JP2021141544A - Prediction device, image forming apparatus, program, and prediction system - Google Patents

Abstract

To solve the program in which: even if an operator determines that an image forming apparatus is in an unfavorable state, an error message is not output when the image forming apparatus does not recognize an abnormality, and thus the operator has difficulty grasping the state of the image forming apparatus.SOLUTION: A prediction device 60 included in an ink jet recording device 1 comprises: an operation detail acquisition unit 61 that acquires, from an operation display unit 52, the operation details of an operation performed on the ink jet recording device 1; a device state prediction unit 63 that, based on the operation details, predicts the device state of the ink jet recording device 1; and a display output unit 64 that outputs the predicted device state so that it can be displayed on the operation display unit 52.SELECTED DRAWING: Figure 4

Description

The present invention relates to a prediction device, an image forming device, a program and a prediction system.

Conventionally, in order to determine an abnormality that has occurred in an image forming apparatus that forms an image on paper and identify the cause of the abnormality, the state of the image forming apparatus is determined by analyzing data acquired from a sensor attached to the image forming apparatus. The method is known. In addition, a method of predicting the cause of an abnormality from an abnormality history collected in time series from an image forming apparatus is known.

For example, Patent Document 1 disclosed as a technique for predicting the cause of an abnormality that has occurred in an image forming apparatus states, "When a serviceman visits, an abnormality accumulated in the storage means for a certain period before a predetermined date and time. The necessity of visiting a serviceman is determined based on the event or pre-abnormal event information and additional information. "

Further, Patent Document 2 describes that "a situation in which an abnormal load detection alarm for the servo shaft is likely to occur is predicted from the operation history of the operator in the machine tool, and a collision with the spindle is warned."

Japanese Unexamined Patent Publication No. 2000-253199 Japanese Unexamined Patent Publication No. 2018-92428

By the way, even if the device does not generate an error message, the operator may determine that the device is in an unfavorable state if the quality of the deliverable output from the device is insufficient. However, in the technique disclosed in Patent Document 1, an error message or the like is not output unless an abnormality that the device cannot withstand use has occurred. Therefore, even if the operator determines that the deliverable is unfavorable, how the device works. It was not shown whether it was in a good condition. Therefore, even if some performance deterioration or deterioration occurs in the device, the operator cannot identify the cause of the performance deterioration or deterioration.

Further, the technique disclosed in Patent Document 2 merely predicts what kind of result the operator's behavior will bring. Therefore, in a specific situation, it is possible to predict the situation corresponding to a specific alarm, but it is not possible to predict the state of the device. In any of the techniques disclosed in Patent Documents 1 and 2, even if the image forming apparatus does not recognize an abnormality, what kind of state the apparatus is in when the operator determines that the condition is unfavorable. It was difficult to grasp.

The present invention has been made in view of such a situation, and an object of the present invention is to enable the operator to grasp the state of the image forming apparatus from the operation of the image forming apparatus.

In order to realize at least one of the above-mentioned objects, the prediction device reflecting one aspect of the present invention includes an operation content acquisition unit for acquiring the operation content of the operation performed on the image forming device from the operation unit, and an operation. It includes a device state prediction unit that predicts the device state of the device based on the contents, and a display output unit that outputs the predicted device state to the display unit in a displayable manner.
The above-mentioned prediction device is one aspect of the present invention, and an image forming device, a program, and a prediction system reflecting one aspect of the present invention are configured in the same manner as the above-mentioned prediction device.

According to the present invention, the device state predicted based on the operation content of the operator is displayed on the display unit, and the operator can easily grasp the state of the image forming device.
Issues, configurations and effects other than those described above will be clarified by the following description of the embodiments.

It is a schematic block diagram which shows the whole structure of the inkjet recording apparatus which concerns on 1st Embodiment of this invention. It is a top view which shows the structure of the head unit which concerns on 1st Embodiment of this invention as seen from the recording material side. It is a block diagram which shows the hardware configuration example of the inkjet recording apparatus which concerns on 1st Embodiment of this invention. It is a block diagram which shows the functional structure example of the inkjet recording apparatus which concerns on 1st Embodiment of this invention. It is a figure which shows the content of the history table at the time of occurrence of the printing defect other than the ink ejection defect which concerns on 1st Embodiment of this invention. It is a figure which shows the content of the history table at the time of the misalignment of each nozzle which concerns on 1st Embodiment of this invention. It is a figure which shows the example of the required print state table which concerns on 1st Embodiment of this invention. It is a figure which shows the example of the apparatus state table which concerns on 1st Embodiment of this invention. It is a figure which shows the modification of the apparatus state table which concerns on 1st Embodiment of this invention. It is a figure which shows the modification of the apparatus state table which concerns on 1st Embodiment of this invention. It is a flowchart which shows the example of the determination process of the apparatus state performed in the control part which concerns on 1st Embodiment of this invention. It is a flowchart which shows an example of the process which the control part which concerns on 1st Embodiment of this invention presents a countermeasure to an operation display part. It is a block diagram which shows the structural example of the prediction system which concerns on the modification of 1st Embodiment of this invention. It is a figure which shows the example of the machine learning which concerns on the 2nd Embodiment of this invention. It is a block diagram which shows the functional structure example of the inkjet recording apparatus which concerns on 2nd Embodiment of this invention. It is a figure which shows the example of the required print state table which concerns on 2nd Embodiment of this invention. It is a figure which shows the example of the apparatus state table which concerns on 2nd Embodiment of this invention. It is a figure which shows the modification of the apparatus state table which concerns on 2nd Embodiment of this invention. It is a block diagram which shows the structural example of the prediction system which concerns on the modification of the 2nd Embodiment of this invention.

Hereinafter, embodiments for carrying out the present invention will be described with reference to the accompanying drawings. In the present specification and the drawings, components having substantially the same function or configuration are designated by the same reference numerals, and redundant description will be omitted.

First, a configuration example of an inkjet recording device as an example of the image forming device according to the present embodiment will be described.
FIG. 1 is a schematic configuration diagram showing an overall configuration of an inkjet recording device.

The inkjet recording device 1 forms (records) an image on the recording material P (recording medium) by ejecting ink from a nozzle 244 provided in the inkjet head 242 shown in FIG. 2, which will be described later. In the present embodiment, an example in which paper is applied as a recording medium will be described, but the present invention is not limited to this, and various other recording media such as a film and a cloth can be applied as the recording medium.

The inkjet recording device 1 is a color inkjet recording device that superimposes four color inks of yellow (Y), magenta (M), cyan (C), and black (Bk). The inkjet recording device 1 includes a paper feeding unit 10, an image forming unit 20, a paper ejection unit 30, and a control unit 40. Then, the inkjet recording device 1 forms the image data input from the terminal device 2 (see FIG. 3) on the recording material P.

The paper feed unit 10 includes a paper feed tray 11 and a recording material supply unit 12. The paper feed tray 11 is a plate-shaped member provided on which the recording material P can be placed. The paper feed tray 11 is provided so as to be movable in the vertical direction according to the number of mounted recording materials P. Then, among the plurality of recording materials P placed on the paper feed tray 11, the uppermost recording material P in the vertical direction is held at a position where it is conveyed by the recording material supply unit 12.

The recording material supply unit 12 has a plurality of (two in this example) rollers 121 and 122, and a transport belt 123. The transport belt 123 is formed in an endless shape in which both ends in the longitudinal direction are connected. The transport belt 123 is stretched on the rollers 121 and 122. Then, when one of the rollers 121 and 122 is rotationally driven, the transport belt 123 circulates between the two rollers 121 and 122. As a result, the recording material P placed on the transport belt 123 is transported.

