Information processing device, information processing method, and developing device

Abstract

To provide a technique for estimating an abnormality in a development process. An information processing device 2 according to an aspect of this disclosure includes: an acquisition unit 100 that acquires vibration data related to vibration of a development device that executes a development process including obtaining a printing plate by removing an uncured part of a target resin plate; and an estimation unit 108 that estimates an abnormality in the development process on the basis of the vibration data.

Description

Information processing device, information processing method, and developing device

[0001] The present disclosure relates to an information processing apparatus, an information processing method, and a developing apparatus.

[0002] Conventionally, techniques for producing printing plates by developing resin plates have been known. For example, Patent Document 1 describes a technique for preventing accumulated substances from photosensitive lithographic printing plate materials, developer processing solution compositions, various additives, etc. from being mixed into and accumulating in the wash water of a post-development wash section, even when the wash water in the post-development wash section is circulated and used for a long period of time. Patent Document 2 describes a technique for sensing the composition of a developer in real time and on the spot. Patent Document 3 describes a technique for suppressing the adhesion of development residue, enabling proper development, and reducing the frequency of changing the developer.

[0003] Japanese Patent Application Publication No. 2006-350076 International Publication No. 2023 / 022177 International Publication No. 2022 / 044570

[0004] However, the techniques described in these prior art documents all relate to maintenance of the developer, and are not intended to estimate abnormalities in the development process.

[0005] The present disclosure provides a technique for estimating an abnormality in a development process.

[0006] An information processing device according to one aspect of the present disclosure includes an acquisition unit that acquires vibration data relating to the vibration of a development device that performs a development process including removing uncured portions of a target resin plate to obtain a printing plate, and an estimation unit that estimates an abnormality in the development process based on the vibration data.

[0007] Another aspect of the information processing method of the present disclosure causes a computer to acquire vibration data relating to the vibration of a developing device that performs a developing process that includes removing uncured portions of a target resin plate to obtain a printing plate, and estimates an abnormality in the developing process based on the vibration data.

[0008] A developing device according to another aspect of the present disclosure is a developing device comprising a brush, a developer tank containing developer, a developing process processing unit that performs a developing process including removing uncured portions of a target resin plate with the brush and developer to obtain a printing plate, and a computer, wherein the computer acquires vibration data related to the vibration of the developing device for the target resin plate and estimates abnormalities in the developing process based on the vibration data.

[0009] According to the present disclosure, it is possible to provide a technique for estimating an abnormality in a development process.

[0010] FIG. 1 is a diagram for explaining an example of the configuration of the developing device 3. FIG. 2 is a diagram for explaining an example of the configuration of the system 1. FIG. 3 is a diagram for explaining an example of time domain vibration data. FIG. 4 is a diagram for explaining an example of a method for converting time domain vibration data into frequency domain vibration data. FIG. 5 is a diagram for explaining an example of a method for converting time domain vibration data into frequency domain vibration data. FIG. 6 is a diagram for explaining an example of a data set generated by the information processing device 2. FIG. 7 is a diagram for explaining an example of the operation of the information processing device 2. FIG. 8 is a diagram for explaining an example of the operation of the information processing device 2. FIG. 9 is a diagram for explaining an example of the operation of the information processing device 2. FIG. 10 is a diagram for explaining an example of the operation of the information processing device 2. FIG. 11 is a diagram for explaining an example of the hardware configuration of the information processing device 2.

[0011] The present disclosure will be described below through embodiments, but the following embodiments do not limit the content of the present disclosure. Furthermore, not all of the configurations described in the embodiments are necessarily essential for solving the problems of the present disclosure.

[0012] <1. Overview> Before describing the developing device 3 according to this embodiment, one aspect of the process for manufacturing a printing plate from a resin plate will be described. First, ultraviolet light is irradiated from the back side of the resin plate (back exposure) to partially harden the back side. Next, a laser is irradiated onto a mask layer formed on the front side of the resin plate to write a desired pattern. This forms a mask layer with the desired pattern on the front side of the resin plate. Next, ultraviolet light is irradiated from the front side of the resin plate through the mask layer (relief exposure) to harden the front side of the resin plate into the desired pattern. In the relief exposure, the hardened portions become the convex portions of the printing plate, and the unhardened portions located below the mask layer become the concave portions of the printing plate.

[0013] The uncured portions on the surface side of the resin plate are then removed by development to obtain a printing plate. Development methods that can be used include, for example, solvent development using an organic solvent as the developer, water development using a developer that uses water as the solvent, and thermal development in which the resin plate is heated and dissolved and then removed with a wiper (such as a nonwoven fabric), but the application of the present disclosure is not particularly limited to these. For example, when solvent or water development is used, the uncured portions are removed using a developer and a brush to form a printing plate with an uneven surface. Finally, the printing plate may be dried and post-exposed to ultraviolet light from the surface side.

[0014] In the above, the portions hardened by ultraviolet irradiation become convex portions on the printing plate. However, the present invention is not limited to this, and the portions irradiated with ultraviolet rays may become concave portions on the printing plate.

[0015] Next, an example of the configuration of a developing device 3 according to this embodiment (hereinafter simply referred to as "developing device 3") will be described with reference to Figure 1. Figure 1(A) is a three-dimensional perspective view of the developing device 3, which uses a roll brush as the developing brush 300. Figure 1(B) is a schematic cross-sectional view in the width direction of the developing device 3, which uses a flat brush, and is a schematic cross-sectional view of the developing process. In addition to the developing brush 300, the developing device 3 is equipped with a resin plate insertion port 302, a developer tank 304, and a conveyor 308.

[0016] The developing brush 300 may be one or more roll brushes. The roll brush may be arranged so that its axial direction is approximately parallel to the width direction of the resin plate. The roll brush may rotate around its axis, may oscillate parallel to the axis, or may perform both rotation and oscillation. As the roll brush moves, the bristles rub against the surface of the resin plate, removing the uncured portions.

[0017] The developing brush 300 may also be a brush with bristle bundles arranged parallel to the plate surface, such as a flat brush or a cup brush. When using these brushes, the developing brush 300 may move along any trajectory on the plate surface. Furthermore, if the brush itself has a rotation axis perpendicular to the plate surface, the brush itself may rotate around that rotation axis. The movement and mechanism of the developing brush 300 can be implemented with reference to, for example, Japanese Patent Application Laid-Open No. 2020-46544, Japanese Patent Application Laid-Open No. 2021-113881, and International Publication No. 2020 / 217944.

[0018] Regardless of which of the above brushes is used, the developing solution may be supplied from above the brush during development, and the movement of the brush may scrape out uncured portions from the surface of the resin plate and discharge them downward together with the developing solution.

[0019] A resin plate including an uncured portion on its surface is inserted into the resin plate insertion port 302. The uncured portion may be, for example, a portion where ultraviolet light that cures the resin plate is masked (blocked / shielded) by a mask layer.

[0020] The developer tank 304 contains a developer 306. The developer 306 may be supplied from above the resin plate via a pipe. As described above, the uncured portions and developer scraped off the surface of the resin plate by the movement of the brush are discharged into the developer tank 304. In this way, the developer 306 is circulated and reused. The uncured portions in the developer, i.e., the resin concentration, increase over time. The developing device 300 may further include a filter system (not shown). The filter system removes the uncured portions contained in the developer 306 to prevent the resin concentration from increasing. The filter system may be installed, for example, at any position inside or outside the developer tank 304.

[0021] Furthermore, the developing device 3 may have a rinse tank (not shown) downstream of the developing step for rinsing the uneven portions of the printing plate after development with a rinse liquid. The rinse liquid may be supplied from above the printing plate via a pipe. The rinse liquid supplied to the uneven portions of the printing plate is scraped out by the rotation of the roll brush and discharged into the rinse tank.

[0022] In addition, the developing device 3 may include a drying device (not shown) that dries the printing plate after the developing process. The printing plate that has undergone the developing process is swollen with the developer and water. In the drying process, the printing plate may be dried by blowing hot air onto it.

[0023] The conveyor 308 is configured to convey the resin plate and the developed printing plate. Examples of such a conveyor 308 include, but are not limited to, a roller conveyor, a belt conveyor, and a chain conveyor. The conveyor 308 may also have a mounting section for the resin plate. The resin plate may be mounted on the conveyor 308 with the surface that has been subjected to the exposure treatment facing upward. The conveyor 308 may also be configured to transport the printing plate for rinsing treatment or drying treatment.

[0024] The resin plate inserted into the resin plate insertion port 302 is developed by the developing brush 300 and the developer 306 as it passes over the developer tank 304 while being transported on the transport machine 308, and then may be rinsed by passing over a rinse tank, and dried by passing through a drying device.

[0025] The developing device 3 may be equipped with a vibration sensor (not shown) that generates vibration data based on measurements of its own vibration (e.g., acceleration, etc.). The vibration sensor may continuously generate vibration data regardless of whether the developing device 3 is performing the resin plate development process. In this case, the vibration sensor may start generating vibration data when the developing device 3 is started, for example, and stop generating vibration data when the developing device 3 is stopped. Alternatively, the vibration sensor may generate vibration data for a predetermined period while the developing device 3 is running.

