the machine chamber 22b. The machine chamber 22b houses the drive unit 4 and the plunger 5. In the following description, the direction in which the pump chamber 21a is disposed relative to the machine chamber 22b is referred to as the "forward direction," and the opposite direction is referred to as the "rearward direction." In other words, the pump chamber 21a is disposed in front of the machine chamber 22b.

[0019] The control device 3 controls the overall operation of the pump 1. The control device 3 includes, for example, a processor such as a central processing unit (CPU) 31, a volatile memory such as a random access memory (RAM) 32 that functions as a work area for the CPU 31, a non-volatile memory such as a read only memory (ROM) 33 that stores various information such as a state determination program, a driver circuit 34 for the motor device 41 (described below), and a storage unit 35. The CPU 31 functions as an actuator control unit 311, an acquisition unit 312, a determination unit 313, a setting unit 314, a change detection unit 315, and a specific operation control unit 316. That is, the control device 3 also includes the actuator control unit 311, the acquisition unit 312, the determination unit 313, the setting unit 314, the change detection unit 315, and the specific operation control unit 316.

[0020] A state determination program runs in the control device 3, and the state determination program cooperates with the hardware resources of the pump 1 to realize the method described below. By causing a processor (CPU 31) included in the control device 3 to execute the state determination program, the state determination program can cause the processor to function as an actuator control unit 311, an acquisition unit 312, a determination unit 313, a setting unit 314, a change detection unit 315, and a specific operation control unit 316, thereby causing the processor to execute the method. Furthermore, by causing a computer to execute the state determination program, the state determination program can cause the computer to function as the control device 3.

[0021] In the present invention, the condition determination program may be stored in the storage unit 35. Alternatively, the condition determination program may be stored in an installable file format or an executable file format on a non-transitory storage medium (e.g., a CD (Compact Disc), a DVD (Digital Versatile Disc), a USB (Universal Serial Bus) memory, etc.) and provided to the pump 1 via a dedicated read-out medium.

[0022] The actuator control section 311 controls the operation of the motor device 41 via the driver circuit 34. The specific operation of the actuator control section 311 will be described later.

[0023] The acquisition unit 312 acquires information (e.g., a control current value) necessary for executing the present method. Specific operations of the acquisition unit 312 will be described later. The acquisition unit 312 also functions as a reference acquisition unit in the present invention. In other words, the acquisition unit 312 is an example of a reference acquisition unit in the present invention.

[0024] The "control current value" is the current value of a current (hereinafter referred to as "control current") used to control the operation of the motor 411, which will be described later. The control current value is supplied from the driver circuit 34 to the motor 411 based on the control of the actuator control unit 311. In the present invention, the control current value includes a normal control current value and a specific current value.

[0025] The "normal control current value" is the control current value when the suction and discharge states of the handled liquid in this pump 1, i.e., the flow state of the handled liquid in this pump 1 (hereinafter collectively referred to as the "flow state"), is normal (hereinafter simply referred to as "normal").

[0026] The "specific current value" is, among the control current values, the control current value when the plunger 5 is at least located at a specific position within one stroke of the plunger 5. In other words, the specific current value includes the control current value that was supplied to the motor 411 when the plunger 5 was located at the specific position. In this embodiment, the specific current value is the control current value that was supplied to the motor 411 when the plunger 5 was located at the specific position.

[0027] The "specific position" is a position of the plunger 5 within one stroke of the plunger 5 at which, when there is an abnormality in the flow state in this pump 1 (hereinafter simply referred to as "abnormality"), a characteristic change corresponding to the abnormality appears in the current value waveform of the control current value for one stroke of the plunger 5. The specific position will be described in detail later.

[0028] The "abnormal flow state" is, for example, an abnormality caused by cavitation occurring in the pump chamber 21a during the suction stroke, a gas lock occurring in the pump chamber 21a during the discharge stroke, leakage of the pumped liquid, or wear or damage to the diaphragm 6. For example, if cavitation occurs in the pumped liquid in the pump chamber 21a due to negative pressure during the suction stroke, normal suction of the pumped liquid is hindered, resulting in an abnormal suction state. Also, for example, if a gas lock occurs due to air in the pump chamber 21a, normal discharge of the pumped liquid is hindered, resulting in an abnormal discharge state. When such an abnormality occurs, a larger or smaller torque is required for the reciprocating motion of the plunger 5, and the control current value increases or decreases.

[0029] The determining unit 313 determines whether the suction state and / or discharge state (flow state) is normal or abnormal based on the specific current value and the reference current value. The specific operation of the determining unit 313 will be described later.

[0030] The "reference current value" is a current value that serves as a reference for determination by the determination unit 313. The reference current value is, among the normal control current values, the normal control current value when at least the plunger 5 is located at a specific position. In other words, the reference current value includes the normal control current value that was supplied to the motor 411 when the plunger 5 was located at a specific position. In this embodiment, the reference current value is, among the normal control current values, the normal control current value that was supplied to the motor 411 when the plunger 5 was located at a specific position.

[0031] FIG. 3( a ) is a schematic diagram showing an example of a normal current waveform, FIG. 3( b ) is a schematic diagram showing an example of an abnormal current waveform, and FIG. 3( c ) is a schematic diagram showing an example of a change in the current waveform. This diagram shows only the outer shell of the current waveform. The vertical axis of FIG. 3 represents the control current value. The horizontal axis of FIG. 3 represents the position of the plunger 5 in angle within one stroke of the plunger 5. In FIG. 3( c ), one stroke of the plunger 5 is represented as one cycle of 360°, the angle when the plunger 5 is at top dead center is represented as 0° / 360°, and the angle when the plunger 5 is at bottom dead center is represented as 180°. In FIG. 3( c ), the normal current waveform and the abnormal current waveform are superimposed, and the area where there is a difference between the two (the difference area: the area indicated by the dashed arrow) is highlighted, and the boundary of the difference area within the stroke is indicated by a dashed line.

[0032] As shown in FIG. 3( a), during the discharge stroke, a high torque is required to push out the pumped fluid drawn into the pump chamber 21a. Therefore, the control current value increases significantly during the discharge stroke. On the other hand, during the suction stroke, such high torque is not required, and the control current value is steadily reduced. As shown in FIG. 3( b), when an abnormality occurs in the suction or discharge state of the pumped fluid, the control current value increases or decreases in response to the abnormality, causing the current waveform to deform from its normal state. This increase or decrease in the control current value (deformation of the current waveform) is a unique increase or decrease (deformation) corresponding to the abnormality. As a result, as shown by the shading in FIG. 3( c), a characteristic change (increase or decrease, deformation) corresponding to the abnormality appears in the differential region between the normal current waveform and the abnormal current waveform. In this embodiment, the control device 3 (determination unit 313) detects an abnormal flow state based on the difference between the normal control current value (reference current value) and the abnormal control current value (specific current value) at a specific position, among the changes in the control current waveform. Here, the specific position is, for example, the position in the differential region where the difference value is the largest, or a position in the vicinity thereof (for example, "θ １ " "θ ２ ").

[0033] 1 and 2. The setting unit 314 sets the reference current value based on the normal control current value. The specific operation of the setting unit 314 will be described later.

[0034] The change detection unit 315 detects changes in the operating conditions (for example, the operating mode, forward speed, backward speed, number of strokes, etc., which will be described later) of the pump 1. The specific operation of the change detection unit 315 will be described later.

[0035] When the determination unit 313 determines that "there is an abnormality in the flow state," the specific operation control unit 316 executes control necessary for a specific operation to resolve the abnormality. The specific operation of the specific operation control unit 316 will be described later.

[0036] The "specific operation" is an operation for resolving an abnormality in the flow state in the pump 1. For example, when the abnormality is a discharge abnormality caused by a gas lock, the specific operation is an operation for increasing the number of strokes of the plunger 5 (high-speed operation). The specific operation includes an operation for prompting the user to resolve the abnormality, such as an operation for turning on a warning light or an operation for sounding an alarm.

[0037] The driver circuit 34 is a circuit that supplies a control current (applies a control voltage) to the motor 411 based on the control of the actuator control unit 311 (e.g., an operation instruction for the motor 411) to control the operation of the motor 411. The driver circuit 34 includes a current detection circuit 341 that detects the value of the control current supplied to the motor 411. The control current value detected by the current detection circuit 341 is transmitted to the acquisition unit 312, for example, and acquired by the acquisition unit 312 in association with information (angle information) indicating the position angle of the motor 411 based on the operation instruction.

[0038] The storage unit 35 stores information (e.g., a reference current value, a control current value, etc.) necessary for the operation of the pump 1. The storage unit 35 is, for example, a non-volatile memory such as an EEPROM (Electrically Erasable Programmable Read-Only Memory) or a flash memory.