Further, the recording material supply unit 12 is a supply device (not shown) that delivers a drive unit (not shown) that rotationally drives the rollers 121 and 122 and the uppermost recording material P mounted on the paper feed tray 11 to the transport belt 123 (not shown). Omitted). Then, the recording material supply unit 12 conveys the recording material P placed on the conveying belt 123 toward the image forming unit 20 and feeds the recording material P to the image forming unit 20.

The image forming unit 20 includes an image forming drum 21, a transfer unit 22, a heating unit 23, a head unit 24, a fixing unit 25, an image reading unit 26, a recording material discharging unit 27, and a recording material reversing unit 28. And to form an image on the recording material P.

The image forming drum 21 is formed in a cylindrical shape. The image forming drum 21 is rotated counterclockwise by a drive motor (not shown). The recording material P supplied from the paper feeding unit 10 is supported on the outer peripheral surface of the image forming drum 21. The image forming drum 21 has three holding regions 211 obtained by dividing the outer peripheral surface into three equal parts. In the holding region 211, a resin sheet or the like is attached to the surface of a stainless steel (SUS) sheet or the like, and the recording material P can be held on the surface of each holding region 211. Then, the image forming drum 21 is rotationally driven to convey the recording material P toward the paper ejection unit 30. Further, a heating unit 23, a head unit 24, a fixing unit 25, and an image reading unit 26 are arranged on the outer peripheral surface of the image forming drum 21 so as to face each other.

The delivery unit 22 is provided between the recording material supply unit 12 of the paper feed unit 10 and the image forming drum 21. The delivery unit 22 has a claw portion 221 and a cylindrical delivery drum 222 and the like. The claw portion 221 supports one end of the recording material P conveyed by the recording material supply unit 12. The delivery drum 222 guides the recording material P supported on the claw portion 221 toward the image forming drum 21. As a result, the recording material P is transferred from the recording material supply unit 12 to the outer peripheral surface of the image forming drum 21 via the transfer unit 22.

A heating unit 23 is arranged on the downstream side of the delivery drum 222 in the transport direction of the recording material P. The heating unit 23 has, for example, a heating wire, and generates heat in response to energization. The heating unit 23 generates heat so that the recording material P supported on the image forming drum 21 and passing in the vicinity of the heating unit 23 reaches a predetermined temperature under the control of the control unit 40.

Further, a temperature sensor (not shown) is provided in the vicinity of the heating unit 23. The temperature sensor detects the temperature near the heating unit 23. Then, the control unit 40 controls the temperature of the heating unit 23 based on the temperature information detected by the temperature sensor.

A head unit 24 is provided on the downstream side of the heating unit 23 in the transport direction of the recording material P. Four head units 24 are provided according to yellow (Y), magenta (M), cyan (C), and black (Bk). The four head units 24 are arranged in the order of yellow, magenta, cyan, and black from the upstream side with respect to the transport direction of the recording material P.

The head unit 24 is set to a length (width) that covers the entire recording material P in a direction (width direction) orthogonal to the transport direction of the recording material P. That is, the inkjet recording device 1 is a one-pass type line head type inkjet recording device. The four head units 24 have the same configuration as each other, except that the colors of the inks to be ejected are different from each other.

Next, a configuration example of the head unit 24 will be described.
FIG. 2 is a plan view showing a state in which the head unit 24 is viewed from the recording material side.

The head unit 24 has a plurality of (16 in this example) inkjet heads 242. Then, the two inkjet heads 242 form a set to form one inkjet module 243. Therefore, the head unit 24 of this example is provided with eight inkjet modules 243.

The eight inkjet modules 243 are arranged in two rows along the transport direction of the recording material P. Then, four inkjet modules 243 in a row are arranged side by side along a direction (width direction) orthogonal to the transport direction of the recording material P. Further, in the eight inkjet modules 243, two rows of inkjet modules 243 are arranged in a staggered pattern along the conveying direction of the recording material P.

The number and arrangement of the inkjet modules 243 are not limited to those described above, and 6 or 10 or more inkjet modules 243 may be arranged.

Further, the inkjet head 242 has a plurality of nozzles 244. Then, the inkjet head 242 ejects ink from the nozzle 244 toward the recording material P. As a result, an image is formed on the recording material P supported on the image forming drum 21.

A fixing portion 25 is arranged on the downstream side of the four head units 24 in the transport direction. As the fixing portion 25, for example, a fluorescent tube that irradiates ultraviolet rays such as a low-pressure mercury lamp is applied. Then, the fixing unit 25 irradiates the recording material P conveyed by the image forming drum 21 with ultraviolet rays to cure the ink discharged on the recording material P. As a result, the fixing portion 25 fixes the image formed on the recording material P.

Fluorescent tubes that emit ultraviolet rays include low-pressure mercury lamps, mercury lamps with an operating pressure of several hundred Pa to 1 MPa, light sources that can be used as germicidal lamps, cold cathode tubes, ultraviolet laser light sources, metal halide lamps, and light emitting diodes. Can be mentioned. Among these, a light source (for example, a light emitting diode) that can irradiate ultraviolet rays with higher illuminance and consumes less power is more desirable.

The fixing portion 25 is not limited to the one that irradiates ultraviolet rays, and may be any one that irradiates energy rays having a property of curing the ink according to the properties of the ink, and the light source is also the wavelength of the energy rays. It is replaced according to such. Further, the fixing portion 25 is not limited to the one that irradiates light such as ultraviolet rays. As the fixing portion, for example, various other methods such as drying the ink by applying heat to the recording material or applying a liquid that causes a chemical change to the ink can be applied.

Further, an image reading unit 26 is arranged on the downstream side of the fixing unit 25 in the transport direction. The image reading unit 26 irradiates the reading target with light from the light source and reads the reflected image. The image reading unit 26 is composed of an in-line sensor in which a plurality of detection elements are arranged along a direction (width direction) orthogonal to the conveying direction of the recording material P, and the recording material P is formed by a head unit 24 and a fixing unit 25. The two-dimensional reflection image formed in the image forming drum 21 and the reflection image on the surface of the image forming drum 21 are read. The image data read by the image reading unit 26 is sent to the control unit 40.

The distance between the detection elements constituting the image reading unit 26 is set wider than the distance between the nozzles 244 of the inkjet head 242. That is, the resolution of the image reading unit 26 is set to be coarser than the resolution of the head unit 24. Further, the image reading unit 26 is formed longer than the width direction of the image forming drum 21, and is formed so that the entire holding region 211 on the surface of the image forming drum 21 can be read in the width direction.

Further, a recording material discharging unit 27 and a recording material reversing unit 28 are provided on the downstream side of the image reading unit 26 in the transport direction. The recording material discharging unit 27 conveys the recording material P conveyed by the image forming drum 21 toward the paper discharging unit 30.

The recording material discharge unit 27 has a cylindrical separation drum 271 and a discharge belt 272. The separation drum 271 separates the recording material P supported on the image forming drum 21 from the outer peripheral surface of the image forming drum 21. Then, the separation drum 271 guides the recording material P to the discharge belt 272 or the recording material reversing portion 28.

The separation drum 271 guides the recording material P to the discharge belt 272 when performing face-up paper discharge in single-sided image formation. Further, the separation drum 271 guides the recording material P to the recording material reversing portion 28 when performing face-down paper ejection and double-sided image formation in single-sided image formation.