[0026] An example of the vibration sensor generating vibration data for a predetermined period of time is when the vibration sensor starts generating vibration data upon insertion of a resin plate into the resin plate insertion port 302. In this case, an objective sensor (not shown) that detects the insertion of the resin plate may be provided in the resin plate insertion port 302. The vibration sensor may receive a detection signal indicating the insertion of the resin plate from the objective sensor and start generating vibration data in response to the detection signal. In this case, the vibration sensor may stop generating vibration data after a predetermined time has elapsed since the resin plate was inserted into the resin plate insertion port 302 (for example, the time from when the resin plate is inserted into the resin plate insertion port 302 to when it is ejected from the conveyor 308, which may be calculated based on the conveying speed of the resin plate by the conveyor 308). Alternatively, the vibration sensor may stop generating vibration data upon detection of the resin plate by an objective sensor provided near the end of the conveyor 308.

[0027] Another example of the vibration sensor generating vibration data for a predetermined period of time is when the vibration sensor starts generating vibration data based on a user's operation to start the development process. That is, the vibration sensor may be configured to start generating vibration data substantially simultaneously with the start of the development process by the user's operation. The "user" may refer to a person using the development device 3, such as an operator of the development device 3. In this case, the development device 3 may be provided with a user interface (not shown). The user interface may be, for example, a touch panel, a display, or a button. When the user interface accepts a user's operation to start the development process, the vibration sensor may receive a control signal from the user interface via a communication interface such as a PLC (Power Line Communications) and start generating vibration data for the development process. In this case, the vibration sensor may stop generating vibration data a predetermined time after the user's operation to start the development process, or may stop generating vibration data in response to detection of a resin plate by an objective sensor provided near the end of the conveyor 308.

[0028] The vibration sensor may generate vibration data during the development process of the target resin plate. The target resin plate may be a resin plate for which an abnormality is estimated. When the developing device 3 executes development processes for multiple resin plates, each of the multiple resin plates may be the target resin plate, or some of the multiple resin plates may be the target resin plate. When the vibration sensor generates vibration data during a predetermined period, the predetermined period may be the period during which the development process of the target resin plate is performed. In this case, the vibration sensor may start generating vibration data based on the insertion of the target resin plate into the resin plate insertion port 302, or may start generating vibration data based on a user's operation when the target resin plate is inserted into the resin plate insertion port 302.

[0029] The developing device 3 may store the vibration data generated by the vibration sensor. The developing device 3 may also store any or all of information related to the start and end of the development process of the target resin plate and the start and stop of the generation of vibration data by the vibration sensor. In one example, the developing device 3 may include a storage device, and the information may be stored at least temporarily in the storage device. In another example, the developing device 3 may transmit the information to the information processing device 2 (described later) and store the information at least temporarily in the storage unit 12 of the information processing device 2. The acquiring unit 100 (described later) of the information processing device 2 may acquire the vibration data stored in this manner. The identifying unit 102 (described later) of the information processing device 2 may be configured to identify time-domain vibration data of the development process of the target resin plate by referring to at least one of the stored information related to the start and end of the development process of the target resin plate and the start and stop of the generation of vibration data.

[0030] Although details of the abnormality will be described later, from the viewpoint of estimating the state (abnormality) of the resin plate, brush, and developer with high accuracy, it is preferable that the vibration data in the developing process include vibrations caused by contact between the target resin plate and the brush and developer. The installation position of the vibration sensor is not particularly limited, but, for example, from the viewpoint of detecting vibrations caused by contact between the developing brush 300 and the resin plate in the presence of developer, it is preferable that the vibration sensor be installed in the area of ​​the conveyor 308 where the developing process is performed, on the device frame of the area where the developing process is performed, on the shaft of the developing brush 300, or on a frame that fixes the shaft of the developing brush 300.

[0031] In this disclosure, the plate fed into the developing device 3 is referred to as a "resin plate," and the plate output from the developing device 3 is referred to as a "printing plate."

[0032] Various abnormalities can occur during the resin plate development process. Possible abnormalities during the development process include, for example, the following (1) to (6): (1) Abnormalities of the developing brush 300 during the development process (e.g., loss of stiffness, wear, hardening, adhesion of resin dust, and density changes). (2) Abnormalities of the resin plate used during the development process (e.g., hardness, thickness, surface roughness, and coefficient of friction. Note that if the recommended grade of resin plate that can be used (or is used) differs depending on the developing device 3, the hardness, thickness, surface roughness, and coefficient of friction of a resin plate other than the recommended grade intended for use in the developing device 3 can be considered abnormal. The recommended grade can be specified based on, for example, the standards, specifications, and requirements to which the resin plate conforms). (3) Abnormalities of the developer 306 during the development process (e.g., temperature, resin concentration, surfactant concentration, and viscosity). (4) Abnormalities related to incorrect settings of development conditions (for example, incorrect settings of development speed, resin plate transport speed, brush rotation speed / oscillation speed, brush contact pressure, and aging deviation, etc.). (5) Abnormalities in the developing device 3 used in the development process (for example, pump failure or malfunction, clogging or pressure increase in the filter system, breakdown of the transport device 308, etc.). (6) Abnormalities in the printing plate obtained by the development process (for example, abnormal development depth, chipping, abnormal surface roughness or friction coefficient, etc.).

[0033] The developer 306 may be an aqueous developer containing a water solvent or a solvent-based developer containing an organic solvent. The aqueous developer may contain at least one of a surfactant, a development accelerator, and a pH adjuster. From the perspective of environmental adaptability, an aqueous developer is preferable to a solvent-based developer. Furthermore, while a solvent-based developer dissolves uncured portions easily as the developer 306, an aqueous developer is relatively poor at dissolving uncured portions, and removal of the uncured portions tends to depend on the developing brush 300. In other words, the frictional force between the developing brush 300 and the resin plate tends to be relatively strong, so the condition (abnormality) of the resin plate, brush, or developer is likely to be reflected in the vibration data. For these reasons, an aqueous developer is preferable for application of the present disclosure because it tends to enable highly accurate anomaly detection based on vibration data.

[0034] The above-described anomalies (1) to (6) can also cause anomalies in the printing plate from which the uncured portions have been removed. It is difficult to comprehensively and completely detect anomalies across the entire surface of a printing plate through visual inspection of the printing plate itself. Abnormalities are often only discovered after the printing plate is used in a printing process and it is discovered that the expected quality of printing cannot be achieved. If an anomaly is discovered in a printing plate, printing plates produced before and after the printing plate may also have the same anomaly, potentially forcing the disposal of these printing plates. In other words, if an anomaly can be predicted at an early stage, the development process can be interrupted, thereby reducing the loss of printing plates. System 1 according to this embodiment (hereinafter simply referred to as "system 1") can solve these problems. The detailed configuration and operation of system 1 are described below.

[0035] In this disclosure, "a state in which an abnormality has occurred in a printing plate" is not limited to a state in which the printing plate cannot be used, but also includes a state in which the printing plate can be used but does not exhibit the expected quality, and a state in which the likelihood of reaching such a state is increasing (i.e., a state in which signs of an abnormality have occurred).

[0036] Furthermore, in the present disclosure, "a state in which an abnormality has occurred in the development process" is not limited to a state in which the development process has stopped, but also includes the abnormalities exemplified in (1) to (6) above, as well as a state in which the efficiency of the development process (e.g., yield and number of printing plates produced per unit time) has decreased, and a state in which the likelihood of these states occurring is increasing.

[0037] Furthermore, the information processing device 2 disclosed in the present invention is not limited to a development method using a developing brush 300 and a developer 306, as in the example developing device 3, and can also be applied to a development device using a thermal development method, for example. In a thermal development device (not shown), for example, a resin plate is placed (fixed) on a support such as a drum or a belt conveyor supported by multiple rolls. The unexposed areas of the resin plate are softened or liquefied by direct heating from the drum or rolls or indirect heating from irradiation with hot air or infrared rays. Furthermore, an absorbent material supported by one or more winding rolls, separate from the support for the resin plate, comes into contact with the heated unexposed areas, removing the unexposed areas and forming a textured structure. The absorbent material is not particularly limited, and examples include paper and nonwoven fabric. In addition, the device may also be equipped with a steam treatment system that draws in, treats (detoxifies), and reuses air containing volatile organic compounds released when the resin plate is heated and liquefied or softened.

[0038] In a thermal development apparatus, various parameters can affect the printing plate, such as the rotation speed of the resin plate (or its support), the take-up speed of the absorbent material, the condition and type of the resin plate and absorbent material, the contact pressure between the resin plate and the absorbent material, the temperature of the resin plate surface, etc. Generally, vibrations are generated by the rotation of the resin plate (or its support) and the take-up of the absorbent material, and characteristic quantities related to these vibrations, such as intensity and frequency, vary depending on development conditions, such as the rotation speed of the resin plate, the take-up speed of the absorbent material, the contact pressure between the resin plate and the absorbent material, the condition and type of the resin plate and absorbent material, and the temperature of the resin plate surface. Therefore, by applying the information processing device 2 disclosed in the present invention, it is possible to detect abnormalities in the resin plate or absorbent material input into the thermal development process, abnormalities related to incorrect setting of development conditions (e.g., rotation speed of the resin plate, winding speed of the absorbent material, contact pressure between the resin plate and the absorbent material, heating temperature of the resin plate, etc.), abnormalities in the thermal development device (not shown) used in the development process (e.g., failure or malfunction of the motor or other driving parts, abnormalities in the steam treatment equipment, etc.), and abnormalities in the printing plate obtained by the development process (e.g., abnormalities in development depth, chipping, abnormalities in surface roughness or friction coefficient, etc.).

