Filtration system

Abstract

The present disclosure provides a filtration system capable of automatically controlling a processing assembly for filtering a slurry liquid by artificial intelligence (AI). The apparatus comprises: an imaging unit which is disposed on a path of a fluid in a filter module and acquires image data of the fluid; a determination unit that determines the state of the processing component on the basis of the image data; and an output unit that outputs a control signal for the processing unit on the basis of a determination result of a state of the processing unit, the state of the processing unit including a processing state of the filter unit, a degraded state of the filter, or an abnormal state of the filter, the determination unit determines a processing state of the filter module, a degradation state of the filter, or an abnormal state of the filter using a determination model in which machine learning is performed using the image data as input data and the processing state of the filter module, the degradation state of the filter, or the abnormal state of the filter as output data.

Description

Technical Field

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

[0002] In order to keep a filtration device operating continuously, it is important to accurately grasp the states of filtration elements such as filter membranes, filters, filter papers, and filter cloths.

[0003] The filtration performance of a filtration element decreases due to the attachment of substances to be separated to the surface and inside of the filtration element caused by repeated filtration. In addition, when the filtration element deteriorates and breaks, etc., the quality of products such as filtrate and filter cake deteriorates. Conventionally, the state of the filtration element has been managed by visual confirmation by an operator, time setting, etc. Therefore, techniques for managing the state of the filtration element have been provided.

[0004] In Patent Document 1, an oil state monitoring method and an oil state monitoring device that can accurately monitor the state of oil to appropriately predict the life of oil used in various machines or equipment are disclosed. In Patent Document 1, light is projected onto a filter that filters and removes oil components from oil, and the color components of the transmitted light are detected to monitor the deterioration state of the oil.

[0005] In Patent Document 2, an important factor calculation device that accurately predicts important factors that are the main causes of the occurrence of fouling of a filter membrane is disclosed. In Patent Document 2, a predictor that predicts the filtration performance of a filter membrane after a unit time step is learned from time series data of a plurality of parameters including a parameter indicating the filtration performance of the filter membrane and a parameter indicating the water quality of a drainage treatment component using the filter membrane, and parameters that contribute to predicting the filtration performance after a unit time step are output according to the prediction period.

[0006] In Patent Document 3, a technique for achieving appropriate operation control of a centrifugal separation system without relying on the judgment of an operator is disclosed. Patent Document 3 discloses including: a learning dataset storage unit 22 that stores a plurality of sets of learning datasets, the learning datasets including input data and output data including control parameters corresponding to the input data, the input data including image data obtained by photographing a separation liquid from a specified viewing angle, and the control parameters including at least one of the supply amount of an additive added to a liquid to be processed PL1, the centrifugal force of a bowl 2, and the differential speed controlled by a differential speed generating device 5; a learning unit 23 that learns a learning model that infers the correlation between input data and output data by inputting a plurality of sets of learning datasets; and a learned model storage unit 24 that stores the learned learned model.

[0007] According to Patent Document 3, it is known that automating the control in a centrifugal separation device is effective in eliminating work load and improving productivity etc.

[0008] Prior Art Documents

[0009] Patent Documents

[0010] Patent Document 1: Japanese Patent Publication No. 5190660

[0011] Patent Document 2: International Publication No. WO2022 / 085802

[0012] Patent Document 3: Japanese Unexamined Patent Application Publication No. 2022 - 021243 Summary of the Invention

[0013] Problems to be Solved by the Invention

[0014] The technical problem of the present invention is to provide a filtration system that can automatically control a processing component for filtering a slurry liquid by artificial intelligence (AI: Artificial Intelligence).

[0015] Solutions to the Problems

[0016] To solve the above - mentioned technical problems, the present invention is a filtration system that can automatically control a processing component for filtering a slurry liquid. The processing component includes a filtration component. The filtration system includes: a photographing unit disposed on a path of a fluid in the filtration component to acquire image data of the fluid; a determination unit that determines a state of the processing component based on the image data; and an output unit that outputs a control signal for the processing component based on a determination result of the state of the processing component. The state of the processing component includes a processing state of the filtration component, a deterioration state of a filter element, or an abnormal state of the filter element. The determination unit uses a determination model to determine the processing state of the filtration component, the deterioration state of the filter element, or the abnormal state of the filter element. The determination model has performed machine learning with the image data as input data and the processing state of the filtration component, the deterioration state of the filter element, or the abnormal state of the filter element as output data.

[0017] Thus, by using a machine - learned model in the filtration device, it is possible to determine the processing state of the filtration component, the deterioration state of the filter element, and the abnormal state of the filter element based on the image data of the fluid, and automatically control the processing component according to their states. By incorporating artificial intelligence (AI) in the filtration device, it is possible to automatically determine the determination criteria related to the state of the processing component that an operator judges based on experience. Therefore, without the need for operations by the operator, it is possible to achieve automatic control of the appropriate processing component corresponding to the state.

