Image forming apparatus, image forming system, and control program

The image forming apparatus optimizes paper conveyance by using read image analysis and machine learning to adjust speed, addressing defects and damage issues across diverse paper types.

JP7582038B2Active Publication Date: 2024-11-13KONICA MINOLTA INC
View PDF 7 Cites 0 Cited by

Patent Information

Application Number
JP2021069591
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Filing Date
2021-04-16
Publication Date
2024-11-13
Estimated Expiration
2041-04-16

AI Technical Summary

Technical Problem

Existing image forming apparatuses struggle to maintain optimal paper conveyance under varying conditions, leading to issues such as transfer misalignment, image defects, and paper damage due to inadequate loop detection and speed control, particularly when using diverse paper types.

Method used

The apparatus incorporates a transfer unit, fixing unit, conveyance speed control, and read image acquisition to analyze defects and damage, adjusting conveyance speed based on read image data and machine learning models to optimize paper transport between the transfer and fixing sections.

Benefits of technology

This approach effectively reduces image defects and paper damage by dynamically adjusting conveyance speed, ensuring proper paper posture and alignment, even under unforeseen conditions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007582038000002
    Figure 0007582038000002
  • Figure 0007582038000003
    Figure 0007582038000003
  • Figure 0007582038000004
    Figure 0007582038000004
Patent Text Reader

Abstract

To appropriately perform sheet conveyance between a transfer unit and a fixing unit to reduce the occurrence of an image defect and damage to a sheet.SOLUTION: An image forming apparatus 10 comprises: a transfer unit 142 that transfers a toner image to a sheet; a fixing unit 143 that fixes the toner image on the sheet by heating the toner image; a conveyance speed control unit 112 for controlling the conveyance speed of the sheet in the transfer unit 142 and / or the fixing unit 143; a read image acquisition unit 19 that acquires read image data on the sheet generated by a reading unit 24 reading an image of the sheet on the downstream side of the fixing unit 143 in a conveyance direction; and a conveyance speed setting unit 112 that changes or updates the conveyance speed based on an image of the acquired read image data.SELECTED DRAWING: Figure 5B
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present invention relates to an image forming apparatus, an image forming system, and a control program. [Background technology]

[0002] In the commercial color printing industry, electrophotographic image forming devices are widely used. In the field of PP (production print) corresponding to the color printing industry, there is a demand for adaptability to a wider variety of paper types than those used in offices. In order to perform high-quality printing on these various types of paper, there are image forming devices that set the transport conditions according to the type of paper used and perform printing.

[0003] However, the combinations of paper types and usage conditions are enormous, and a great deal of development work is required to design control that assumes every possible combination. Conventionally, manual design has been relied upon, and control design is performed so that paper transport can be performed within a normal range under certain conditions, i.e., worst conditions and representative conditions, among the possible combination conditions. However, with this method, there is a risk that optimal transport will not be performed under unexpected usage conditions. For example, if there is a difference in the transport speed between the transfer unit and the fixing unit, the paper may be pulled or an excessive loop may be formed, resulting in poor image quality such as transfer misalignment and image rubbing, or paper damage such as paper wrinkles and scratches.

[0004] In the image forming device disclosed in Patent Document 1, a loop detection sensor that detects paper loops is placed between the transfer unit and the fixing unit, and the amount of paper loop is controlled to be constant by correcting the speed of the drive motor of the fixing unit based on the detection results at predetermined time intervals. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2008-158076 A Summary of the Invention [Problem to be solved by the invention]

[0006] However, in the image forming device of Patent Document 1, one loop detection sensor measures the loop of the paper at one measurement position, so the posture of the paper at positions other than the measurement position cannot be detected. Although the loop (the passing height of the paper on the transport path) at the measurement position can be detected, the correct posture of the paper cannot be maintained in the transport path at other positions, so there is still a risk of poor image quality or paper damage.

[0007] The present invention has been made in consideration of the above-mentioned circumstances, and aims to provide an image forming apparatus, an image forming system, and a control program that can properly transport paper between a transfer unit and a fixing unit, thereby reducing the occurrence of image defects and paper damage. [Means for solving the problem]

[0008] The above object of the present invention can be achieved by the following means.

[0009] (1) a transfer unit that transfers a toner image formed on an image carrier by an imaging unit onto a sheet of paper; a fixing section for fixing the toner image on the paper by heating; a conveying speed control unit that controls a conveying speed of the paper in the transfer unit and / or the fixing unit; a read image acquisition unit that is located downstream of the fixing unit in a conveying direction and that acquires read image data of the paper generated by a reading unit that reads an image of the paper; a conveying speed setting unit that changes or updates the conveying speed based on an image of the acquired read image data.

[0010] (2) The image forming apparatus according to (1) above, wherein the conveying speed setting unit analyzes image defects and / or paper damage as the image analysis.

[0011] (3) the image defect includes at least one of image rubbing, transfer misalignment, color misalignment, image expansion / contraction, gloss unevenness, and image staining; The image forming apparatus according to (2) above, wherein the paper damage includes at least one of paper wrinkles, waviness, scratches, and folds.

[0012] (4) The image forming apparatus according to (2) or (3) above, wherein the transport speed setting unit analyzes the image defects and the paper damage by comparing with a pre-stored image pattern.

[0013] (5) An image forming device described in any of (2) to (4) above, wherein the conveying speed setting unit changes the speed control information based on the presence or absence of the image defect and / or the paper damage, or the position at which the image defect and / or the paper damage occurred on the paper.

[0014] (6) An image forming apparatus according to any one of (1) to (5) above, wherein the conveying speed control unit controls the conveying speed based on speed control information for controlling the drive speed of the drive unit of the transfer unit and / or the fixing unit, the speed control information being stored in advance in a memory unit.

[0015] (7) An image forming apparatus as described in (6) above, wherein the speed control information is set to information for controlling the drive speed of the drive unit depending on the transport position within the paper relative to the transfer unit and / or the fixing unit.

[0016] (8) An image forming apparatus as described in (6) or (7) above, wherein the speed control information is set as a speed control table in which a speed control value of the drive unit for each transport position within the paper relative to the transfer unit and / or the fixing unit is set.

[0017] (9) the read image acquisition unit acquires the physical properties of the paper measured by a media detection device; An image forming apparatus according to any one of (1) to (8) above, wherein the conveying speed control unit controls the drive unit of the transfer unit and / or the fixing unit using speed control information corresponding to the physical properties of the paper, the speed control information being stored in a memory unit.

[0018] (10) An image forming apparatus as described in any one of (1) to (9) above, wherein in a paper passing mode in which images are formed on both sides of a sheet of paper, the conveying speed control unit controls the drive units of the transfer unit and / or the fixing unit using speed control information corresponding to each of the front and back sides of the sheet of paper, which is stored in a memory unit.

[0019] (11) The image forming apparatus according to any one of (1) to (10), wherein a conveying speed control unit uses speed control information to control a relative speed between the transfer unit and the fixing unit.

[0020] (12) The image forming apparatus according to any one of (1) to (10) above, wherein the conveying speed control unit uses speed control information to keep the speed of the transfer unit constant and to control the speed of the fixing unit.

[0021] (13) a transport position in the paper with respect to the transfer unit and / or the fixing unit, a speed control value of the drive unit for each of the transport positions, and The image forming apparatus according to (8) above, wherein the conveying speed setting unit determines speed control information using a learning model generated by machine learning based on the analysis results of the image of the read image data.

[0022] (14) The image forming apparatus according to (13) above, further comprising machine learning based on information on the physical properties of the paper.

[0023] (15) The image forming apparatus according to (14) above, wherein the physical properties include at least one of stiffness, moisture content, and grain direction of the paper.