[0039] 4 is a schematic diagram showing an example of information (reference current values) stored in the storage unit 35. The figure shows, for example, two reference current values ​​"I １ " "I ２ In this embodiment, the reference current value "I １ " is a reference current value corresponding to an abnormality in the suction process, and the reference current value "I ２ " is a reference current value corresponding to an abnormality in the discharge process. １ " and plunger position (specific position) "θ １ " are stored in the storage unit 35 in association with each other. Here, the plunger position is calculated based on, for example, the angle information and the reduction ratio between the drive gear 42 and the driven gear 43, which will be described later.

[0040] In the present invention, the RAM 32 may function as the storage unit 35 .

[0041] 1 and 2 will be referred to in the following description. The drive unit 4 reciprocates the plunger 5. The drive unit 4 includes a motor device 41, a drive gear 42, a driven gear 43, a crankshaft 44, and two bearings 45 and 46.

[0042] The motor device 41 generates rotational power to reciprocate the plunger 5. The motor device 41 is a known stepping motor and is an example of an electromagnetic actuator according to the present invention. The motor device 41 includes a motor 411 and an encoder 413. The encoder 413 acquires the position angle of the rotation shaft 412 of the motor 411. A signal (angle information) indicating the position angle acquired by the encoder 413 is transmitted to the control device 3, and the control device 3 (actuator control unit 311) controls the control current value supplied to the motor 411 via the driver circuit 34. In other words, the operation of the motor 411 is controlled by the control device 3 in a closed loop. Therefore, the motor 411 is controlled so as to always rotate accurately, and the power consumption (control current value) of the motor 411 is appropriately controlled according to the load on the motor 411. The motor device 41 is supported by the housing 2 with the rotation shaft 412 oriented vertically.

[0043] The drive gear 42 and the driven gear 43 reduce the rotational power from the motor device 41 at a predetermined reduction ratio and transmit the reduced power to the crankshaft 44. The drive gear 42 is attached to a rotary shaft 412, and the driven gear 43 is attached to the crankshaft 44 so as to mesh with the drive gear 42.

[0044] The crankshaft 44 converts the rotational power from the motor device 41 into reciprocating power in the front-rear direction. The crankshaft 44 is rotatably supported by bearings 45 and 46 so that its axial direction (longitudinal direction) is aligned with the up-down direction.

[0045] In the present invention, the drive unit 4 may include a rotation shaft and an eccentric cam instead of the crankshaft 44. In this case, the plunger 5 is reciprocated by the eccentric cam and does not need to include a bearing 51, which will be described later.

[0046] The bearing 45 rotatably supports the upper end of the crankshaft 44, and the bearing 46 rotatably supports the lower end of the crankshaft 44. The bearings 45 and 46 are supported by the housing 2.

[0047] The plunger 5 reciprocates in the front-to-rear direction in response to the rotation of the crankshaft 44, thereby reciprocating the diaphragm 6. The plunger 5 is housed in the machine chamber 22b so as to extend along the front-to-rear direction, and is rotatably attached to the crankshaft 44 via a bearing 51 provided in the plunger 5. The plunger 5 is disposed behind the diaphragm 6 (on the rear surface side of the diaphragm 6).

[0048] The operation of the plunger 5 is controlled by the control device 3 via the motor device 41. That is, when the control device 3 rotates the motor 411, the crankshaft 44 rotates, and the plunger 5 reciprocates in the forward and backward directions. Here, when the plunger 5 is positioned at its rearmost position, it corresponds to the bottom dead center of the plunger 5 (the 180° position in FIG. 3 ). When the plunger 5 moves forward from the bottom dead center, the diaphragm 6 is pushed forward, and the pumped fluid in the pump chamber 21a is discharged to the discharge flow path 21c. At this time, the pump 1 is performing a discharge stroke. When the plunger 5 is positioned at its forwardmost position, it corresponds to the top dead center of the plunger 5 (the 0° / 360° position in FIG. 3 ). When the plunger 5 moves backward from the top dead center, the diaphragm 6 is pulled back, and the pumped fluid is drawn from the suction flow path 21b into the pump chamber 21a. At this time, the pump 1 is performing a suction stroke. That is, when one reciprocating movement (one stroke) of the plunger 5 is defined as one cycle, the pump 1 periodically repeats a discharge stroke and a suction stroke. The forward direction of the plunger 5 is the forward direction in the present invention.

[0049] In this embodiment, the pump 1 operates in one of the standard mode, high-viscosity mode, and low-pulsation mode based on the magnitude relationship between the forward speed and the backward speed of the plunger 5. In other words, the operation modes of the pump 1 include the standard mode, high-viscosity mode, and low-pulsation mode.

[0050] Figure 5 is a schematic diagram illustrating the operating modes of the pump 1, with (a) showing the standard mode, (b) showing the high-viscosity mode, and (c) showing the low-pulsation mode. The diagram shows the elapsed time for each of the suction stroke and discharge stroke during one reciprocation (one cycle) of the plunger 5. The vertical axis of the diagram represents the stroke amount of the plunger 5, and the horizontal axis represents the elapsed time. In the following explanation of the operating modes, reference will also be made to Figures 1 and 2 as appropriate.

[0051] 5A, the "standard mode" is an operating mode in which the time "To" for the plunger 5 to move from the bottom dead center to the top dead center (the discharge stroke time) is set to be the same as the time "Tr" for the plunger 5 to move from the top dead center to the bottom dead center (the suction stroke time). That is, in the standard mode, the forward speed "Vo" is set to be the same as the return speed "Vr" of the plunger 5.

[0052] As shown in FIG. 5B, the "high-viscosity mode" is an operating mode in which the discharge stroke time "To" is set to be shorter than the suction stroke time "Tr." That is, in the high-viscosity mode, the forward speed "Vo" is set to be faster than the return speed "Vr" of the plunger 5. Generally, when a reciprocating pump sucks a highly viscous liquid, if the return speed is fast, the required amount of liquid is not sucked into the pump chamber 21a, resulting in the formation of a mass of air bubbles in the pump chamber 21a. In the present pump 1 operating in the high-viscosity mode, the pumped liquid is slowly sucked in, thereby suppressing the generation of air bubbles in the pump chamber 21a.

[0053] As shown in FIG. 5( c), the “low pulsation mode” is an operating mode in which the discharge stroke time “To” is set to be longer than the suction stroke time “Tr.” That is, in the low pulsation mode, the forward speed “Vo” is set to be slower than the return speed “Vr” of the plunger 5. Generally, in a reciprocating pump, the discharge stroke and the suction stroke are alternately repeated, resulting in intermittent pumping of the pumped fluid, which causes pulsation. Here, the forward speed “Vo” increases as the plunger 5 moves away from the bottom dead center and decreases as the plunger 5 approaches the top dead center. Therefore, the flow velocity of the pumped fluid flowing through the discharge pipe L2 also accelerates and decelerates. Pulsation increases as the acceleration and deceleration of the flow velocity of the pumped fluid increases, and is suppressed as the acceleration and deceleration decrease. In the pump 1 operating in the low pulsation mode, the pumped fluid is discharged slowly, which slowly accelerates and decelerates the flow velocity of the pumped fluid, thereby suppressing pulsation.

[0054] Here, the control current value increases or decreases depending on the output (load) of the motor 411. That is, as the discharge pressure value increases, the output (load) of the motor 411 increases, and the control current value increases. On the other hand, as the discharge pressure value decreases, the output (load) of the motor 411 decreases, and the control current value decreases. The control current value also increases as the stroke count (i.e., the rotational speed of the motor 411) increases and decreases as the stroke count decreases. Furthermore, as the stroke count decreases, the torque of the motor 411 increases and decreases. During the discharge stroke, the motor 411 requires a relatively high torque to discharge the pumped fluid. Therefore, as the stroke count increases, the control current value tends to increase in order to obtain the required torque, and as the stroke count decreases, the control current value tends to decrease. That is, even for the same discharge pressure value, the control current value required to discharge the pumped fluid at that discharge pressure value increases as the stroke count increases and decreases as the stroke count decreases. On the other hand, during the suction stroke, high torque like that required during the discharge stroke is not required, and the control current value tends to be steadily smaller. This tendency is also apparent in FIG. 7, which will be described later.

[0055] Here, when the rotational speed of the motor 411 increases or decreases during the discharge stroke, the forward speed during the discharge stroke also increases or decreases. Therefore, during the discharge stroke, the control current value increases as the forward speed increases and decreases as the forward speed decreases. Therefore, when the forward speed changes, the control current value corresponding to the discharge pressure value also changes. As described above, in this embodiment, the operating modes of the pump 1 include the standard mode, the high-viscosity mode, and the low-pulsation mode, and each of these modes has a different forward speed. Therefore, even if the position of the plunger 5 is the same, the control current value differs for each operating mode.