[0018] In addition, the present invention is a filtration system that can automatically control a processing component for filtering a slurry liquid. Among them, the processing component includes a cleaning component, and the filtration system has: a photographing unit configured on the path of the fluid in the cleaning component to acquire image data of the fluid; a determination unit that determines the state of the processing component based on the image data; and an output unit that outputs a control signal for the processing component based on the determination result of the state of the processing component. And the state of the processing component includes the processing state of the cleaning component, the deterioration state of the filter element, or the abnormal state of the filter element. The determination unit uses a determination model to determine the processing state of the cleaning component, the deterioration state of the filter element, or the abnormal state of the filter element. Among them, the determination model uses the image data as input data and performs machine learning with the processing state of the cleaning component, the deterioration state of the filter element, or the abnormal state of the filter element as output data.

[0019] In this way, by using a machine learning-complete model in the filtration device, it is possible to determine the processing state of the cleaning component, the deterioration state of the filter element, and the abnormal state of the filter element based on the image data of the fluid, and automatically control the processing component according to their states. By incorporating artificial intelligence (AI), the filtration device can automatically determine the determination criteria related to the state of the processing component that operators judge based on experience. Therefore, without the operations performed by the operator, it is possible to achieve appropriate automatic control of the processing component corresponding to the state.

[0020] In a preferred embodiment of the present invention, the processing component includes a filtration component, the fluid is a filtrate, the determination unit determines the processing state of the filtration component, and the output unit outputs a control signal for the filtration component based on the determination result of the processing state of the filtration component.

[0021] In a preferred embodiment of the present invention, the control signal for the filtration component includes at least one of a control signal related to the operation of the filtration component, a control signal related to the supply amount of the slurry liquid, a control signal related to the dewatering process, and a control signal related to the supply amount of the filter aid liquid.

[0022] In a preferred embodiment of the present invention, the control signal for the filtration component is a control signal for pressurizing or depressurizing the internal pressure of the filtration chamber on at least one of the supply path side and the discharge path side.

[0023] In a preferred embodiment of the present invention, the control signal for the filtration component is a control signal for changing the opening and closing time of the supply valve of the slurry liquid or the filter aid liquid.

[0024] By adopting such a configuration, it is possible to determine the processing state of the filtration component and achieve appropriate automatic control corresponding to each state.

[0025] In a preferred embodiment of the present invention, the processing component includes a cleaning component, the fluid is cleaning drainage, the determination unit determines the processing state of the cleaning component, and the output unit outputs a control signal for the cleaning component based on the determination result of the processing state of the cleaning component.

[0026] By adopting such a configuration, the processing state of the cleaning component can be determined, and appropriate automatic control corresponding to this state can be achieved.

[0027] In a preferred embodiment of the present invention, the processing component includes a filtering component, a cleaning component, a replacement component, or a warning notification component, the fluid is filtrate, the determination unit determines the deterioration state of the filter element, and the output unit outputs at least one of a control signal for the cleaning component, a control signal for the replacement component, and a warning signal for the warning notification component based on the determination result of the deterioration state of the filter element.

[0028] In a preferred embodiment of the present invention, the processing component includes a cleaning component, a replacement component, or a warning notification component, the fluid is cleaning drainage, the determination unit determines the deterioration state of the filter element, and the output unit outputs at least one of a control signal for the cleaning component, a control signal for the replacement component, and a warning signal for the warning notification component based on the determination result of the deterioration state of the filter element.

[0029] By adopting such a configuration, the deterioration state of the filter element can be determined, and appropriate automatic control corresponding to this state can be achieved.

[0030] In a preferred embodiment of the present invention, the processing component includes a filtering component, a cleaning component, or a warning notification component, the fluid is filtrate or cleaning drainage, the determination unit determines the abnormal state of the filter element, and the output unit outputs a control signal related to operation to the filtering component or a warning signal related to the abnormal state of the filter element to the warning notification component based on the determination result of the abnormal state of the filter element.

[0031] By adopting such a configuration, the abnormal state of the filter element can be determined, and the abnormal state can be quickly responded to.

[0032] In a preferred embodiment of the present invention, the processing component includes a filtering component, the fluid is the slurry liquid, the photographing unit is arranged on the supply path of the slurry liquid to obtain image data of the slurry liquid, and the output unit outputs a control signal related to the supply amount of the filter aid liquid to the filtering component based on the determination result of the processing state of the filtering component.

[0033] By adopting such a configuration, it is possible to control the supply amount of the filter aid liquid appropriately according to the state of the slurry liquid.

