Filtration system

TH2501002687APending Publication Date: 2026-10-12MITSUBISHI KAKOKI KAISHA LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
TH2501002687
Authority / Receiving Office
TH · TH
Patent Type
Applications
Current Assignee / Owner
Priority Date
2022-10-28
Filing Date
2023-09-19
Publication Date
2026-10-12

AI Technical Summary

Technical Problem

Current filtration systems rely on manual visual inspection and time settings to manage filter media conditions, leading to decreased filtration performance and product quality due to adhesion and potential damage, which can result in operational inefficiencies and quality issues.

Method used

An automated filtration system utilizing artificial intelligence (AI) and machine-learned models to determine the processing state, deterioration, and abnormal states of filter media from image data, enabling automatic control of filtration processes, including adjustments to pressure and supply of liquids, and triggering necessary actions such as filter medium replacement.

Benefits of technology

The system reduces operational load, improves product quality by preventing defects, and enhances productivity through precise control, reducing energy consumption and resource wastage by optimizing filtration and cleaning processes.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

Invention details;
Need to check novelty before this filing date? Find Prior Art

Description

Filtration System

[0001] The present invention relates to a filtration system capable of automatically controlling a treatment system for filtering a slurry liquid.

[0002] To ensure the continuous operation of filtration equipment, it is important to accurately understand the condition of filter media such as filtration membranes, filters, filter paper, and filter cloth.

[0003] The filtering performance of filter media deteriorates due to the adhesion of target substances to the surface and interior of the filter media during repeated filtration. Furthermore, if the filter media deteriorates and breaks, it can lead to a decline in the quality of products such as filtrate and cake. Traditionally, the condition of filter media has been managed by visual inspection by workers or by setting a time. Therefore, technologies for managing the condition of filter media have been developed.

[0004] Patent Document 1 discloses an oil condition monitoring method and an oil condition monitoring device that can accurately monitor the condition of oil used in various machines or equipment in order to appropriately predict the lifespan of oil. Patent Document 1 discloses that the deterioration state of the oil is monitored by filtering the oil, projecting light onto a filter that removes the oil, and detecting the color components of the transmitted light.

[0005] Patent Literature 2 discloses an important factor calculation device that accurately predicts important factors that cause fouling of a filtration membrane. Patent Literature 2 discloses a device that learns a predictor that predicts the filtration performance of a filtration membrane after a unit time step from time series data of multiple parameters, including parameters that indicate the filtration performance of the filtration membrane and parameters that indicate the water quality of a wastewater treatment system in which the filtration membrane is used, and outputs parameters that contribute to predicting the filtration performance after a unit time step from the prediction period.

[0006] Patent Literature 3 discloses a technology for achieving suitable operational control of a centrifuge system without relying on the judgment of an operator. Patent Literature 3 discloses a technology including: a training dataset storage unit 22 that stores a plurality of training datasets, each of which includes input data including image data of the separated liquid captured from a predetermined angle of view, and output data including control parameters associated with the input data, the control parameters including at least one of the amount of additive added to the treated liquid PL1, the centrifugal force of the bowl 2, and the differential speed controlled by the differential speed generator 5; a learning unit 23 that inputs a plurality of training datasets to learn a learning model that infers a correlation between the input data and the output data; and a trained model storage unit 24 that stores the trained trained model.

[0007] According to Patent Document 3, it is understood that automating the control of a centrifugal separator is effective in reducing operational load and improving productivity.

[0008] Japanese Patent No. 5190660 International Publication No. 2022 / 085802 Japanese Patent Application Laid-Open No. 2022-021243

[0009] An object of the present invention is to provide a filtration system that can automatically control a processing system for filtering a slurry liquid using artificial intelligence (AI).

[0010] In order to solve the above-mentioned problems, the present invention provides a filtration system capable of automatically controlling a processing system for filtering a slurry liquid, wherein the processing system includes a filtration system, and is arranged on a fluid path within the filtration system and comprises: an imaging unit that acquires image data of the fluid; a judgment unit that judges the state of the processing system based on the image data; and an output unit that outputs a control signal to the processing system based on the judgment result of the state of the processing system, wherein the state of the processing system includes the processing state of the filtration system, the deterioration state of the filter media, or the abnormal state of the filter media, and the judgment unit judges the processing state of the filtration system, the deterioration state of the filter media, or the abnormal state of the filter media using a machine-learned judgment model that uses the image data as input data and the processing state of the filtration system, the deterioration state of the filter media, or the abnormal state of the filter media as output data.

[0011] In this way, by using a machine learning model in a filtration device, the processing state of the filtration system, the deterioration state of the filter media, and the abnormal state of the filter media can be determined from image data of the fluid, and the processing system can be automatically controlled according to these states.By incorporating artificial intelligence (AI), the filtration device can automatically determine the criteria for the processing system state that operators previously determined based on experience, making it possible to achieve automatic control of the processing system appropriately according to the state without the need for operator operation.

[0012] The present invention also provides a filtration system capable of automatically controlling a processing system for filtering a slurry liquid, wherein the processing system includes a cleaning system, and is arranged on a fluid path within the cleaning system and comprises an imaging unit that acquires image data of the fluid, a judgment unit that judges the state of the processing system based on the image data, and an output unit that outputs a control signal to the processing system based on the judgment result of the state of the processing system, wherein the state of the processing system includes the processing state of the cleaning system, the deterioration state of the filter media, or the abnormal state of the filter media, and the judgment unit judges the processing state of the cleaning system, the deterioration state of the filter media, or the abnormal state of the filter media using a machine-learned judgment model that uses the image data as input data and the processing state of the cleaning system, the deterioration state of the filter media, or the abnormal state of the filter media as output data.

[0013] In this way, by using a machine learning model in the filtration device, the processing state of the cleaning system, the deterioration state of the filter media, and the abnormal state of the filter media can be determined from image data of the fluid, and the processing system can be automatically controlled according to these states.By incorporating artificial intelligence (AI), the filtration device can automatically determine the judgment criteria for the processing system state that operators previously judged based on experience, making it possible to achieve automatic control of the processing system appropriately according to the state without the need for operator operation.

[0014] In a preferred embodiment of the present invention, the treatment system includes a filtration system, the fluid is filtrate, the determination unit determines the treatment state of the filtration system, and the output unit outputs a control signal to the filtration system based on the determination result of the treatment state of the filtration system. In a preferred embodiment of the present invention, the control signal to the filtration system includes at least one of a control signal related to the operation of the filtration system, a control signal related to the supply amount of slurry liquid, a control signal related to the deliquification process, and a control signal related to the supply amount of filter aid liquid. In a preferred embodiment of the present invention, the control signal to the filtration system is a control signal for increasing or decreasing the pressure inside the filter chamber on at least one of the supply path side and the discharge path side. In a preferred embodiment of the present invention, the control signal to the filtration system is a control signal for changing the opening and closing time of a supply valve for the slurry liquid or the filter aid liquid. With this configuration, the treatment state of the filtration system can be determined and appropriate automatic control according to each state can be achieved.