The discharge belt 272 is formed in an endless shape like the transport belt 123 of the recording material supply unit 12. The discharge belt 272 is rotatably supported by a plurality of rollers. The discharge belt 272 sends the recording material P delivered by the separation drum 271 to the paper discharge unit 30.

The recording material reversing unit 28 has a plurality of reversing rollers 281,282 and a reversing belt 283. When performing face-down paper ejection, the recording material reversing section 28 reverses the front and back of the recording material P guided by the separation drum 271 and conveys it to the recording material discharging section 27. As a result, the recording material P is conveyed to the paper discharging unit 30 in a state where the surface on which the image formed by the recording material discharging unit 27 is formed faces downward in the vertical direction.

Further, when performing double-sided image formation, the recording material reversing portion 28 reverses the front and back sides of the recording material P guided by the separation drum 271 and conveys it to the outer peripheral surface of the image forming drum 21 again. As a result, the recording material P is conveyed by the image forming drum 21, and again passes through the heating unit 23, the head unit 24, the fixing unit 25, and the image reading unit 26.

The paper ejection unit 30 stores the recording material P sent from the image forming unit 20 by the recording material discharging unit 27. The paper ejection unit 30 has a flat plate-shaped paper ejection tray 31. Then, the paper ejection unit 30 places the recording material P on which the image is formed on the paper ejection tray 31.

[Configuration of inkjet recording device]
Next, a hardware configuration example of the inkjet recording device 1 will be described.
FIG. 3 is a block diagram showing a hardware configuration example of the inkjet recording device 1.

The inkjet recording device 1 includes a control unit 40. The control unit 40 is, for example, a CPU (Central Processing Unit) 41, a RAM (Random Access Memory) 42 used as a work area of the CPU 41, and a ROM (Read Only Memory) 43 for storing a program or the like executed by the CPU 41. And have. Further, the control unit 40 has a storage unit 44 including a hard disk drive (HDD) or the like as a large-capacity storage device. The storage unit 44 stores information for controlling the inkjet recording device 1 of a control program (for example, an image processing program) and various data (for example, a learning model learned by machine learning or the like). Therefore, the ROM 43 and the storage unit 44 record programs, data, and the like necessary for the CPU 41 to operate, and are non-computer readable and store the programs executed by the computer that operates the inkjet recording device 1. It is used as an example of a transient storage medium.

Further, the inkjet recording device 1 includes an image forming drum 21, a recording material discharging unit 27, a recording material reversing unit 28, and the like, a transport unit 51 for transporting the recording material P, an operation display unit 52, and an input / output interface 53. have.

The CPU 41 of the control unit 40 is connected to the heating unit 23, the head unit 24, the fixing unit 25, the image reading unit 26, the RAM 42, the ROM 43, and the storage unit 44 via the system bus 54, respectively, and controls the entire device. Further, the CPU 41 is connected to the transport unit 51, the operation display unit 52, and the input / output interface 53 via the system bus 54, respectively.

The operation display unit 52 is a touch panel composed of a display such as a liquid crystal display (LCD) or an organic ELD (Electro Luminescence Display). The operation display unit 52 displays information about the inkjet recording device 1, such as an instruction menu for the operator, information about the ejection detection operation of the nozzle 244, and information about the acquired image data. Further, the operation display unit 52 is provided with a plurality of keys and has a role as an input unit for receiving input of various instructions, characters, numbers and the like by the operator's key operation. As a result, the operator can operate the inkjet recording device 1 through the operation display unit 52. The operation display unit 52 may be divided into an operation unit and a display unit. In this case, the operator operates the inkjet recording device 1 through the operation unit, and the display unit displays information about the inkjet recording device 1.

The input / output interface 53 is connected to a terminal device 2 such as a PC (personal computer) or a facsimile machine. Then, the input / output interface 53 receives the image data from the terminal device 2. The input / output interface 53 outputs the received image data to the control unit 40. Then, the control unit 40 performs image processing such as shading correction, image density adjustment, and image compression on the image data received from the input / output interface 53, if necessary.

Further, the head unit 24 receives the image data image-processed by the control unit 40, and forms a predetermined image on the recording material P based on the image data. Specifically, the head unit 24 drives the head drive unit 241 to eject ink from the inkjet head 242 to a predetermined position.

The image formed on the recording material P by the head unit 24 is read by the image reading unit 26, and the image data is sent to the control unit 40. Further, when performing the ejection defect determination operation of the nozzle 244, the control unit 40 determines the nozzle 244 in which the ejection defect has occurred based on the image data sent from the image reading unit 26. Then, the control unit 40 corrects the head unit 24 by increasing the amount of ink ejected from the nozzle 244 adjacent to the nozzle 244 in which the ejection defect has occurred, for example.

[Configuration of inkjet recording device]
Next, an example of the functional configuration of the inkjet recording device 1 will be described.
FIG. 4 is a block diagram showing a functional configuration example of the inkjet recording device 1.

The inkjet recording device 1 has a built-in prediction device 60 that determines the printing state requested by the operator and predicts the device state of the inkjet recording device 1. The prediction device 60 includes a control unit 40 and a storage unit 44. In FIG. 3, the storage unit 44 is included in the control unit 40, but in FIG. 4, the control unit 40 and the storage unit 44 are described separately for convenience of explanation. Further, the prediction device 60 may be configured to be detachable from the inkjet recording device 1.

The control unit 40 includes an operation content acquisition unit 61, a request print state determination unit 62, a device state prediction unit 63, and a display output unit 64.
Further, the storage unit 44 includes a history table 71 and a device status table 73. However, the device status table 73 includes the request print status table 72. Therefore, the device state table 73 stores the operation content and the device state in association with each other.

When the operator performs an operation on the inkjet recording device 1 through the operation display unit 52, the operation content acquisition unit 61 acquires the operation content of the operation performed on the inkjet recording device 1 from the operation display unit 52, and records the operation content in the history table. Write to 71. The operation content includes the history including the operation on the inkjet recording device 1 performed by the operator and the degree of operation, as well as the operation result of the inkjet recording device 1 operated by the operation such as the printing state of the inkjet recording device 1. Is done.

The operation history is information in which the operation contents are managed in chronological order. The operation history includes, for example, test chart printing, maintenance operation, proof printing, change of print setting, and the like. Test chart printing includes, for example, defective nozzle detection, unevenness correction, position adjustment, head connection correction, head voltage adjustment, and the like. Further, the maintenance operation includes, for example, cleaning of the head (printing portion), cleaning of the transport portion, and the like. Further, the change of the print setting includes, for example, tone curve correction, print position adjustment, front and back position adjustment, and the like.

The degree of operation includes, for example, a correction amount when the position is adjusted, the number of defective nozzles when a defective nozzle is detected, a change amount when the print position is changed in the print setting, and the like.
Further, the printed state includes, for example, the presence / absence of white / black streaks in the printed matter, the presence / absence of density unevenness, the presence / absence of misalignment of printing, the presence / absence of stains, and the like.

The request print state determination unit 62 reads the history of the operation content from the history table 71 in which the operation content is stored, and determines the request print state requested by the operator through the operation display unit 52. For example, when it is found from the operation history that the same operation has been performed a plurality of times in a predetermined period, the request print state determination unit 62 performs a process of determining the request print state. Therefore, the request print state determination unit 62 refers to the request print state table 72, and sets the print state requested by the operator (referred to as “operator request print state”) associated with the operation content acquired from the history table 71. judge. As shown in FIG. 7, which will be described later, the request print status table 72 stores the operation contents of the operator and the operator request print status in association with each other.