[0024] (16) During machine learning of the learning model, The image forming device according to any one of (13) to (15) above, which analyzes the image for image defects and paper damage, and evaluates the occurrence of an image defect or paper damage as abnormal if it is detected, and evaluates the occurrence of an image defect or paper damage as normal if it is not detected, and gives a positive reward if the evaluation is normal, and a negative reward if the evaluation is abnormal.

[0025] (17) Further, a loop detection sensor is provided in the conveying path between the transfer unit and the fixing unit to detect a loop of the paper, The image forming apparatus according to any one of (13) to (16) above, further comprising machine learning based on the output of the loop detection sensor.

[0026] (18) An image forming apparatus according to any one of (1) to (17) above; a reading unit that is located downstream in a conveying direction of the image forming apparatus and that reads an image on a sheet of paper formed by the image forming apparatus; An image forming system comprising:

[0027] (19) A control program for controlling a paper conveying speed in an image forming apparatus including a transfer unit that transfers a toner image formed on an image carrier by an image creating unit onto a paper sheet, and a fixing unit that heats the toner image on the paper sheet to fix it, the control program comprising: a first step of acquiring read image data of the paper, the read image data being generated by a reading unit that reads an image of the paper, downstream of the fixing unit in a conveying direction; and a second step of changing or updating the conveying speed based on an image of the read image data acquired in the first step.

[0028] (20) The control program according to (19) above, further comprising: controlling the conveying speed based on speed control information for controlling a drive speed of a drive unit of the transfer unit and / or the fixing unit.

[0029] (21) The control program according to (20) above, wherein the speed control information is set with information for controlling the drive speed of the drive unit depending on the transport position within the paper relative to the transfer unit and / or the fixing unit.

[0030] (22) The control program described in (20) or (21) above, wherein the speed control information is set as a speed control table in which a speed control value of the drive unit for each transport position within the paper relative to the transfer unit and / or the fixing unit is set.

[0031] (23) The control program described in (22) above, in which in the second step, the speed control information is determined using a learning model generated by machine learning based on the paper transport position relative to the transfer unit and / or the fixing unit, the speed control value of the drive unit for each transport position, and the analysis results of the image of the read image data. Effect of the Invention

[0032] The image forming apparatus according to the present invention includes a transfer unit that transfers a toner image formed on an image carrier by an image creating unit onto a sheet of paper, and a fixing unit that fixes the toner image on the sheet of paper by heating the toner image, The image forming apparatus includes a conveying speed control unit that controls the conveying speed of the paper in the transfer unit and / or the fixing unit, a read image acquisition unit that acquires read image data of the paper generated by a reading unit that reads an image of the paper downstream of the fixing unit in the conveying direction, and a conveying speed setting unit that changes or updates the conveying speed based on an image of the acquired read image data. This makes it possible to appropriately convey the paper between the transfer unit and the fixing unit and reduce the occurrence of image defects and paper damage. [Brief description of the drawings]

[0033] [Figure 1] FIG. 1 is a schematic diagram illustrating a configuration of an image forming system. [Diagram 2] FIG. 1 is a block diagram of an image forming system. [Diagram 3]5A and 5B are schematic diagrams showing start and end timings of paper transport control. [Figure 4] 13A and 13B are schematic diagrams illustrating sections corresponding to positions in the transport direction of a sheet, and a speed control table. [Figure 5A] 5 is a flowchart showing a process of changing a speed control table in the first embodiment. [Figure 5B] 5B is a subroutine flowchart showing the process of step S22 in FIG. 5A. [Figure 6] 11 is a schematic diagram showing the position where an image abnormality occurs and a change section of the speed control table. FIG. [Figure 7] 1 is a table showing the correspondence between each defect classification included in image defects and paper damage and the occurrence position on paper. [Figure 8] 13 is a diagram illustrating an example of an operation screen for receiving permission from a user in a modified example. [Figure 9] FIG. 11 is a schematic configuration diagram of an image forming system according to a second embodiment. [Figure 10] FIG. 2 is a block diagram showing the functions of a machine learning device. [Figure 11A] 1 is a flowchart showing a learning process of the machine learning device. [Figure 11B] 11B is a subroutine flowchart showing the reward granting process in step S34 of FIG. 11A. [Figure 12] 11 is a table showing behavior information. [Figure 13A] This is an example of a Q table. [Figure 13B] 1 is a table showing patterns within a state. [Figure 14] 1 is a table showing examples of paper properties, image forming conditions, and usage states. [Figure 15] 5B is a subroutine flowchart showing the process of step S22 in FIG. 5A using a learning model in the second embodiment. [Figure 16] 13 is a table showing intra-state patterns in another modified example. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0034] Hereinafter, an embodiment of the present invention will be described with reference to the attached drawings. In the description of the drawings, the same elements are given the same reference numerals, and duplicated description will be omitted. Furthermore, the dimensional ratios of the drawings are exaggerated for the convenience of explanation, and may differ from the actual ratios. Furthermore, in this embodiment, "paper" includes printing paper and various films. In particular, printing paper includes paper made using mechanical pulp and / or chemical pulp derived from plants. Furthermore, paper types (paper types) include gloss paper (also called coated paper), matte paper, plain paper, fine paper, high gloss paper, etc.

[0035] Fig. 1 is a schematic diagram showing an image forming system 1000 according to an embodiment of the present invention. Fig. 2 is a block diagram showing a hardware configuration of the image forming system 1000.

[0036] (1) Overall structure 1 and 2, the image forming system 1000 includes an image forming apparatus 10, an image inspection apparatus 20, a post-processing apparatus 30, and an external media detection apparatus 40. Along the transport direction of the paper 90, the image forming apparatus 10, the image inspection apparatus 20, and the post-processing apparatus 30 are connected in this order from the upstream side.

[0037] (2) Image forming device 10 The image forming apparatus 10 includes a control unit 11 , a storage unit 12 , a paper transport unit 13 , an image forming unit 14 , an operation display unit 15 , a sensor 16 and a communication unit 19 .

[0038] The control unit 11 has a CPU (Central Processing Unit) and a memory. The CPU is a control circuit composed of a multi-core processor and the like that controls the above-mentioned units and executes various arithmetic processing according to a program, and each function of the image forming apparatus 10 is realized by the CPU executing the corresponding program. The memory is a high-speed accessible main storage device that temporarily stores programs and data as a working area. For example, the memory may be a dynamic random access memory (DRAM), a synchronous dynamic random access memory (SDRAM), or a static random access memory (SRAM).

[0039] The control unit 11 controls the entire image forming apparatus 10. The control unit 11 functions as a conveying speed control unit 111 and a conveying speed setting unit 112. The control unit 11 of the image forming apparatus 10 also controls the entire image forming system 1000 by cooperating with the control units 21 and 31 of the other devices.

[0040] The storage unit 12 is a large-capacity auxiliary storage device that stores various programs including an operating system and various data. For example, a hard disk, a solid state drive, a flash memory, a ROM (Read Only Memory), etc. are used as the storage. The storage unit 12 stores a speed control table as speed control information, in which a speed control value of the drive unit 148 is set corresponding to each transport position (transport section) in the paper 90 nipped by the transfer unit 142 and the fixing unit 143. In this speed control table, drive speeds corresponding to each transport position in the paper of the drive motors of the transfer unit 142 and the fixing unit 143 (hereinafter referred to as the transfer motor and the fixing motor, respectively) among the multiple drive motors included in the drive unit 148 are set. The speed control table is changed and updated by the transport speed setting unit 112. In addition, there are multiple types of speed control tables (see FIG. 4, etc. described later) stored in the storage unit 12 corresponding to the physical properties or type of paper and the paper passing mode (single-sided / double-sided). Here, the physical properties of the paper are measured by a media detection device 40, which will be described later, or are set by the user via the operation display unit 15. The physical properties of the paper are stored in association with the paper stored in the paper feed tray 131. The physical properties of the paper include stiffness, moisture content, and grain direction of the paper.