[0015] In a preferred embodiment of the present invention, the processing system includes a cleaning system, the fluid is a cleaning wastewater, the determining unit determines a processing state of the cleaning system, and the output unit outputs a control signal to the cleaning system based on the determination result of the processing state of the cleaning system. With this configuration, it is possible to determine the processing state of the cleaning system and realize appropriate automatic control according to the state.

[0016] In a preferred embodiment of the present invention, the treatment system includes a filtration system, a cleaning system, a replacement system, or a warning system, the fluid is filtrate, the determination unit determines the deterioration state of the filter media, and the output unit outputs at least one of a control signal to the cleaning system, a control signal to the replacement system, and a warning signal to the warning system based on the determination result of the deterioration state of the filter media. In a preferred embodiment of the present invention, the treatment system includes a cleaning system, a replacement system, or a warning system, the fluid is cleaning wastewater, the determination unit determines the deterioration state of the filter media, and the output unit outputs at least one of a control signal to the cleaning system, a control signal to the replacement system, and a warning signal to the warning system based on the determination result of the deterioration state of the filter media. With this configuration, the deterioration state of the filter media can be determined and appropriate automatic control can be achieved according to the state.

[0017] In a preferred embodiment of the present invention, the processing system includes a filtration system, a cleaning system, or a warning system, the fluid is filtrate or cleaning wastewater, the determination unit determines whether the filter medium is in an abnormal state, and the output unit outputs a control signal regarding operation to the filtration system or a warning signal regarding the abnormal state of the filter medium to the warning system based on the determination result of the abnormal state of the filter medium. With this configuration, it is possible to determine whether the filter medium is in an abnormal state and quickly respond to the abnormal state.

[0018] In a preferred embodiment of the present invention, the processing system includes a filtration system, the fluid is the slurry liquid, the imaging unit is disposed on a supply path of the slurry liquid to acquire image data of the slurry liquid, and the output unit outputs a control signal regarding the supply amount of filter aid liquid to the filtration system based on the determination result of the processing state of the filtration system. With this configuration, it is possible to control the supply amount of filter aid liquid appropriately according to the state of the slurry liquid.

[0019] In a preferred embodiment of the present invention, the imaging unit is disposed on the path of the fluid at a field angle that is approximately horizontal with respect to the ground plane. With this configuration, the fluid level, the fluid volume, and the like can be imaged at a suitable field angle, thereby improving the accuracy of the determination.

[0020] In a preferred embodiment of the present invention, the apparatus further includes a control mode switching unit for switching the treatment system to be controlled, the treatment system including a filtration system, a cleaning system, and a replacement system, the determination unit determining the treatment system state including the treatment state of the filtration system, the treatment state of the cleaning system, and the deterioration state of the filter media, the output unit outputting control signals to the filtration system, the cleaning system, and the replacement system based on the determination result of the treatment system state, and the control mode switching unit switching the control mode based on the determination result of the treatment system state or the output of the control signal. In a preferred embodiment of the present invention, the treatment system includes a warning notification system, the determination unit further determining the treatment system state including an abnormal state of the filter media, and the output unit outputting a warning signal corresponding to the control mode to the warning notification system based on the determination result of the deterioration state of the filter media or the abnormal state of the filter media. With this configuration, a series of operations in the filtration device including the filtration system, the cleaning system, and the replacement system can be automatically controlled.

[0021] In a preferred embodiment of the present invention, the filtrate discharge path of the filtration system also serves as a discharge path for cleaning wastewater from the cleaning system, and the image capturing unit is disposed in the filtrate discharge path. With this configuration, the operation of the filtration device including the filtration system and the cleaning system can be automatically controlled by disposing a single image capturing unit.

[0022] According to the present invention, a filtration system can be provided in which a processing system for filtering a slurry liquid can be automatically controlled by artificial intelligence (AI).

[0023] A block diagram of a filtration system according to this embodiment is shown. A schematic diagram of a filtration system according to this embodiment is shown. A schematic diagram of a filtration system according to this embodiment is shown. A schematic diagram of an imaging unit according to this embodiment is shown. A schematic diagram of an imaging unit according to this embodiment is shown. An example of image data acquired by the imaging unit according to this embodiment is shown. A block diagram of a model generation device according to this embodiment is shown. An example configuration of a determination model according to this embodiment is shown. An example configuration of a dataset according to this embodiment is shown. A flowchart of machine learning processing according to this embodiment is shown. An example configuration of a control signal table according to this embodiment is shown. Each process in a filtration system according to this embodiment is shown. A block diagram of a filtration system according to embodiment 2 is shown. Each process in a conventional filtration device is shown.

[0024] The filtration system of the present invention will be described below with reference to the drawings. Note that the embodiment shown below is an example of the present invention, and the present invention is not limited to the embodiment below, and various configurations can be adopted.

[0025] A filtration device separates a slurry liquid, which is a solid-liquid mixture to be filtered, into a cake, which is a solid product, and filtrate by passing the slurry liquid through a filter material such as a filtration membrane, filter, filter paper, or filter cloth.

[0026] FIG. 14 shows the steps of each treatment using a conventional filtration device.

[0027] The filtration process is a process in which the slurry liquid is filtered to separate it into a cake and filtrate. In the filtration process, an operator performs operations on the filtration device, such as starting filtration, setting and adjusting the supply amount of the slurry liquid, setting and adjusting the filter chamber pressure, and ending filtration. The filtration process may also include a deliquifying process in which the cake is deliquified. In the deliquifying process, an operator performs operations on the filtration device, such as starting deliquifying, setting and adjusting the filter chamber pressure, and ending deliquifying.

[0028] The cake washing process is a process in which a washing liquid is supplied to the filter chamber after filtration to wash the cake. In the cake washing process, an operator performs operations on the filtration device, such as starting cake washing, setting and adjusting the amount of washing liquid supplied, setting and adjusting the filter chamber pressure, and ending washing.

[0029] The filter media replacement process is a process for cleaning or replacing the filter media when it has deteriorated (for example, fouling) or when an abnormality has occurred (for example, breakage of the filter media, filtration leakage, etc.). In the filter media replacement process, workers perform operations such as starting the filter media cleaning or filter media replacement and completing the filter media replacement. If the filter media replacement process is not performed using an automatic filtration device, opening the filter chamber frame, installing the filter media, and closing the filter chamber frame are performed manually. Note that in the filter media replacement process, cake accumulated on the filter media is discharged.

[0030] When the filter material replacement is completed in the filter material replacement step, the process returns to the filtration step and repeats the process.

[0031] In the above-mentioned treatment process, workers visually inspected the state of the filtrate and the state of the washing wastewater and determined whether to terminate filtration or washing. Alternatively, workers controlled each process by setting a timer for the filtration time or washing time. Furthermore, the amount and timing of adding filter aid liquid were also controlled by workers visually inspecting the process or by setting a timer.

[0032] The filtration system of the present invention aims to automatically control the above-mentioned filtration process, cake washing process, and filter media replacement process.