[0039] 2, the functional configuration of the system 1 will be described. The system 1 includes a developing device 3, an information processing device 2, and a communication network 4. The information processing device 2 and the developing device 3 are configured to be able to communicate with each other via the communication network 4.

[0040] [Information Processing Device 2] The information processing device 2 is a device that estimates an abnormality in the development process. The information processing device 2 may be a cloud server device, or may be a device built into the development device 3 or externally connected thereto.

[0041] The information processing device 2 includes a control unit 10, a storage unit 12, a network interface unit 14, and a bus 16. The control unit 10, the storage unit 12, and the network interface unit 14 are electrically connected via the bus 16.

[0042] (Control unit 10) The control unit 10 can function as an acquisition unit 100, an identification unit 102, a conversion unit 104, a counting unit 105, a generation unit 106, an estimation unit 108, and an output unit 110 by executing various programs stored in the memory unit 12 described below.

[0043] Acquisition Unit 100: The acquisition unit 100 acquires vibration data related to the vibration of the developing device 3. In one embodiment, the acquisition unit 100 acquires at least a portion of the vibration data generated by a vibration sensor included in the developing device 3. In one example, the acquisition unit 100 continuously (sequentially) acquires vibration data from the vibration sensor. In another example, the acquisition unit 100 acquires accumulated vibration data when the vibration data generated and stored by the vibration sensor reaches a predetermined capacity. In another example, the acquisition unit 100 acquires vibration data accumulated at predetermined intervals (e.g., the time interval between when a predetermined number of resin plates are inserted into the resin plate insertion slot 302 and when the predetermined number of resin plates are subsequently ejected from the conveyor 308). Note that, as described above, the vibration data generated by the vibration sensor is at least temporarily stored in, for example, a storage device included in the developing device 3 and the storage unit 12 of the information processing device 2, and the acquisition unit 100 can acquire the vibration data from these devices or configurations.

[0044] In one embodiment, the vibration data acquired by the acquisition unit 100 includes vibration data resulting from contact between the resin plate and at least one of the developing brush 300 and the developer 306. That is, the vibration data may reflect the state of the resin plate, the state of the developing brush 300, and the state of the developer 306. For example, if the resin plate has an abnormality or damage as described in (2) above, the vibration data acquired during the development process of that resin plate will differ from the vibration data acquired during the development process of a normal resin plate. Furthermore, for example, if the resin concentration (the concentration of unexposed resin dispersed or dissolved in the developer 306) or surfactant concentration of the developer 306 is excessive, the vibration data acquired during the development process while that developer 306 is being used will differ from the vibration data acquired during the development process using a developer 306 whose resin concentration or surfactant concentration is within the acceptable range. Furthermore, the vibration data acquired by the acquisition unit 100 may also include vibration data from all elements constituting the developing device 300, such as pumps, motors, blowers, chains, rollers, and belts. For example, if a pump that supplies developer from the developer tank 304 to the developer brush is clogged with resin and the flow rate is reduced, this abnormality can be reflected in the vibration data.

[0045] Vibration data can generally be expressed in the time domain and the frequency domain. Time-domain vibration data represents how vibration changes over time and can usually be expressed as a function of vibration level (e.g., acceleration or speed) with time as a variable. On the other hand, frequency-domain vibration data represents time-domain vibration data broken down into various frequency components and can usually be expressed as a function of amplitude with frequency as a variable. Analysis of time-domain vibration data tends to make it easier to capture characteristics such as vibration amplitude (magnitude), period (time over which vibration is repeated), and waveform (shape of vibration). Analysis of frequency-domain vibration data also tends to make it easier to identify problems occurring at a certain frequency based on the strength of vibration at a specific frequency. Hereinafter, frequency-domain vibration data will be referred to as "frequency-domain vibration data." In contrast, time-domain vibration data will be referred to as "time-domain vibration data." The acquisition unit 100 may acquire time-domain vibration data. Furthermore, the frequency-domain vibration data may be obtained as an output of the conversion unit 104, which will be described later.

[0046] Hereinafter, when there is no particular distinction between the time domain vibration data and the frequency domain vibration data, or when the time domain vibration data and the frequency domain vibration data are referred to collectively, they will be referred to as "vibration data".

[0047] In one embodiment, the acquisition unit 100 acquires time-domain vibration data identified by the identification unit 102, which will be described later. As will be described in detail later, the identification unit 102 identifies time-domain vibration data of the development process of the target resin plate. That is, in one embodiment, the acquisition unit 100 acquires time-domain vibration data of the development process of the target resin plate.

[0048] In the present disclosure, "obtaining information" includes making the information processable in the control unit 10. "Obtaining information" may mean receiving the information from another device, reading the information from the storage unit 12, or obtaining the information as a result of a predetermined process.

[0049] The developing device 3 sequentially takes in resin plates through the resin plate insertion port 302 and transports multiple resin plates, including the target resin plate, on the transporter 308. In this case, if the vibration sensor of the developing device 3 is configured to continuously generate vibration data while the developing device 3 is running, for example, it is assumed that it is impossible to identify which part of the series of vibration data corresponds to which resin plate based on the vibration data alone. Even if the vibration sensor of the developing device 3 is configured to start generating vibration data in response to the insertion of a resin plate, given that the developing device 3 may develop multiple resin plates simultaneously, the configuration of this vibration sensor still makes it impossible to identify which part of the vibration data corresponds to which resin plate. The identifying unit 102 identifies time-domain vibration data for the development process of the target resin plate from the vibration data related to the vibration of the developing device 3 in this example.

[0050] In one embodiment, the identifying unit 102 identifies time-domain vibration data for the development process of a target resin plate based on a list in which the identification information of each of multiple resin plates is associated with the development time of the resin plate. For example, time-domain vibration data for the development processes of a first resin plate, a second resin plate, and a third resin plate is available. The list includes information such as "(1) First resin plate: development process from time point 1 to time point 2, (2) Second resin plate: development process from time point 3 to time point 4, and (3) Third resin plate: development process from time point 5 to time point 6." If the target resin plate is the second resin plate, the identifying unit 102 identifies the time-domain vibration data from time point 3 to time point 4 from the entire time-domain vibration data. The identification information of the resin plate may be, for example, the order information of the resin plates inserted into the developing device 3 (e.g., first plate, second plate, third plate, etc.). The order information may be counted by a job counter provided in the developing device 3.

[0051] The development time may be the start time of development, or may be a combination of the start time of development and the end time of development. If the time required for development is known, the end time of development may be determined based on the start time of development. For example, as described in paragraphs

[0026] -

[0027] , the development time may be determined based on information such as the time when the resin plate is inserted into the resin plate insertion port 302 and the time when the development process is started based on an operation on the user interface.

[0052] An example of the operation of the identification unit 102 will be described with reference to FIG. 3 . The graph in FIG. 3 is an example of time-domain vibration data in the developing process of the nth to n+2th resin plates. The list includes information indicating that the developing process of the nth resin plate started at time t1, the developing process of the n+1th resin plate started at time t2, and the developing process of the n+2th resin plate started at time t3. In the example of FIG. 3 , the time required for the developing process of one resin plate is T. Note that the time T required for the developing process may vary depending on the resin plate. FIG. 3 also illustrates an example in which the vibration sensor continuously generates vibration data regardless of whether the developing device 3 is performing the resin plate developing process.

[0053] The time period from time t2 to time t1+T corresponds to the time period during which the developing process for the nth and (n+1)th resin plates is being performed in the developing device 3. The time period from time t2+T to time t3 corresponds to the time period during which the developing process is not being performed and the device is in a standby state. In this standby state, vibration data including vibrations of a pump, blower, etc. may be generated. If the vibration sensor starts generating vibration data based on the start of the developing process, vibration data need not be generated during the standby time period from time t2+T to time t3 in FIG. 3 .

[0054] If the nth resin plate is the target resin plate, the identification unit 102 identifies time-domain vibration data s1[t] from time t1 to time t1+T. If the n+1th resin plate is the target resin plate, the identification unit 102 identifies time-domain vibration data s2[t] from time t2 to time t2+T. If the n+2th resin plate is the target resin plate, the identification unit 102 identifies time-domain vibration data s3[t] from time t3 to time t3+T.

[0055] - Transformation Unit 104 - The transformation unit 104 transforms the time domain vibration data into frequency domain vibration data. In one embodiment, the transformation unit 104 applies a Fourier transform to the time domain vibration data. The transformation unit 104 may perform the Fourier transform using an FFT (Fast Fourier Transform) or a DFT (Discrete Fourier Transform). In the following description, it is assumed that the transformation unit 104 transforms the time domain vibration data into frequency domain vibration data using an FFT.