[0041] The paper transport section 13 includes a plurality of paper feed trays 131, a transport path 132 in which a plurality of transport rollers are arranged, and a transport path 133. The paper feed tray 131 stores a plurality of sheets of paper 90, and feeds and sends the sheets to the transport path 132 one by one.

[0042] (Image forming unit 14) The image forming unit 14 includes an image creating unit 141, a transfer unit 142, a fixing unit 143, a drive unit 148, a pressure unit 149, and the like.

[0043] The image forming unit 141 includes writing units and drum units corresponding to the basic colors of yellow (Y), magenta (M), cyan (C), and black (K), as well as an intermediate transfer belt 1a. Each drum unit includes a cylindrical photosensitive drum, a charging electrode, a developing device, a cleaning unit, etc. The developing device of each drum unit contains a two-component developer containing toner of a color corresponding to the basic color.

[0044] The toner images of each color formed by the drum unit are transferred in sequence to the surface of intermediate transfer belt 1a at the transfer nips with the respective primary transfer rollers, and after being superimposed, are transferred to paper 90 conveyed to the transfer nip with transfer unit 142, which is the secondary transfer unit. In the example shown in FIG. 1, intermediate transfer belt 1a functions as an image carrier, and the toner image formed on its surface is transferred to paper 90. Note that in a configuration in which intermediate transfer belt 1a is not provided and a toner image formed on one or more photosensitive drums is directly transferred to paper, the photosensitive drums function as image carriers.

[0045] In the image creating unit 141, the start timing of exposure (writing) by the writing unit and the timing of paper conveyance by the registration roller immediately upstream of the transfer unit 142 are synchronized to coincide with the start timing of image formation. The leading edge of the paper 90 is abutted against the registration roller, which is stopped by a clutch or the like, to temporarily stop it, and then it is synchronized with the leading edge position of the image on the intermediate transfer belt 1a and restarted at a predetermined timing to start conveying the paper 90, thereby aligning the toner image with the paper at the transfer nip position.

[0046] A foam roller, a solid roller, or the like is used for the transfer unit 142. The transfer unit 142 forms a transfer nip by pressing against an opposing roller that is in contact with the intermediate transfer belt 1a.

[0047] The fixing unit 143 includes a heating roller with a built-in heater, and a pressure roller. The fixing unit 143 includes a temperature sensor that measures the temperature of the heating roller, and a heater that heats the heating roller. Based on the temperature information of the temperature sensor, the power supply to the heater is controlled so that the heating roller reaches the control temperature. The fixing control temperature of the fixing unit 143 is controlled according to the basis weight or thickness of the paper. For example, the fixing control temperature is set to a higher temperature for thick paper than for plain paper. The pressure roller abuts against the heating roller with a predetermined pressure, and the transferred toner image on the paper 90 conveyed to the fixing nip formed by both rollers is heated and pressurized. As a result, an image is formed on the paper 90. In addition, when the image forming apparatus 10 or the image forming system 1000 is turned on, if the temperature of the fixing unit 143 is a cold start close to room temperature, the elapsed time from the power ON is measured. This timer is used as the heat storage time, which is the elapsed time from controlling the fixing temperature. It is determined that the start is a cold start if the power is turned on after a predetermined time has elapsed since the power was turned off, or if the temperature of the fixing unit 143 is equal to or lower than a predetermined temperature when the power is turned on.

[0048] There are two paper passing modes: single-sided mode and double-sided mode, and in the single-sided mode, the paper 90 that has passed through the fixing unit 143 and has an image formed on its front side is conveyed directly to the downstream device. On the other hand, in the double-sided mode, the paper 90 that has passed through the fixing unit 143 is conveyed to the conveying path 133 for double-sided conveying, and after being inverted by the switchback path, it again passes through the conveying path 132 and an image is formed on the back side by the image forming unit 14.

[0049] (Drive unit 148, pressure unit 149) As described above, the driving unit 148 is composed of a plurality of driving motors, including a transfer motor for the transfer unit 142 and a fixing motor for the fixing unit 143, which are driven independently. The driving speed of the transfer motor and / or the fixing motor is controlled according to a speed control value set in a speed control table. For example, in this embodiment, the transfer motor that drives the transfer unit 142 rotates at a constant speed during image formation, and the driving speed is not changed during image formation. On the other hand, the speed of the fixing motor that drives the fixing unit 143 is changed at a predetermined period (corresponding to section 1 to section E described later) during image formation according to the setting of the speed control table. When the clock frequency of the imaging unit 141 is changed due to vertical magnification adjustment, the driving speed of the transfer unit 142 is adjusted in synchronization with the clock frequency, and thereafter, the transfer unit 142 is driven at a constant speed at the adjusted driving speed.

[0050] Pressurizing unit 149 changes the pressure of the nip of transfer unit 142 and / or fixing unit 143. Specifically, pressurizing unit 149 is composed of a drive source constituted by a solenoid or a motor, an arm, a cam, etc., and changes the pressure applied to the rotation shaft of the roller of transfer unit 142 and the rotation shaft of the pressure roller of fixing unit 143. For example, the fixing pressure in the fixing nip is changed to strong, medium, or weak depending on whether the thickness of paper 90 is thick paper, plain paper, or thin paper. Also, in the case of embossed paper with many projections and recesses, the transfer pressure is set to a higher pressure than for plain paper.

[0051] The operation display unit 15 is equipped with a touch panel, numeric keypad, start button, stop button, etc., and is used to display various information and input various instructions. Through the operation display unit 15, the user can set paper information such as the size and type of paper stored in each paper feed tray. The physical properties of the paper can also be input. For example, the grain direction of the paper can be input as a physical property.

[0052] The sensor 16 includes a plurality of sensors 161-16x arranged in the device. The sensors 161 and 163 are optical sensors and are also called paper detection sensors. The sensors 161 and 163 detect the presence (or absence) of the paper 90 at the location where the transport path 133 is installed, and send the detection result to the control unit 11. Among the paper detection sensors, the sensor 161 is particularly called a registration sensor. Although only two paper sensors are shown as representatives, many are arranged along each transport path of the image forming system 1000 and near the branching points. The sensor 162 is a loop detection sensor (hereinafter, referred to as the loop detection sensor 162), which is composed of an actuator and an optical sensor, and outputs a plurality of levels of signals according to the passing position (passing height) of the paper 90 in the transport path 133. The sensor 164 is arranged inside the main body of the image forming device 10, and detects temperature and humidity (relative humidity).

[0053] The communication unit 19 is an interface for communicating with the image inspection device 20, the post-processing device 30, and other devices via a network. The communication unit 19 functions as a read image acquisition unit, and acquires read image data from the image inspection device 20 and measurement results (paper physical properties) from the media detection device 40.

[0054] (Image inspection device 20) 1 and 2, the image inspection device 20 includes a control unit 21, a storage unit 22, a paper transport unit 23, a reading unit 24, and a communication unit 29. These are connected to each other via signal lines such as a bus for exchanging signals.