[0033] <Embodiment 1> Fig. 1 shows a block diagram of a filtration system 1 according to this embodiment. The filtration system 1 includes a control unit 11, an imaging unit 12, a determination unit 13, an output unit 14, a first control mode switching unit 15, a storage unit (DB) serving as a database, and a control panel 60. Each component is connected to and controlled by the control unit 11. The output unit 14 is connected to a processing system 10 and controls the operation of the processing system 10 by outputting a control signal. The processing system 10 includes a filtration system 20 that controls the filtration process, a cleaning system 30 that controls the cleaning process, a filter media replacement system 40 that controls the filter media replacement process, and an alarm notification system 50 that controls notification of a deterioration state of the filter media (e.g., fouling) or an abnormal state of the filter media (e.g., filter media breakage, filtration leakage, etc.). In this embodiment, the filtration system 1 is configured as a filtration device.

[0034] The deterioration state of the filter medium includes at least "no deterioration of filter medium" and "deteriorated deterioration of filter medium." The deterioration state of the filter medium may further include a graded level of deterioration (such as slight deterioration and severe deterioration).

[0035] The abnormal state of the filter medium includes at least "no abnormality" and "abnormality present." The abnormal state of the filter medium may include types of abnormality such as breakage of the filter medium and filtration leakage.

[0036] The processing system 10 represents a device configuration that executes processes including filtration, cake washing, device cleaning, filter media replacement, and warning notification in the filtration system 1 (1A to 1D). The processing system 10 operates components based on input of control signals to realize various processes.

[0037] The control unit 11 is configured by a calculation device such as a CPU (Central Processing Unit). The control unit 11 may be connected to the processing system 10 and control the mechanism based on a control signal. The control unit 11 is connected to a control panel 60 and can accept operation input via the control panel 60. The control unit 11 may perform display processing on the control panel 60.

[0038] The imaging unit 12 is disposed on the path of the fluid in the filtration device and acquires image data of the fluid.

[0039] The determination unit 13 determines the state of the processing system 10 based on the image data acquired by the imaging unit 12 .

[0040] The output unit 14 outputs a control signal to the processing system 10 based on the determination result of the state of the processing system 10 by the determination unit 13. The processing system 10 is controlled based on the control signal output by the output unit 14.

[0041] The first control mode switching unit 15 switches the control mode based on the determination result of the state of the processing system 10 by the determination unit 13 or the output of a control signal by the output unit 14. The control mode indicates whether the processing system 10 to be controlled is the filtration system 20, the cleaning system 30, or the exchange system 40.

[0042] The storage unit DB stores a determination model used in the determination process of image data by the determination unit 13. The storage unit DB stores a control signal table used to determine the control signal to be output by the output unit 14. The storage unit DB also stores various data such as image data and a program including various commands to be executed by the control unit 11. This program may be stored and installed on a non-transitory computer-readable recording medium such as a CD-ROM, flash memory, or SSD memory.

[0043] 2(a), (b), and (c) are diagrams showing the configuration of a device for controlling the filtration process of the slurry filtration systems 1A to 1C.

[0044] 2(a), the filtration system 20 is configured to control the filtration process of a slurry filtration system 1A that filters a slurry liquid F1 in a filter chamber R. The filtration system 20 includes a first supply tank UT1 that supplies the slurry liquid F1, a supply path 101 that supplies the slurry liquid F1 from the first supply tank UT1 to the filter chamber R, a discharge path 102 that discharges the filtrate F2 filtered in the filter chamber R, a first discharge tank DT1 that stores the filtrate F2, a first imaging unit 12A that is installed in the supply path 101 between the first supply tank UT1 and the filter chamber R and that acquires image data of the slurry liquid F1, and a second imaging unit 12B that is installed in the discharge path 102 between the filter chamber R and the first discharge tank DT1 and that acquires image data of the filtrate F2. The slurry filtration system 1A also includes a filter chamber pressure control unit (not shown).

[0045] The first supply tank UT1 is provided with a supply valve (not shown), and the supply amount of the slurry liquid F1 is controlled by the opening and closing time of the supply valve. The filter chamber R is provided with a filter medium (not shown) between the supply path 101 side and the discharge path 102 side, and the slurry liquid F1 is filtered by the differential pressure between the supply path 101 side and the discharge path 102 side.

[0046] The filter chamber pressure control unit (not shown) is composed of a pressure pump (not shown) or a pressure valve (not shown) that controls the internal pressure of at least one of the filter chambers R on the supply path 101 side and the discharge path 102 side. Furthermore, when a mechanism for centrifuging by rotating the filter chamber R is employed, the filter chamber pressure control unit (not shown) is composed of an actuator (not shown) that controls the number of rotations or rotation speed. Furthermore, when a mechanism for separating the slurry liquid F1 by squeezing with a press or the like is employed, the filter chamber pressure control unit (not shown) is composed of an actuator (not shown) that controls the pressing pressure and pressing time.

[0047] As shown in FIG. 2B, the filtration system 20 may supply filter aid liquid F3 to the slurry filtration system 1B when filtering the slurry liquid F1 in the filter chamber R. The filter aid liquid F3 is supplied through a filter aid liquid supply path 103 that connects a second supply tank UT2 that supplies the filter aid liquid F3 to a supply path 101. The second supply tank UT2 is equipped with a supply valve (not shown), and the amount of filter aid liquid F3 supplied is controlled by its opening and closing time. The filter aid liquid supply path 103 for the filter aid liquid F3 from the second supply tank UT2 to the filter chamber R may be provided separately so as to supply the filter aid liquid F3 to the filter chamber R without being connected to the supply path 101 for the slurry liquid F1 from the first supply tank UT1 to the filter chamber R.

[0048] The first imaging unit 12A acquires image data of the slurry liquid F1 passing through the supply path 101. The second imaging unit 12B acquires image data of the filtrate F2 or the filtrate F4 containing the filter aid liquid F3 passing through the discharge path 102.

[0049] Furthermore, as shown in Figure 2 (c), in the filtration system 20, in the slurry filtration system 1C, the first imaging unit 12A may be arranged in the supply path 101a of the mixed liquid F5 of the slurry liquid F1 and the filter aid liquid F3, and may be configured to acquire image data of the mixed liquid F5.

[0050] FIG. 3 is a diagram showing the configuration of a device for controlling the cleaning process of the slurry filtration system 1D.

[0051] As shown in FIG. 3, the cleaning system 30 in the slurry filtration system 1D includes a third supply tank UT3 that supplies cleaning liquid F6, a filtration chamber R, a third imaging unit 12C installed in the discharge path 102, and a second discharge tank DT2 that stores used cleaning waste liquid F7.

[0052] The cleaning system 30 further includes a filter chamber pressure control unit (not shown). The third supply tank UT3 includes a supply valve (not shown), and the amount of cleaning liquid supplied is controlled by the opening and closing time of the valve. The filter chamber R, the filter chamber pressure control unit (not shown), part of the supply path 101, and part of the discharge path 102 may have the same configuration as the filtration system 20.

[0053] In this embodiment, the cleaning process includes a cake cleaning process for cleaning the cake and an apparatus cleaning process for cleaning the inside of the filtration apparatus. The cleaning liquid F6 cleans the cake (not shown) or the filtration chamber R, and is discharged as cleaning waste liquid F7 through the discharge path 102 into the second discharge tank DT2.