[0056] 1 and 2 will be referred to in the following description. The diaphragm 6 draws pumped fluid into the pump chamber 21a and discharges the pumped fluid from the pump chamber 21a by reciprocating. The diaphragm 6 is housed in the pump chamber 21a. The outer edge of the diaphragm 6 is fixed to the housing 2 with both surfaces of the diaphragm 6 facing forward and backward.

[0057] The operation unit 7 is, for example, an operation panel that accepts operations by a user. The operation unit 7 is configured to allow the user to input and select operating conditions of the pump 1 (such as an operating mode, forward speed, backward speed, and number of strokes).

[0058] The connection unit 8 is, for example, a terminal block to which a cable (not shown, same below) is connected to an external device (not shown, same below) of the pump 1. The connection unit 8 is, for example, disposed on the rear surface of the housing 2. The connection unit 8 includes, for example, an input terminal 8a and an output terminal 8b.

[0059] The suction pipe L1 is a path for pumped liquid to be sucked into the pump chamber 21a. The suction pipe L1 is connected to the suction flow path 21b. The discharge pipe L2 is a path for pumped liquid to be discharged from the pump chamber 21a. The discharge pipe L2 is connected to the discharge flow path 21c.

[0060] The suction valve V1 is connected to the suction pipe L1 and is a one-way valve that allows the pumped fluid to flow only to the suction flow path 21b. The discharge valve V2 is connected to the discharge pipe L2 and is a one-way valve that allows the pumped fluid to flow only to the discharge pipe L2.

[0061] Operation of the Reciprocating Pump (1) Next, the operation (method) of the pump 1 will be described below. In the following description, reference will be made to Figures 1 to 3 as appropriate.

[0062] Fig. 6 is a flowchart showing an example of the operation of the pump 1. In the following description of the operation, Fig. 6 will be referred to as appropriate.

[0063] First, the change detection unit 315 determines whether or not there has been a change in the operating conditions (e.g., operating mode, number of strokes, discharge pressure, etc.) of the pump 1 (ST101: change determination step). The operating conditions are changed, for example, by the user operating the operation unit 7. The change detection unit 315 detects the change in the operating conditions based on, for example, a signal indicating a change in the operating conditions (change signal) from the operation unit 7, and determines that "the operating conditions have been changed."

[0064] When the change detection unit 315 determines that "the operating conditions have been changed" ("Y" in ST101), the acquisition unit 312 acquires normal control current values ​​(ST102: control current value acquisition step). Specifically, the acquisition unit 312 acquires, for example, the control current values ​​transmitted from the current detection circuit 341 of the driver circuit 34 as normal control current values ​​for one stroke. At this time, the acquisition unit 312 also acquires, for example, angle information corresponding to each control current value in association with the control current value. The acquired control current values ​​and angle information are, for example, associated with each other and stored in the storage unit 35. On the other hand, when the change detection unit 315 determines that "the operating conditions have not been changed" ("N" in ST101), the operation proceeds to processing (ST104).

[0065] In the process (ST101) immediately after the start of operation of the pump 1, the change detection unit 315 determines that "the operating conditions have been changed," and the processes (ST102, ST103) are executed.

[0066] In addition, in the process (ST102), the acquisition unit 321 may acquire control current values ​​for several strokes and set the average of the control current values ​​as the normal control current value.

[0067] 7 is a schematic diagram showing an example of a normal control current value. This diagram shows, as an example of a normal control current value, the control current value corresponding to one stroke (suction stroke and discharge stroke) of the plunger 5 among the current value waveforms. In this diagram, the pump 1 is operating at a discharge pressure of 0.5 MPa. The two-dot chain line in this diagram indicates a specific position, and the dashed line in this diagram indicates a reference current value.

[0068] Next, the setting unit 314 sets a reference current value based on the normal control current value (ST103: reference current value setting step). Specifically, the setting unit 314 identifies the control current value supplied to the motor 411 when the plunger 5 was located at the specific position, for example, based on specific position information indicating the specific position and angle information stored in the storage unit 35, and sets the identified control current value as the reference current value. The reference current value is stored in, for example, the storage unit 35. At this time, if a previously set reference current value is stored in the storage unit 35, the storage unit 35 updates the reference current value with the newly set control current value and stores it.

[0069] Here, when the abnormalities to be judged in the present invention are set in advance, the specific position information is, for example, written directly into the status judgment program or is stored in advance together with the status judgment program in the ROM 33. Furthermore, when the abnormalities are selectable by the user, the specific position information corresponding to each abnormality is, for example, stored in advance in the ROM 33.

[0070] In the present invention, the specific position information may be input by the user operating the operation unit 7 and stored in the storage unit 35, for example.

[0071] Next, the acquisition unit 312 acquires the specific current value (ST104: specific current value acquisition step). Specifically, the acquisition unit 312 acquires, as the specific current value, the most recent control current value supplied to the motor 411 when the plunger 5 was located at the specific position, for example, based on the specific position information and angle information. The acquired specific current value is stored in, for example, the storage unit 35.

[0072] Next, the determination unit 313 calculates a difference between the reference current value and the specific current value (ST105: difference value calculation step), and compares the difference with a predetermined threshold value to determine whether or not an abnormality exists at the specific position (ST106: state determination step). The predetermined threshold value is, for example, set in advance for each corresponding abnormality and stored in the storage unit 35.

[0073] In the present invention, the predetermined threshold value may be stored as a fixed value, or may be stored as a predetermined ratio by which the difference value is multiplied. In the latter case, the threshold value increases or decreases as the control current value increases or decreases depending on the operating conditions.

[0074] When the difference value is less than the predetermined threshold value ("Y" in ST106), the determination unit 313 determines that "the flow state is normal (no abnormality)" (ST107).

[0075] Next, the determination unit 313 determines whether a specific action, which will be described later, is being executed (ST108). When the determination unit 313 determines that "a specific action is being executed" ("Y" in ST108), the specific action control unit 316 ends the specific action (ST109: specific action end step), and the operation returns to the process (ST101). On the other hand, when the determination unit 313 determines that "a specific action is not being executed" ("N" in ST108), the operation returns to the process (ST101).

[0076] On the other hand, when the difference value is equal to or greater than the predetermined threshold value ("N" in ST106), the determination unit 313 determines that "the flow state is abnormal (there is an abnormality)" (ST110).

[0077] FIG. 8 is a schematic diagram showing an example of a control current value during an abnormality. This diagram shows, as an example of a control current value during an abnormality, the control current value corresponding to one stroke of the plunger 5 (the suction stroke and the discharge stroke) of the current waveform. In this diagram, the pump 1 operates at a discharge pressure of 0.5 MPa. The two-dot chain line in this diagram indicates the specific position, and the dashed line in this diagram indicates the specific current value. This diagram also shows a current waveform during an abnormality (gas lock) as an example of a control current value during an abnormality. As shown in FIGS. 7 and 8 , when a gas lock occurs, the current waveform during the discharge stroke is significantly deformed. There is a difference between the reference current value at the specific position and the specific current value, and the difference value in the difference region (the region indicated by a circle in FIG. 8 ) is large.

[0078] Next, the determination unit 313 determines whether a specific action, which will be described later, is being executed (ST111). When the determination unit 313 determines that "a specific action is not being executed" ("N" in ST111), the specific action control unit 316 executes the control necessary for the specific action (ST112: specific action execution step). As a result, the specific action is started. Next, this operation returns to the process (ST104).

[0079] On the other hand, when the determination unit 313 determines that "a specific action is being performed" ("Y" in ST111), the determination unit 313 determines whether the number of times the specific action has been performed has reached a predetermined number (ST113). Here, the "predetermined number of times" is set in advance for each specific action and is stored in the ROM 33 together with the state determination program, for example.

[0080] When the determination unit 313 determines that "the number of executions is less than the predetermined number" ("N" in ST113), this operation returns to the process (ST104). On the other hand, when the determination unit 313 determines that "the number of executions has reached the predetermined number" ("Y" in ST113), the abnormality has not been resolved even after the specific operation has been performed the predetermined number of times, so the control device 3 (actuator control unit 311 and specific operation control unit 316) stops this operation (liquid delivery) (ST114).

[0081] In this way, the pump 1 stores the control current value (reference current value) for a specific position during normal operation and determines whether the flow state is normal or abnormal based on the difference between the control current value (specific current value) and the control current value for the specific position during operation. In other words, the pump 1 monitors the flow state based on the control current value. Therefore, the pump 1 can monitor the flow state in the pump 1 without using an abnormality detection device (e.g., a pressure gauge, a flow meter, etc.) that comes into contact with the pumped liquid. As a result, the pump 1 does not require the costs associated with conventional pumps equipped with pressure gauges, such as the machining required to install a pressure gauge, the design and installation of seals, the discharge piping, and the installation of wiring for the pressure gauge. Furthermore, the pump 1 can easily determine (monitor) the flow state in the pump 1 even when the pumped liquid is corrosive.