[0055] The control unit 21 and the storage unit 22 have the same configuration as the control unit 11 and the storage unit 12 described above. The control unit 21 cooperates with the control unit 11 of the image forming apparatus 10 to perform image adjustment, image inspection, and setting of the speed control table of the image forming system 1000. The paper transport unit 23 includes a transport path 231, a plurality of transport rollers arranged on the transport path 231, and a drive motor (not shown) for driving the transport rollers. The transport path 231 is connected to the upstream transport path 132, receives the paper 90 on which an image is formed by the image forming apparatus 10, and sends it to the downstream post-processing device 30. The storage unit 22 (or the storage unit 12) stores image data such as a color chart on which multiple color patches for various evaluations are arranged, an inspection chart on which multiple grid images and register marks images for detecting misalignment are arranged, and a "transport evaluation chart" consisting of uniform density of single-color halftone (e.g., monochrome) used to change the speed control table, and horizontal thin lines of each color for detecting color misalignment or image expansion / contraction.

[0056] (Reading unit 24) The reading unit 24 is made up of first and second image reading units 241 and 242. As both image reading units have the same configuration, the first image reading unit 241 will be described below as a representative. The first image reading unit 241 is disposed above the transport path 231, and the second image reading unit 242 is disposed below the transport path 231. In the double-sided mode, when an image is formed on both sides of the paper 90 by the image forming apparatus 10, both sides are simultaneously read (in one pass) by the first and second reading units 241 and 242.

[0057] The two first and second image reading units 241 and 242 have the same configuration. The first image reading unit 241 includes a sensor array, a lens optical system, an LED (Light Emitting Diode) light source, and a housing for accommodating these. The sensor array is a color line sensor in which multiple optical elements (e.g., CCDs (Charge Coupled Devices)) are arranged in a line along the main scanning direction, and the reading area in the width direction corresponds to the full width of the paper 90. The optical system is composed of multiple mirrors and lenses. Light from the LED light source passes through the document glass and irradiates the surface of the paper 90 passing through the reading position on the transport path 231. The image at this reading position is guided by the optical system and formed on the sensor array. For example, the resolution of the first image reading unit 241 is 100 to several hundred dpi.

[0058] The read image data obtained by the first image reading unit 241 or the second image reading unit 242 is analyzed by the control unit 21 or the control unit 11 and used for various determination processes.

[0059] For example, as a "color judgment", the image forming apparatus 10 reads the paper 90 on which the image of the color chart is formed, and analyzes the color data of each color patch using the read image data obtained, and sends the result data to the image forming apparatus 10. The image forming apparatus 10 adjusts the LUT for color conversion (conversion from print data to CMYK signals of the writing unit), gamma correction, screen correction, and density balance adjustment, or adjusts the image formation conditions of the image forming unit 14, based on the result data. Also, as an "image position judgment", the image forming apparatus 10 analyzes the position of each register mark image using the read image data obtained by reading the paper 90 on which the image of the inspection chart is formed, and sends the result data to the image forming apparatus 10. The image forming apparatus 10 adjusts parameters for two-dimensional position correction, such as skew and shift, based on the result data.

[0060] Furthermore, when judging the paper transport in the transport path 132 between the transfer unit 142 and the fixing unit 143, the image inspection device 20 reads the paper 90 on which the image of the "transport evaluation chart" is formed by the image forming device 10, and sends the read image data obtained by the read image data to the image forming device 10. The control unit 11 of the image forming device 10 analyzes the behavior of the paper based on the obtained read image data, and changes and updates the speed control table used for speed control of the drive unit 148.

[0061] (Post-processing device 30) 1 and 2, post-processing device 30 includes control unit 31, memory unit 32, paper transport unit 33, post-processing unit 34, and communication unit 39, and performs processing on paper 90 by cooperating with other upstream devices. Control unit 31 and memory unit 32 include configurations similar to those of control units 11 and 21, and memory units 12 and 22 described above, respectively. Paper transport unit 33 includes transport paths 331 and 332, a plurality of transport rollers arranged on transport paths 331 and 332, a drive motor (not shown) that drives these, and paper output trays 333 and 334. For example, paper output tray 333 is where paper 90 for actual printing, which will be described later, is output, and paper output tray 334 is where paper 90 for pre-printing, which will be described later, is output.

[0062] The post-processing section 34 performs stapling processing. It may also perform various other post-processing such as punching processing, booklet formation processing, etc. For example, the post-processing section 34 has a stack section for stacking sheets of paper and a staple section, and after stacking a plurality of sheets of paper 90 in the stack section, performs side-stitching processing using staples in the staple section.

[0063] (Media detection device 40) The media detection device 40 is composed of a paper thickness detection section, a basis weight detection section, a surface property detection section, a moisture content detection section, a stiffness detection section, and the like (none of which are shown). In the case of an external type as shown in FIG. 1, a user inserts a hand-carried paper 90 into the paper passage area of ​​the measuring device, and operates a measurement start button on the device main body or the operation display section 15 of the image forming device 10 to measure the paper properties of the paper 90. Note that, as a modified example, a built-in type media detection device may be used. For example, a built-in type media detection device is configured to use the conveying path (e.g., conveying path 132) of the image forming device 10 as the measurement area. Also, although an example in which the paper type is determined on the image forming device 10 side will be described in this embodiment, it may be configured to be determined on the media detection device 40 side.

[0064] The paper thickness detection unit includes a pair of transport rollers, at least one of which moves depending on the thickness of the paper 90 passing through the nip, and a measurement unit that measures the axis distance of the pair of transport rollers. This measurement unit is composed of, for example, an actuator, an encoder, and a light emitter / receiver. The paper thickness detection unit outputs the measurement result of the paper thickness of the paper 90 (measurement value 1).

[0065] The basis weight detection unit is a sensor that detects the basis weight of the paper 90, and includes a light-emitting unit and a light-receiving unit, and measures the basis weight based on the attenuation of light that passes through the paper 90. For example, the basis weight sensor has a light-emitting unit disposed below the paper passage area (paper passage path) where the paper is inserted, and a light-receiving unit disposed above it, and detects the basis weight of the paper 90 based on the intensity of light that passes through the paper 90 and is received by the light-receiving unit, and outputs the measurement result (measurement value 2).

[0066] The surface property detection unit includes a housing, a light emitting unit, a collimating lens, and a plurality of light receiving units, and optically detects the specular reflected light and the diffuse reflected light from the paper surface as described below. An opening (measurement area) is provided in the guide plate above the paper passing area, and this opening becomes the irradiation area of ​​the light receiving unit. The paper 90 inserted up to the opening is pressed by a pressing mechanism lifted from below the paper passing area. In this state, the light emitting unit irradiates the irradiation light, which is made approximately parallel by the collimating lens, at an incident angle of 75° with respect to the reference surface. The wavelength of the irradiated light is, for example, 465 nm. The plurality of light receiving units receive the specular reflected light and the diffuse reflected light. For example, they are arranged at three locations with reflection angles of 30 degrees (for diffuse reflected light), 60 degrees (for diffuse reflected light), and 75 degrees (for specular reflected light), or at two locations with 60 degrees and 75 degrees. The surface property detection unit outputs the signal of this light receiving unit as the measurement result of the paper physical property (measurement value 3).

[0067] The moisture content detection unit includes, for example, a moisture content sensor that optically detects the amount of light absorbed by OH groups using a near-infrared method. This moisture content sensor uses the property that light of a predetermined wavelength in the near-infrared region is irradiated onto the paper 90 and the light absorption rate changes depending on the moisture content of the paper 90. As another example, the moisture content may be measured by measuring the change in the amount of light reflected from inside the paper using reflected light separated by a polarizing filter. For example, the moisture content is output from the moisture content detection unit as a measurement result of the paper physical property.