[0056] In one embodiment, the converter 104 divides the time-domain vibration data during the development process of the target resin plate into a predetermined number of divisions and converts each division into the frequency domain. If the development process of one resin plate takes a long time, the length of the time-domain vibration data during the development process also increases. Therefore, the frequency-domain vibration data obtained by applying FFT or the like to the time-domain vibration data increases in number while maintaining a constant frequency range. In other words, if the development process of one resin plate takes a long time, the resolution of the frequency-domain vibration data increases. If the resolution of the frequency-domain vibration data becomes excessively high, noise in the frequency-domain vibration data may increase. In response to this, the converter 104 divides the time-domain vibration data into a predetermined number of divisions and converts each division into the frequency domain, thereby preventing the resolution of the frequency-domain vibration data from becoming excessively high, thereby suppressing noise. The length of the time-domain vibration data can also be referred to as the window length. Furthermore, increasing the resolution of the frequency-domain vibration data can also be referred to as reducing the bandwidth corresponding to each element of the frequency-domain vibration data.

[0057] An example of the operation of the conversion unit 104 will be described with reference to Fig. 4-5. Fig. 4 shows how the time domain vibration data s1[t] of the graph in Fig. 3 is divided. In the example of Fig. 4, the conversion unit 104 divides the time domain vibration data s1[t] into four pieces: time domain vibration data s11[t], time domain vibration data s12[t], time domain vibration data s13[t], and time domain vibration data s14[t].

[0058] 5 shows how frequency domain vibration data is obtained based on the divided time domain vibration data. In the example of FIG. 5, the conversion unit 104 converts the time domain vibration data s11[t], time domain vibration data s12[t], time domain vibration data s13[t], and time domain vibration data s14[t] into frequency domain vibration data g11[f], frequency domain vibration data g12[f], frequency domain vibration data g13[f], and frequency domain vibration data g14[f], respectively. Also, in the example of FIG. 5, the conversion unit 104 averages these frequency domain vibration data to obtain frequency domain vibration data g1[f]. Since the frequency domain vibration data g1[f] is determined based on the time domain vibration data s1[t] in the development process of the nth resin plate, it can also be said that it is frequency domain vibration data corresponding to the development process of the nth resin plate.

[0059] 4-5 shows an example in which the division number is "4", but the division number may be other values. In one embodiment, the division number of the time domain vibration data is determined based on the estimation accuracy of abnormalities in the development process when each of a plurality of division numbers is set as the division number of the time domain vibration data. Details will be described later with reference to FIG. 9.

[0060] Furthermore, the time-domain vibration data converted into frequency-domain vibration data by the conversion unit 104 does not necessarily represent vibration data for the entire time period from the beginning to the end of the development process of the target resin plate. That is, before the division process described with reference to FIGS. 4-5 , the time-domain vibration data of the target resin plate may be cut based on a certain criterion and used. For example, in the case of the time-domain vibration data from t1 to t1+T in FIG. 4 , the time-domain vibration data from t1 to t1+T' (0<T'<T) may be cut in advance and converted by the conversion unit 104. T' can be set arbitrarily. Furthermore, although the end point t1+T of the development process has been described, the start point t1 may also be cut in a similar manner.

[0061] - Aggregation Unit 105 - The aggregation unit 105 generates a dataset to be input into a model, which will be described later. In one embodiment, the aggregation unit 105 generates a dataset including a plurality of time-domain vibration data acquired by the acquisition unit 100. In one embodiment, the aggregation unit 105 generates a dataset including a plurality of frequency-domain vibration data obtained by the conversion unit 104. In one embodiment, the aggregation unit 105 generates a dataset including development process data, which will be described later. The dataset generated by the aggregation unit 105 may be used as at least one of learning data and estimation data.

[0062] An example of a data set generated by the tallying unit 105 is shown in FIG. 6. In the table of FIG. 6, frequency domain vibration data and development process data corresponding to the development process of each resin plate are associated with each resin plate identification number. Note that the development process data items shown in FIG. 6 (time required for development, feed speed, brush type, brush oscillation speed, and developer temperature) are merely examples, and the development process data may include other items.

[0063] According to the table of FIG. 6 , for example, the frequency domain vibration data for a resin plate with an identification number of “1” is indicated as g1[1], g1[2], g1[3], etc. for each frequency component, and the development process data for that resin plate is indicated as p11, p12, p13, etc. for each item. Similarly, for example, the frequency domain vibration data for a resin plate with an identification number of “2” is indicated as g2[1], g2[2], g2[3], etc. for each frequency component, and the development process data for that resin plate is indicated as p21, p22, p23, etc. for each item. Note that the values ​​associated with the development process data items (p11, p12, p13, etc.) are merely examples, and at least some of these values ​​may be common. For example, if a resin plate with an identification number of “1” and a resin plate with an identification number of “2” are developed using the same brush, p13 and p23 may be common.

[0064] -Generation Unit 106- The generation unit 106 generates a model. In one embodiment, the model is constructed by learning vibration data in a development process of a resin plate other than the target resin plate. Having the model learn the vibration data may include having the model learn the feature quantities of the vibration data. Hereinafter, the vibration data that is learned by the model will be referred to as learning data.

[0065] In one embodiment, the training data includes frequency-domain vibration data obtained by the conversion unit 104. That is, the training data may include frequency-domain vibration data obtained by dividing time-domain vibration data in the development process of another printing plate into a predetermined number of divisions and converting each division into the frequency domain (see FIGS. 4-5 ). In another embodiment, the training data includes time-domain vibration data acquired by the acquisition unit 100 and / or data or statistics obtained by processing the time-domain vibration data using a general data preprocessing method (e.g., normalization, standardization, background correction, extraction of lag features obtained by detecting autocorrelation, etc.).

[0066] Furthermore, the learning data may be further included in development process data that can identify the processing details of the development process for the target resin plate. The development process data includes, for example, information regarding the time required for development, the feed speed of the conveying unit 308, the type of developing brush 300, the rotation speed of the developing brush 300, the oscillation speed of the developing brush 300, other parameters related to the movement of the developing brush 300, the temperature of the developer 306, and the size of the resin plate. The development process data may be setting values ​​for an operating unit that controls the developing brush 300, the conveying unit 308, the developer supply unit, etc. of the developing device 3. The setting values ​​may be input by the user to the operating unit. The development process data may also be actual measurement values ​​measured by various sensors installed in the developing device 3. The development process data may also include both setting values ​​and actual measurement values.

[0067] In one embodiment, the generator 106 may generate a model based on the data set generated by the aggregation unit 105 .

[0068] The model may be based on unsupervised learning or supervised learning.

[0069] The unsupervised learning model may be, for example, a model based on an algorithm such as OCSVM (One Class Support Vector Machine), MT (Mahalanobis-Taguchi method), PCA (Principal Component Analysis), Auto Encoder, or k-nearest neighbor algorithm. The generation unit 106 may apply a kernel function when determining the boundary surface of the model. For example, when applying OCSVM, the generation unit 106 may input all learning data to the unsupervised learning model, assuming that the data corresponds to a normal printing plate or development process. Alternatively, if the anomaly rate is expected to be large enough to be non-negligible, the generation unit 106 may perform model learning by assuming an anomaly rate v (≠0). The anomaly rate v may be used as a hyperparameter to select a model with the highest anomaly detection accuracy. The anomaly detection model obtained in this manner can estimate the presence or absence of an anomaly in the development process (whether an anomaly exists or not).

[0070] The model based on supervised learning may be a model based on an algorithm such as a neural network, a tree classification, an SVM, etc. The generation unit 106 may cause the model to learn information that associates, as labels, the degree and / or type of an abnormality in the development process that corresponds to the learning data.

[0071] In one embodiment, the generation unit 106 may generate a model by learning first learning data related to vibration data in a development process of a resin plate having a printing surface of a first shape and second learning data related to vibration data in a development process of a resin plate having a printing surface of a second shape different from the first shape. The vibration data may vary depending on the shape of the printing surface. This configuration makes it possible to generate a general-purpose model that is not dependent on the shape of the printing surface.

[0072] The difference between the first shape and the second shape is defined by, for example, image quality, line frequency (lpi), development depth, the presence or absence of halftone dots, and halftone dot area ratio. The image quality is the area ratio of the convex portions in the relief structure of the printing plate (i.e., the ratio of the area of ​​the convex portions on the printing surface to the total area of ​​the printing plate (i.e., the sum of the area of ​​the convex portions and the area of ​​the concave portions on the printing surface)), and can take a value between 0% and 100%. The image quality may also be the area ratio of the concave portions in the relief structure of the printing plate. From the viewpoint of improving the generalization performance of the model, it is preferable that the image quality of the first shape and the second shape differ by 10% or more. By training the model with training data related to printing plates with somewhat different image quality ratios, the model can more easily distinguish between vibration data features caused by differences in the shape of the printing plate and not by abnormalities in the development process and vibration data features caused by abnormalities in the development process. The line frequency can take a value of approximately 50 to 250 lpi. When multiple line rulings exist within the same printing plate, the highest line ruling is used, and it is preferable that the difference between the first and second shapes be 10 lpi or more. The development depth can take a value of approximately 0.2 to 5.0 mm. When multiple development depths exist within the same printing plate, the deepest development depth is used, and it is preferable that the difference between the first and second shapes be 0.2 mm or more. In addition, it is preferable that one of the first and second shapes does not contain halftone dots, and the other contains halftone dots. The above-mentioned preferable conditions regarding the difference between the first and second shapes may be used alone or in combination.