[0034] In a preferred embodiment of the present invention, the photographing unit is disposed on the path of the fluid at a viewing angle that is substantially horizontal with respect to the ground plane.

[0035] By adopting such a configuration, it is possible to photograph the liquid surface, liquid volume, etc. of the fluid at an appropriate viewing angle, thereby improving the accuracy of determination.

[0036] In a preferred embodiment of the present invention, the filtration system further includes a control mode switching unit for switching the control mode of the processing components to be controlled. The processing components include a filtration component, a cleaning component, and a replacement component. The determination unit determines the states of the processing components including the processing state of the filtration component, the processing state of the cleaning component, and the deterioration state of the filter element. The output unit outputs control signals for the filtration component, the cleaning component, and the replacement component based on the determination result of the states of the processing components. The control mode switching unit switches the control mode based on the determination result of the states of the processing components or the output of the control signals.

[0037] In a preferred embodiment of the present invention, the processing components include a warning notification component. The determination unit also determines the states of the processing components including the abnormal state of the filter element. The output unit outputs a warning signal corresponding to the control mode to the warning notification component based on the determination result of the deterioration state or the abnormal state of the filter element.

[0038] By adopting such a configuration, it is possible to automatically control a series of operations in the filtration device including the filtration component, the cleaning component, and the replacement component.

[0039] In a preferred embodiment of the present invention, the filtrate discharge path of the filtration component also serves as the cleaning liquid discharge path of the cleaning component, and the photographing unit is disposed on the filtrate discharge path.

[0040] By adopting such a configuration, it is possible to automatically control the operations of the filtration device including the filtration component and the cleaning component by arranging one photographing unit.

[0041] Advantages of the Invention

[0042] According to the present invention, it is possible to provide a filtration system that can automatically control the processing components for filtering slurry liquid by artificial intelligence (AI). BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 It is a block diagram showing the filtration system of the present embodiment.

[0044] Figure 2Shows a schematic diagram of the filtration system of this embodiment.

[0045] Figure 3 Shows a schematic diagram of the filtration system of this embodiment.

[0046] Figure 4 Shows a schematic diagram of the photographing unit of this embodiment.

[0047] Figure 5 Shows a schematic diagram of the photographing unit of this embodiment.

[0048] Figure 6 Shows an example of the image data obtained by the photographing unit of this embodiment.

[0049] Figure 7 Shows a block diagram of the model generation device of this embodiment.

[0050] Figure 8 Shows a configuration example of the determination model of this embodiment.

[0051] Figure 9 Shows a configuration example of the data set of this embodiment.

[0052] Figure 10 Shows a flowchart of the machine learning process of this embodiment.

[0053] Figure 11 Shows a configuration example of the control signal table of this embodiment.

[0054] Figure 12 Shows each step in the filtration system of this embodiment.

[0055] Figure 13 Shows a block diagram of the filtration system of Embodiment 2.

[0056] Figure 14 Shows each step in the conventional filtration device. Detailed implementation mode

[0057] Hereinafter, with reference to the drawings, the filtration system of the present invention will be described. It should be noted that the following-described embodiment is an example of the present invention, and the present invention is not limited to the following embodiment, and various configurations can be adopted.

[0058] The filtration device separates the solid-liquid mixture, that is, the slurry liquid, which is the object to be filtered, into a solid formation, that is, a filter cake and a filtrate, by passing it through a filtration membrane or a filter, a filter paper or a filter cloth, etc.

[0059] Figure 14 Shows the steps of each process using an existing filtration device.

[0060] The filtration step is a step of filtering the slurry liquid to separate it into a filter cake and a filtrate. In the filtration step, the operator performs operations such as starting the filtration of the filtration device, setting or adjusting the supply amount of the slurry liquid, setting or adjusting the pressure in the filtration chamber, and ending the filtration. It should be noted that the filtration step may also include a liquid drainage step of performing liquid drainage treatment on the filter cake. In the liquid drainage step, the operator performs operations such as starting the liquid drainage of the filtration device, setting or adjusting the pressure in the filtration chamber, and ending the liquid drainage.

[0061] The filter cake cleaning step is a step of supplying a cleaning liquid to the filtration chamber after the filtration is completed to clean the filter cake. In the filter cake cleaning step, the operator performs operations such as starting the cleaning of the filter cake of the filtration device, setting or adjusting the supply amount of the cleaning liquid, setting or adjusting the pressure in the filtration chamber, and ending the cleaning.