[0054] The replacement system 40 shows a device configuration for controlling the filter media replacement process of the filtration system 1. The replacement system 40 includes replacement filter media, a replacement filter media rack for storing the replacement filter media, a filter media supply mechanism for supplying the replacement filter media from the replacement filter media rack to the filter chamber, a filter media discharge mechanism for discharging used filter media from the filter chamber, and a filter media cleaning mechanism for cleaning the filter media.

[0055] The warning notification system 50 is configured to control warning notifications regarding the deterioration state of the filter media and abnormal states of the filter media in the filtration system 1. The warning notification system 50 is connected to the control panel 60 and displays the warning notifications on a display. The warning notification system 50 may also be configured to include an alarm or lamp and output the warning notification as a voice, an alarm sound, or a lit or flashing lamp.

[0056] In this embodiment, the filtration system 1 may be configured to automate at least one system selected from the filtration system 20, the cleaning system 30, the replacement system 40, and the warning notification system 50. The filtration system 1 may be configured to include, for example, a processing system including the filtration system 20 and the cleaning system 30, but not include the replacement system 40.

[0057] The control panel 60 is configured as a touch panel or the like that functions as a display unit and an input unit. The control panel 60 may be configured as a display that serves as a display unit that displays the status of the filtration system 1, and an operation panel that serves as an input unit that operates the filtration system 1. The control panel 60 may be configured as a computer terminal that is externally connected to the filtration system 1. The input unit of the control panel 60 accepts input of operation information and transmits the operation information to the control unit 11, and the control unit 11 controls each component based on the received operation information.

[0058] Here, the second imaging unit 12B and the third imaging unit 12C are disposed in the filtrate discharge path and the washing wastewater discharge path, respectively. Note that if the filtrate discharge path of the filtration system 20 also serves as the washing liquid discharge path of the washing system 30, at least one of the second imaging unit 12B and the third imaging unit 12C may be configured to be disposed in the filtrate discharge path. In other words, the second imaging unit 12B (third imaging unit 12C) monitors the two fluids, the filtrate and the washing wastewater, and acquires image data thereof.

[0059] 4A and 4B show schematic diagrams of the imaging unit 12. Fig. 4A shows a front view of the imaging unit 12. Fig. 4B shows a top view of the imaging unit 12.

[0060] As shown in Fig. 4(a), the imaging unit 12 includes an imaging device 121 such as a camera, a sight glass 122, and a window 123 provided in the sight glass 122. The imaging device 121 is disposed facing the direction of the window 123 (dotted arrow) and acquires image data of the fluid F flowing inside the path. The imaging device 121 is connected to the control unit 11. Fig. 4(a) shows an example of horizontal piping in which the fluid F passes from left to right through piping provided on the left and right sides of the sight glass 122, but the position and direction of the piping provided in the sight glass 122 are not limited thereto.

[0061] The imaging unit 12 is disposed, for example, at an angle of view that is substantially horizontal with respect to the horizontal. "Substantially horizontal" refers to an angle of view ranging from -20 degrees to +20 degrees with respect to the horizontal, preferably from -10 degrees to +10 degrees, and more preferably from -5 degrees to +5 degrees. In this embodiment, the fluid F is a liquid, and it is preferable to dispose the imaging unit 12 so that it can acquire image data including characteristic quantities such as the liquid level and flow rate.

[0062] The imaging unit 12 preferably further includes a reference light source L. The light source L is provided for the purpose of maintaining a constant amount of light in the surrounding environment of the imaging unit 12. Because the amount of light in the surrounding environment affects the appearance of the color of the fluid imaged by the imaging device 121, maintaining a constant amount of light during imaging can improve the accuracy of color-based judgment. The imaging unit 12 may be provided with a cover or the like to block external light from reaching the imaging device 121 and the sight glass 122. As an example of the arrangement of the light source L, as shown in FIG. 4( b), the sight glass 122 may have a window 123B on the back, and the reference light source L may be provided as a backlight to cover the window 123B. Note that the light source L may be provided to irradiate light at a predetermined angle relative to the front window 123A.

[0063] Fig. 5 shows a schematic diagram of the imaging unit 12 in the case of a vertical pipe. In Fig. 5, the fluid F passes from top to bottom through the pipes provided above and below the sight glass 122. In particular, the imaging units 12B and 12C, which are disposed in the discharge path from the filter chamber R (not shown), can acquire image data showing the tendency of an increase or decrease in the amount of fluid F by using vertical pipes. For example, as shown in Fig. 5(a), the amount of fluid F tends to increase relatively in the initial stage of discharge of the fluid F from the filter chamber R because the pressure inside the filter chamber is low.

[0064] On the other hand, as shown in Figure 5(b), the amount of fluid F tends to decrease relatively because the pressure inside the filter chamber increases as the discharge proceeds. The imaging units 12B and 12C are preferably provided within a range of 10 m from the filter chamber R, and more preferably within a range of 5 m from the filter chamber R. Furthermore, the imaging units 12B and 12C are preferably provided directly below the filter chamber R.

[0065] 6 shows examples of image data of the fluid F acquired by the imaging unit 12. Figures 6(a) to 6(c) show image data of a transparent fluid F. Figures 6(d) and 6(e) show image data of a colored fluid F. The determination unit 13 can extract feature amounts contained in the image data and determine the state of the processing system 10.

[0066] 6(a) to 6(c) show states with different amounts of transparent fluid F. The image data in Fig. 6(d) and 6(e) show states with different amounts of colored fluid F. The determination unit 13 can determine the state of the processing system 10 from the feature amounts of these different states of fluid F.

[0067] The image data (Images A to E) in Figures 6(a) to (e) were input into the judgment model, and the test results were "Image A: Transparent Sky," "Image B: Transparent Medium," "Image C: Transparent Supply," "Image D: Colored Sky," and "Image E: Color Supply." From these results, it can be seen that the judgment unit 13 using the judgment model can extract the amount and color feature values ​​of fluid F. Note that the judgment model can also judge the amount of fluid F in a more step-by-step manner, and can also judge the lightness, luminance, density, and color values ​​related to the color of fluid F in a more step-by-step manner.

[0068] 6 shows an example of image data of fluid F passing through a horizontal pipe, but image data of fluid F passing through a vertical pipe may also be used, as shown in Fig. 5. In this embodiment, the position and direction of the pipe are not limited as long as the image data can extract feature amounts including at least the amount and color of fluid F.

[0069] The determination unit 13 can determine the state of the processing system 10 of the filtration system 1 based on the feature quantities of the image data of the fluid F as described above.

[0070] In this embodiment, the state of the treatment system 10 includes the treatment state of the filtration system 20, the treatment state of the cleaning system 30, the deterioration state of the filter media, and the abnormal state of the filter media.

[0071] The processing state of the filtration system 20 includes a filtration processing state and a deliquification processing state. The filtration processing state includes before filtration, during filtration, completed filtration, shortage of slurry liquid, excess of slurry liquid, shortage of filter aid liquid, excess of filter aid liquid, etc. The filtration processing state may also include shortage or excess of slurry liquid or filter aid liquid. The deliquification processing state includes before deliquification, during deliquification, completed deliquification, etc.