[0073] An example of a model generated by the generation unit 106 will be described with reference to FIG. 7 . FIGS. 7A and 7B are conceptual diagrams in which feature quantities of training data are plotted on a plane. FIG. 7A shows a boundary surface when training data relating to vibration data in the development process of a resin plate having a printing surface of a single shape is input to the model. FIG. 7B shows a boundary surface when first training data relating to vibration data in the development process of a resin plate having a printing surface of a first shape and second training data relating to vibration data in the development process of a resin plate having a printing surface of a second shape different from the first shape are input to the model. The estimation unit 108, which will be described later, estimates that the development process of the target resin plate is normal if the estimation data is plotted inside the boundary surface, and estimates that an abnormality exists in the development process of the target resin plate if the estimation data is plotted outside the boundary surface. Furthermore, although not shown in the figure, not only a boundary surface corresponding to "normal" but also a boundary surface corresponding to "abnormal" may be determined, and boundary surfaces corresponding to one or more types of abnormality (for example, a boundary surface corresponding to "damage to the developing brush 300" and a boundary surface corresponding to "abnormality in the developer 306") may be determined. Note that, although Fig. 7 plots the feature amounts of the learning data on a plane for convenience of explanation, the feature amounts of the learning data may be three-dimensional or more.

[0074] Similarly, the generation unit 106 may generate a model configured by learning n-th learning data related to vibration data in the development process of a resin plate having a printing surface of an n-th shape, where n = 1, 2, ..., N (where N is 3 or more). The larger N is, the more likely it is that the effect of the shape of the printing surface on the vibration data can be treated as being statistically sufficiently small, making it possible to generate a more versatile model.

[0075] The generating unit 106 may generate one or more of a model for estimating the presence or absence of an abnormality, a model for estimating the degree of the abnormality, and a model for estimating the type of the abnormality.

[0076] In the present disclosure, "generating a model" may mean either making the model available to the control unit 10 or inputting learning data into the model.

[0077] - Estimation Unit 108 - The estimation unit 108 estimates an abnormality in the development process based on the vibration data acquired by the acquisition unit 100. The estimation unit 108 estimating an abnormality in the development process may include detecting an abnormality in the development process of the target resin plate. The abnormality in the development process of the target resin plate may be any one of an abnormality in the target resin plate, an abnormality in the developing device 3, and an abnormality in the printing plate.

[0078] In one embodiment, the estimation unit 108 estimating an abnormality in the development process includes estimating the presence or absence of an abnormality or the degree of an abnormality in the development process of the target resin plate.

[0079] In one embodiment, the estimation unit 108 estimating the abnormality in the development process includes estimating the type of abnormality in the development process of the target resin plate. The type of abnormality in the development process of the target resin plate may be any of the type of abnormality in the target resin plate, the type of abnormality in the developing device 3, and the type of abnormality in the printing plate. The estimation unit 108 estimating the type of abnormality in the developing device 3 may include estimating whether the abnormality is an abnormality in the developing brush 300, an abnormality in the developer 306, or an abnormality in the process settings of the development process of the target resin plate.

[0080] In one embodiment, the estimation unit 108 estimates an abnormality in the development process based on development process data that can identify the processing details of the development process for the target resin plate. The development process data includes, for example, information on the time required for development, the feed speed of the conveying unit 308, the type of developing brush 300, the rotation speed of the developing brush 300, the oscillation speed of the developing brush 300, other parameters related to the movement of the developing brush 300, the temperature of the developer 306, and the size of the resin plate. The development process data may be set values ​​for an operating unit that controls the developing brush 300, the conveying unit 308, the developer supply unit, etc. of the developing device 3. The set values ​​may be input by a user to the operating unit. The development process data may also be actual measurements measured by various sensors installed in the developing device 3. The development process data may also include both set values ​​and actual measurements.

[0081] In one embodiment, the estimation unit 108 estimates an abnormality in the development process based on the time domain vibration data acquired by the acquisition unit 100 or the frequency domain vibration data obtained by the conversion unit 104, and the model generated by the generation unit 106. Hereinafter, data input to the model for estimating an abnormality in the development process will be referred to as estimation data.

[0082] 7 again, an example of the operation of the estimation unit 108 will be described. When the estimation data is plotted inside the boundary surface, it is estimated that the development process is normal (i.e., no abnormality exists). On the other hand, when the estimation data is plotted outside the boundary surface, it is determined that an abnormality exists in the development process.

[0083] The estimation unit 108 may estimate the degree of abnormality based on the distance between the estimation data and the center of gravity of the plurality of learning data. Specifically, the estimation unit 108 may determine that the greater the distance between the estimation data and the center of gravity of the plurality of learning data, the greater the degree of abnormality, and that the smaller the distance, the smaller the degree of abnormality. Note that the distance may be Euclidean distance or Mahalanobis distance based on the standard deviation of the plurality of learning data.

[0084] The estimation unit 108 may estimate the degree of abnormality based on the distance between the estimation data and a boundary surface determined based on a plurality of learning data. Specifically, the estimation unit 108 may determine that the degree of abnormality is greater as the distance between the estimation data and the boundary surface is greater, and that the degree of abnormality is smaller as the distance is smaller.

[0085] If the model generated by the generation unit 106 includes information regarding boundary surfaces corresponding to each of one or more types of abnormality, the estimation unit 108 can estimate the type of abnormality based on which of the one or more types of abnormality the data for estimation belongs to. For example, if a boundary surface corresponding to "damage to the developing brush 300" and a boundary surface corresponding to "abnormality in the developer 306" are defined in the model, if the data for estimation belongs to the boundary surface corresponding to "damage to the developing brush 300", the estimation unit 108 may estimate that "damage to the developing brush 300" is the type of abnormality, and if the data for estimation belongs to the boundary surface corresponding to "abnormality in the developer 306", the estimation unit 108 may estimate that "abnormality in the developer 306" is the type of abnormality.

[0086] In the case where the generation unit 106 generates a model for estimating the presence or absence of an abnormality, a model for estimating the degree of the abnormality, and a model for estimating the type of the abnormality, the estimation unit 108 may first input estimation data to the model for estimating the presence or absence of an abnormality, and if it is estimated that an abnormality exists based on the input data, input the estimation data to the model for estimating the degree of the abnormality or the model for estimating the type of the abnormality. In other words, the estimation unit 108 may estimate the presence or absence of an abnormality in the development process and the degree or type of the abnormality in the development process of the target resin plate in two stages.

[0087] In one embodiment, the estimation unit 108 may estimate an abnormality in the development process based on the data set generated by the aggregation unit 105 .

[0088] - Output Unit 110 - Based on the estimation by the estimation unit 108 that there is an abnormality in the printing plate, the output unit 110 outputs a notification regarding the abnormality. The notification may be displayed on a display screen provided in the developing device 3, or may be displayed on a display screen of a terminal device used by an operator of the developing device 3, etc. The notification may be output by voice. The notification may include information regarding at least one of the presence or absence of an abnormality, the degree of the abnormality, and the type of the abnormality. The notification may be a control signal for controlling the developing device 3. The control signal may include a signal to interrupt the development process.

[0089] An example of the relationship between the components of the acquisition unit 100, the identification unit 102, the conversion unit 104, the aggregation unit 105, the generation unit 106, the estimation unit 108, the output unit 110, and the development device 3 will be described with reference to FIG.

[0090] First, the vibration data generated by the vibration sensor is stored in the storage unit 12 via the communication interface of the developing device 3, the communication network 4, and the network interface unit 14 of the information processing device 2 (S1-S3). Next, the acquisition unit 100 acquires the vibration data series stored in the storage unit 12 at a predetermined timing (S4). An example of the vibration data series is a series of vibration data from the start of s1[t] to the end of s3[t] in FIG. 3.

[0091] Next, the acquiring unit 100 passes the vibration data series to the identifying unit 102 (S5). The identifying unit 102 identifies time-domain vibration data in the development process of each of the multiple resin plates and returns the data to the acquiring unit 100 (S6). Examples of the multiple time-domain vibration data are s1[t], s2[t], and s3[t] in FIG. 3.

[0092] Next, the acquisition unit 100 passes the plurality of time domain vibration data to the conversion unit 104 (S7). The conversion unit 104 converts each of the plurality of time domain vibration data into frequency domain vibration data and passes it to the aggregation unit 105 (S8). An example of frequency domain vibration data is g1[t] in FIG. 5 .

[0093] The aggregation unit 105 passes the plurality of frequency domain data sets to the generation unit 106 as a training data set (S9). The generation unit 106 generates a model based on the training data set and passes the model to the estimation unit 108 (S10). The aggregation unit 105 also passes the other plurality of frequency domain data sets to the estimation unit 108 as an estimation data set (S11). The estimation unit 108 estimates an abnormality in the development process based on the model and the estimation data set and passes the estimation result to the output unit 110 (S12). The output unit 110 outputs a notification from the user interface of the developing device 3 to confirm the estimation result (S13).