[0082] In the process (ST106), the determination unit 313 may determine that "there is no abnormality" when the difference value is equal to or less than a predetermined threshold, and may determine that "there is an abnormality" when the difference value is greater than the predetermined threshold.

[0083] Summary (1) According to the embodiment described above, the pump 1 includes a diaphragm 6, a plunger 5, a pump chamber 21a, a motor device 41, an actuator control unit 311, an acquisition unit 312, a memory unit 35, and a determination unit 313. The memory unit 35 stores, as a reference current value, a normal control current value, among the normal control current values, when at least the plunger 5 is positioned at a specific position. The actuator control unit 311 performs closed-loop control of the operation of the motor device 41. The acquisition unit 312 acquires, as a specific current value, a control current value, among the control current values, when at least the plunger 5 is positioned at a specific position. The determination unit 313 determines whether the flow state is normal or abnormal based on the specific current value and the reference current value. The specific position corresponds to a point where, when there is an abnormality in the flow state, a characteristic change corresponding to the abnormality appears in the current value waveform of the control current value for one stroke of the plunger 5. According to this configuration, the abnormality in the flow state appears in the control current value at the specific position. Therefore, the pump 1 electrically acquires a specific current value and a reference current value and can determine whether or not there is an abnormality in the flow state based on the difference between the specific current value and the reference current value. In other words, the pump 1 can monitor the flow state based on the control current value. Therefore, the pump 1 can monitor the flow state in the pump 1 without using an abnormality detection device (e.g., a pressure gauge, a flow meter, etc.) that comes into contact with the pumped liquid. As a result, the pump 1 does not require the costs associated with conventional pumps equipped with pressure gauges, such as the machining required to install a pressure gauge, the seal members, the discharge piping design and installation, and the wiring installation required for the pressure gauge. Furthermore, the pump 1 can easily monitor (determine) the flow state in the pump 1 even when the pumped liquid is corrosive.

[0084] Furthermore, according to the embodiment described above, the pump 1 includes a setting unit 314 that sets a reference current value based on the normal control current value acquired by the acquisition unit 312. With this configuration, the pump 1 can set an appropriate reference current value that corresponds to an abnormality in the environment in which the pump 1 is used.

[0085] Furthermore, according to the embodiment described above, the pump 1 includes a change detection unit 315 that detects a change in the operating conditions of the pump 1. When the change detection unit 315 detects a change in the operating conditions, the acquisition unit 312 acquires a new normal control current value. The storage unit 35 updates and stores the reference current value based on the acquired new normal control current value. With this configuration, the pump 1 can automatically update the reference current value in response to changes in the operating conditions in the usage environment of the pump 1. As a result, the pump 1 can appropriately set the reference current value in response to the control current value, which increases or decreases depending on the operating conditions.

[0086] Furthermore, according to the embodiment described above, the pump 1 includes a specific operation control unit 316 that, when the determination unit 313 determines that an abnormality has occurred, executes control necessary for a specific operation to resolve the abnormality. With this configuration, the pump 1 can automatically execute a specific operation to resolve the abnormality even if an abnormality occurs.

[0087] In the present invention, the reference current value is not limited to the normal control current value as long as it includes at least the normal control current value when the plunger 5 is positioned at the specific position. That is, for example, the reference current value may be the normal control current value constituting the current value waveform within the differential region, the normal control current value constituting the current value waveform within a half-stroke region corresponding to the suction stroke or discharge stroke, or the normal control current value constituting the current value waveform within a full-stroke region. The full-stroke region, half-stroke region, and differential region are examples of the predetermined range in the present invention. In this case, the reference current value may be all of the normal control current values ​​constituting the current value waveform, or may be their average value. Similarly to the normal control current value, the specific current value may also be set to match the reference current value as long as it includes the control current value when the plunger 5 is positioned at the specific position. That is, for example, the acquisition unit 312 may acquire a current value waveform belonging to the same region as the reference current value as the specific current value. In this case, the specific current value may be, for example, all of the control current values ​​constituting the current value waveform within the same region as the reference current value, or the average value thereof. In this way, when the reference current value and the specific current value are stored as waveforms within a predetermined range, the pump 1 can easily monitor the flow condition even when, for example, the specific position fluctuates greatly or the change (difference value) at the specific position is small (for example, when the discharge pressure value is small).

[0088] In the present embodiment, the reference current value may be set in advance and stored, for example, together with the state determination program in the ROM 33. In this case, the pump 1 does not need to include the setting unit 314 and the change detection unit 315.

[0089] Furthermore, in this embodiment, the determination unit 313 may determine not only the presence or absence of an abnormality but also the type of abnormality. That is, for example, the storage unit 35 may store a plurality of reference current values ​​corresponding to a plurality of abnormalities, the acquisition unit 312 may acquire a plurality of specific current values ​​corresponding to the plurality of abnormalities, and the determination unit 313 may determine the presence or absence of an abnormality for each abnormality.

[0090] Furthermore, in this embodiment, the number of abnormalities determined by the determination unit 313 may be only "1" or may be multiple. In the former case, the determination unit 313 may determine the presence or absence of an abnormality in the flow state only in the discharge process or the suction process. In the latter case, the determination unit 313 may determine only the presence or absence of an abnormality, and may also determine the type of abnormality as described above.

[0091] Furthermore, in this embodiment, the horizontal axis of the current value waveform may be represented by the time elapsed from the start of operation of the pump 1, instead of the angle. In this case, the specific position may be represented by the time, instead of the angle.

[0092] Reciprocating Pump (2) Next, another embodiment of this pump (hereinafter referred to as the "second embodiment") will be described below, focusing on differences from the previously described embodiment (hereinafter referred to as the "first embodiment"). The second embodiment differs from the first embodiment in that it uses a learning model to estimate the presence or absence of an abnormality. In the following description of the second embodiment, for convenience of explanation, the same components as those in the first embodiment and components having common functions are assigned the same reference numerals as those in the first embodiment, and detailed explanations will be omitted. In the following description, FIG. 1 will be referenced as appropriate.

[0093] Configuration of Reciprocating Pump (2) FIG. 9 is a functional block diagram showing a second embodiment of the pump.

[0094] This pump 1A includes a housing 2, a control device 3A, a drive unit 4, a plunger 5, a diaphragm 6, an operating unit 7, a connection unit 8, a suction pipe L1, a discharge pipe L2, a suction valve V1, and a discharge valve V2.

[0095] The control device 3A controls the overall operation of the pump 1A. The control device 3A includes, for example, a CPU 31A, a RAM 32, a non-volatile memory such as a ROM 33A that stores various information such as a state estimation program, a driver circuit 34, and a storage unit 35A. The CPU 31A functions as an actuator control unit 311, an acquisition unit 312, a determination unit 313, a specific operation control unit 316, and an estimation unit 317. That is, the control device 3A also includes the actuator control unit 311, the acquisition unit 312, the determination unit 313, the specific operation control unit 316, and the estimation unit 317. The acquisition unit 312 also functions as a current value acquisition unit, a speed acquisition unit, and a stroke number acquisition unit according to the present invention. The acquisition unit 312 is an example of the current value acquisition unit, the speed acquisition unit, and the stroke number acquisition unit according to the present invention.

[0096] A state estimation program runs in the control device 3A, and the state estimation program cooperates with the hardware resources of the pump 1A to implement a state estimation method, which will be described later. By causing a processor (CPU 31A) included in the control device 3A to execute the state estimation program, the state estimation program causes the processor to function as an actuator control unit 311, an acquisition unit 312, a determination unit 313, a specific operation control unit 316, and an estimation unit 317, thereby causing the processor to execute the state estimation method. By causing a computer to execute the state estimation program, the state estimation program causes the computer to function as the control device 3A.

[0097] The estimation unit 317 inputs the specific current value, the movement speed (forward speed and backward speed), and the number of strokes into a learning model M1 described later, and estimates whether the flow state is normal or abnormal. Specific operations of the estimation unit 317 will be described later.

[0098] The storage unit 35A stores information (such as the learning model M1) necessary for the operation of the pump 1A. The storage unit 35A is, for example, a non-volatile memory such as an EEPROM or a flash memory.

[0099] FIG. 10 is a schematic diagram showing an example of information (learning model M1) stored in the storage unit 35A.