[0062] The filter element replacement step is a step of cleaning or replacing the filter element when the filter element deteriorates (such as dirt, etc.) or when the filter element has an abnormality (such as filter element breakage, filtration leakage, etc.). In the filter element replacement step, the operator performs operations such as starting the cleaning or replacement of the filter element and ending the replacement of the filter element. In the case of a non-automatic filtration device, the filter element replacement step is manually performed by opening the filtration chamber frame, setting the filter element, and closing the filtration chamber frame, etc. It should be noted that it is set that in the filter element replacement step, the filter cake accumulated on the filter element is discharged.

[0063] After the filter element replacement is completed in the filter element replacement step, return to the filtration step and repeat the process.

[0064] In the above processing steps, the operator visually confirms the state of the filtrate and the state of the cleaning drainage to judge whether to end the filtration and cleaning. Or, the operator manages each step by setting the filtration time and the cleaning time using a timer. In addition, regarding the input amount and input timing of the filter aid liquid, judgments based on the operator's visual inspection and management based on timer settings are also performed.

[0065] The purpose of the filtration system of the present invention is to automatically control the above-mentioned filtration step, filter cake cleaning step, and filter element replacement step.

[0066] <Embodiment 1>

[0067] Figure 1The block diagram showing the filtration system 1 of the present embodiment. The filtration system 1 includes a control unit 11, a photographing 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 the control unit 11 and controlled thereby. The output unit 14 is connected to the processing component 10 and controls the operation of the processing component 10 by outputting a control signal. The processing component 10 includes: a filtration component 20 for controlling the filtration step, a cleaning component 30 for controlling the cleaning step, a replacement component 40 for controlling the filter element replacement step, and a warning notification component 50 for controlling the notification of the deterioration state (such as dirt, etc.) of the filter element or the abnormal state (such as filter element breakage, filtration leakage, etc.) of the filter element. In the present embodiment, the filtration system 1 is configured as a filtration device.

[0068] The deterioration state of the filter element includes at least no deterioration of the filter element and deterioration of the filter element. It should be noted that the deterioration state of the filter element may further include the degree of the stagewise deterioration state (small deterioration, large deterioration, etc.).

[0069] The abnormal state of the filter element includes at least no abnormality and an abnormality. The abnormal state of the filter element may also include filter element breakage, filtration leakage, etc. as categories of abnormalities.

[0070] The processing component 10 shows a device configuration for executing processes including filtration, filter cake cleaning, device cleaning, filter element replacement, and warning notification in the filtration system 1 (1A to 1D). The processing component 10 causes the components to operate based on the input of a control signal to implement various processes.

[0071] The control unit 11 is constituted by an arithmetic device based on a CPU (Central Processing Unit). The control unit 11 can be connected to the processing component 10 to control the mechanism based on a control signal. The control unit 11 can be connected to the control panel 60 to receive an operation input via the control panel 60. The control unit 11 can also perform display processing on the control panel 60.

[0072] The photographing unit 12 is arranged on the path of the fluid in the filtration device and acquires image data of the fluid.

[0073] The determination unit 13 determines the state of the processing component 10 based on the image data acquired by the photographing unit 12.

[0074] The output unit 14 outputs a control signal for the processing component 10 based on the determination result of the state of the processing component 10 obtained by the determination unit 13. The processing component 10 is controlled based on the control signal output by the output unit 14.

[0075] The first control mode switching unit 15 switches the control mode based on the determination result of the state of the processing component 10 obtained by the determination unit 13 or the output of the control signal by the output unit 14. The control mode indicates which one of the filtering component 20, the cleaning component 30, and the replacement component 40 is the processing component 10 to be controlled.

[0076] The storage unit DB stores the determination model used in the determination process of the image data by the determination unit 13. The storage unit DB stores a control signal table for determining the control signal output by the output unit 14. In addition, the storage unit DB stores various data such as image data, and a program including various instructions executed by the control unit 11. This program can also be stored in a computer-readable non-transitory recording medium such as a CD-ROM (Compact Disc Read-Only Memory), a flash memory, or an SSD (Solid State Disk) memory for installation.

[0077] Figure 2 of (a), Figure 2 of (b), Figure 2 of (c) are diagrams showing the device configuration for controlling the filtration steps of the slurry filtration systems 1A to 1C.

[0078] As Figure 2 shown in (a) of, the filtering component 20 is composed of a device for controlling the filtration step of the slurry filtration system 1A that filters the slurry liquid F1 in the filtration chamber R. The filtering component 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 filtration chamber R; a discharge path 102 that discharges the filtrate F2 filtered in the filtration chamber R; a first discharge tank DT1 that stores the filtrate F2; a first imaging unit 12A that is clamped to the supply path 101 between the first supply tank UT1 and the filtration chamber R to acquire image data of the slurry liquid F1; and a second imaging unit 12B that is clamped to the discharge path 102 between the filtration chamber R and the first discharge tank DT1 to acquire image data of the filtrate F2. It should be noted that the slurry filtration system 1A includes a filtration chamber pressure control unit (not shown).