[0094] (Storage Unit 12) The storage unit 12 stores various types of information that must be stored for the operation of the information processing device 2. The storage unit 12 stores various programs that the control unit 10 executes.

[0095] (Network Interface Unit 14 ) The network interface unit 14 realizes communication with other devices or systems via the communication network 4 .

[0096] [Communication Network 4] The communication network 4 realizes mutual communication between the information processing device 2 and the developing device 3. The communication network 4 realizes communication based on, for example, the TCP / IP protocol.

[0097] 3. Operation An example of the operation of the information processing device 2 will be described with reference to FIGS.

[0098] First Embodiment An example of the operation of the information processing device 2 according to the first embodiment will be described with reference to FIGS.

[0099] 9 is a flowchart showing an example of the operation of the information processing device 2 when generating a model. First, the information processing device 2 sets the value of a variable n, which is a variable used for repeated processing, to "1" (S100). Next, the information processing device 2 acquires time-domain vibration data in the development process of the nth resin plate (S102). The time-domain vibration data in the development process of the nth resin plate corresponds to the time-domain vibration data s1[t] in FIG. 3-4 described above. At this point, the value of the variable n is "1," so the information processing device 2 acquires time-domain vibration data in the development process of the first resin plate.

[0100] Next, the information processing device 2 divides the acquired time domain vibration data into M parts (S104), where M is the number of divisions into which the time domain vibration data is divided in the developing process of one resin plate. In the example shown in FIG. 4, M is four.

[0101] Next, the information processing device 2 sets the value of variable m, which is another variable used in the repeated processing, to "1" (S106). Next, the information processing device 2 applies FFT to the m-th divided portion of the time-domain vibration data divided in S104 (S108). At this point, since variable m is "1", the information processing device 2 applies FFT to the first divided portion of the time-domain vibration data. The first divided portion of the time-domain vibration data corresponds to the time-domain vibration data s11[t] in FIGS. 4-5 described above. Furthermore, the result of applying FFT to the first divided portion of the time-domain vibration data corresponds to the frequency-domain vibration data g11[f] in FIG. 5 described above.

[0102] Next, the information processing device 2 determines whether the value of the variable m is equal to the division number M (S110). If the value of the variable m is not equal to the division number M (NO in S110), the variable m is incremented (S112) and an FFT is applied to the next division portion of the time-domain vibration data (S108). The information processing device 2 repeatedly executes the processes of S108-S112 to apply an FFT to each of the first to M-th division portions of the time-domain vibration data. The results of applying the FFT to each of the first to M-th division portions correspond to the frequency-domain vibration data g11[f], frequency-domain vibration data g12[f], frequency-domain vibration data g13[f], and frequency-domain vibration data g14[f] in FIG. 5 described above. Hereinafter, the processes of S104-S112 will be referred to as "FFT processing fp."

[0103] If the value of the variable m is equal to the division number M (YES in S110), the information processing device 2 averages the results of applying FFT to each of the first to Mth divisions of the time-domain vibration data to obtain frequency-domain vibration data corresponding to the developing process of the nth resin plate (S114). The frequency-domain vibration data corresponding to the developing process of the nth resin plate corresponds to the frequency-domain vibration data g1[f] in FIG. 5 described above.

[0104] Next, the information processing device 2 determines whether the value of the variable n is equal to the learning amount N (S116). The learning amount N can also be referred to as the number of frequency domain vibration data to be learned. If it is determined that the value of the variable n is not equal to the learning amount N (NO in S116), the variable n is incremented (S118) and time domain vibration data for the developing process of the next resin plate is obtained (S102). The information processing device 2 repeatedly executes the processes of S102-S118 to obtain frequency domain vibration data corresponding to the developing process of the first resin plate through frequency domain vibration data corresponding to the developing process of the Nth resin plate.

[0105] If it is determined that the value of the variable n is equal to the learning amount N (YES in S116), the information processing device 2 inputs the frequency domain vibration data corresponding to the developing processes of the first to Nth resin plates into the model as learning data (S120).Then, the information processing device 2 ends the process.

[0106] The learning amount N may be determined in advance. The learning amount N is preferably equal to or greater than 50 resin plates (N=100), more preferably equal to or greater than 100 resin plates (N=100), and even more preferably equal to or greater than 200 resin plates (N=200). Furthermore, from the viewpoint of reducing the load on data processing and shortening the period for collecting learning data, it is preferable that the learning amount N be equal to or less than 1000 resin plates (N=1000).

[0107] FIG. 10 is a flowchart showing an example of the operation of the information processing device 2 when estimating an abnormality in the development process using a model. The information processing device 2 first acquires time-domain vibration data in the development process of the target resin plate (S200). Next, the information processing device 2 performs FFT processing fp on the acquired time-domain vibration data to acquire frequency-domain vibration data corresponding to the development process of the target resin plate (S104-S112). Next, the information processing device 2 inputs the frequency-domain vibration data corresponding to the development process of the target resin plate into the model as estimation data (S202). If the information processing device 2 estimates that an abnormality exists in the development process (YES in S204), it outputs a notification indicating this (S206). If the information processing device 2 estimates that no abnormality exists in the development process (NO in S204), it terminates the process. The information processing device 2 can repeatedly execute the processes of S200-S206. In another embodiment, steps S200-S112 are first repeated for multiple target resin plates to generate multiple estimation data sets (data sets), which are then input into the model, thereby enabling abnormalities in multiple target resin plates to be estimated simultaneously.

[0108] Second Embodiment In the first embodiment, the time domain vibration data in the resin plate development process is described as being divided by the division number M. In the second embodiment, an example of a method for determining the division number M will be described.

[0109] The information processing device 2 according to the second embodiment executes a learning phase and a test phase for a plurality of division numbers M. In the learning phase, a model is generated. In the test phase, the accuracy of anomaly estimation using the generated model is tested. Thereafter, the information processing device 2 determines, from the plurality of division numbers M, the division number M that provides the highest accuracy of anomaly estimation. Hereinafter, an example of the operation of the information processing device 2 according to the second embodiment will be described with reference to FIG. 11 .

[0110] First, the information processing device 2 executes a learning phase (S100-S120) with an initial value of the division number M. The learning phase may be the same as the operation described with reference to FIG. 9, and therefore will not be described below.

[0111] Next, the information processing device 2 starts the test phase. First, the information processing device 2 sets the value of a variable k, which is another variable used in the repeated processing, to "1" (S300).

[0112] Next, the information processing device 2 acquires time-domain vibration data for the development process of the kth resin plate from the dataset of time-domain vibration data for the development process including the abnormality (S302). The dataset of time-domain vibration data for the development process including the abnormality is, for example, time-domain vibration data for the development process of each of 100 resin plates, and is time-domain vibration data for five of the 100 resin plates in which an abnormality was actually found in the printing plate after development of the resin plate (for example, by visual inspection). In this case, ideally, it is estimated that an abnormality exists in five printing plates from the dataset of time-domain vibration data for the development process including the abnormality.

[0113] Next, the information processing device 2 performs FFT processing fp with the division number M on the time domain vibration data in the developing process of the kth resin plate.

[0114] Next, the information processing device 2 determines whether the value of the variable k is equal to the test amount K (S306). The test amount K can also be referred to as the number of frequency domain vibration data to be tested. If it is determined that the value of the variable k is not equal to the test amount K (NO in S306), the information processing device 2 increments the variable k (S308) and acquires time domain vibration data in the development process of the next resin plate from the data set of time domain vibration data in the development process including the abnormality (S302).

[0115] The test amount K may be the number of data in a data set of time-domain vibration data of the development process including an abnormality. For example, if the data set of time-domain vibration data of the development process including an abnormality is time-domain vibration data in the development process of 100 resin plates, the test amount K may be "100".

[0116] If it is determined that the value of the variable k is equal to the test amount K (YES in S306), the information processing device 2 inputs each of the multiple frequency domain vibration data obtained by repeating S302-S308 into the model as estimation data (S310). This enables the information processing device 2 to estimate an abnormality in the development process of the first to Kth resin plates.

[0117] Next, the information processing device 2 calculates the abnormality estimation rate (S311). The abnormality estimation rate may be, for example, any of (1) to (3) below. Note that these are merely examples, and other calculation methods may be used. (1) The ratio between the number of data points in a data set of time-domain vibration data for the development process that includes an abnormality and the number of printing plates that are estimated to have an abnormality. For example, if the data set of time-domain vibration data for the development process that includes an abnormality is time-domain vibration data for each development process of 100 resin plates, and it is estimated that two of the 100 resin plates have an abnormality after development of the resin plates, the abnormality estimation rate may be calculated as 2 / 100 = 2%. (2) The ratio between the number of printing plates in which an abnormality was actually found and the number of printing plates that are estimated to have an abnormality. For example, if, as a result of the development process of 100 resin plates, abnormalities are actually found in five printing plates, and it is estimated that two of the 100 resin plates have abnormalities after development of those resin plates, the abnormality estimation rate may be calculated as 2 / 5 = 40%. (3) The ratio between the number of data points in the data set of time-domain vibration data of the development process including abnormalities and the sum of the following (3-A) and (3-B). (3-A) The number of printing plates that were estimated to have abnormalities and for which abnormalities were actually found. (3-B) The number of printing plates that were estimated to have no abnormalities and for which no abnormalities were actually found.