[0068] The stiffness detection unit detects the physical properties corresponding to the stiffness of the paper. For example, when the paper is conveyed along a curved conveying path, the force with which the paper presses against one of the outer guide plates that make up the conveying path, or the amount of displacement, is mechanically measured. The stiffness detection unit outputs the stiffness as the measurement result of the paper physical properties. The measurement result is converted into a stiffness value (mN) and output (see FIG. 14(a) described later). In addition, by rotating the paper 90 by 90 degrees and conveying it along the conveying path, the stiffness in both the vertical and horizontal directions can be obtained, and the grain direction (longitudinal / horizontal) can be determined.

[0069] The image forming apparatus 10 also performs paper type discrimination based on the measurement results (measurements 1 to 3 above) obtained from the media detection device 40. A learning model (also called a trained model) based on machine learning used for paper type discrimination is stored in the storage unit 12 of the image forming apparatus 10. This is a learning model generated by supervised learning using teacher data, with the detection output of the media detection device 40 for the paper 90 as an input value and the paper type information for the paper 90 set by the user as a correct answer label. The control unit 11 of the image forming apparatus 10 obtains a basis weight equivalent value, a paper thickness equivalent value, and a surface property measurement value using the measurement results (measurements 1 to 3) obtained by measuring the paper 90 with the paper thickness detection unit, basis weight detection unit, and surface property detection unit of the media detection device 40. In this case, the basis weight conversion value is calculated from the first and second basis weights by coefficients and a formula determined by the surface property measurement value of the measurement value 3 and the ambient environment information of the image forming system 1000 (temperature, humidity (for example, detected by the temperature and humidity sensor 164)). Here, the basis weight difference = first basis weight - second basis weight. The basis weight detection unit is equipped with multiple LEDs that irradiate light of different wavelengths. The first basis weight is calculated by using a first LED that outputs irradiation light of a wavelength (750 nm to 900 nm) and the amount of light transmitted through the paper 90 of this irradiation light. The second basis weight is calculated by using a second LED that outputs irradiation light of a wavelength (400 nm to 470 nm) and the amount of light transmitted through the paper 90 of this irradiation light. For paper type discrimination, a learning model is used to input the surface property measurement value, basis weight difference, and density (=basis weight M / paper thickness), and the "paper type score" of each candidate paper type is obtained as an output. The control unit 11 determines the paper type with the highest paper type score as the paper type, or presents it to the user as a candidate.

[0070] (Paper transport control) Next, the paper transport control will be described with reference to FIG. 3 and FIG. 4. In the embodiment described below, the transport speed control unit 111 controls the drive speed of the fixing motor when the paper 90 is transported, keeps the speed of the transfer unit 142 constant, and controls only the speed of the fixing unit 143. FIG. 3 is a schematic diagram showing the start and end timing of the paper transport control. FIG. 3 is a partial enlarged view of FIG. 1, showing the periphery of the transport path 132 between the transfer unit 142 and the fixing unit 143. The transport path 132 includes an upper guide plate 132a and a lower guide plate 132b. A hole is provided at the approximate center in the width direction of the lower guide plate 132b, and the tip of the actuator of the loop detection sensor 162 protrudes upward from below the hole with a weak biasing force. This actuator moves up and down by being pushed by the paper 90 according to the height of the paper 90 passing through, and detects the loop amount (passing height) of the paper 90 with two, three, or four levels of resolution according to the up and down movement.

[0071] 3A is a diagram showing the start timing of paper transport control. The start timing is before the leading edge of paper 90 reaches the fixing nip, that is, when the leading edge of paper 90 reaches a position a predetermined distance a upstream from the fixing nip. The leading edge position and timing of paper 90 can be calculated from the driving time (amount of rotation) after the registration roller is restarted.

[0072] 3(b) is a diagram showing the timing at which paper transport control ends. The end timing is a predetermined time after the rear end of paper 90 passes through the transfer nip, that is, when the rear end of paper 90 reaches a position downstream of the transfer nip by a processing distance b. The leading edge position and timing of paper 90 can be calculated from information on the length of paper 90 in the transport direction (hereinafter simply referred to as "paper length") and the driving time from when sensor 163 is driven after the leading edge of paper 90 reaches it, or the driving time from when the registration roller is restarted.

[0073] FIG. 4 is a schematic diagram for explaining the sections corresponding to the position of the paper in the transport direction, and the speed control table. Starting from the leading edge of the paper, a number of sections 0 to E are provided at predetermined intervals of 1 mm to several tens of mm. In the example shown in FIG. 4, the predetermined intervals correspond to 10 mm on the paper 90. The start of section 0 corresponds to the start timing of FIG. 3(a), and the end of section E corresponds to the end timing of FIG. 3(b). The number of sections (=E+1) depends on the length of the paper 90. For example, if the length is 420 mm, the number of sections is 42. If the length of the paper 90 used is not exactly divisible by the predetermined intervals, the trailing edge of the paper will be located midway through section E.

[0074] The speed control table shown in FIG. 4 describes the speed control value of the fixing unit 143 in each section, and is stored in the storage unit 12. There are a plurality of speed control tables, each corresponding to a printing condition. The printing conditions include the physical properties of the paper. As described above, the physical properties of the paper include the stiffness, moisture content, and grain direction of the paper. Of these paper physical properties, the stiffness, moisture content, and grain direction of the paper may be paper properties obtained by measurement by the media detection device 40. The drive speed of the fixing motor is controlled according to the speed control value set in this speed control table. V0 is a reference speed, for example, 200 to 500 mm / sec. At the reference speed V0, the conveying speed of the fixing unit 143 is the same as the conveying speed of the transfer unit 142. As shown in FIG. 4, in section 0, the fixing motor is controlled by the fixing speed control value "V0 x 0.983", and in section 1, the fixing motor is controlled by the same fixing speed control value "V0 x 0.983". The drive speed can be controlled in units of 0.1%. During image formation, the transfer motor that drives the transfer unit 142 is controlled to a constant drive speed that provides a constant transport speed (process speed). The fixing motor of the fixing unit 143 is controlled to a constant speed at a reference speed V0 before section 0, that is, from when the registration is restarted until the paper 90 reaches the position in Fig. 3(a), and after the paper 90 reaches the position in Fig. 3(b), or after section E.

[0075] (Speed ​​control table change processing) FIG. 5A is a flowchart showing the process of changing the speed control table in the first embodiment.

[0076] (Step S21) When the image forming apparatus 10 receives a print job from a user via the operation display unit 15 or the like, it starts the print job (YES). The received print job data includes print data and a job ticket. The job ticket describes the printing conditions, which include paper information such as paper type information, basis weight, paper size, and paper physical properties, and a paper feed mode (single-sided / double-sided). This paper information may also be linked to the paper physical properties measured by the media detection device 40 and / or the paper type (paper type) and basis weight determined from these paper physical properties, and used as the paper information.

[0077] (Step S22) The control unit 11 executes pre-printing under the printing conditions of the print job and updates the speed control table. The update process will be described later.

[0078] (Step S23) The image forming apparatus 10 performs actual printing for the print job started in step S21. At this time, the conveying speed control unit 111 controls the conveying of the paper 90 using the speed control table updated in step S22.

[0079] (Speed ​​control table update process) (Step S221) FIG. 5B is a subroutine flowchart showing the process of step S22 in FIG. 5A.

[0080] In step S221, the control unit 11 obtains a speed control table corresponding to the printing conditions from the storage unit 12. For example, there are multiple speed control tables according to the printing conditions of paper type and basis weight or stiffness, and from among them, a speed control table with matching or closest conditions is obtained.

[0081] (Step S222) The control unit 11 performs pre-printing using image data of the "transport evaluation chart." In the pre-printing, image formation is started under the printing conditions of the print job. In addition, in the pre-printing, the transport speed control unit 111 controls the transport of the paper 90 during image formation using the speed control table acquired in step S221.