[0072] The processing status of the cleaning system 30 includes a processing status of cake washing, a processing status of apparatus washing, a processing status of filter media washing, and a processing status of deliquor washing depending on the object to be cleaned, and the processing status is determined for each object to be cleaned. The processing status of the cleaning system 30 includes before cleaning, during cleaning, cleaning completed, lack of cleaning liquid, excess of cleaning liquid, etc. The processing status of the cleaning system 30 may also include a shortage or excess of cleaning liquid.

[0073] The deliquor processing state indicates the deliquor state of the filtrate in the cake in the processing state of the filtration system 20, and indicates the deliquor state of the washing liquid in the cake in the processing state of the washing system 30.

[0074] As a specific example, when the fluid F in the image data is image data of filtrate, the determination unit 13 can determine the processing state related to the progress of filtration from the fact that a stable flow of filtrate is observed when the filtration state is stable and a decrease in filtrate is observed due to the accumulation of filter cake. Furthermore, the determination unit 13 can determine the deterioration state of the filter media from the fact that the amount of filtrate decreases in accordance with deterioration (fouling, etc.) of the filter media. Furthermore, when the fluid F in the image data is image data of washing wastewater, the determination unit 13 can determine the processing state related to the progress of cleaning from the fact that the transparency of the color of the washing wastewater increases as cleaning progresses.

[0075] In this embodiment, the determination unit 13 determines the state of the processing system 10 using a determination model stored in the storage unit DB. The determination model represents a trained model that has been machine-learned using a dataset. The determination model is generated by an external model generation device 7 and stored in the storage unit DB.

[0076] 7 is a block diagram of a model generation device 7 that performs processing to generate a determination model. The model generation device 7 includes a dataset acquisition unit 71, a model generation unit 72, a dataset storage unit 73, and a model storage unit 74. The dataset acquisition unit 71 is preferably configured to acquire, as a dataset, image data acquired by the imaging unit 12 illustrated in FIG. 3.

[0077] A general-purpose computer can be used as the model generation device 7. The model generation device 7 includes, as hardware components, an arithmetic unit such as a CPU, a main storage unit such as a RAM, an auxiliary storage unit, a communication unit, an input / output unit, and the like.

[0078] The filtration system 1 and the model generating device 7 are configured to be capable of wired or wireless data communication. Alternatively, the filtration system 1 may be configured as a filtration system including the filtration system 1 and the model generating device 7. Alternatively, the filtration system 1 may be configured to include the functional components (71-74) of the model generating device 7.

[0079] In this embodiment, the machine learning algorithm may employ a neural network. As shown in FIG. 8 , the neural network N has an input layer N1, an intermediate layer N2, and an output layer N3. Each layer is composed of a plurality of neurons having an activation function. The input layer N1 receives input data from a dataset. The input layer N1 is composed of a plurality of neurons according to the input data and outputs a calculation result for the input data to the intermediate layer N2. The intermediate layer N2 is composed of one or more layers, each having a plurality of neurons. The intermediate layer N2 receives input of the calculation result from the input layer N1 and further outputs the calculation result for the input to an adjacent layer in the intermediate layer N2 or to the output layer N3. The output layer N3 outputs an estimated value according to the input from the intermediate layer N2. The accuracy of determining output data relative to input data can be improved by adjusting the coefficients of each neuron so as to reduce the error between the estimated value of the output layer N3 and the output data of the dataset.

[0080] The machine learning algorithm is not limited to a neural network, and a regression analysis model, a support vector machine, a k-nearest neighbor method, a decision tree model, etc. may also be adopted.

[0081] FIG. 9 shows an example of the configuration of a dataset used for machine learning of a determination model.

[0082] In this embodiment, the data set is configured with image data as input data and the state of the processing system 10 as output data, as shown in FIG. 9(a).

[0083] 9B, the data set may be configured with image data as input data and the state of the fluid as output data. The state of the fluid includes the amount, color, etc. of the fluid F.

[0084] 9(c), the data set may be configured with image data as input data and control signals for the processing system as output data. The control signals are signals output from the output unit 14 to the processing system 10, and correspond to the control of each of the filtration system 20, cleaning system 30, replacement system 40, and warning notification system 50.

[0085] In this embodiment, the data set may be configured with a combination of multiple image data along a time series as input data, and at least one selected from the state of the processing system 10, the state of the fluid, and a control signal for the processing system as output data. A determination model machine-learned using a combination of image data along a time series as input data can output output data corresponding to changes in the image data by accepting input of image data in the order of the combinations.

[0086] In this case, the output data may correspond to a time series (time difference) of the multiple image data and may be configured as at least one data selected from the state of the processing system 10 after a predetermined time has elapsed, the state of the fluid, and a control signal for the processing system. For example, if the data set uses image data every five seconds as input data and the state of the processing system 10 five seconds later as output data, the judgment model can output a predicted value of the state of the processing system 10 as the judgment result. By using the judgment model as described above, even if there is a time lag between the acquisition of image data and the control of the processing system 10, the judgment unit 13 can correct the time lag and control the processing system 10.

[0087] FIG. 10 is a flowchart showing the process of generating a decision model in the model generating device 7.

[0088] Step S11: The dataset acquisition unit 71 acquires a dataset and stores it in the dataset storage unit 73. The dataset acquisition unit 71 preferably sets image data acquired by the imaging unit 12 as input data for the dataset. The dataset acquisition unit 71 accepts input from an operator of output data corresponding to the image data set as input data. The dataset storage unit 73 may be configured to pre-store list data of the state of the processing system 10, the state of the fluid, and the control signal that can be selected as output data for the dataset, and the dataset acquisition unit 71 may be configured to accept selection of output data corresponding to the image data from the list data and determine it as a dataset.

[0089] Step S12: The model generation unit 72 executes machine learning processing of the model using the datasets stored in the dataset storage unit 73. The number of datasets used in the machine learning processing is not particularly limited.

[0090] Step S13: By completing the machine learning process in step S12, the model generation unit 72 generates a judgment model, stores it in the model storage unit 74, and ends the process.

[0091] The judgment model stored in the model storage unit 74 is stored in the storage unit DB of the filtration system 1, and the judgment unit 13 can use the judgment model to judge the state of the processing system 10 based on the image data.

[0092] A plurality of judgment models may be generated depending on the type of fluid, such as a slurry liquid, a filter aid liquid, or a cleaning liquid, and the type of filtration device used in the filtration system 1. The control panel 60 can determine a judgment model to be used by the judgment unit 13 from the plurality of judgment models stored in the memory unit DB by receiving instruction input related to the type of fluid or filtration device.

[0093] The output unit 14 outputs a control signal to the processing system 10 based on the determination result of the state of the processing system 10 by the determination unit 13. The output unit 14 outputs a control signal based on a control signal table that indicates the correspondence between the determination result of the state of the processing system 10 and the control signal to the processing system 10.

[0094] FIG. 11( a) shows an example of the data structure of a control signal table. The control signal table has a control mode, a judgment result, and a control signal. The control signal table shown in FIG. 11( a) shows the judgment result based on the judgment model machine-learned using the dataset of FIG. 9( a) and the corresponding control signal. The control signal table may also show the judgment result regarding the fluid state based on the judgment model machine-learned using the dataset of FIG. 9( b) and the corresponding control signal. Note that the control signal table can be omitted when using a judgment model machine-learned using the dataset of FIG. 9( c), and the output unit 14 outputs the control signal, which is the judgment result by the judgment unit 13, to the processing system 10.