[0118] Next, the information processing device 2 determines whether to try another division number M (S312). For example, the information processing device 2 may set the number of trials for another division number M in advance and make this determination based on whether that number of trials has been performed. The values ​​(combinations) of the other division numbers M to be tried can be determined in advance (generated based on some rule or formula), or another division number M can be generated randomly for each trial. Alternatively, an optimization algorithm such as a genetic algorithm or gradient descent method can be used to determine the next value of the other division number M to be tried. If another division number M is to be tried (YES in S312), the information processing device 2 changes the division number M (S314) and executes the learning phase and test phase again. If another division number M is not to be tried (NO in S312), the information processing device 2 determines the division number M with the highest anomaly estimation rate (S316). For example, if the learning phase and test phase are executed with the division numbers M set to "4," "8," and "16," and the anomaly estimation rate is highest when the division number M is set to "8," the information processing device 2 determines the division number M to be "8."

[0119] The division number M at which the abnormality detection rate increases may differ depending on the shape of the printing surface, the contents of the developing process, etc. With this configuration, an appropriate division number M can be determined depending on any situation.

[0120] 12, an example of a hardware configuration in which the above-described information processing device 2 is realized by a computer 70 will be described. Note that the functions of each device can also be realized by dividing them into multiple devices.

[0121] As shown in FIG. 12, the computer 70 includes a processor 700 , a storage device 702 , an input I / F 704 , a data I / F 706 , a communication I / F 708 , and a display device 710 .

[0122] The processor 700 controls various processes in the computer 70 by executing programs stored in the storage device 702. For example, each functional unit included in the control unit 10 of the information processing device 2 can be realized by the processor 700 executing the programs stored in the storage device 702.

[0123] The storage device 702 is a storage medium such as a RAM (Random Access Memory), etc. The RAM temporarily stores the program code of the program executed by the processor 700 and data required when the program is executed.

[0124] The storage device 702 may also be a non-volatile storage medium such as a hard disk drive (HDD) or flash memory. The storage device 702 stores an operating system and various programs for implementing the above-described configurations. The storage medium storing the various programs may be a non-transitory computer-readable medium. The storage device 702 may also store tables for registering various pieces of information and a database for managing the tables. Such programs and data are loaded into the storage device 702 as needed and referenced by the processor 700.

[0125] The input I / F 704 is a device for receiving input from a user. Specific examples of the input I / F 704 include a camera, a button, a microphone, a keyboard, a mouse, a touch panel, various sensors, and a wearable device. The input I / F 704 may be connected to the computer 70 via an interface such as a USB (Universal Serial Bus).

[0126] The data I / F 706 is a device for inputting data from outside the computer 70. A specific example of the data I / F 706 is a drive device for reading data stored in various storage media. The data I / F 706 may be provided outside the computer 70. In this case, the data I / F 706 is connected to the computer 70 via an interface such as a USB.

[0127] The communication I / F 708 is a device for performing data communication with devices external to the computer 70 via the communication network 4, either wired or wirelessly. The communication I / F 708 may be provided external to the computer 70. In this case, the communication I / F 708 is connected to the computer 70 via an interface such as a USB.

[0128] The display device 710 is a device for displaying various types of information. Specific examples of the display device 710 include a liquid crystal display, an organic EL (Electro-Luminescence) display, and a display of a wearable device. The display device 710 may be provided outside the computer 70. In this case, the display device 710 is connected to the computer 70 via, for example, a display cable. Furthermore, when a touch panel is used as the input I / F 704, the display device 710 can be configured as an integrated part of the input I / F 704.

[0129] Furthermore, the components of the devices included in the information processing device 2 described in the above embodiment are assumed to realize predetermined processing in cooperation with other hardware by the processor 700 executing a program stored in the storage device 702. In other words, these components are assumed to be software or firmware, as well as corresponding hardware, and in both of these concepts, they are also referred to as "functions," "means," "parts," "processing circuits," "units," or "modules," and can be interpreted as such.

[0130] In the above embodiment, the developing device 3 is an AWP (Asahi Kasei Corporation) ＴＭ As the resin plate, a flexographic printing plate (AWP) manufactured by Asahi Kasei Corporation can be used. ＴМ -DEW (1.14 mm, BL size)) can be used. As the vibration sensor, a Keyence vibration sensor (sensor head GH-313A, amplifier unit GA-245) can be used.

[0131] 5. Modifications In the above embodiment, the model is described as being constructed by inputting learning data, but this is not limited to this. The model may be determined based on, for example, the shape of the printing surface, the number of rotations of the developing brush 300, the oscillation speed of the developing brush 300, the concentration of the developer 306, etc. The model may include thresholds for each frequency component (or components in a predetermined frequency band) of the frequency domain vibration data. In other words, the information processing device 2 may infer that there is an abnormality in the printing plate when a component in a predetermined frequency band in the frequency domain vibration data, which is the estimation data, exceeds a threshold.

[0132] In the above embodiment, vibration data from the development process of multiple resin plates that satisfy a first condition and vibration data from the development process of multiple resin plates that satisfy a second condition may be input as learning data to the model. In this case, a normal label may be associated with the vibration data from the development process of multiple resin plates that satisfy the first condition, and an abnormal label may be associated with the vibration data from the development process of multiple resin plates that satisfy the second condition. For example, a resin plate satisfying the first condition may mean that the resin plate conforms to a first standard and has a first size (width, thickness, length), etc. In contrast, for a resin plate to satisfy the second condition may mean that the resin plate conforms to a second standard that is different from the first standard and has a second size that is different from the first size, etc.

[0133] 6. Example: AWP resin plate manufactured by Asahi Kasei Corporation (registered trademark) ＴＭThe anomaly detection technology of the present invention was applied to the development process, in which a 1.14 mm BL-size DEW was developed using a CrystalCleanConnect (hereinafter, CCC) machine manufactured by the same company to obtain a printing plate. A Keyence (registered trademark) vibration sensor (sensor head GH-313A, amplifier unit GA-245) was used, and the generated vibration data was stored in the company's data logger NR-X100. Regardless of whether the development process was being performed, vibration data was generated continuously while the developing device was running and stored in the data logger. The vibration sensor sampling interval (frequency) was 1 ms. The data logger was connected to a computer and the vibration data was transferred to the computer. Vibration data during the development process of each resin plate was identified using the method described below. The time-domain vibration data was converted to frequency-domain vibration data, and an anomaly detection model was then generated. First, a correspondence table was created between each resin plate's identification number and the development time point. The identification number was determined by the value of a job counter, which increments by one each time a resin plate is inserted into the developing device. The development time includes the development start time, and the development start time here is defined as the time when the resin plate set on the transport bar begins to move toward the development zone in the developing device. To identify the vibration data for the development process, vibration data for 8 minutes from the development start time was defined as the vibration data for the development process of each resin plate and acquired as reference data for generating an anomaly detection model. Next, the reference data for the development process of each resin plate (hereinafter referred to as time-domain vibration data) was Fourier-transformed to obtain frequency-domain vibration data. Before applying the Fourier transform, the time-domain vibration data was divided into 6-second intervals, and the divided data was then Fourier-transformed and re-added to obtain the average value (average value at each frequency). This operation was performed on the time-domain vibration data of each resin plate to obtain frequency-domain vibration data. Next, an anomaly detection model was generated using One Class SVM. An anomaly detection model was prepared by training using frequency domain vibration data of resin plates for which the development process had been performed normally, and training using two groups of reference data: (i) only printing plates with an image rate of 0% (20 plates), and (ii) printing plates with multiple image rates ranging from 0% to 90% (120 plates).These models were fed with vibration data from resin plates that had undergone normal development and vibration data from resin plates with abnormalities, excluding the reference data, to evaluate their anomaly detection accuracy. The anomalies that occurred included brush anomalies (resin residue accumulation on the brush and subsequent adhesion of resin residue to the printing plate) and resin plate anomalies (use of expired resin plates, and abnormal development depth due to use of resin plates of a different grade than recommended). First, the prediction accuracy of model (i) was 65%, exceeding the 60% performance standard established as the application standard. Furthermore, the prediction accuracy of model (ii) was 79%. This indicates that generalization performance was improved by using vibration data from resin plates with various image rates as reference data for learning.

[0134] 7. Embodiments of the Present Disclosure The present disclosure includes, for example, the following embodiments.

[0135] [Appendix 1] An information processing device 2 includes an acquisition unit 100 that acquires vibration data related to vibrations of a developing device 3 that performs a development process including removing uncured portions of a target resin plate to obtain a printing plate, and an estimation unit 108 that estimates an abnormality in the development process based on the vibration data.

[0136] [Supplementary Note 2] The information processing device 2 according to Supplementary Note 1, further comprising a conversion unit 104 that converts time-domain vibration data into frequency-domain vibration data, and the estimation unit 108 estimates an abnormality in the development process based on the frequency-domain vibration data converted by the conversion unit 104.