[0100] The "learning model M1" is a trained machine learning algorithm (i.e., a learning model) that has been trained to output whether the flow state is "normal" or "abnormal" when a control current value (e.g., a specific current value) "In" and predetermined operating conditions (forward speed "Vo", return speed "Vr", and number of strokes "Sn") are input. The learning model M1 is, for example, generated in advance by a machine learning device and stored in the memory unit 35A.

[0101] Here, machine learning by a machine learning device is performed by, for example, having a known machine learning algorithm (e.g., a neural network having an input layer, multiple intermediate layers, and an output layer) learn training data by machine learning. The "training data" is, for example, information that serves as input data for the machine learning algorithm and information that serves as output data associated with the input data.

[0102] The "input data" is an explanatory variable in machine learning. In the second embodiment, the input data are predetermined operating conditions (forward speed, return speed, and stroke count), a discharge pressure value under the same operating conditions, and a specific current value (control current value at a specific position) under the same operating conditions. As described above, as the movement speed (mainly forward speed) of the plunger 5 increases or decreases, the control current value also increases or decreases. Furthermore, as the stroke count increases or decreases, the control current value also increases or decreases. Therefore, there is a correlation between the operating conditions and the control current value. Furthermore, as the discharge pressure value increases, the output (load) of the motor 411 increases, and the control current value increases. On the other hand, as the discharge pressure value decreases, the output (load) of the motor 411 decreases, and the control current value decreases. Therefore, there is a correlation between the discharge pressure value and the control current value. The control current value (specific current value) and the discharge pressure value can be obtained, for example, by operating a test device simulating the pump 1A, equipped with a pressure gauge capable of measuring the discharge pressure value, under the same operating conditions for a predetermined period of time.

[0103] The "output data" is a target variable in machine learning, and in the second embodiment, is information indicating the operating state ("normal" or "abnormal") when the operation of the motor 411 is controlled by the input operating conditions and control current value (i.e., input data). In other words, the output data is information indicating the flow state ("normal" or "abnormal"). Numeric values ​​such as "0" and "1" are assigned to the "normal" and "abnormal" discharge pressures. One piece of output data is associated with one corresponding piece of input data to constitute one piece of learning data.

[0104] In the second embodiment, the input data used for machine learning may not be a specific current value, but may be all or part (e.g., a part corresponding to the differential region) of the control current value (current value waveform) for one stroke of the plunger 5. In this case, the same type of information (all or part of the control current value) as the input data used for machine learning is input to the learning model M1.

[0105] In addition, in the second embodiment, any one of the forward speed, backward speed, and number of strokes (for example, the number of strokes) may be calculated from the remaining two (forward speed, backward speed) and used for machine learning.

[0106] Furthermore, in the second embodiment, the machine learning algorithm used for machine learning is not limited to a neural network as long as the learning model M1 generated by machine learning can output a flow state. That is, for example, the machine learning algorithm may be a random forest, a decision tree, a support vector machine, or the like.

[0107] 10, the learning model M1 generated in this manner is capable of outputting whether the flow state is "normal: 0" or "abnormal: 1" by inputting the (current) control current value "In" and predetermined operating conditions (forward speed "Vo", return speed "Vr", and number of strokes "Sn"). In other words, the learning model M1 has been machine-trained to output whether the flow state is "normal: 0" or "abnormal: 1" when the control current value "In", forward speed "Vo", return speed "Vr", and number of strokes "Sn" are input as input data.

[0108] Operation of the Reciprocating Pump (2) Next, the operation (state estimation method) of the pump 1A will be described below. In the following description, Figures 1 and 9 will be referred to as appropriate.

[0109] FIG. 11 is a flowchart showing an example of the operation of the pump 1A.

[0110] First, the acquisition unit 312 acquires the current predetermined operating conditions (forward speed, return speed, and stroke count) of the pump 1A (ST201: operating condition acquisition step). Specifically, the forward speed and return speed are, for example, set in advance for each operating mode and stored in the storage unit 35A as parameters of the operating conditions corresponding to the operating mode. The stroke count is, for example, input by the user operating the operation unit 7 and stored in the storage unit 35A as a parameter of the operating conditions. The acquisition unit 312 acquires the current predetermined operating conditions from the storage unit 35A.

[0111] Next, the acquisition unit 312 acquires the specific current value (ST202: specific current value acquisition step) in the same manner as in the process (ST104).

[0112] Next, the estimation unit 317 inputs the specified operating conditions and specific current values ​​acquired by the acquisition unit 312 into the learning model M1 and estimates the flow state based on the output of the learning model M1 (ST203: flow state estimation step).

[0113] When the estimation unit 317 estimates that the "flow state is normal" ("Normal" in ST203), the determination unit 313 determines whether or not a specific action is being performed (ST204). When the determination unit 313 determines that the "specific action is being performed" ("Y" in ST204), the specific action control unit 316 ends the specific action (ST205: specific action end step), and the operation returns to the process (ST201). When the determination unit 313 determines that the "specific action is not being performed" ("N" in ST204), the operation returns to the process (ST201).

[0114] On the other hand, when the estimation unit 317 estimates that "the flow state is abnormal" ("abnormal" in ST203), the determination unit 313 determines whether or not a specific action is being performed (ST206). When the determination unit 313 determines that "a specific action is not being performed" ("N" in ST206), the specific action control unit 316 executes the control necessary for the specific action (ST207: specific action execution step). As a result, the specific action is started. Next, this operation returns to the process (ST201).

[0115] On the other hand, when the determination unit 313 determines that "a specific action is being performed" ("Y" in ST206), the determination unit 313 determines whether the number of times the specific action has been performed has reached a predetermined number (ST208).

[0116] When the determination unit 313 determines that "the number of executions is less than the predetermined number" ("N" in ST208), this operation returns to the process (ST201). On the other hand, when the determination unit 313 determines that "the number of executions has reached the predetermined number" ("Y" in ST208), the abnormality has not been resolved even after the specific operation has been performed the predetermined number of times, so the control device 3 (actuator control unit 311 and specific operation control unit 316) stops this operation (liquid delivery) (ST209).

[0117] In this manner, in the present pump 1A, the estimator 317 estimates the presence or absence of a flow abnormality based on the predetermined operating conditions and the control current value (specific current value). In other words, the present pump 1A detects a flow abnormality based on the predetermined operating conditions and the control current value. Therefore, the present pump 1A can monitor the flow abnormality without using an abnormality detection device (e.g., a pressure gauge, a flow meter, etc.) that comes into contact with the pumped liquid. As a result, the present pump 1A does not require the costs associated with conventional pumps equipped with pressure gauges, such as the machining required to install a pressure gauge, the seal members, the discharge piping design and installation, and the wiring installation required for the pressure gauge. Furthermore, the present pump 1A can easily estimate the presence or absence of a flow abnormality even if the pumped liquid is a corrosive liquid. Furthermore, the present pump 1 according to the first embodiment does not require a reference current value, which is essential.

[0118] Summary (2) According to the second embodiment described above, the pump 1A includes an acquisition unit 312, a memory unit 35A, and an estimation unit 317. The acquisition unit 312 acquires the specific current value, forward speed, return speed, and stroke count. The memory unit 35A stores a learning model M1. The estimation unit 317 inputs the specific current value, forward speed, return speed, and stroke count into the learning model M1 to estimate whether the flow state is normal or abnormal. With this configuration, the pump 1A can monitor the flow state in the pump 1A without using an abnormality detection device that comes into contact with the pumped liquid.

[0119] In the second embodiment, the information (predetermined operating conditions) other than the specific current value among the machine learning input data of the learning model M1 may be set appropriately according to the usage environment of the pump 1A, and is not limited to the forward speed, return speed, and number of strokes. That is, for example, when the pump 1A is used under fixed operating conditions, the machine learning input data may be only the specific current value. Furthermore, when the operating mode of the pump 1A is fixed, the machine learning input data may be only the specific current value and number of strokes (or the forward speed and / or return speed). In these cases, the information input to the learning model M1 may be information corresponding to the machine learning input data.

[0120] Furthermore, in the second embodiment, if the estimated abnormality is an abnormality occurring only in the discharge process, the input data for machine learning may not include the return speed. Similarly, if the estimated abnormality is an abnormality occurring only in the suction process, the input data for machine learning may not include the forward speed. In these cases, the information input to the learning model M1 may be information corresponding to the input data for machine learning.

[0121] Furthermore, in the second embodiment, the input data for machine learning of the learning model M1 may include other information (e.g., discharge pressure values, etc.).

[0122] Furthermore, in the second embodiment, the storage unit 35A may store multiple types of learning models M1 corresponding to multiple abnormalities, respectively.