The device state prediction unit 63 predicts the device state of the inkjet recording device 1 based on the operation contents stored in the history table 71. For example, the device state prediction unit 63 performs a process of predicting the device state of the inkjet recording device 1 when it is found from the operation history that the same operation has been performed a plurality of times in a predetermined period. The device state indicates, for example, a state in which the inkjet head 242 or nozzle 244 of the inkjet recording device 1 is distorted, streaks due to poor ejection, density unevenness, or the like is generated. In addition, the device state includes countermeasures that are recommended to be applied to the inkjet recording device 1 according to the device state.

Therefore, the device state prediction unit 63 refers to the device state table 73 and predicts the device state associated with the operation content acquired from the history table 71. Further, the device state prediction unit 63 refers to the device state table 73 and predicts the operation content acquired from the history table 71 and the device state associated with the request print state acquired from the request print state table 72. You can also. As shown in FIG. 8 to be described later, the device status table 73 stores the operator's operation contents recorded in the request print status table 72, the operator request print status, and the device status in association with each other.

The image reading unit 26 analyzes the read image by reading the area where the image is printed on the recording material P and the test chart printed on the recording material P with the in-line sensor. Therefore, the request print state determination unit 62 and the device state prediction unit 63 may use the pixel value (sensing value) of the image read by the image reading unit 26 in each process. Then, the history table 71 may store the sensing value detected by the detection device other than the image reading unit 26. For example, the temperature inside the device detected by the thermometer, the transfer timing, the size and amount of the recording material P supplied to the image forming unit 20, and the like may be stored in the history table 71 as sensing values.

The display output unit 64 outputs the device state predicted by the device state prediction unit 63 to the operation display unit 52 so as to be displayable. The operation display unit 52 displays the predicted device status. As described above, since the device state includes the countermeasures, the operation display unit 52 can also display the countermeasures.

Even if there is only one operator-requested print state, the device state may differ depending on the operation history. For example, even if the streaks cannot be seen from the reading result of the printed matter, the streaks may be seen as a whole when the operator visually recognizes the printed matter. In this case, the operator's request state is "the streaks cannot be seen", but the device states of the inkjet recording apparatus 1 that created the printed matter so that the streaks can be seen are various. Therefore, the operation display unit 52 displays the countermeasures according to the device state predicted based on the operation history and the operator request printing state. For example, when the implementation of head cleaning is presented as a countermeasure, the operator performs head cleaning. If a streak is visible at the joint of the heads, adjustment of the joint position of the heads is presented as a countermeasure. In addition, methods such as changing the printing speed and resolution, changing the amount of ink adhered per unit area, and processing with a dither mask are presented as countermeasures. By implementing these presented countermeasures, the inkjet recording apparatus 1 can create a printed matter that matches the printing state required by the operator and enhance the satisfaction of the operator.

Here, the operation and result of the image forming apparatus 1 when a printing defect occurs will be described.
FIG. 5 is a diagram showing the contents of the history table 71 when a printing defect other than an ink ejection defect occurs. The history table 71 is configured with a table summarizing the operations of the operator or the operation of the inkjet recording device 1 and the results. As an example, the history table 71 stores the history of the operation contents for six times.

For example, in the first operation, it is assumed that the inkjet recording device 1 prints a test chart on the recording material P, detects a defective nozzle, and complements the defective nozzle. With a defective nozzle in which a part of the nozzle 244 shown in FIG. 2 is clogged with ink or contains air bubbles, ink ejection is poor. Therefore, a process for preventing printing chipping is performed by ejecting ink from another nozzle 244 around the defective nozzle at a position where the defective nozzle should originally eject ink. This process is referred to as "complementing defective nozzles" in the following description. In the first operation shown in FIG. 5, since the number of defective nozzles is small, it is shown that the job of printing the test chart has also been completed normally.

In the second operation, as a result of the inkjet recording device 1 executing the job A of several pages and printing the image specified for the job A, it is shown that the job A has been completed normally. However, since the operator is not satisfied with the quality of the image printed on the recording material P, the test chart is printed again.

Therefore, in the third operation, the inkjet recording device 1 prints the test chart again, and in the fourth operation, the inkjet recording device 1 again executes the job A of several pages. Even after the fourth operation, the operator is not satisfied with the quality of the image printed on the recording material P, so the fifth and sixth operations similar to the third and fourth operations are repeated. Since the same operation is repeated many times in this way, it is assumed that the operator does not judge that the quality of the image printed on the recording material P is good. For example, in corporate color printing of a company, even if the processing of the inkjet recording device 1 itself is normally completed, if the operator recognizes slight color unevenness or streaks on the printed matter, the operator determines that the printed matter is of poor quality. I often do it. In this case, the operator-requested print state is a state in which white streaks / black streaks do not occur in the printed matter.

Therefore, the request print state determination unit 62 determines the operator request print state with reference to the history table 71 and the request print state table 72. Then, the device state prediction unit 63 predicts the device state with reference to the device state table 73, so that a device state (for example, occurrence of white streaks / black streaks) indicating that there is a printing defect other than a discharge defect can be obtained. .. For the determination according to the present embodiment, a determination process using an algorithm based on a rule base (a combination of defective nozzle detection and a certain job is N times or more) is used. Then, the device state predicted by the device state prediction unit 63 is output to the operation display unit 52 by the display output unit 64, and the operator can confirm the device state.

Next, the detection result when the position shift occurs for each nozzle will be described.
FIG. 6 is a diagram showing the contents of the history table 71A when the position shift occurs in each nozzle. The history table 71A is a table in which contents different from the history table 71 shown in FIG. 5 are recorded. As an example, the history table 71A stores the history of the operation contents for seven times.

For example, it is assumed that the control unit 40 detects a defective nozzle in the first operation and complements the defective nozzle. At this time, since the number of defective nozzles was small, the print result was not bad and the job was completed normally. As shown in the history table 71A, as a result of performing the second to sixth operations, the jobs are normally completed in each case. However, in the 3rd and 4th operations, job A is printed after head maintenance is performed, and in the 5th and 6th operations, job A is printed after head unit position adjustment is performed. ing. Then, in the seventh operation, the defective nozzle is detected again, and the defective nozzle is complemented.

As shown in the history table 71A, since the position of each head is adjusted at the fifth time, it is presumed from the operation history that there is a problem in ink landing rather than a bad nozzle condition. As a countermeasure, it is necessary to adjust the position of each nozzle or adjust the head voltage. Therefore, if the adjustment method is presented to the operator, the operator can make appropriate adjustments so that printing can be performed with the quality required by the operator without performing extra operations.

The history table 71 shown in FIG. 5 and the history table 71A shown in FIG. 6 are described separately because the contents of defects are different, but the contents of the history tables 71 and 71A are continuous in one history table 71. May be recorded.

FIG. 7 is a diagram showing an example of the request print state table 72. The request print state determination unit 62 determines what the operator request print state is based on the latest N operation histories acquired from the history table 71. Here, the item of defective nozzle detection shown in the history table 71 represents a process automatically executed by the inkjet recording apparatus 1. Items from density correction (cleaning of inlay sensor and white plate, etc.) to proof printing represent processing executed by the operator giving an operation instruction through the operation display unit 52.

In the request print state table 72, the operator request print state is set according to the operation of the inkjet recording device 1 and the operation of the operator. For example, No. As shown in the record 1, when the defective nozzle is detected twice or more and the head connection correction is performed once or more, the operator-requested printing state is "I want to reduce streaks". The head connection correction is a correction process for preventing image quality deterioration due to a joint between inkjet heads 242.