[0137] [Supplementary Note 3] The information processing device 2 according to Supplementary Note 2, wherein the conversion unit 104 divides the vibration data in the time domain in the development process of the target resin plate into a predetermined number of divisions, and converts each of the divisions into data in the frequency domain.

[0138] [Appendix 4] The information processing device 2 described in any one of Appendices 1 to 3, further comprising an identification unit 102 that identifies time-domain vibration data of the development process of the target resin plate from time-domain vibration data related to the vibration of the developing device 3, and the acquisition unit 100 acquires the vibration data identified by the identification unit 102.

[0139] [Supplementary Note 5] The information processing device 2 according to any one of Supplementary Notes 1 to 4, wherein the vibration data is generated based on measurements of a vibration sensor that operates in response to the start of a development process for the target resin plate.

[0140] [Supplementary Note 6] The information processing device 2 according to any one of Supplementary Notes 1 to 5, wherein the developing step is a step of removing uncured portions using at least one of a brush and a developer, and the vibration data includes vibration data in the developing step.

[0141] [Supplementary Note 7] The information processing device 2 according to Supplementary Note 6, wherein the developer contains a water solvent.

[0142] [Supplementary Note 8] The information processing device 2 according to any one of Supplementary Notes 1 to 7, wherein the estimation unit 108 estimates one or more of the presence or absence, degree, and type of an abnormality in the development process.

[0143] [Appendix 9] An information processing device 2 described in any one of Appendices 1 to 8, wherein the estimation unit 108 estimating an abnormality in the development process includes estimating the type of abnormality in at least one of the target resin plate, the printing plate, and the development equipment used in the development process of the target resin plate, including the development device 3.

[0144] [Appendix 10] The types of abnormalities estimated by the estimation unit 108 include at least one of the following: (1) an abnormality in the brush (developing brush 300) used in the development process of the target resin plate; (2) an abnormality in the developer 306 used in the development process of the target resin plate; (3) an abnormality in the target resin plate; (4) an abnormality in the processing settings of the development process of the target resin plate; (5) an abnormality in the developing device 3 used in the development process of the target resin plate; and (6) an abnormality in the printing plate obtained by the development process of the target resin plate, as described in any one of Appendices 1 to 9.

[0145] [Supplementary Note 11] The information processing device 2 according to any one of Supplementary Notes 1 to 10, further comprising: an output unit 110 that outputs a notification regarding the abnormality based on the estimation unit 108 estimating that there is an abnormality.

[0146] [Supplementary Note 12] The information processing device 2 according to any one of Supplementary Notes 1 to 11, wherein the estimation unit 108 estimates an abnormality further based on development process data that can identify the processing content of the development process of the target resin plate.

[0147] [Appendix 13] The information processing device 2 described in any one of Appendices 1 to 12, wherein the estimation unit 108 estimates an abnormality based on vibration data and a model constructed by learning reference data related to vibration data in the development process of another resin plate different from the target resin plate.

[0148] [Supplementary Note 14] The information processing device 2 according to Supplementary Note 13, wherein the reference data is determined based on vibration data obtained by dividing time domain vibration data in the development process of another printing plate into a predetermined number of divisions and converting each division into frequency domain vibration data.

[0149] [Supplementary Note 15] The information processing device 2 according to Supplementary Note 14, wherein the predetermined number of divisions is determined based on an estimation accuracy of an abnormality when each of a plurality of division numbers is set as a division number of the vibration data in the time domain.

[0150] [Appendix 16] An information processing device 2 described in any one of Appendices 13 to 15, wherein the model is constructed by learning first reference data related to vibration data in a development process of a resin plate having a printing surface of a first shape, and second reference data related to vibration data in a development process of a resin plate having a printing surface of a second shape different from the first shape.

[0151] [Appendix 17] The information processing device 2 described in Appendix 16, wherein the first shape and the second shape have an image ratio, which indicates the ratio of the area of ​​one of the convex portions or concave portions on the printing surface to the sum of the area of ​​the convex portions and the area of ​​the concave portions on the printing surface, that differs by 10% or more.

[0152] [Appendix 18] An information processing method that causes a computer 70 to acquire vibration data related to vibrations of a developing device 3 that performs a development process including removing uncured portions of a target resin plate to obtain a printing plate, and estimates an abnormality in the development process based on the vibration data.

[0153] [Appendix 19] A developing device 3 including a brush (developing brush 300), a developing solution tank 304 containing developing solution 306, a developing process processing unit that performs a developing process including removing uncured portions of a target resin plate with the brush and the developing solution 306 to obtain a printing plate, and a computer 70, wherein the computer 70 acquires vibration data relating to the vibration of the developing device 3 of the target resin plate, and estimates an abnormality in the developing process based on the vibration data.

[0154] 1...system, 2...information processing device, 3...developing device, 4...communication network, 10...control unit, 12...storage unit, 70...computer, 100...acquisition unit, 102...identification unit, 104...conversion unit, 106...generation unit, 108...estimation unit, 110...output unit, 300...developing brush, 304...developing solution tank, 306...developing solution

Claims

1. An information processing device comprising: an acquisition unit that acquires vibration data relating to the vibration of a developing device that performs a development process including removing uncured portions of a target resin plate to obtain a printing plate; and an estimation unit that estimates an abnormality in the development process based on the vibration data.

2. The information processing device according to claim 1, further comprising a conversion unit that converts time domain vibration data into frequency domain vibration data, and the estimation unit estimates an abnormality in the development process based on the frequency domain vibration data converted by the conversion unit.

3. The information processing device according to claim 2, wherein the conversion unit divides the vibration data in the time domain in the developing process of the target resin plate into a predetermined number of divisions and converts each division into the frequency domain.

4. The information processing device according to claim 1, further comprising an identification unit that identifies time domain vibration data of the development process of the target resin plate from time domain vibration data related to the vibration of the developing device, and the acquisition unit acquires the vibration data identified by the identification unit.

5. The information processing device according to claim 1, wherein the vibration data is generated based on measurements of a vibration sensor that operates in response to the start of a development process for the target resin plate.

6. The information processing device according to claim 1, wherein the developing step is a step of removing the unhardened portion using at least one of a brush and a developing solution, and the vibration data includes vibration data in the developing step.

7. The information processing device according to claim 6, wherein the developer contains a water solvent.

8. The information processing device according to claim 1, wherein the estimation unit estimates one or more of the presence or absence, degree, and type of abnormality in the development process.

9. An information processing device as described in claim 1, wherein the estimation unit's estimation of an abnormality in the development process includes estimating the type of abnormality in at least one of the target resin plate, the printing plate, and the development equipment used in the development process of the target resin plate, including the development device.

10. The information processing device of claim 1, wherein the type of abnormality estimated by the estimation unit includes at least one of the following: (1) an abnormality in a brush used in the developing process of the target resin plate; (2) an abnormality in a developer used in the developing process of the target resin plate; (3) an abnormality in the target resin plate; (4) an abnormality in the processing settings of the developing process of the target resin plate; (5) an abnormality in a developing device used in the developing process of the target resin plate; and (6) an abnormality in a printing plate obtained by the developing process of the target resin plate.

11. The information processing device according to claim 1, further comprising an output unit that outputs a notification regarding the abnormality based on the estimation unit estimating that the abnormality exists.

12. The information processing device according to claim 1, wherein the estimation unit estimates the abnormality further based on development process data that can identify the processing details of the development process of the target resin plate.

13. The information processing device described in claim 1, wherein the estimation unit estimates the abnormality based on the vibration data and a model constructed by learning reference data related to vibration data in the development process of a resin plate other than the target resin plate.

14. An information processing device according to claim 13, wherein the reference data is determined based on vibration data obtained by dividing time domain vibration data in the development process of the other printing plate into a predetermined number of divisions and converting each division into frequency domain vibration data.

15. The information processing device according to claim 14, wherein the predetermined number of divisions is determined based on the accuracy of estimating the abnormality when each of a plurality of division numbers is set as the number of divisions of the vibration data in the time domain.

16. An information processing device as described in claim 13, wherein the model is constructed by learning first reference data relating to vibration data in a development process of a resin plate having a printing surface of a first shape, and second reference data relating to vibration data in a development process of a resin plate having a printing surface of a second shape different from the first shape.

17. An information processing device according to claim 16, wherein the first shape and the second shape have an image ratio, which indicates the ratio of the area of ​​one of the convex portions or concave portions on the printing surface to the sum of the area of ​​the convex portions and the area of ​​the concave portions on the printing surface, that differs by 10% or more.

18. An information processing method that causes a computer to perform the following: acquiring vibration data related to the vibration of a developing device that performs a developing process that includes removing uncured portions of a target resin plate to obtain a printing plate; and estimating an abnormality in the developing process based on the vibration data.

19. A developing device comprising: a brush; a developer tank containing a developer; a developing process processing section that performs a developing process including removing unhardened portions of a target resin plate with the brush and the developer to obtain a printing plate; and a computer, wherein the computer performs the following: acquiring vibration data related to vibrations of the developing device for the target resin plate; and estimating an abnormality in the developing process based on the vibration data.

Images