[0123] Reciprocating Pump (3) Next, yet another embodiment of this pump (hereinafter referred to as the "third embodiment") will be described below, focusing on differences from the first and second embodiments described above. The third embodiment differs from the first and second embodiments in that it uses a learning model to predict the presence or absence of future abnormalities. In the following description of the second embodiment, for convenience of explanation, the same components as those in the first embodiment and components having common functions are assigned the same reference numerals as those in the first embodiment, and detailed description thereof will be omitted. In the following description, FIG. 1 will be referenced as appropriate.

[0124] Configuration of the Reciprocating Pump (3) FIG. 12 is a functional block diagram showing a third embodiment of the present pump.

[0125] This pump 1B includes a housing 2, a control device 3B, a drive unit 4, a plunger 5, a diaphragm 6, an operation unit 7, a connection unit 8, a suction pipe L1, a discharge pipe L2, a suction valve V1, and a discharge valve V2.

[0126] The control device 3B controls the overall operation of the pump 1B. The control device 3B includes, for example, a CPU 31B, a RAM 32, a non-volatile memory such as a ROM 33B that stores various information such as a state prediction program, a driver circuit 34, and a storage unit 35B. The CPU 31B functions as an actuator control unit 311, an acquisition unit 312, a specific operation control unit 316, a calculation unit 318, and a prediction unit 319. That is, the control device 3B also includes the actuator control unit 311, the acquisition unit 312, the specific operation control unit 316, the calculation unit 318, and the prediction unit 319. The acquisition unit 312 also functions as a current value acquisition unit, a speed acquisition unit, and a stroke number acquisition unit according to the present invention. The acquisition unit 312 is an example of the current value acquisition unit, the speed acquisition unit, and the stroke number acquisition unit according to the present invention.

[0127] A state prediction program runs in the control device 3B, and the state prediction program cooperates with the hardware resources of the pump 1B to realize a state prediction method, which will be described later. By causing a processor (CPU 31B) included in the control device 3B to execute the prediction program, the state prediction program causes the processor to function as an actuator control unit 311, an acquisition unit 312, a specific operation control unit 316, a calculation unit 318, and a prediction unit 319, thereby causing the processor to execute the state prediction method. By causing a computer to execute the state prediction program, the state prediction program causes the computer to function as the control device 3B.

[0128] The calculation unit 318 calculates the amount of change between a reference specific current value and the latest specific current value among the multiple specific current values ​​acquired at a predetermined sampling interval. The specific operation of the calculation unit 318 will be described later.

[0129] The "predetermined sampling interval" is indicated, for example, by the number of strokes of the plunger 5, and in this embodiment, it is several thousand to several tens of thousands of strokes.

[0130] The "reference specific current value" is, for example, a specific current value obtained at a time when there is no obvious abnormality in the flow state of the handled liquid (for example, immediately after the installation of this pump 1B, immediately after maintenance of this pump 1B, etc.).

[0131] In the present invention, the calculation unit 318 may calculate the amount of change between two specific current values ​​consecutively acquired at a predetermined sampling interval. In this case, the reference specific current value is the specific current value acquired immediately before the latest specific current value. In this case, the input data for machine learning also includes a similar amount of change.

[0132] The prediction unit 319 inputs the amount of change, the movement speed (forward speed and backward speed), and the number of strokes into a learning model M2 described below, and predicts whether or not an abnormality in the flow state will occur in a predetermined future period (hereinafter referred to as the "prediction period"). Specific operations of the prediction unit 319 will be described later.

[0133] The "prediction period" is, for example, a few hours, one day, several days, one week, etc. The prediction period is set as a fixed period during machine learning, which will be described later, for example.

[0134] The storage unit 35B stores information (such as the learning model M2) necessary for the operation of the pump 1B. The storage unit 35B is, for example, a non-volatile memory such as an EEPROM or a flash memory.

[0135] FIG. 13 is a schematic diagram showing an example of information (learning model M2) stored in the storage unit 35B.

[0136] The "learning model M2" is a trained machine learning algorithm (i.e., a learning model) that has been trained to output whether or not an abnormality in the flow state will occur after a prediction period has elapsed (i.e., a predicted flow state) when the amount of change "ΔI" and predetermined operating conditions (forward speed "Vo", return speed "Vr", and number of strokes "Sn") are input. The learning model M2 is generated in advance by, for example, a machine learning device, and is stored in the storage unit 35B, for example, in association with information indicating the prediction period corresponding to the learning model M2.

[0137] Here, abnormalities in the suction or discharge state include abnormalities that occur over time and in stages. The timing of occurrence of such abnormalities, which increase or decrease the control current value over time from the initial to final stages of the abnormality, can be predicted to some extent by tracking the amount of change in the control current value. Furthermore, the progression of the abnormality varies depending on the operating conditions. For example, if the abnormality is related to deterioration of the diaphragm 6, the progression will accelerate as the number of strokes increases. Therefore, the amount of change and predetermined operating conditions are used as machine learning input data for the machine learning algorithm of the third embodiment.

[0138] The output data of the machine learning of the machine learning algorithm of the third embodiment is whether or not an abnormality occurs in the suction state or the discharge state after the prediction period has elapsed.

[0139] 13, the learning model M2 generated in this manner is capable of outputting whether the predicted flow state is "normal prediction: 0" or "abnormal prediction: 1" by inputting the amount of change "ΔI" and predetermined operating conditions (forward speed "Vo", return speed "Vr", and number of strokes "Sn"). In other words, the learning model M2 has been machine-trained to output whether the predicted flow state is "normal prediction: 0" or "abnormal prediction: 1" when the amount of change "ΔI", forward speed "Vo", return speed "Vr", and number of strokes "Sn" are input as input data.

[0140] Operation of the Reciprocating Pump (3) Next, the operation (prediction method) of the pump 1B will be explained below. In the following explanation, FIG. 12 will be referred to as appropriate.

[0141] FIG. 14 is a flowchart showing an example of the operation of the pump 1B.

[0142] First, the acquisition unit 312 acquires the current predetermined operating conditions (forward speed, return speed, and number of strokes) of the pump 1B (ST301: operating condition acquisition step), similar to the process (ST201).

[0143] Next, the acquiring unit 312 acquires the specific current value at a predetermined sampling interval (ST302: specific current value acquiring step). At this time, the specific current value that serves as the reference is stored as a reference specific current value in, for example, the storage unit 35B.

[0144] Next, the calculation unit 318 calculates the amount of change between the reference specific current value and the latest specific current value (ST303). Here, the amount of change may be the difference between the two specific current values, or the rate of change in current value between the two specific current values.

[0145] Next, the prediction unit 319 inputs the amount of change and the predetermined operating conditions (forward speed, return speed, and number of strokes) into the learning model M2, and predicts whether or not an abnormality will occur in the suction state or discharge state (flow state) during the prediction period based on the output of the learning model M2 (ST304). Here, the prediction period is set during machine learning of the learning model M2.

[0146] When the prediction unit 319 predicts that "an abnormality exists," i.e., that "the flow state is abnormal" ("abnormal" in ST304), the specific operation control unit 316 executes the control required for the specific operation (ST305: specific operation execution step). At this point, no abnormality has occurred in the flow of the handled fluid. Therefore, in the third embodiment, the specific operation is an operation to display a warning light or an operation to sound an alarm. Such an operation is an operation to prevent the occurrence of an abnormality in the future, and can be included in the specific operation in a broad sense. Next, this operation returns to the process (ST302).

[0147] On the other hand, when the prediction unit 319 predicts that "there is no abnormality", that is, that "the flow state is normal" ("normal" in ST304), the operation returns to the process (ST302).

[0148] In this manner, in the present pump 1B, the prediction unit 319 predicts the presence or absence of an abnormality in the flow state during the prediction period based on the amount of change and the predetermined operating conditions. In other words, the present pump 1B predicts the occurrence of an abnormality in the flow state based on the amount of change and the operating conditions. Therefore, the present pump 1B can monitor the flow state of the pumped liquid in the present pump 1B without using an abnormality detection device (e.g., a pressure gauge, a flow meter, etc.) that comes into contact with the pumped liquid. As a result, the present pump 1B does not require the costs associated with conventional pumps equipped with pressure gauges, such as the machining required to install a pressure gauge, the seal members, the discharge piping design and installation, and the wiring installation required for the pressure gauge. Furthermore, the present pump 1B can easily estimate the presence or absence of an abnormality in the suction state and / or discharge state, even if the pumped liquid is corrosive. Furthermore, the present pump 1 in the first embodiment does not require a reference current value, which is essential.

[0149] Summary (3) According to the embodiment described above, the pump 1B includes an acquisition unit 312, a memory unit 35B, a calculation unit 318, and a prediction unit 319. The acquisition unit 312 acquires a specific current value, a forward speed, a return speed, and a stroke count. The memory unit 35B stores a learning model M2. The calculation unit 318 calculates the amount of change between a reference specific current value and the most recent specific current value among multiple specific current values ​​acquired at a predetermined sampling interval. The prediction unit 319 inputs the amount of change, the forward speed, the return speed, and the stroke count into the learning model M2 to estimate whether an abnormality will occur in the flow state after the prediction period has elapsed. This configuration allows the pump 1B to monitor the flow state of the pumped liquid without using an abnormality detection device that comes into contact with the pumped liquid.