[0082] (Step S223) In the single-sided mode, the image inspection device 20 uses the first reading unit 241 downstream of the fixing unit 143 to read the image of the paper 90 after it has passed through the fixing unit 143, and generates read image data. In the double-sided mode, the image inspection device 20 uses the first and second reading units 241 and 242 to read both sides of the paper 90, and generates two read image data. The control unit 11 acquires the read image data generated by the image inspection device 20 via the communication unit 19. The paper 90 on which the image of the transport evaluation chart has been formed is discharged to, for example, a paper output tray 334 serving as a sub-tray.

[0083] (Step S224) The conveying speed setting unit 112 analyzes the read image data. This analysis determines whether or not image defects and / or paper damage have occurred. Image defects include at least one of image rubbing, transfer misalignment, color misalignment, image expansion / contraction, uneven gloss, and image stains. Paper damage includes at least one of paper wrinkles, waving, paper scratches, and paper folds. The conveying speed setting unit 112 compares pattern images corresponding to image defects and paper damage, respectively, which are stored in advance in the storage unit 12, and determines that image defects or paper damage have occurred if the similarity is high (hereinafter, image defects and paper damage may be collectively referred to as "image abnormalities"). Waving is determined by the occurrence of periodic out-of-focus of the read image.

[0084] If there is an image defect or paper damage (YES), the process proceeds to step S225. If there is neither an image defect nor paper damage (NO), the process ends and returns to the process of FIG. 5A.

[0085] (Step S225) The transport speed setting unit 112 changes the speed control table when an image abnormality (image defect or paper damage) occurs or based on the location of the occurrence, and updates the contents of the storage unit 22 with the changed speed control table.

[0086] Fig. 6 is a schematic diagram showing the occurrence position of an image abnormality and the change section of the speed control table. Fig. 7 is a table showing the correspondence between each defect classification included in the image defect and paper damage and the occurrence position on the paper.

[0087] For example, as shown in Fig. 6, if wrinkles occur on the leading edge of the transport evaluation chart, the speed control values ​​for sections 1 and 2 corresponding to the position where the wrinkles occurred are changed. Also, if image rubbing occurs on the trailing edge, the speed control values ​​for sections E-5 to E-3 are changed. As shown in the table in Fig. 7, the relationship between the abnormality classification, the occurrence member, the occurrence position on the paper, and the correction of the fixing speed is preset in the storage unit 12, and the speed control table for the position (section) corresponding to the occurrence position is changed according to the abnormality classification that has occurred.

[0088] In this way, the image forming apparatus according to the first embodiment includes a conveying speed control unit that controls the conveying speed of the paper in the transfer unit and / or the fixing unit, a read image acquisition unit that acquires the read image data of the paper generated by the reading unit that reads the image of the paper downstream of the fixing unit in the conveying direction, and a conveying speed setting unit that changes or updates the conveying speed based on the image of the acquired read image data. This makes it possible to appropriately convey the paper between the transfer unit and the fixing unit, thereby reducing the occurrence of image defects and paper damage. More specifically, the image forming apparatus includes a conveying speed control unit that controls the driving speed of the driving unit of the fixing unit during image formation based on a speed control table for controlling the conveying speed of the paper in the fixing unit stored in the storage unit, and a conveying speed setting unit that analyzes the image of the acquired read image data and changes the speed control table based on the analysis result. This makes it possible to appropriately convey the paper between the transfer unit and the fixing unit, thereby reducing the occurrence of image defects and paper damage.

[0089] (Modification) In the first embodiment, when the image forming apparatus 10 determines that an image abnormality has occurred, the image forming apparatus 10 automatically changes the speed control value in the speed control table. In contrast, in the modified example described below, the image forming apparatus 10 receives permission from the user before changing the speed control value in the speed control table.

[0090] Fig. 8 is an example of an operation screen 151 that accepts permission from a user in the modified example. In the modified example, the operation display unit 15 functions as a setting acceptance unit. When changing the speed control table in step S255 in Fig. 5A, the control unit 11 displays the operation screen 151 on the operation display unit 15. Then, when the "Yes" button is operated on the operation screen 151 and permission from the user is obtained, the control unit 11 changes the speed control table. On the other hand, when the "No" button is operated on the operation screen 151, no change is made and the speed control table is restored to the previous one.

[0091] In this manner, in the modified example, the speed control table is changed only after receiving permission from the user, which makes it possible to prevent the speed control table from being changed against the user's will.

[0092] Second embodiment FIG. 9 is a schematic diagram showing an image forming system 10000 including a machine learning device 50 and an image forming apparatus 10 according to a second embodiment. FIG. 10 is a block diagram showing the functions of the machine learning device 50. The image forming apparatus 10 and the machine learning device 50 are connected via a network. The machine learning device 50 learns behavior related to paper transport within the apparatus, particularly paper transport between the transfer unit 142 and the fixing unit 143, as described below, and generates a machine learning model (hereinafter, simply referred to as a learning model). The generated learning model is sent to the image forming apparatus 10. The image forming apparatus 10 uses the acquired learning model to change a speed control table for transport control.

[0093] The machine learning device 50 shown in FIG. 9 includes a control unit 51, a storage unit 52, and a network I / F. The control unit 51 is composed of a plurality of CPUs, a plurality of GPUs (Graphics Processing Units), a RAM, a ROM, and the like, and functions as a learning block 510 and a state control block 550. The storage unit 52 stores a learning model. The machine learning device 50 may be an on-premise server, or may be a cloud server using a commercial cloud service. As another example, the machine learning device 50 may be integrated with the image forming system 1000 or the image forming device 10. For example, in this example, an engine control SoC (System-on-a-Chip) in the control unit 31 of the image forming device 10 functions as a machine learning device, and a learning model is generated by machine learning in the image forming device 10.

[0094] The machine learning device 50 calculates the reward when a certain action is taken in a certain state according to a predetermined rule, and calculates the action value (Q value) according to a predetermined formula to optimize the total reward, and updates the Q table. In this way, the machine learning device learns actions, and determines an action based on the learning result (selects the action with the highest action value).

[0095] Here, the learning coefficient is η, the time discount rate is γ, and the reward at time t is R t Then, the action value (Q(s t ,a t ) can be calculated, for example, by the following Q-learning formula (1):

[0096]

number

[0097] (Generating learning models) Hereinafter, a machine learning method for the learning model used in this embodiment will be described with reference to Fig. 10 and Fig. 11A to Fig. 14. Fig. 11A and Fig. 11B are flowcharts showing the learning process executed by the machine learning device 50. Fig. 12 to Fig. 14 are diagrams showing a table-type learning model according to one embodiment.

[0098] As shown in FIG. 10, a control unit 51 of a machine learning device 50 functions as a learning block 510 and a state control block 550.

[0099] The learning block 510 includes a decision making unit 511, a state observation unit 512, a reward calculation unit 513, and a learning unit 514. The state control block 550 includes an observation information generation unit 551. In an example shown in FIG. 10, the state control block 550 of the machine learning device 50 may acquire information on the use state of the device and image formation conditions stored in the storage unit 12 from the image forming device 10, and thereby reproduce (simulate) the paper conveying state of the image forming device 10. In the example shown below, a time equivalent to one print is used as the time step. For example, it is 0.5 to several seconds. In the present embodiment shown in FIG. 10, FIG. 11A, etc., one time step corresponds to one episode.

[0100] (Step S31) See Fig. 11A. In this step, the decision-making unit 511 determines and outputs behavior information (also called speed instruction) related to drive control for the fixing motor that drives the fixing unit 143 based on the state at the time step at the current time point (current cycle). Hereinafter, this behavior information is also called an action. The behavior information is a fixing speed value corresponding to the speed control table (see Fig. 4, etc.).