[0079] In addition, 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 its opening and closing time. The filtration chamber R is provided with a filter element (not shown) between the supply path 101 side and the discharge path 102 side, and the slurry liquid F1 is filtered by the pressure difference between the supply path 101 side and the discharge path 102 side.

[0080] The filtration chamber pressure control section (not shown) is composed of a pressure pump (not shown) and a pressure valve (not shown) that control the internal pressure of the filtration chamber R on at least one of the supply path 101 side and the discharge path 102 side. In addition, for example, in the case of adopting a mechanism that performs centrifugal separation by rotating the filtration chamber R, the filtration chamber pressure control section (not shown) is composed of an actuator (not shown) that controls its rotational speed or rotation speed, etc. In addition, for example, in the case of adopting a mechanism that separates the slurry liquid F1 by pressing or the like, the filtration chamber pressure control section (not shown) is composed of an actuator (not shown) that controls the pressing pressure and pressing time, etc.

[0081] As Figure 2 shown in (b) of, in the slurry filtration system 1B, the filtration module 20 may also be configured to supply the filter aid liquid F3 when filtering the slurry liquid F1 in the filtration chamber R. Regarding the filter aid liquid F3, a filter aid liquid supply path 103 is provided that connects the second supply tank UT2 that supplies the filter aid liquid F3 to the supply path 101 to supply the filter aid liquid F3. The second supply tank UT2 is equipped with a supply valve (not shown), and the supply amount of the filter aid liquid F3 is controlled by its opening and closing time. It should be noted that the filter aid liquid supply path 103 for the filter aid liquid F3 from the second supply tank UT2 to the filtration chamber R may be separately provided in such a way that it does not connect to the supply path 101 for the slurry liquid F1 from the first supply tank UT1 to the filtration chamber R and supplies it to the filtration chamber R.

[0082] The first imaging unit 12A acquires image data of the slurry liquid F1 passing through the supply path 101. In addition, the second imaging unit 12B acquires image data of the filtrate F2 passing through the discharge path 102 or the filtrate F4 including the filter aid liquid F3.

[0083] Moreover, as Figure 2 shown in (c) of, the filtration module 20 may also be configured such that in the slurry filtration system 1C, the first imaging unit 12A is disposed on the supply path 101a of the mixed liquid F5 of the slurry liquid F1 and the filter aid liquid F3 and acquires image data of the mixed liquid F5.

[0084] Figure 3 It is a diagram showing the device configuration for controlling the cleaning step of the slurry filtration system 1D.

[0085] As Figure 3 shown, in the slurry filtration system 1D, the cleaning module 30 includes a third supply tank UT3 that supplies the cleaning liquid F6, a filtration chamber R, a third imaging unit 12C disposed on the discharge path 102, and a second discharge tank DT2 that stores the used cleaning drainage liquid F7.

[0086] The cleaning component 30 also includes a filter chamber pressure control section (not shown). The third supply tank UT3 is equipped with a supply valve (not shown), and the supply amount of the cleaning liquid is controlled by its opening and closing time. The filter chamber R, the filter chamber pressure control section (not shown), a part of the supply path 101, and a part of the discharge path 102 can adopt the same structure as that of the filter component 20.

[0087] In addition, in this embodiment, the cleaning step includes a filter cake cleaning step for cleaning the filter cake and a device cleaning step for cleaning the interior of the filtration device. The cleaning liquid F6 cleans the filter cake (not shown) or the filter chamber R, and is discharged as cleaning drainage F7 to the second discharge tank DT2 through the discharge path 102.

[0088] The replacement component 40 represents the device structure for controlling the filter element replacement step of the filtration system 1. The replacement component 40 includes a replacement filter element, a replacement filter element bracket for accommodating the replacement filter element, a filter element supply mechanism for supplying the replacement filter element from the replacement filter element bracket to the filter chamber, a filter element discharge mechanism for discharging the used filter element from the filter chamber, and a filter element cleaning mechanism for cleaning the filter element.

[0089] The warning notification component 50 represents the structure for controlling the warning notification related to the deterioration state and abnormal state of the filter element in the filtration system 1. The warning notification component 50 is connected to the control panel 60 to display the warning notification on the display. In addition, the warning notification component 50 can also be a structure equipped with an alarm, a lamp, and outputting the warning notification as voice, alarm sound, lighting, or flashing of the lamp.

[0090] In this embodiment, the filtration system 1 can also be configured such that at least one component selected from the filter component 20, the cleaning component 30, the replacement component 40, and the warning notification component 50 is automated. For example, the filtration system 1 can also adopt a structure including a processing component that includes the filter component 20 and the cleaning component 30, but does not include the replacement component 40.