[0150] In the third embodiment, the information (predetermined operating conditions) other than the amount of change among the input data for machine learning of the learning model M2 may be set appropriately according to the usage environment of the pump 1B, and is not limited to the forward speed, return speed, and number of strokes. That is, for example, when the pump 1B is used under fixed operating conditions, the input data for machine learning may be only the amount of change. Furthermore, when the operating mode of the pump 1B is fixed, the input data for machine learning may be only the amount of change and the number of strokes (or the forward speed and / or return speed). In these cases, the information input to the learning model M2 may be information corresponding to the input data for machine learning.

[0151] Furthermore, in the third embodiment, if the estimated abnormality is an abnormality occurring only in the discharge process, the input data for machine learning may not include the return speed. Similarly, if the estimated abnormality is an abnormality occurring only in the suction process, the input data for machine learning may not include the forward speed. In these cases, the information input to the learning model M2 may be information corresponding to the input data for machine learning.

[0152] Furthermore, in the third embodiment, the input data for machine learning of the learning model M2 may include other information (e.g., discharge pressure values, etc.).

[0153] Furthermore, in the third embodiment, the memory unit 35B may store multiple types of learning models M2 corresponding to each of multiple abnormalities, or may store multiple types of learning models M2 corresponding to each of multiple prediction periods.

[0154] Other Embodiments: In the present invention, the electromagnetic actuator is not limited to a stepping motor. For example, the electromagnetic actuator may be a known solenoid (capable of closed-loop control) equipped with a position sensor that detects the position of the moving axis, or a servo motor.

[0155] In the present invention, the operation modes of the pumps 1, 1A, and 1B are not limited to the standard mode, the high viscosity mode, and the low pulsation mode, and the operation modes of the pump 1 do not necessarily include any of these operation modes.

[0156] Furthermore, in the present invention, the pumps 1, 1A, and 1B may include, instead of the current detection circuit 341 of the driver circuit 34, another current detection circuit capable of detecting a control current value.

[0157] Furthermore, in the present invention, the abnormality in the flow state is not limited to gas lock or cavitation, as long as it affects the flow of the pumped liquid in the pump 1, 1A, or 1B and causes a characteristic change in the control current value. That is, for example, the abnormality in the flow state may be an abnormality in the pumped liquid itself (e.g., freezing of the pumped liquid or a change in viscosity) or an abnormality caused by something outside the pump chamber 21a (e.g., leakage or clogging in the suction flow path 21b or the discharge flow path 21c, or wear and tear on the drive gear 42, the driven gear 43, or the bearings 45 and 46).

[0158] Embodiments of the Present Invention Next, embodiments of the present invention that can be understood from the above-described embodiments will be described below, using the terms and symbols used in the embodiments.

[0159] A first embodiment of the present invention is a reciprocating pump (e.g., reciprocating pump 1) that sucks and discharges pumped fluid by reciprocating a diaphragm (e.g., diaphragm 6), and includes the diaphragm, a pump chamber (e.g., pump chamber 21a) in which the diaphragm is accommodated, a plunger (e.g., plunger 5) that reciprocates the diaphragm, an electromagnetic actuator (e.g., motor device 41) that reciprocates the plunger, an actuator control unit (e.g., actuator control unit 311) that performs closed-loop control of the operation of the electromagnetic actuator, and a control current value used to control the operation of the electromagnetic actuator that is at least set when the plunger is located at a specific position within one stroke of the plunger. a storage unit (e.g., storage unit 35) that stores, as a reference current value, the normal control current value when the plunger is positioned at the specific position, among the normal control current values ​​that are the control current values ​​when the suction and discharge states of the pumped fluid are normal; and a determination unit (e.g., determination unit 313) that determines whether the suction and / or discharge states are normal or abnormal based on the specific current value and the reference current value, wherein the specific position corresponds to a point at which, when an abnormality exists, a characteristic change corresponding to the abnormality appears in a current waveform of the control current value for one stroke of the plunger. With this configuration, the pump can monitor the flow state in the pump without using an abnormality detection device (e.g., a pressure gauge, a flow meter, etc.) that comes into contact with the pumped fluid.

[0160] A second embodiment of the present invention is the reciprocating pump of the first embodiment, further comprising a reference acquisition unit (e.g., acquisition unit 312) that acquires the normal control current value, and a setting unit (e.g., setting unit 314) that sets the reference current value based on the normal control current value acquired by the reference acquisition unit. With this configuration, the pump can set an appropriate reference current value that corresponds to an abnormality in the usage environment of the pump.

[0161] A third aspect of the present invention is the reciprocating pump of the second aspect, further comprising a change detection unit (e.g., change detection unit 315) that detects a change in the operating conditions of the reciprocating pump, the reference acquisition unit acquires the normal control current value when the change detection unit detects a change in the operating conditions, and the storage unit updates and stores the reference current value based on the acquired normal control current value. With this configuration, the pump can appropriately set the reference current value in accordance with the control current value, which increases or decreases depending on the operating conditions.

[0162] A fourth aspect of the present invention is the reciprocating pump of the first aspect, further comprising a specific operation control unit (e.g., specific operation control unit 316) that, when the determination unit determines that an abnormality exists in the suction state or the discharge state, executes control necessary for a specific operation to resolve the abnormality. With this configuration, even if an abnormality occurs, the pump can automatically execute the specific operation to resolve the abnormality.

[0163] A fifth aspect of the present invention is a reciprocating pump according to any one of the first to fourth aspects, wherein the memory unit stores, as the reference current value, a current value waveform of the normal control current value used to control the plunger during a stroke within a predetermined range including the specific position, and the acquisition unit acquires, as the specific current value, a current value waveform of the control current value used to control the plunger during a stroke within the predetermined range. With this configuration, the pump can easily monitor the flow state even when the specific position fluctuates significantly or when the change in the specific position is small.

[0164] A sixth embodiment of the present invention is a reciprocating pump (for example, reciprocating pump 1A) that sucks and discharges pumped fluid by reciprocating a diaphragm (for example, diaphragm 6), and includes the diaphragm, a pump chamber (for example, pump chamber 21a) in which the diaphragm is accommodated, a plunger (for example, plunger 5) that reciprocates the diaphragm, an electromagnetic actuator (for example, motor device 41) that reciprocates the plunger, an actuator control unit (for example, actuator control unit 311) that performs closed-loop control of the operation of the electromagnetic actuator, and a current value acquisition unit (for example, acquisition unit 312) that acquires, as a specific current value, the control current value when at least the plunger is located at a specific position within one stroke of the plunger, among the control current values ​​used to control the operation of the electromagnetic actuator, and the plunger. a speed acquisition unit (e.g., acquisition unit 312) that acquires the movement speed of the plunger; a stroke number acquisition unit (e.g., acquisition unit 312) that acquires the number of strokes of the plunger; a memory unit (e.g., memory unit 35A) that stores a trained learning model (e.g., learning model M1) that has been machine-learned to output whether the suction state or the discharge state of the pumped fluid is normal or abnormal when the specific current value, the movement speed, and the stroke number are input; and an estimation unit (e.g., estimation unit 317) that inputs the specific current value, the movement speed, and the stroke number into the learning model and estimates whether the suction state or the discharge state is normal or abnormal, wherein the specific position is a position that corresponds to a point at which a characteristic change corresponding to the abnormality appears in the current value waveform of the control current value for one stroke of the plunger when the abnormality exists. According to this configuration, the flow state of the pump can be monitored without using an abnormality detection device (for example, a pressure gauge, a flow meter, etc.) that comes into contact with the pumped liquid.