[0101] FIG. 12 is a table showing behavior information. In the figure, the drive speed of the fixing motor that drives the fixing unit 143 is set to 0.0% of the reference speed V0, and can take multiple actions a1 to a61 in increments of 0.1% in the range of -3.0% to +3.0% (deceleration to acceleration). This action can be selected by referring to the current state based on the Q table shown in FIG. 13A. In the early stage of learning, the values ​​in the Q table may be random values ​​or may be predetermined values. For example, in the latter case, values ​​equally divided by the number of possible actions (1 / 61 for 61 actions) are entered as the predetermined values. In the initial state where learning has not progressed, actions may be selected randomly at a certain rate using the ε-greedy method. For example, a fixed value of ε is used (for example, any value between 0 and 1, and between 0.1 and 0.3). Alternatively, a formula that is set so that ε decreases as the learning progresses, for example, so that it is halved each time the number of learning times reaches a certain value, or so that the current state (s t ) is the action value Q(s t, a t ) by the maximum value of the above formula.

[0102] (Step S32) The state control block 550 controls the fixing unit 143 based on the action information received in step S31. Specifically, the fixing motor is driven at a fixing speed value based on the speed instruction.

[0103] (Step S33) The observation information generating unit 551 generates each piece of information (hereinafter, referred to as observation information) in the environment of the conveying path between the transfer unit 142 and the fixing unit 143, such as (1) position information of the paper 90 being conveyed, (2) speed state of the fixing unit 143, (3) paper physical properties, (4) image forming conditions, (5) usage state of the image forming apparatus, and (6) read image. The generated observation information is passed to a state observing unit 512 in the learning block 510.

[0104] Fig. 14 is a table showing examples of observed information. As examples of paper physical properties, Fig. 14(a) shows stiffness, and Fig. 14(b) shows moisture content. As an example of image forming conditions, Fig. 14(c) shows fixing pressure, and as an example of the usage state of the image forming apparatus, Fig. 14(d) shows temperature and humidity, which are information on the surrounding environment of the apparatus. The paper physical properties can be measured using the results of measurements by the media detection device 40. The temperature and humidity can be detected by the sensor 164.

[0105] Fig. 13B is a table showing the patterns within one state. Each state corresponds to a section (see Fig. 4) corresponding to the position of the paper 90 and the result of determining whether there is an image abnormality (poor image quality or paper damage) in that section. The total number of states depends on the paper size. For example, for A3 size paper, there are 42 sections corresponding to a length of 420 mm in the transport direction, and by multiplying this by the presence / absence of an abnormality, the total number of states becomes 84.

[0106] (Step S34) The reward calculation unit 513 calculates the reward value using the skew state. Fig. 11B is a subroutine flowchart showing the reward value granting process in step S34.

[0107] (Step S341) The remuneration calculation unit 513 acquires the scanned image of the paper for the current cycle from the state observation unit 512.

[0108] (Step S342) Remuneration calculation unit 513 analyzes the read image and determines whether there is an image defect or paper damage. If there is an image defect or paper damage (YES), processing proceeds to step S343; if there is neither (NO), processing proceeds to step S344.

[0109] (Step S343) The reward calculation unit 513 here gives a negative reward value, ie reward value=−1.

[0110] (Step S344) Here, the reward calculation unit 513 gives a positive reward value, that is, reward value = + 1. After the processing of the subroutine shown in Fig. 11B, the process returns to the processing shown in Fig. 11A.

[0111] (Step S35) In step S35 shown in Fig. 11A, the learning unit 514 updates the learning model from the observation information, the action information, and / or the reward value. For example, in the example shown in Fig. 10, the action is learned by calculating the action value (Q value) using the reward value and the above-mentioned Q-learning formula (1), and the Q table is updated. The above processing is the learning of one episode.

[0112] As another learning model, a neural network type learning model may be applied. For example, the observed information (st: pattern in one state) at a certain timing (e.g., the previous time step) is used as an input, and the output (a t : each action (behavioral information) and its Q(s t ,a t ) is obtained. The learning unit 514 then obtains this Q(s t ,a t ) is "R t+1 + γmaxQ(s t+1 ,a t+1 The parameters are adjusted so that the difference E approaches "(the difference E is small)". For example, by performing a process called back-propagation, the parameters are adjusted and updated so that the error in the comparison results is small, and learning is performed.

[0113] (Step S36) If the control unit 51 has not completed learning a predetermined number of times (e.g., tens of thousands of times) (NO), the process returns to step S31 and repeats the subsequent processes. If the learning has been completed a predetermined number of times (YES), the process proceeds to step S37.

[0114] (Step S37) The control unit 51 of the machine learning device 50 outputs the updated learning result to the storage unit 52, updates the learning model, and ends the learning process (END). This learning model is sent to the image forming apparatus 10 and is also stored in the storage unit 12 thereof.

[0115] (Speed ​​control table change process in the second embodiment) Fig. 15 is a diagram showing a process of changing the speed control table using a learning model in the second embodiment. This figure is a subroutine flowchart showing the process of step S22 in Fig. 5A. In the second embodiment, the main routine flowchart is the same as that of the first embodiment shown in Fig. 5A, and therefore a description thereof will be omitted.

[0116] (Steps S521 to S523) Here, the control unit 11 performs the same processes as steps S221 to S223 in Fig. 5B. Specifically, the control unit 11 obtains a speed control table corresponding to the printing conditions from the storage unit 12, and performs pre-printing in which an image of a transport evaluation chart is printed using the speed control table. Then, the image forming apparatus 10 obtains, via the communication unit 19, which is an obtaining unit, the read image data generated by the image inspection device 20 reading the output of the image inspection device 20.

[0117] (Step S524) The control unit 11 obtains a speed control table using the learning model stored in the storage unit 12, the read image data, the speed control table used in step S522, the physical properties of the paper, the image forming conditions, and the use state of the device. The obtained speed control table is used to update the speed control table before updating stored in the storage unit 12. This speed control table is generated according to a combination of printing conditions (physical properties of the paper, image forming conditions, and the use state of the device). For example, in the example shown in FIG. 14, 1296 patterns (=16×3×3×9) are generated for the paper physical properties (16 levels of stiffness×3 levels of moisture content), image forming conditions (3 levels), and use state (9 levels). This updated speed control table is used to control the transport of the paper 90 in the actual printing (step S23 in FIG. 5A).

[0118] In this way, the image forming apparatus according to the second embodiment uses the learning model stored in the storage unit to determine the speed control information, i.e., the speed control table that defines the speed control value of the drive unit for each transport position. This makes it possible to appropriately transport paper between the transfer unit and the fixing unit, thereby reducing the occurrence of image defects and paper damage.

[0119] (Other variations) In the second embodiment, as shown in Fig. 13B, the states correspond to the paper position and the presence or absence of an image abnormality (bad image or paper damage), but in addition to this, the output of the loop detection sensor 162 may be added. Fig. 16 is a table showing patterns within the states in such another modified example. By adding the output of the loop detection sensor 162 to the input in this way, a learning model that can output a more accurate speed control table can be generated.

[0120] The image forming device 10, the image forming system 1000 including the image forming device 10, and the machine learning device 50 described above are the main configurations described in explaining the features of the above-mentioned embodiment, and are not limited to the above-mentioned configurations, and can be modified in various ways as described below within the scope of the claims. For example, they can be configured as the following modified examples. Furthermore, the configurations of general image forming devices 10 and the like are not excluded.