[0091] The control panel 60 is configured as a touch panel or the like that functions as a display unit and an input unit. It should be noted that the control panel 60 can also be composed of a display for displaying the state of the filtration system 1 and an operation panel as an input unit for operating the filtration system 1. The control panel 60 can also be configured to be externally connected to a computer terminal of the filtration system 1. The input unit of the control panel 60 receives the input of operation information and sends the operation information to the control unit 11, and the control unit 11 controls each component based on the received operation information.

[0092] Here, the second photographing unit 12B and the third photographing unit 12C are respectively arranged in the filtrate discharge path and the cleaning liquid discharge path. It should be noted that in the case where the filtrate discharge path of the filtration component 20 also serves as the cleaning liquid discharge path of the cleaning component 30, the second photographing unit 12B and the third photographing unit 12C may also be configured such that at least any one of them is arranged in the filtrate discharge path. That is, the second photographing unit 12B (the third photographing unit 12C) monitors the two fluids of the filtrate and the cleaning liquid discharge, and acquires their image data.

[0093] Figure 4 Schematic diagram showing the photographing unit 12. Figure 4 (a) thereof shows the front view of the photographing unit 12. Figure 4 (b) thereof shows the top view of the photographing unit 12.

[0094] As Figure 4 As shown in (a) of, the photographing unit 12 includes a photographing device 121 constituted by a camera or the like, an observation window 122, and a window 123 provided on the observation window 122. The photographing device 121 is arranged in the direction of the window 123 (dashed arrow), and acquires the image data of the fluid F flowing inside the path. It should be noted that the photographing device 121 is set to be connected to the control unit 11. In Figure 4 In (a) of, although an example is shown in which the fluid F passes through the horizontal pipe inside the left and right pipes provided on the observation window 122 from the left direction to the right direction, the position and direction of the pipe provided on the observation window 122 are not limited thereto.

[0095] The photographing unit 12 is arranged, for example, with a field of view angle that is substantially horizontal with respect to the ground plane. Substantially horizontal means a field of view angle of -20 degrees to +20 degrees with respect to the ground plane, preferably a field of view angle of -10 degrees to +10 degrees, and more preferably a field of view angle in the range of -5 degrees to +5 degrees. In the present embodiment, preferably, the fluid F is a liquid, and the photographing unit 12 is arranged such that image data including characteristic quantities such as the liquid level and the flow rate can be acquired.

[0096] In addition, preferably, the photographing unit 12 further includes a reference light source L. The light source L is provided for the purpose of keeping the light quantity in the surrounding environment of the photographing unit 12 constant. Since the light quantity in the surrounding environment affects the appearance of the color of the fluid photographed by the photographing device 121, the accuracy of determination based on color can be improved by keeping the light quantity constant during photographing. The photographing unit 12 may also be provided with a cover or the like for blocking external light against the photographing device 121 and the observation window 122. As a configuration example of the light source L, as Figure 4 As shown in (b) of, the observation window 122 has a window 123B on the back surface, and the reference light source L can be set as a backlight so as to cover the window 123B. It should be noted that the light source L may also be set to irradiate the front surface window 123A at a specified angle.

[0097] Figure 5 Schematic diagram of the photographing unit 12 in the case of using vertical piping. In Figure 5 , the fluid F passes through the piping above and below the observation window 122 from the upper direction to the lower direction. In particular, by using vertical piping, the photographing units 12B and 12C arranged in the discharge path starting from the filtration chamber R (not shown) can acquire image data indicating the tendency of increase or decrease in the liquid volume of the fluid F. For example, as shown in Figure 5 (a), at the initial stage when the fluid F is discharged from the filtration chamber R, the pressure in the filtration chamber is low, so the liquid volume of the fluid F relatively shows an increasing tendency.

[0098] On the other hand, as shown in Figure 5 (b), as the discharge progresses, the pressure in the filtration chamber becomes high, so the liquid volume of the fluid F relatively shows a decreasing tendency. The photographing units 12B and 12C are preferably arranged within a range of 10 m from the filtration chamber R, and more preferably within a range of 5 m from the filtration chamber R. In addition, the photographing units 12B and 12C are preferably arranged directly below the filtration chamber R.

[0099] Figure 6 Examples of the image data of the fluid F acquired by the photographing unit 12 are shown. Figure 6 (a) to Figure 6 (c) represent the image data of the transparent fluid F. Figure 6 (d), Figure 6 (e) represent the image data of the colored fluid F. The determination unit 13 can extract the feature amounts included in the image data to determine the state of the processing component 10.