[0165] A seventh embodiment of the present invention is a reciprocating pump (for example, reciprocating pump 1B) that sucks and discharges pumped fluid by reciprocating a diaphragm (for example, diaphragm 6), and includes: the diaphragm; a pump chamber (for example, pump chamber 21a) in which the diaphragm is accommodated; a plunger (for example, plunger 5) that reciprocates the diaphragm; an electromagnetic actuator (for example, motor device 41) that reciprocates the plunger; an actuator control unit (for example, actuator control unit 311) that performs closed-loop control of the operation of the electromagnetic actuator; and a current value acquisition unit (for example, acquisition unit 312) that acquires, as a specific current value, the control current value when at least the plunger is located at a specific position within one stroke of the plunger, among the control current values ​​used to control the operation of the electromagnetic actuator; and a plurality of the specific current values ​​acquired at predetermined sampling intervals. The change amount between the specific current value that is a reference and the latest specific current value is calculated. a speed acquisition unit (e.g., acquisition unit 312) that acquires the movement speed of the plunger; a stroke number acquisition unit (e.g., acquisition unit 312) that acquires the number of strokes of the plunger; a memory unit (e.g., memory unit 35B) that stores a trained learning model (e.g., learning model M2) that has been machine-learned to output whether or not an abnormality will occur in the suction state or discharge state of the pumped fluid after a predetermined future period has elapsed when the specific current value, the amount of change, the movement speed, and the number of strokes are input; and a prediction unit (e.g., prediction unit 319) that inputs the specific current value, the amount of change, the movement speed, and the number of strokes into the learning model and predicts whether or not the abnormality will occur in the suction state or the discharge state within the period, wherein the specific position is a position corresponding to a point at which a characteristic change corresponding to the abnormality appears in the current value waveform of the control current value for one stroke of the plunger when the abnormality occurs. According to this configuration, the flow state of the pump can be monitored without using an abnormality detection device (for example, a pressure gauge, a flow meter, etc.) that comes into contact with the pumped liquid.

[0166] An eighth embodiment of the present invention is a method for monitoring a flow state of a pumped liquid, which is executed by a reciprocating pump having a diaphragm, a pump chamber in which the diaphragm is accommodated, a plunger that reciprocates the diaphragm, an electromagnetic actuator that reciprocates the plunger, an actuator control unit that performs closed-loop control of the operation of the electromagnetic actuator, and a memory unit that stores a reference current value among control current values ​​used to control the operation of the electromagnetic actuator, and which determines whether a suction state and / or a discharge state of the pumped liquid are normal or abnormal, and the reference current value is a normal control current value that is the control current value when the suction state and the discharge state are normal, at least when the plunger is within one stroke of the plunger. The flow state monitoring method includes a step of the reciprocating pump acquiring, as a specific current value, at least the control current value when the plunger was positioned at the specific position among the control current values ​​(e.g., a specific current value acquisition step: ST104), and a step of the reciprocating pump determining whether the suction state and / or the discharge state is normal or abnormal based on the specific current value and the reference current value (e.g., a state determination step: ST106). With this configuration, the pump can monitor the flow state without using an abnormality detection device (e.g., a pressure gauge, a flow meter, etc.) that comes into contact with the pumped liquid.

[0167] 1 Reciprocating pump 21a Pump chamber 311 Actuator control unit 312 Acquisition unit (reference acquisition unit, current value acquisition unit, speed acquisition unit, stroke number acquisition unit) 313 Determination unit 314 Setting unit 315 Change detection unit 316 Specific operation control unit 35 Memory unit 41 Motor device (electromagnetic actuator, stepping motor) 411 Motor 5 Plunger 6 Diaphragm 7 Operation unit 1A Reciprocating pump 317 Estimation unit 35A Memory unit M1 Learning model 1B Reciprocating pump 318 Calculation unit 319 Prediction unit 35B Memory unit M2 Learning model

Claims

1. A reciprocating pump that sucks in and discharges pumped fluid by reciprocating a diaphragm, comprising: the diaphragm; a pump chamber in which the diaphragm is housed; a plunger that reciprocates the diaphragm; an electromagnetic actuator that reciprocates the plunger; an actuator control unit that performs closed-loop control of the operation of the electromagnetic actuator; an acquisition unit that acquires, as a specific current value, among control current values ​​used to control the operation of the electromagnetic actuator, at least the control current value when the plunger is located at a specific position within one stroke of the plunger; a storage unit that stores, as a reference current value, the normal control current value, which is the control current value when the suction state and discharge state of the pumped fluid are normal, at least the normal control current value when the plunger is located at the specific position; and a judgment unit that judges whether the suction state and / or the discharge state are normal or abnormal based on the specific current value and the reference current value. the specific position is a position corresponding to a point at which a characteristic change corresponding to the abnormality appears in a current value waveform of the control current value for one stroke of the plunger when the abnormality exists.

2. A reciprocating pump as described in claim 1, comprising: a reference acquisition unit that acquires the normal control current value; and a setting unit that sets the reference current value based on the normal control current value acquired by the reference acquisition unit.

3. A reciprocating pump as described in claim 2, further comprising: a change detection unit which detects a change in operating conditions of the reciprocating pump; wherein the reference acquisition unit acquires the normal control current value when the change detection unit detects a change in the operating conditions; and the memory unit updates and stores the reference current value based on the acquired normal control current value.

4. The reciprocating pump according to claim 1, further comprising: a specific operation control unit that, when the determination unit determines that an abnormality exists in the suction state or the discharge state, executes control necessary for a specific operation to eliminate the abnormality.

5. A reciprocating pump as claimed in any one of claims 1 to 4, wherein the memory unit stores the current value waveform of the normal control current value used to control the plunger in a stroke of a specified range including the specific position as the reference current value, and the acquisition unit acquires the current value waveform of the control current value used to control the plunger in a stroke of the specified range as the specific current value.

6. A reciprocating pump that sucks in and discharges a pumped liquid by reciprocating a diaphragm, comprising: the diaphragm; a pump chamber in which the diaphragm is housed; a plunger that reciprocates the diaphragm; an electromagnetic actuator that reciprocates the plunger; an actuator control unit that performs closed-loop control of the operation of the electromagnetic actuator; a current value acquisition unit that acquires, as a specific current value, at least the control current value when the plunger is located at a specific position within one stroke of the plunger, among the control current values ​​used to control the operation of the electromagnetic actuator; a speed acquisition unit that acquires the moving speed of the plunger; a stroke number acquisition unit that acquires the number of strokes of the plunger; and a memory unit that stores a trained learning model that has been machine-learned to output whether the suction state or discharge state of the pumped liquid is normal or abnormal when the specific current value, the moving speed, and the stroke number are input. an estimation unit that inputs the specific current value, the moving speed, and the number of strokes into the learning model to estimate whether the suction state or the discharge state is normal or abnormal, wherein the specific position is a position corresponding to a point at which a characteristic change corresponding to the abnormality appears in a current value waveform of the control current value for one stroke of the plunger when the abnormality is present.

7. A reciprocating pump which sucks in and discharges pumped fluid by reciprocating a diaphragm, comprising: the diaphragm; a pump chamber in which the diaphragm is housed; a plunger which reciprocates the diaphragm; an electromagnetic actuator which reciprocates the plunger; an actuator control unit which performs closed-loop control of the operation of the electromagnetic actuator; a current value acquisition unit which acquires, as a specific current value, at least the control current value when the plunger is located at a specific position within one stroke of the plunger, among control current values ​​used to control the operation of the electromagnetic actuator; a calculation unit which calculates, among a plurality of specific current values ​​acquired at a predetermined sampling interval, an amount of change between the specific current value serving as a reference and the latest specific current value; a speed acquisition unit which acquires the moving speed of the plunger; and a stroke number acquisition unit which acquires the number of strokes of the plunger. a memory unit that stores a trained learning model that has been machine-learned to output whether or not an abnormality will occur in the suction state or discharge state of the handled fluid after a predetermined future period has elapsed when the specific current value, the amount of change, the movement speed, and the number of strokes are input; and a prediction unit that inputs the specific current value, the amount of change, the movement speed, and the number of strokes into the learning model and predicts whether or not the abnormality will occur in the suction state or the discharge state within the period, wherein the specific position is a position corresponding to a point at which a characteristic change corresponding to the abnormality appears in the current value waveform of the control current value for one stroke of the plunger when the abnormality occurs.

8. A method for monitoring a flow state of a handled fluid for determining whether a suction state and / or a discharge state of the handled fluid are normal or abnormal, the method being executed by a reciprocating pump comprising: a diaphragm; a pump chamber in which the diaphragm is accommodated; a plunger that reciprocates the diaphragm; an electromagnetic actuator that reciprocates the plunger; an actuator control unit that performs closed-loop control of the operation of the electromagnetic actuator; and a memory unit that stores a reference current value among control current values ​​used to control the operation of the electromagnetic actuator, wherein the reference current value is, among normal control current values ​​which are the control current values ​​when the suction state and the discharge state are normal, the normal control current value when at least the plunger is located at a specific position within one stroke of the plunger, and the specific position is a position corresponding to a point at which a characteristic change corresponding to the abnormality appears in the current value waveform of the control current value for one stroke of the plunger when the abnormality exists, and the flow state monitoring method comprises: A flow state monitoring method comprising: a step of the reciprocating pump acquiring, as a specific current value, among the control current values, at least the control current value when the plunger was located at the specific position; and a step of the reciprocating pump determining whether the suction state and / or the discharge state is normal or abnormal based on the specific current value and the reference current value.

Images