In addition, No. As shown in the record 3, when the density correction is performed twice or more and the proof printing is performed twice or more, the operator-requested printing state is "I want to reduce the density unevenness". In addition, No. As shown in the record of 4, when the position adjustment is performed twice or more and the head connection correction is performed twice or more, the operator-requested printing state is "I want to reduce the color shift".

FIG. 8 is a diagram showing an example of the device state table 73. The device state prediction unit 63 predicts the device state of the inkjet recording device 1 based on the latest N operation histories acquired from the history table 71.

In the device state table 73, the device state of the inkjet recording device 1 is set according to the operation of the inkjet recording device 1, the operation of the operator, and the printing state requested by the operator. For example, No. As shown in the record of 1, when the defective nozzle is detected twice or more and the head connection correction is performed once or more, the operator request printing state is "I want to reduce streaks" and the device state is "Head". White streaks / black streaks occur at the joint position. " Therefore, the display output unit 64 displays the device state predicted by the device state prediction unit 63 on the operation display unit 52. When the operator confirms on the operation display unit 52 that the device state is "white streaks / black streaks generated at the head connection position", the operator can respond according to the device state.

<Modification example of device status table>
Various forms are assumed for the device state table 73. Here, a configuration example of the device state tables 73A to 73B according to the modified example will be described. The device state prediction unit 63 can predict the device state of the inkjet recording device 1 based on the latest N operation histories acquired from the history table 71 in any of the device state tables 73A to 73B. ..

FIG. 9 is a diagram showing a configuration example of the device state table 73A. The device state table 73A is set corresponding to the history table 71 when a printing defect other than the ejection defect shown in FIG. 5 occurs.

The device status table 73A has a configuration in which the device status table 73 shown in FIG. 8 is provided with a countermeasure field. In the present embodiment, it is assumed that the device state also includes countermeasures.
Countermeasure fields store countermeasures according to the device status.

No. As shown in the record 1, when the defective nozzle is detected twice or more, the head connection correction is performed once or more, and the proof printing is performed twice or more, the operator-requested printing state is "reduce streaks". I want to. " Then, since the device state is "white streaks / black streaks occur at the head connection position", as a countermeasure, "(1) prompt detection with a higher threshold for detecting defective nozzles, (2) head cleaning. There are three types: "Prompt, (3) Prompt head replacement". Therefore, the display output unit 64 displays any or all of the three countermeasures stored in the countermeasure field on the screen of the operation display unit 52 as a countermeasure that allows the operator to eliminate the defect of the inkjet recording device 1. Let me. Since the countermeasures to be taken when a printing defect other than the ejection defect occurs in this way, the operator can take an appropriate countermeasure based on the countermeasures.

Note that, for example, adjustment methods and maintenance already performed by the operator may be excluded from the countermeasures stored in the countermeasure field. Further, when the display output unit 64 presents the same countermeasure to the operation display unit 52, the content may be changed and presented so as to raise the threshold value or strengthen the cleaning.

The operator can set and change the sensitivity (including the threshold value) of the item to be determined to be abnormal through the operation display unit 52. That is, the sensitivity when the device state prediction unit 63 predicts the device state is changed to the sensitivity required by the operation display unit 52 based on the countermeasures displayed on the operation display unit 52. For example, the operator may change the setting so as to lower the sensitivity for posters and the like in which color unevenness and streaks are likely to cause quality abnormalities, and to increase the sensitivity for printed matter containing only characters in which color unevenness and the like are unlikely to occur. Further, for example, in the operation history stored in the history table 71 shown in FIG. 5, the head connection position may shift or the operator may request high image quality, so that the operator raises the detection threshold value. You may take action.

FIG. 10 is a diagram showing a configuration example of the device state table 73B. The device state table 73B is a table set corresponding to the history table 71A when the position shift occurs in units of nozzles shown in FIG. The device status table 73B also has a configuration in which the device status table 73 shown in FIG. 8 is provided with a countermeasure field.

No. As shown in the record of 1, when the defective nozzle is detected twice or more, the head unit position adjustment is performed once or more, the head cleaning is performed once or more, and the proof printing is performed twice or more. , The operator request printing state is "I want to reduce the misalignment on the transport side". Then, the device state is "a landing position shift occurs at a specific nozzle (group)". Therefore, the countermeasures are "(1) prompting the head voltage adjustment, (2) prompting the position adjustment of the head nozzle unit, (3) prompting the cleaning of the image reading unit, and (4) prompting the head replacement" 4 There is one. Therefore, the display output unit 64 displays any or all of the four countermeasures stored in the countermeasure field on the screen of the operation display unit 52 as a countermeasure that allows the operator to eliminate the defect of the inkjet recording device 1. Display it. Since the countermeasures when the positional deviation occurs for each nozzle are shown in this way, the operator can take appropriate countermeasures based on the countermeasures.

Next, the device state determination process performed by the control unit 40 will be described.
FIG. 11 is a flowchart showing an example of device state determination processing performed by the control unit 40.

First, the operation content acquisition unit 61 of the control unit 40 waits for an operation by the operator (S1). When the operator performs an operation through the operation display unit 52 (S2), the operation content acquisition unit 61 acquires the operation content from the operation display unit 52, and stores the operation content and the operation result of the inkjet recording device 1 in the history table 71. (S3).

Next, the request print state determination unit 62 refers to the history table 71 and the request print state table 72, and determines whether or not the operation content and the operation result satisfy the operator request print state (S4). In the request print state determination unit 62, for example, the operation content and operation result such as the number of times of detecting a defective nozzle and the number of times of head connection correction are linked to the operator request print state specified in the request print state table 72. Determine if the result is met. If the operation content and the operation result do not satisfy the operator-requested printing state (NO in S4), the process returns to step S1 and the process is repeated.

If the operation content and the operation result satisfy the operator request print state (YES in S4), the request print state determination unit 62 outputs the operator request print state to the device state prediction unit 63. The device state prediction unit 63 predicts the device state corresponding to the operator-requested print state with reference to the device state table 73, stores the predicted device state in the RAM 42 (S5), and ends this process.

Next, a process of presenting a countermeasure to the operation display unit 52 will be described.
FIG. 12 is a flowchart showing an example of a process in which the control unit 40 presents a countermeasure to the operation display unit 52.

First, the device state prediction unit 63 stores the device state predicted in accordance with the operation content and operation result and the operator-requested print state in the RAM 42 (S11). However, if the device state is already stored in the RAM 42 in step S5 of FIG. 11, the process of step S11 is skipped.

Next, the device state prediction unit 63 refers to the device state table 73 and determines whether or not the operator can solve the unfavorable output result (S12). For example, the device state prediction unit 63 refers to the countermeasure field corresponding to the device status field of the device status table 73, stores the countermeasure that can be solved by the operator in the countermeasure field, and confirms whether the countermeasure can be solved by the operator. ..

If the operator can solve the problem (YES in S12), the display output unit 64 determines the content to be displayed on the operation display unit 52. Then, the display output unit 64 displays the determined content on the screen of the operation display unit 52 (S13), and ends this process.

On the other hand, if the operator cannot solve the problem (NO in S12), the device state prediction unit 63 directly notifies the service center of the defect of the inkjet recording device 1, for example. Alternatively, the display output unit 64 determines the content to be displayed on the operation display unit 52. Then, the display output unit 64 displays a message prompting the contact to the service center on the screen of the operation display unit 52 (S14), and ends this process.