[0121] (Variation 2) In the speed control table (see FIG. 4, etc.) used in the first embodiment, the speed control value of the drive unit 148 for each transport position of the paper 90 is set, but the present invention is not limited to this, and a uniform speed value may be set over the entire surface of the paper 90. As another example, the speed control table may be described with a speed relative to the transfer speed. For example, the transfer speed has been described as being the same as the reference speed V0, but when the transfer speed is slightly changed from the reference speed due to vertical magnification adjustment or the like, the value in the speed control table is shifted by the amount of this change. Furthermore, in the first and second embodiments, the speed control table describes the speed control value of the fixing unit 143, but the speed control value of the transfer unit 142, or the speed control values ​​of both the transfer unit 142 and the fixing unit 143 may be described.

[0122] (Variation 3) In the second embodiment, the fixing pressure is shown as an example of the image forming condition. However, other image forming conditions may be used, such as the transfer pressure set by the pressure unit 149, the paper passing mode, and the fixing control temperature.

[0123] (Variation 4) In the second embodiment, ambient environment information is used as an example of the usage state of the device, but other usage states may also be used, such as the amount of usage (corresponding to the amount of wear) of components such as the heating roller and pressure roller of fixing unit 143, or the heat storage time of fixing unit 143.

[0124] (Variation 5) In the first and second embodiments, the speed control table is changed and updated by analyzing the image of the transport evaluation chart, but the present invention is not limited to this, and the speed control table may be changed and updated by analyzing the image of the actual print. In this case, if the four sides of the paper are cut based on the register mark image as the final product, an image similar to the transport evaluation chart may be printed on the outside of the paper that is not used in the actual print, and the speed control table may be changed and updated by analyzing this image. In addition, when the image of the actual print is analyzed to change and update the speed control table, the speed control table may be changed and updated by statistically processing (e.g. averaging) the analysis results of several sheets or several tens of sheets.

[0125] The above-mentioned machine learning device and the means and methods for performing various processes in the image forming device can be realized by either a dedicated hardware circuit or a programmed computer. The above-mentioned program may be provided by a computer-readable recording medium such as a USB memory or a DVD (Digital Versatile Disc)-ROM, or may be provided online via a network such as the Internet. In this case, the program recorded in the computer-readable recording medium is usually transferred to and stored in a storage unit such as a hard disk. The above-mentioned program may be provided as a standalone application software, or may be incorporated into the software of the device as a function. [Explanation of symbols]

[0126] 1000 Image forming system 10 Image forming device 11 Control section 111 Conveyor speed control section 112 Conveyor speed setting unit 12 Storage section 13 Paper transport section 14 Image forming section 141 Image creation section 142 Transcription section 143 Fixing part 148 Drive unit 149 Pressurizing section 15 Operation display section 161~16x Sensor 19 Communications Department 20. Image inspection equipment 21 Control section 22 Memory section 23 Paper transport section 24 Image reading unit 29 Communications Department 30 Aftertreatment device 31 Control Unit 32 Storage section 33 Paper transport section 34 Post-processing section 39 Communications Department 40 Media detection device 50 Machine Learning Device

Claims

1. a transfer unit that transfers a toner image formed on the image carrier by the image forming unit onto a sheet of paper; a fixing section for fixing the toner image on the paper by heating; a conveying speed control unit that controls a conveying speed of the paper in the fixing unit; a read image acquisition unit that is located downstream of the fixing unit in a conveying direction and that acquires read image data of the paper generated by a reading unit that reads an image of the paper; a conveying speed setting unit that changes or updates the conveying speed based on an image of the acquired read image data, the conveying speed control unit controls the conveying speed based on speed control information for controlling a drive speed of a drive unit of the fixing unit, the speed control information being stored in advance in a storage unit; The image forming apparatus, wherein the speed control information is set as a speed control table in which a speed control value of the drive unit for each transport position in the paper with respect to the fixing unit is set.

2. The image forming apparatus according to claim 1 , wherein the transport speed setting unit analyzes image defects and / or paper damage as the image analysis.

3. The image defect includes at least one of image rubbing, transfer misalignment, color misalignment, image expansion / contraction, gloss unevenness, and image staining, The image forming apparatus according to claim 2 , wherein the damage to the paper includes at least one of wrinkles, ripples, scratches, and folds in the paper.

4. 4. The image forming apparatus according to claim 2, wherein the transport speed setting section analyzes the image defects and the paper damage by comparing with a pre-stored image pattern.

5. 5. The image forming apparatus according to claim 2, wherein the transport speed setting section changes the speed control information based on the presence or absence of the image defect and / or the paper damage, or the position of the image defect and / or the paper damage on the paper.

6. the read image acquisition unit acquires the physical properties of the paper measured by a media detection device; 6. The image forming apparatus according to claim 1, wherein the conveying speed control unit controls a drive unit of the fixing unit using speed control information corresponding to the physical properties of the paper, the speed control information being stored in a memory unit.

7. 7. The image forming apparatus according to claim 1, wherein in a paper passing mode in which images are formed on both sides of a sheet of paper, the conveying speed control unit controls the drive unit of the fixing unit using speed control information corresponding to each of the front and back sides of the sheet of paper, the speed control information being stored in a memory unit.

8. A transport position in the paper relative to the fixing unit, a speed control value of the drive unit for each of the transport positions, and 8. The image forming apparatus according to claim 1, wherein the conveying speed setting unit determines speed control information using a learning model generated by machine learning based on an analysis result of the image of the read image data.

9. The image forming apparatus according to claim 8 , further comprising machine learning based on information on physical properties of the paper.

10. 10. The image forming apparatus according to claim 9, wherein the physical property includes at least one of stiffness, moisture content, and grain direction of the paper.

11. During machine learning of the learning model, The image forming apparatus according to any one of claims 8 to 10, wherein the image is analyzed by analyzing image defects and paper damage, and if an image defect or paper damage is detected, the image is evaluated as abnormal, and if no image defect or paper damage is detected, the image is evaluated as normal, and a positive reward is given if the evaluation is normal, and a negative reward is given if the evaluation is abnormal.

12. Further, a loop detection sensor is provided in the transport path between the transfer unit and the fixing unit to detect a loop of the paper, The image forming apparatus according to claim 8 , further comprising machine learning based on an output of the loop detection sensor.

13. An image forming apparatus according to any one of claims 1 to 12, a reading unit that is located downstream in a conveying direction of the image forming apparatus and that reads an image on a sheet of paper formed by the image forming apparatus; An image forming system comprising:

14. A control program for controlling a conveyance speed of a sheet in an image forming apparatus including a transfer unit that transfers a toner image formed on an image carrier by an image forming unit onto a sheet, and a fixing unit that heats the toner image on the sheet to fix it, the control program comprising: a first step of acquiring read image data of the paper, the read image data being generated by a reading unit that reads an image of the paper, downstream of the fixing unit in a conveying direction; a second step of changing or updating the conveying speed based on an image of the read image data acquired in the first step, In the second step, the conveying speed is controlled based on speed control information for controlling a drive speed of a drive unit of the fixing unit, the speed control information being stored in a storage unit in advance; A control program for causing a computer to execute a process in which the speed control information is set as a speed control table in which a speed control value of the drive unit for each transport position within the paper relative to the fixing unit is set.

15. The control program according to claim 14, wherein in the second step, the speed control information is determined using a learning model generated by machine learning based on the paper transport position relative to the fixing unit, the speed control value of the drive unit for each transport position, and the analysis results of the image of the read image data.

Citation Information

Patent Citations

  • Sheet conveying controller of image forming apparatus

    JP2008158076A

  • Image forming apparatus

    JP2014137432A

  • Image forming apparatus and image forming method

    JP2019098734A

  • Image forming apparatus and image forming program

    JP2019135509A

  • Image formation device and image formation method

    JP2020148977A