[0100] Figure 6 (a) to Figure 6 (c) of the image data represent the states with different amounts of the transparent fluid F. Figure 6 (d), Figure 6 (e) of the image data represent the states with different amounts of the colored fluid F. The determination unit 13 can determine the state of the processing component 10 based on the feature amounts of these different states of the fluid F.

[0101] Regarding Figure 6 (a) to Figure 6 (e) of the image data (designated as image A to image E) input to the determination of the determination model, the test results are "image A: transparent empty", "image B: transparent medium", "image C: transparent supply", "image D: color empty", "image E: color supply". From these results, it can be understood that the determination unit 13 using the determination model can extract the feature amounts of the amount and color of the fluid F. It should be noted that the determination model can further determine the amount of the fluid F in stages and further determine the lightness, brightness, concentration, and color value associated with the color of the fluid F in stages.

[0102] Note that, regarding the image data, an example of the image data showing Figure 6 the fluid F passing through the horizontal pipe is shown, but it can also be, as Figure 5 shown, the image data of the fluid F passing through the vertical pipe. In the present embodiment, the image data only needs to be able to extract the characteristic quantities including at least the amount and color of the fluid F, and there is no limitation on the position and direction of the pipe.

[0103] The determination unit 13 can determine the state of the processing component 10 of the filtration system 1 based on the characteristic quantities of the image data of the fluid F as described above.

[0104] In the present embodiment, the state of the processing component 10 includes the processing state of the filtration component 20, the processing state of the cleaning component 30, the deterioration state of the filter element, and the abnormal state of the filter element.

[0105] The processing state of the filtration component 20 includes the processing state of filtration and the processing state of liquid separation. The processing state of filtration includes before filtration, during filtration, filtration completed, insufficient slurry liquid, excessive slurry liquid, insufficient filter aid liquid, excessive filter aid liquid, etc. Note that the processing state of filtration can also include the insufficient amount and excessive amount of the slurry liquid and the filter aid liquid. The processing state of liquid separation includes before liquid separation, during liquid separation, liquid separation completed, etc.

[0106] In addition, the processing state of the cleaning component 30 includes, according to the cleaning object, the processing state of cake cleaning, the processing state of device cleaning, the processing state of filter element cleaning, and the processing state of liquid separation, and determines the processing state according to the cleaning object. The processing state of the cleaning component 30 includes before cleaning, during cleaning, cleaning completed, insufficient cleaning liquid, excessive cleaning liquid, etc. The processing state of the cleaning component 30 can also include the insufficient amount and excessive amount of the cleaning liquid.

[0107] The processing state of liquid separation represents the liquid separation state of the filtrate in the filter cake in the processing state of the filtration component 20, and represents the liquid separation state of the cleaning liquid in the filter cake in the processing state of the cleaning component 30.

[0108] As a specific example, when the fluid F in the image data is the image data of the filtrate, a stable filtrate flow is observed when the filtration state is stable, and a decrease in the filtrate is observed due to the accumulation of the filter cake, so the determination unit 13 can determine the processing state related to the progress of filtration. In addition, due to the deterioration (such as fouling) of the filter element, the filtrate volume decreases, so the determination unit 13 can determine the deterioration state of the filter element. In addition, when the fluid F in the image data is the image data of the cleaning drain liquid, as the cleaning progresses, the transparency of the color of the cleaning drain liquid increases, so the determination unit 13 can determine the processing state related to the progress of cleaning.

[0109] In the present embodiment, the determination unit 13 determines the state of the processing component 10 using a determination model stored in the storage unit DB. The determination model represents a learned completion model obtained by performing machine learning on a data set. The determination model is generated by an external model generation device 7 and stored in the storage unit DB.

[0110] Figure 7 The block diagram of the model generation device 7 that performs the generation process of the determination model is shown. The model generation device 7 includes a data set acquisition unit 71, a model generation unit 72, a data set storage unit 73, and a model storage unit 74. Preferably, the data set acquisition unit 71 is configured to acquire the image data obtained by the Figure 3 illustrated imaging unit 12 as the data set.

[0111] The model generation device 7 can use a general-purpose computer. As hardware components, the model generation device 7 includes an arithmetic device based on a CPU or the like, a main storage device based on a RAM (Random Access Memory), an auxiliary storage device, a communication device, an input / output device, and the like.

[0112] The filtering system 1 and the model generation device 7 are configured to be capable of performing data communication based on wire or wireless. In addition, it may be configured as a filtering system including the filtering system 1 and the model generation device 7. In addition, the filtering system 1 may also adopt a configuration having the functional components (71 - 74) of the model generation device 7.