The prediction device 60 according to the first embodiment described above is an inkjet recording device 1 for the operator based on a combination of a plurality of actions taken by the operator even if the inkjet recording device 1 itself is not determined to be abnormal. It is determined that some unfavorable state such as performance deterioration or deterioration has occurred. At this time, the request print state determination unit 62 of the control unit 40 determines the operator request print state, and the device state prediction unit 63 predicts the device state. However, the device state prediction unit 63 can also predict the device state directly from the operation content. Then, the display output unit 64 displays the device state predicted by the device state prediction unit 63 or the countermeasures on the operation display unit 52. Therefore, the operator can know the current state of the inkjet recording device 1 according to the displayed device state. In addition, the operator can change the inkjet recording device 1 to a preferable state by operating the inkjet recording device 1 according to the displayed countermeasures.

Conventionally, even if the inkjet recording device is not determined to be abnormal, if the operator determines that the inkjet recording device 1 is in an unfavorable state due to performance deterioration, output deterioration, etc., the inkjet recording device is the cause. It could not be identified and no countermeasures could be shown. However, in the inkjet recording device 1 according to the present embodiment, the actions (maintenance, various adjustments) taken by the operator are stored in the history table 71 as an operation history. The device state prediction unit 63 can predict the device state not only from the operation history for the abnormal event stored in the history table 71 but also from the operation history loved for the normal event. Therefore, the display output unit 64 can present a countermeasure to the operator based on the device state predicted by the device state prediction unit 63.

The display output unit 64 may display only the device state predicted by the device state prediction unit 63 on the operation display unit 52. When such a display is made, if the device state recognized by the operator and the device state predicted by the prediction device 60 are different, the operator can easily notice an error in his / her own judgment, which is effective. It is also possible to take various measures.

Further, in the conventional inkjet recording device, parts cannot be inspected or replaced until an actual abnormality occurs. After an abnormality occurred, the time for checking and replacing parts tended to be longer when the paper and equipment could not be used. However, in the inkjet recording device 1 according to the present embodiment, it is possible to notify the operator of the device state before an actual abnormality occurs. Therefore, the operator can easily grasp the state of the inkjet recording device 1 before an abnormality requiring maintenance occurs in the inkjet recording device 1. Therefore, since the operator can take appropriate measures for the inkjet recording device 1 before the abnormality becomes large, the abnormality that actually occurs can be suppressed to a small size, and the time for inspection and replacement of parts can be reduced.

Further, when the operator is not satisfied with the execution result of the job A executed by the inkjet recording apparatus 1, for example, another job B may be executed once, and then the job A may be executed again. Even when a plurality of different jobs are executed consecutively, all the operation histories are stored in the history table 71, so that the request print state determination unit 62 and the device state prediction unit 63 execute. The history can be extracted from the history table 71 for each job to be performed, and the request print state can be determined and the device state can be predicted.

[Forecasting system configuration example]
Next, the prediction system according to the modified example of the first embodiment will be described. The amount of data in the history table 71, the request print state table 72, and the device state table 73 according to the first embodiment described above becomes enormous. Therefore, these tables may be configured as a prediction system provided with a management server capable of connecting these tables to the inkjet recording device 1 via the communication network N.
FIG. 13 is a block diagram showing a configuration example of the prediction system 100.

The prediction system 100 includes an inkjet recording device 1 and a management server 101 that is communicably connected to the inkjet recording device 1 via a communication network N such as a LAN (Local Area Network) or the Internet. It is assumed that a plurality of inkjet recording devices 1 are connected to the management server 101.

The management server 101 includes a storage unit 110 configured in an HDD or the like capable of recording a large amount of data. The storage unit 110 includes a history table 111 and a device status table 113. However, the device status table 113 includes the request print status table 112. The data recorded in the history table 111, the request print status table 112, and the device status table 113 are the same as the data recorded in the history table 71, the request print status table 72, and the device status table 73 shown in FIG. 4, respectively. Is.

Further, the operation content acquisition unit 61, the request print state determination unit 62, the device state prediction unit 63, and the display output unit 64 of the control unit 40 of the inkjet recording device 1 according to the modification are the history table 111 and the request print state table 112, respectively. The process of accessing the device state table 113 and writing or reading necessary data is the same as the process described with reference to FIGS. 4 to 12.

By configuring such a prediction system 100, the inkjet recording device 1 does not need to store a huge amount of data in its own device. Further, the management server 101 makes it possible to refer to the request print state table 112 and the device state table 113 when the control unit 40 of the other inkjet recording device 1 accesses the storage unit 110. As a result, even the control unit 40 of the other inkjet recording device 1 determines the already set operator-requested printing state by accessing the management server 101, predicts the device state, and presents a countermeasure. It becomes possible.

[Second Embodiment]
Next, the inkjet recording apparatus according to the second embodiment of the present invention will be described. The control unit of the inkjet recording device according to the first embodiment described above determines various states with reference to a table in which values are set in advance, but controls of the inkjet recording device according to the second embodiment. The unit determines various states by machine learning.

First, an overview of machine learning will be given.
FIG. 14 is a diagram showing an example of machine learning.

In machine learning, an input layer, an intermediate layer and an output layer are used. Data X _t of the time series input to the input layer are outputted to the intermediate layer. In the intermediate layer, for example, a learning model using RNN (Recurrent Neural Network), GRU (Gated Recurrent Unit), and LSTM (Long Short Term Memory) is created. At this time, the data Y _t-1 is repeatedly calculated inside the intermediate layer. Then, the data Y _t-1 output from the intermediate layer is output to the output layer. This data Y _t-1 is used as a result.

FIG. 15 is a block diagram showing a functional configuration example of the inkjet recording apparatus 1A according to the second embodiment.

The inkjet recording device 1A incorporates a prediction device 60A that determines the printing state requested by the operator and predicts the device state. The prediction device 60A includes a control unit 40 and a storage unit 44. The prediction device 60A may be configured to be removable from the inkjet recording device 1A.

The control unit 40A includes an operation content acquisition unit 61, a request print state determination unit 62A, a device state prediction unit 63A, and a display output unit 64.
The storage unit 44A includes a history table 81, a device state table 83, a request print state learning model 84, and a device state learning model 85. The device status table 83 includes a request print status table 82.

Similar to the first embodiment, the operation content acquisition unit 61 acquires the operation content of the operator from the operation display unit 52 and writes the operation content in the history table 81. The history of the operation contents recorded in the history table 81 is the same as that of the history table 71 (see FIG. 4) according to the first embodiment.

The request print state determination unit 62A reads the history of the operation content from the history table 81 in which the operation content is stored, and determines the request print state requested by the operator through the operation display unit 52. For example, the request print state determination unit 62A performs a process of determining the request print state when it is found from the operation history that the same operation has been performed a plurality of times in a predetermined period. The request print state determination unit 62A writes the request print state learning model 84 created by machine learning that inputs the operation content and outputs the request print state to the storage unit 44A. Then, the request print state determination unit 62A determines the request print state corresponding to the operation content acquired from the history table 81 by using the request print state learning model 84 read from the storage unit 44A.

The request print state table 82 is a table configured to explain the operation content that is input to the request print state learning model 84 and the request print state that is output. As shown in FIG. 16 described later, the request print status table 82 is represented by associating the operator's operation history with the operator request print status.