[0095] The control mode indicates which process the filtration system 1 is performing: the filtration system 20, the cleaning system 30, or the replacement system 40. The control mode includes a filtration mode related to the control of the filtration system 20, a cake cleaning mode related to the control of cake cleaning in the cleaning system 30, a replacement mode related to the control of the replacement system 40, and an equipment cleaning mode related to the control of equipment cleaning in the cleaning system 30.

[0096] The filtration mode may be further classified into a filtration mode and a filtrate draining mode, the cake washing mode may be further classified into a cake washing mode and a wash liquid draining mode, and the replacement mode may be further classified into a filter medium replacement mode and a filter medium cleaning mode.

[0097] The control signal indicates a signal for controlling the processing system 10 corresponding to the control mode. A plurality of control signals may be set for one determination result. For example, the control signal related to the operation of the filtration system 20 corresponding to the determination result of filtration completion may include a control signal for controlling the pressure inside the filter chamber by the filter chamber pressure control unit and a control signal for controlling the opening and closing times of the supply valves for the slurry liquid and the filter aid liquid.

[0098] The filtration mode control signal corresponds to the determination result of the treatment state of the filtration system 20, and includes control signals related to operation / stop of the filtration system 20, increase / decrease in the amount of slurry liquid supplied, increase / decrease in the amount of filter aid liquid supplied, operation / stop of cake deliquor treatment, etc. Upon receiving the control signal output from the output unit 14, the filtration system 20 controls the filter chamber pressure control unit, the supply valves of the slurry liquid supply tank and the filter aid liquid supply tank, etc.

[0099] Specifically, the filter chamber pressure control unit increases or decreases the pressure inside the filter chamber on at least one of the supply path side and the discharge path side, thereby controlling the operation / stop of the filtration system 20 and the operation / stop of the cake deliquification process. Also, the supply valve of the slurry liquid supply tank or the filter aid liquid supply tank controls the supply amount of the slurry liquid or the filter aid liquid by changing the opening and closing time of the supply valve.

[0100] The control signal for the cake washing mode corresponds to the processing state of the washing system 30, and includes signals for controlling the operation and stop of the washing system 30, the increase / decrease in the amount of washing liquid supplied, the operation and stop of the cake deliquor process, etc. Upon receiving the control signal output from the output unit 14, the washing system 30 controls the filter chamber pressure control unit, the supply valve of the washing liquid supply tank, etc.

[0101] Specifically, the filter chamber pressure control unit increases or decreases the pressure inside at least one of the filter chambers on the supply path side and the discharge path side, thereby controlling the operation / stop of the cleaning system 30 and the operation / stop of the cake deliquor treatment. Also, the supply valve of the cleaning liquid supply tank controls the supply amount of the cleaning liquid by changing the opening and closing time of the supply valve.

[0102] The control signal for the replacement mode corresponds to the deterioration state of the filter media and includes a signal for controlling the operation and stop of the replacement system 40. When the replacement system 40 receives the control signal output from the output unit 14, it controls the filter media cleaning mechanism, filter media supply mechanism, or filter media discharge mechanism. The control signal for the replacement mode can be set as a control signal for a different mechanism depending on the degree of deterioration of the filter media. The deterioration state of the filter media may be determined in the filtration mode or the cake cleaning mode, and the mode may be switched to the replacement mode depending on the determination result.

[0103] The control signal for the equipment cleaning mode corresponds to the processing state of the cleaning system 30, and includes signals for controlling the continuation or stop of operation of the cleaning system 30, and the increase / decrease in the amount of cleaning liquid supplied. A control signal for starting equipment cleaning in the cleaning system 30 in the equipment cleaning mode is output from the output unit 14 upon receiving an equipment cleaning instruction operation from the control panel 60. Upon receiving the control signal output from the output unit 14, the cleaning system 30 controls the filter chamber pressure control unit, the supply valve of the cleaning liquid supply tank, etc. The equipment cleaning instruction operation is executed by an operator, for example, when replacing the object to be filtered by the filtration system 1, such as the slurry liquid.

[0104] The output unit 14 outputs a warning signal to the warning notification system 50 based on the determination result of the processing system 10 by the determination unit 13. Fig. 11(b) shows an example of the data configuration of the warning signal table. The warning signal table has a control mode, a determination result, and a warning signal. The control mode in the warning signal table is either a filtration mode or a cake washing mode, and a warning signal corresponding to each control mode is set.

[0105] The warning signal corresponds to a deteriorated state of the filter media or an abnormal state of the filter media, and includes a signal for controlling the warning notification of the warning notification system 50. When the warning notification system 50 receives the warning signal output from the output unit 14, it controls the display on the control panel 60, the voice or alarm sound of the alarm, the lighting of a lamp, etc.

[0106] Specifically, when the control panel 60 receives a warning signal indicating an abnormality in the filter media, it displays a notification regarding the abnormality. Also, when the control panel 60 receives a warning signal indicating deterioration of the filter media, it displays a notification regarding the deterioration. An alarm or lamp may warn of the abnormality or deterioration by sound or lighting pattern corresponding to the warning signal.

[0107] The output unit 14 may output a control signal to stop the filtration system 20 or the cleaning system 30 in response to the determination result of the abnormal state of the filter media. This makes it possible to automate an emergency stop operation in response to an abnormality such as breakage of the filter media.

[0108] The first control mode switching unit 15 switches the control mode indicated in the processing system 10 to be controlled in the filtration system 1. The first control mode switching unit 15 determines which control mode to switch to, among the filtration mode, the cake washing mode, and the replacement mode, based on the determination result of the state of the processing system 10 by the determination unit 13 or the output of a control signal by the output unit 14. The first control mode switching unit 15 normally switches between the filtration mode, the cake washing mode, and the replacement mode in that order, but it is also possible to switch from the filtration mode to the replacement mode if, for example, a determination result indicating a deteriorated or abnormal state of the filter media is obtained during filtration.

[0109] The first control mode switching unit 15 switches to the next control mode based on the determination result of the filtration completion, cleaning completion, or filter media replacement completion state. The first control mode switching unit 15 also switches to the next control mode based on the output of a predetermined control signal such as the stoppage of the filtration system 20, the stoppage of the cleaning system 30, or the stoppage of the replacement system 40.

[0110] The first control mode switching unit 15 may switch the control mode to the warning notification mode based on the determination result of the abnormal state of the filter medium. When the abnormal state is resolved, the first control mode switching unit 15 switches from the warning notification mode to the original control mode.

[0111] The first control mode switching unit 15 switches the control mode to the apparatus cleaning mode when it receives an apparatus cleaning instruction operation from the control panel 60. The first control mode switching unit 15 switches back to the original control mode based on the determination result of cleaning completion or the output of a control signal to stop operation.

[0112] 12 shows a flowchart of a series of controls performed by the filtration system 1. The filtration system 1 receives input of basic settings such as the type of slurry liquid to be filtered, the total amount or operating time, and the type of filter media to be used via the control panel 60, and then starts processing.