The device state prediction unit 63A predicts the device state based on the operation content acquired from the history table 81 and the request print state stored in the request print state table 82. For example, the device state prediction unit 63A performs a process of predicting the device state when it is found from the operation history that the same operation has been performed a plurality of times in a predetermined period. The device state prediction unit 63A writes the device state learning model 85 created by machine learning, which inputs the operation content and outputs the device state, to the storage unit 44A. Then, the device state prediction unit 63A predicts the device state corresponding to the operation content acquired from the history table 81 by using the device state learning model 85 read from the storage unit 44A.

The device state table 83 is a table configured to explain the operation content and the required print state input to the device state learning model 85, and the device state to be output. As shown in FIG. 17, which will be described later, the device state table 83 is represented by associating the operation history of the operator, the operator-requested printing state, and the device state of the inkjet recording device 1.

The display output unit 64 determines the display content to be displayed on the operation display unit 52 based on the determination result of the device state. The determined display content is displayed on the operation display unit 52.

FIG. 16 is a diagram showing an example of the request print state table 82.

The request print state determination unit 62A creates a request print state table 82 in which the operation history of the operator and the operator request print state are linked as time series data for creating the request print state learning model 84. The operator request print state is set in advance as teacher data. Therefore, in machine learning, after the operator operates the inkjet recording device 1 several times, it is necessary for the operator to set, for example, that the operator-requested printing state is "want to reduce streaks". .. Each time these operations are performed, the request print state determination unit 62A inputs the operation content to the input layer, performs machine learning, and creates the request print state learning model 84.

When the machine learning is completed, the request print state determination unit 62A uses the request print state learning model 84, and what is the operator request print state based on the latest N operation histories acquired from the history table 81. Can be determined. At this time, the request print state determination unit 62A refers to the request print state table 82.

It will be described as assuming that the request print status table 82 stores the time when the operation for a maximum of 5 times is performed, the operation history, and the operator request print status.
No. As shown in the record 1, defective nozzle detection is performed after 0 minutes, which represents the time at the moment when the first operation is performed, and 4 minutes later, defective nozzle detection is performed as the second operation. Further, 7 minutes later, the head connection position correction is performed as the third operation, 10 minutes later, proof printing is performed as the fourth operation, and 12 minutes later, proof printing is performed as the fifth operation. Then, the request print state determination unit 62A is No. It can be seen that the operator request print state of one record is "I want to reduce streaks".

In addition, No. The operation history shown in the record 2 is No. It is different from the operation history shown in the record 1. However, the request print state determination unit 62A can create the request print state learning model 84 to obtain the No. Like the record of No. 1, No. It can be seen that the operator request print status of the record 2 is "I want to reduce streaks".

After that, when the same operation history is recorded in the history table 81, the request print state determination unit 62A uses the data combining the operator's operation history and the operator request print state to create the request print by machine learning. The state learning model 84 can be used to determine the operator-requested print state.

FIG. 17 is a diagram showing an example of the device state table 83.

The device state prediction unit 63A creates a device state table 83 in which the operator's operation history, the operator request print state, and the device state are linked as time-series data for creating the device state learning model 85. When the device state prediction unit 63A performs machine learning, when the operator operates the inkjet recording device 1 several times and the operator request printing state is "want to reduce streaks", for example, the device state is ". It is necessary for the operator to set that "white streaks / black streaks occur at the head connection position". Each time these operations are performed, the device state prediction unit 63A inputs the operation content and the operator request print state to the input layer, performs machine learning, and creates the device state learning model 85.

When the machine learning is completed, the device state prediction unit 63A uses the device state learning model 85, and based on the latest N operation histories acquired from the history table 81, what is the device state of the inkjet recording device 1A. Predict. At this time, the device state prediction unit 63A refers to the request print state table 82 and the device state table 83.

The device status table 83 will be described as storing the times when the operations for up to four times are performed, the operation history, the operator request print status, and the device status.
No. The 1st to 4th time and operation history shown in 1 record and the operator request print state are the 1st to 4th time and operation history shown in the request print state table 82 of FIG. 16 and the operator request print state. The same is true. Then, the device state prediction unit 63A is No. It can be seen that the device state of the record 1 is "white streaks / black streaks generated at the head connection position".

In addition, No. The operation history shown in the record 2 is No. It is different from the operation history shown in the record 1. However, the device state prediction unit 63A can create the device state learning model 85 to obtain the No. Like the record of No. 1, No. It can be seen that the device state of the record 2 is "white streaks / black streaks occur at the head connection position".

After that, when the same operation history is recorded in the history table 81, the device state prediction unit 63A creates the device state prediction unit 63A by machine learning using the data combining the operator operation history, the operator request print state, and the device state. The device state learning model 85 can be used to predict the device state.

Therefore, the display output unit 64 displays the device state predicted by the device state prediction unit 63A on the operation display unit 52. When the operator confirms on the operation display unit 52 that the device state is "white streaks / black streaks generated at the head connection position", the operator can respond according to the device state.

<Modification example of device status table>
Various forms are assumed for the device state table 83. Here, a configuration example of the device state table 83A according to the modified example will be described. When the control unit 40 predicts the device state, the device state can be predicted by using the parameters of the inkjet recording device 1 in addition to the operator's operation history and the operator-requested printing state.
FIG. 18 is a diagram showing a configuration example of the device state table 83A.

The device state table 83A has a configuration in which the device state table 73 shown in FIG. 8 is provided with a head temperature field.
In the head temperature field, for example, three head temperatures of standard, standard + 5 ° C. or higher, and standard −5 ° C. or lower are stored. The head temperature is an example of a device parameter representing the state quantity of each part included in the inkjet recording device 1A. Other device parameters include the number of missing nozzles when a defective nozzle was detected last time, the temperature and pressure of the head, and the pressure of the degassing module. The degassing module is a device for degassing the air entering the nozzle 244 of the inkjet head 242 shown in FIG.

No. As shown in the record 1, when the defective nozzle is detected twice or more and the head cleaning is performed twice or more, the operator-requested printing state is "I want to reduce streaks". Here, if the head temperature is standard, the device state is "streak generation due to nozzle bending". No. 2 and No. In all three records, the number of defective nozzles detected, the number of head cleanings, and the operator-requested printing status are No. It is the same as the record of 1. However, No. In the second record, the head temperature is standard + 5 ° C. or higher, and the device state is "streak generation due to nozzle bending". On the other hand, No. In the record of 3, when the head temperature is standard −5 ° C. or lower, the device state is “streak generation due to decrease in discharge temperature”. Since the device state changes depending on the head temperature in this way, the countermeasures may also differ. Therefore, the display output unit 64 displays the device state on the operation display unit 52, so that the operator can grasp the device state and take an appropriate action.

In the inkjet recording apparatus 1A according to the second embodiment described above, the request print state determination unit 62A determines the request print state from the operation history using the request print state learning model 84. Further, the device state prediction unit 63A uses the device state learning model 85 to predict the device state from the operation history and the requested print state. The required print state learning model 84 and the device state learning model 85 are created by machine learning. Therefore, even if the operator does not define the contents of the request print state table 82 and the device state table 83 in advance, the contents of each table are recorded almost automatically. Then, when the operator recognizes that the inkjet recording device 1A is in an unfavorable state, the control unit 40A can predict the operator-requested printing state and the device state.

As in the first embodiment, the device status table 83 may store countermeasures according to the device status determined by the operation history and the requested print status. Therefore, the display output unit 64 can display the countermeasure on the operation display unit 52, and the operator can confirm the countermeasure and take an appropriate countermeasure.

[Forecasting system configuration example]
Next, the prediction system according to the modified example of the second embodiment will be described. The amount of data in the history table 81, the request print state table 82, and the device state table 83 according to the second embodiment described above becomes enormous. Therefore, these tables may be provided on a management server that can be connected to the inkjet recording device 1A via the communication network N.
FIG. 19 is a block diagram showing a configuration example of the prediction system 100A.

The prediction system 100A includes an inkjet recording device 1A and a management server 101A that is communicably connected to the inkjet recording device 1A via a communication network N such as a LAN or the Internet. It is assumed that a plurality of inkjet recording devices 1A are connected to the management server 101A.

The management server 101A includes a storage unit 130 configured in an HDD or the like capable of recording a large amount of data. The storage unit 130 includes a history table 131, a device state table 133, a request print state learning model 134, and a device state learning model 135. The device status table 133 includes the request print status table 132. The data recorded in the history table 131, the request print status table 132, and the device status table 133 are the same as the data recorded in the history table 81, the request print status table 82, and the device status table 83 shown in FIG. 15, respectively. Is.

Further, the operation content acquisition unit 61, the request print state determination unit 62A, the device state prediction unit 63A, and the display output unit 64 of the control unit 40A of the inkjet recording device 1A according to the modification are the history table 131 and the request print state table 132, respectively. The process of accessing the device status table 133 and writing or reading necessary data is the same as the process described with reference to FIGS. 15 to 18. Further, the process of machine learning by the request print state determination unit 62A using the request print state learning model 134 and the process of machine learning by the device state prediction unit 63A using the device state learning model 135 are both shown in FIG. This is the same as the processing of the request print state determination unit 62A and the device state prediction unit 63A.

By configuring such a prediction system 100A, the inkjet recording device 1A does not need to store a huge amount of data in its own device. Further, the management server 101A makes it possible to refer to the request print state table 132 and the device state table 133 when the control unit 40A of the other inkjet recording device 1A accesses the storage unit 130. Therefore, even the control unit 40A of the other inkjet recording device 1A determines the already set operator-requested printing state by accessing the management server 101A, predicts the device state, and presents a countermeasure. It becomes possible.

In the first and second embodiments described above, the inkjet recording devices 1 and 1A are mentioned as an example of the image forming device, but an electrophotographic image forming method in which toner is adhered to a recording material to form an image. The present invention may be applied to an apparatus, an offset printing machine, or the like.

Further, the prediction devices 60 and 60A may transmit the device state predicted by the device state prediction units 63 and 63A to an external support center or the like. Then, the maintenance staff of the support center can grasp the device state of the inkjet recording devices 1, 1A, and perform maintenance, parts replacement, and the like before an abnormality actually occurs in the inkjet recording devices 1, 1A.

Further, the present invention is not limited to the above-described embodiments, and it goes without saying that various other application examples and modifications can be taken as long as the gist of the present invention described in the claims is not deviated.
For example, each of the above-described embodiments describes in detail and concretely the configurations of the apparatus and the system in order to explain the present invention in an easy-to-understand manner, and is not necessarily limited to those including all the described configurations. Further, it is possible to replace a part of the configuration of the embodiment described here with the configuration of another embodiment, and further, it is possible to add the configuration of another embodiment to the configuration of one embodiment. It is possible. It is also possible to add, delete, or replace a part of the configuration of each embodiment with another configuration.
In addition, the control lines and information lines indicate those that are considered necessary for explanation, and do not necessarily indicate all the control lines and information lines in the product. In practice, it can be considered that almost all configurations are interconnected.

1 ... Inkjet recording device, 20 ... Image forming unit, 26 ... Image reading unit, 40 ... Control unit, 44 ... Storage unit, 52 ... Operation display unit, 60 ... Prediction device, 61 ... Operation content acquisition unit, 62 ... Request printing Status determination unit, 63 ... Device status prediction unit, 64 ... Display output unit, 70 ... Storage unit, 71 ... History table, 72 ... Request print status table, 73 ... Device status table, 84 ... Request print status learning model, 85 ... Device state learning model

Claims (13)

An operation content acquisition unit that acquires the operation content of the operation performed on the image forming apparatus from the operation unit, and an operation content acquisition unit.
A device state prediction unit that predicts the device state of the image forming device based on the operation content, and
A prediction device including a display output unit that outputs the predicted device state to the display unit in a displayable manner. The device state includes countermeasures taken by the image forming device according to the device state.
The prediction device according to claim 1, wherein the operation content includes an operation result of the image forming apparatus operated by the operation. A device status table for storing the operation content and the device status in association with each other is provided.
The prediction device according to claim 2, wherein the device state prediction unit predicts the device state associated with the operation content by referring to the device state table. It is provided with a request print state determination unit that reads the history of the operation content from the history table in which the operation content is stored and determines the request print state requested by the operator through the operation unit.
The prediction device according to claim 3, wherein the device state prediction unit predicts the device state based on the operation content and the required print state. A request print status table for storing the operation content and the request print status in association with each other is provided.
In the device state table, the operation content recorded in the request print state table, the request print state, and the device state are recorded in association with each other.
The request print state determination unit determines the request print state associated with the operation content with reference to the request print state table.
The prediction device according to claim 4, wherein the device state prediction unit predicts the device state associated with the operation content and the required print state with reference to the device state table. The sensitivity when the device state prediction unit predicts the device state is changed to the sensitivity required by the operation unit based on the countermeasure displayed on the display unit. The prediction device according to one item. It is equipped with a device state learning model created by machine learning that inputs the operation content and outputs the device state.
The prediction device according to claim 2, wherein the device state prediction unit predicts the device state corresponding to the operation content by using the device state learning model. It is provided with a request print state determination unit that reads the history of the operation content from the history table in which the operation content is stored and determines the request print state requested by the operator through the operation unit.
The prediction device according to claim 7, wherein the device state prediction unit predicts the device state based on the operation content and the required print state by using the device state learning model. It is equipped with a request print state learning model created by machine learning that inputs the operation content and outputs the request print state.
The prediction device according to claim 8, wherein the required print state determination unit determines the required print state corresponding to the operation content by using the required print state learning model. The prediction device according to claim 8, wherein the device state prediction unit further predicts the device state using the parameters of the image forming device. An image forming part that forms an image on the recording material,
An operation unit that operates the image forming device and
A display unit that displays information about the image forming apparatus and
A prediction device for predicting the device state of the image forming device is provided.
The prediction device is
An operation content acquisition unit that acquires the operation content of the operation performed on the image forming apparatus from the operation unit, and an operation content acquisition unit.
A device state prediction unit that predicts the device state based on the operation content,
An image forming apparatus having a display output unit that outputs the predicted device state to the display unit in a displayable manner. The procedure for acquiring the operation content of the operation performed on the image forming apparatus from the operation unit, and
A procedure for predicting the device state of the image forming apparatus based on the operation content, and
A program for causing a computer constituting the prediction device to execute a procedure for outputting the predicted device status to a display unit so that it can be displayed. An image forming apparatus and a management server to which the image forming apparatus is connected via a network are provided.
The image forming apparatus
An operation content acquisition unit that acquires the operation content of the operation performed on the image forming apparatus from the operation unit, and an operation content acquisition unit.
A device state prediction unit that predicts the device state of the image forming device based on the operation content, and
It has a display output unit that outputs the predicted device state to the display unit so that it can be displayed.
The management server
It has a history table that stores the operation contents acquired by the operation content acquisition unit as a history.
The device state prediction unit is a prediction system that predicts the device state with reference to the history table.

Images