[0113] <Filtration Step> Step S21: The filtration system 20 performs filtration by controlling the filter chamber pressure control unit, the slurry liquid supply valve, and the filter aid liquid supply valve. At this time, the control mode is set to the filtration mode.

[0114] Step S22: The imaging unit 12 acquires image data of the slurry liquid and the filtrate. The determination unit 13 inputs the image data into a determination model and determines the state of the processing system 10, including the processing state of the filtration system 20, the deterioration state of the filter media, and the abnormal state of the filter media. Based on the determination results regarding the supply rates of the slurry liquid and the filtrate, the output unit 14 outputs control signals to the filtration system 20 to control the supply rates of these liquids, thereby continuing the filtration process.

[0115] Step S23: If the determination unit 13 determines that the filtration processing status is filtration completed (YES in S23), the output unit 14 outputs a control signal to the filtration system 20 to stop operations related to the filtration processing. The first control mode switching unit 15 switches the control mode to the filtrate drainage mode. If the determination unit 13 determines that the filtration processing status is filtering in progress (NO in S23), it repeatedly executes the first determination processing of step S22 until it determines that filtration is completed.

[0116] Step S24: If the determination unit 13 determines that the deliquification processing status is deliquification completed (YES in S24), the output unit 14 outputs a control signal to the filtration system 20 to stop operations related to the deliquification processing. The first control mode switching unit 15 switches the control mode to the cake washing mode. If the determination unit 13 determines that the deliquification processing status is deliquifying (NO in S24), the first determination processing of step S22 is repeatedly executed until it determines that deliquification is completed.

[0117] <Cake Washing Step> Step S25: The washing system 30 controls the filter chamber pressure control unit and the washing liquid supply valve to perform a cake washing process.

[0118] Step S26: The imaging unit 12 acquires image data of the waste cleaning fluid. The determination unit 13 inputs the image data into a determination model to determine the state of the processing system 10, including the processing state of the cleaning system 30, the deterioration state of the filter media, and the abnormal state of the filter media. Based on the determination result regarding the supply amount of cleaning fluid, the output unit 14 outputs a control signal to the cleaning system 30 to control the supply amount, thereby continuing the cleaning process.

[0119] Step S27: If the determination unit 13 determines that the processing status of the cleaning system 30 is cleaning completed (YES in S27), the output unit 14 outputs a control signal to the cleaning system 30 to stop operations related to the cleaning process. The first control mode switching unit 15 switches the control mode to a cleaning liquid removal mode. If the determination unit 13 determines that the processing status of the cleaning system 30 is cleaning in progress (NO in S27), the second determination process of step S26 is repeatedly executed until it is determined that cleaning is completed.

[0120] Step S28: If the determination unit 13 determines that the draining process status is draining complete (YES in S28), the output unit 14 outputs a control signal to the cleaning system 30 to stop operations related to the draining process. The first control mode switching unit 15 switches the control mode to the replacement mode. If the determination unit 13 determines that the draining process status is draining (NO in S28), the second determination process of step S26 is repeatedly executed until draining is determined to be complete.

[0121] <Filter Media Replacement Step> Step S29: The determination unit 13 determines whether or not the filter media is deteriorated as a deterioration state of the filter media during the filtration step or the cake washing step. When the determination unit 13 determines that the filter media is not deteriorated (NO in S29), it determines that filter media replacement is not necessary and completes the filter media replacement step.

[0122] Step S30: If the determination unit 13 determines that the filter media has deteriorated (YES in S29), the output unit 14 outputs a control signal to the replacement system 40 regarding the operation of the replacement system 40 to replace or clean the filter media. Whether to replace or clean the filter media is determined depending on the degree of deterioration of the filter media. The first control mode switching unit 15 switches the control mode to the filtration mode and completes the process. Note that the determination unit 13 may be configured to immediately execute step S30 when it determines that the filter media has deteriorated in the filtration step or the cake washing step.

[0123] Step S31: After the filter medium replacement step, the filtration system 1 executes a cake discharge step of discharging cake accumulated on the filter medium. Note that the cake discharge step may be omitted if the filter medium replacement step is configured to discharge cake together with the filter medium.

[0124] Step S32: When the cake discharge process is completed, the first control mode switching unit 15 switches the control mode to the filtration mode, and the process returns to step S21, where the process can be repeated from the filtration process (YES in S32). The filtration system 1 completes the process when it receives an instruction to complete filtration or when it completes the predetermined filtration process (NO in S32).

[0125] As described above, the present disclosure makes it possible to automatically control a series of processes for filtering a slurry liquid using a filtering device.

[0126] The filtration system 1 can reduce the operational load by automatically controlling a series of processes. It can also reduce safety risks by preventing poor cleaning when handling chemicals. It can also improve product quality by preventing poor drainage, maintaining filtrate coolness, and preventing poor cake washing. It can also reduce energy loss by preventing excessive drainage operation. It can also reduce resource usage by preventing excessive cake washing and excessive equipment washing.

[0127] Second Embodiment Hereinafter, a second embodiment will be described, which is different from the filtration system 1. Note that the same components as those in the first embodiment will be denoted by the same reference numerals and the description thereof will be omitted.

[0128] 13 shows a block diagram of a filtration system 1 in embodiment 2. In embodiment 2, the filtration system 1 includes a determination device 100 and a filtering device 200.

[0129] The determination device 100 includes a first control unit 110, an imaging unit 12, a determination unit 13, an output unit 14, and a storage unit DB serving as a database, and each component is connected to and controlled by the first control unit 110. In this embodiment, the output unit 14 is connected to the filtration device 200 and controls the operation of the filtration device 200 by outputting a control signal.

[0130] The imaging unit 12 is disposed on the path of the fluid within the filtering device 200 and acquires image data of the fluid.

[0131] The determining unit 13 determines the state of the processing system 10 in the filtering device 200 based on the image data acquired by the imaging unit 12 .

[0132] In the second embodiment, the output unit 14 outputs a control signal to the filtering device 200 based on the determination result of the state of the processing system 10 by the determination unit 13. The processing system 10 is controlled based on the control signal output by the output unit 14.

[0133] A general-purpose computer can be used as the determination device 100. The determination device 100 includes, as hardware components, an arithmetic unit such as a CPU, a main storage unit such as a RAM, an auxiliary storage unit, a communication unit, an input / output unit, and the like.

[0134] The filtration device 200 includes a control panel 60 and a processing system 10. In the second embodiment, an existing filtration device can be used as the filtration device 200. Furthermore, the control panel 60 can be a programmable logic controller (PLC) or the like.

[0135] The control panel 60 includes a signal input unit 61, a second control unit 62, and a signal output unit 63. The signal input unit 61 receives input of a control signal output from the output unit 14 of the determination device 100. The second control unit 62 includes a signal processing unit 64 and a second control mode switching unit 65 as functional components, and executes processing based on the control signal received as input by the signal input unit 61. The signal output unit 63 is connected to the processing system 10, obtains a control signal as a processing result by the second control unit 62, and outputs the control signal to the processing system 10, thereby controlling the operation of the processing system 10.

[0136] The signal processing unit 64 receives and processes the control signal of the output unit 14. The signal processing includes converting the control signal into a signal to be output to the processing system 10 and determining the output destination.

[0137] The second control mode switching unit 65 switches the control mode indicated for the processing system 10 to be controlled in the filtering device 200. The second control mode switching unit 65 switches the control mode based on the determination result of the state of the processing system 10 by the determination unit 13 or the output of a control signal by the signal output unit 63. The second control mode switching unit 65 may output the switched control mode to the determination device 100.

[0138] As in the first embodiment, the processing system 10 includes a filtration system 20 that controls the filtration process, a cleaning system 30 that controls the cleaning process, a filter media replacement system 40 that controls the filter media replacement process, and a warning notification system 50 that controls notification of a deteriorated or abnormal state of the filter media. The operations of the filtration system 20, cleaning system 30, replacement system 40, and warning notification system 50 are controlled based on control signals output from a signal output unit 63.

[0139] The filtration system 1 is not limited to the above-described embodiment, and various configurations can be adopted as long as the same overall operational effects can be achieved.

[0140] The present invention can be used in various types of filtration devices, such as pressure filtration devices, vacuum filtration devices, centrifugal filtration devices, and gravity filtration devices.

[0141] REFERENCE SIGNS LIST 1 Filtration system 7 Model generation device 11 Control unit 12 Imaging unit 13 Determination unit 14 Output unit 15 First control mode switching unit 20 Filtration system 30 Cleaning system 40 Replacement system 50 Warning notification system 60 Control panel DB Memory unit 100 Determination device 110 First control unit 200 Filtration device 61 Signal input unit 62 Second control unit 63 Signal output unit 64 Signal processing unit 65 Second control mode switching unit F Fluid F1 Slurry liquid F2 Filtrate F3 Filter aid liquid F4 Filtrate containing filter aid liquid F5 Mixed liquid F6 Cleaning liquid F7 Cleaning waste liquid

Claims

DEPCT681. An automated control filtration system, a treatment unit for filtering viscous liquids, a treatment unit which includes a filtration unit, a filtration system consisting of: an imaging unit positioned on the fluid path in the filtration unit and configured to obtain fluid image data; an identification unit configured to identify the state of the treatment unit based on the image data; and an output unit configured to output control signals for the treatment unit based on the identification results of the treatment unit's state, where the treatment unit's state includes the filtration unit's treatment state, the filtration medium's deterioration state, or the filtration medium's abnormal state, and the identification unit identifies the treatment unit's state, the filtration medium's deterioration state, or the filtration medium's abnormal state using a machine-learned identification model with image data as input and with the treatment unit's state, the filtration medium's deterioration state, or the filtration medium's abnormal state as output.2.The filtration system with automated control capabilities includes a treatment unit for filtering viscous liquids. This treatment unit comprises a cleaning unit and a filtration system consisting of: an imaging unit positioned along the fluid path in the cleaning unit and configured to obtain fluid image data; an identification unit configured to identify the state of the treatment unit based on the image data; and an output unit configured to output control signals for the treatment unit based on the identification results of the treatment unit's state. The treatment unit's state includes the treatment state of the cleaning unit, the deterioration state of the filtration medium, or the abnormal state of the filtration medium. The identification unit identifies the treatment state of the cleaning unit, the deterioration state of the filtration medium, or the abnormal state of the filtration medium using a machine-learned identification model with image data as input and the treatment state of the cleaning unit, the deterioration state of the filtration medium, or the abnormal state of the filtration medium as output.

4. Filtration system under claim 1, whereby the treatment unit includes the filtration unit, the fluid is the filtered liquid, the identification unit identifies the treatment status of the filtration unit, and the output unit outputs control signals for the filtration unit based on the identification results of the filtration unit's treatment status.

5. Filtration system under claim 3, whereby the control signals for the filtration unit include at least one or more of the control signals related to the operation of the filtration unit, control signals related to the dispensing volume of viscous liquid, control signals related to the treatment of liquid removal, or control signals related to the dispensing volume of filtration aid liquid.

6. Filtration system under claim 3, whereby the control signals for the filtration unit are control signals for increasing or decreasing the internal pressure of the filter chamber on at least one side of the supply or discharge path.

7. Filtration system under claim 3, whereby the control signals for the filtration unit are control signals for changing the opening and closing times of the supply valves for viscous liquid or filtration aid liquid.Filtration system under claim 2, where the treatment unit includes a cleaning unit, the fluid is the cleaning waste, the identification unit identifies the treatment status of the cleaning unit, and the output unit outputs control signals for the cleaning unit based on the identification results of the cleaning unit's treatment status.

8. Filtration system under claim 1, where the treatment unit includes a filtration unit, a cleaning unit, a replacement unit, or an alarm unit, the fluid is the filtered liquid, the identification unit identifies the deterioration status of the filtration medium, and the output unit outputs at least one or more control signals for the cleaning unit, control signals for the replacement unit, or alarm signals for the alarm unit based on the identification results of the deterioration status of the filtration medium. 9.Filtration system under claim 2, where the treatment unit includes a cleaning unit, a replacement unit, or an alarm unit, the fluid is the cleaning waste, the identification unit identifies the deterioration status of the filtration medium, and the output unit outputs at least one or more control signals for the cleaning unit, control signals for the replacement unit, or alarm signals for the alarm unit based on the results of the identification of the deterioration status of the filtration medium.

10. Filtration system under claim 1 or 2, where the treatment unit includes a filtration unit, a cleaning unit, or an alarm unit, the fluid is the filtered liquid or cleaning waste, the identification unit identifies the abnormal status of the filtration medium, and the output unit outputs control signals related to the operation to the filtration unit or outputs alarm signals related to the abnormal status of the filtration medium to the alarm unit, based on the results of the identification of the abnormal status of the filtration medium.11.Filtration system according to claim 1, where the treatment unit includes the filtration unit, the fluid is a viscous liquid, the imaging unit is positioned on the fluid supply path and obtains image data of the viscous liquid, and the output unit sends control signals related to the amount of filtration fluid supplied to the filtration unit based on the results of the identification of the treatment status of the filtration unit.

12. Filtration system according to claim 1 or 2, where the imaging unit is positioned on the fluid path at an angle of view that is substantially horizontal relative to the horizontal plane.13.The filtration system under claim 1 or 2, which includes additional components: a mode-switching control unit configured to switch the treatment units to be controlled, where the treatment units include the filtration unit, the cleaning unit, and the replacement unit; an identification unit that indicates the status of the treatment units, including the treatment status of the filtration unit, the treatment status of the cleaning unit, and the deterioration status of the filtration media; an output unit that outputs control signals for the filtration unit, the cleaning unit, and the replacement unit based on the identification results of the treatment unit status; and a mode-switching control unit that switches the control mode based on the identification results of the treatment unit status or the output of the control signal.

14. The filtration system under claim 13, where the treatment unit includes an alarm unit; an identification unit that further indicates the status of the treatment units, including the abnormal status of the filtration media; and an output unit that outputs the corresponding alarm signal to the alarm unit based on the identification results of the deterioration status of the filtration media or the abnormal status of the filtration media.15.The filtration system under claim 13, where the filtered liquid discharge path of the filtration unit also serves as the waste discharge path from the cleaning unit, and the imaging unit is positioned in the filtered liquid discharge path;