[0113] In the present embodiment, the algorithm of machine learning may adopt a neural network. As Figure 8 shown, 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 the input of the input data of the data set. The input layer N1 is composed of a plurality of neurons corresponding to the input data, and outputs the calculation result for the input data to the intermediate layer N2. The intermediate layer N2 is composed of one or more layers, and each layer has a plurality of neurons. The intermediate layer N2 receives the input of the calculation result from the input layer N1, and further outputs the calculation result for the input to the adjacent layer or the output layer N3 within the intermediate layer N2. The output layer N3 outputs an estimated value in response to the input from the intermediate layer N2. By adjusting the coefficients of each neuron in such a way that the error between the estimated value of the output layer N3 and the output data of the data set becomes smaller, the determination accuracy of the output data with respect to the input data can be improved.

[0114] It should be noted that the algorithm of machine learning 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.

[0115] Figure 9Shows a configuration example of a data set for machine learning of a determination model.

[0116] In the present embodiment, as Figure 9 (a) of shows, the data set is configured such that image data is used as input data and the state of the processing component 10 is used as output data.

[0117] In addition, as Figure 9 (b) of shows, the data set can also be configured such that image data is used as input data and the state of the fluid is used as output data. The state of the fluid includes states such as the amount and color of the fluid F.

[0118] In addition, as Figure 9 (c) of shows, the data set can also be configured such that image data is used as input data and the control signal for the processing component is used as output data. The control signal is a signal output by the output unit 14 to the processing component 10, and represents a signal corresponding to the control of each of the filtering component 20, the cleaning component 30, the replacement component 40, and the warning notification component 50.

[0119] In the present embodiment, the data set can be configured such that a combination of a plurality of image data in time series is used as input data, and at least one selected from the state of the processing component 10, the state of the fluid, and the control signal for the processing component is used as output data. A determination model that has performed machine learning using a combination of image data in time series as input data can output output data corresponding to the change in the image data by receiving the input of the image data in the order of the combination.

[0120] In addition, at this time, the output data can also be configured as at least one data selected from the state of the processing component 10, the state of the fluid, and the control signal for the processing component after a predetermined time corresponding to the time series (time difference) of the plurality of image data. For example, when the image data every 5 seconds is used as input data, the data set further uses the state of the processing component 10 after 5 seconds as output data. Thus, the determination model can output the predicted value of the state of the processing component 10 as the determination result. Even when there is a time lag from the acquisition of the image data to the control of the processing component 10, the determination unit 13 can control the processing component 10 on the basis of correcting the time lag by using the determination model as described above.

[0121] Figure 10 Represents a flowchart related to the generation process of the determination model in the model generation device 7.

[0122] Step S11: The dataset acquisition unit 71 acquires a dataset and stores it in the dataset storage unit 73. Preferably, the dataset acquisition unit 71 sets the image data acquired by the imaging unit 12 as the input data of the dataset. The dataset acquisition unit 71 accepts input from the operator regarding the output data corresponding to the image data set as the input data. The dataset storage unit 73 can be configured to pre-store list data of the states of the processing component 10, the states of the fluid, and control signals that can be selected as the output data of the dataset. The dataset acquisition unit 71 can be configured to determine the dataset by accepting the selection of the output data corresponding to the image data from the list data.

[0123] Step S12: The model generation unit 72 performs machine learning processing of the model using the dataset stored in the dataset storage unit 73. The number of datasets for machine learning processing is not particularly limited.

[0124] Step S13: The model generation unit 72 completes the machine learning processing based on Step S12, thereby generating a determination model, storing it in the model storage unit 74, and ending the processing.

[0125] The determination model stored in the model storage unit 74 is stored in the storage unit DB of the filtration system 1. Therefore, the determination unit 13 can use the determination model to determine the state of the processing component 10 based on the image data.

[0126] Multiple determination models can be generated according to the types of fluids such as the slurry liquid, filter aid liquid, and cleaning liquid used in the filtration system 1, and the types of filtration devices. The control panel 60 can determine the determination model to be used by the determination unit 13 from the multiple determination models stored in the storage unit DB by accepting an instruction input related to the types of fluids and filtration devices.

[0127] The output unit 14 outputs a control signal for the processing component 10 based on the determination result of the state of the processing component 10 obtained by the determination unit 13. The output unit 14 outputs a control signal based on a control signal table indicating the correspondence between the determination result representing the state of the processing component 10 and the control signal for the processing component 10.

[0128] Figure 11 (a) shows an example of the data configuration of the control signal table. The control signal table has a control mode, a determination result, and a control signal. Figure 11 The control signal shown in (a) shows the determination result obtained by the determination model that has undergone machine learning using the dataset in Figure 9 (a) of and the corresponding control signal. The control signal table can also show the determination result related to the state of the fluid obtained by the determination model that has undergone machine learning using the dataset in Figure 9 (b) of and the corresponding control signal. It should be noted that when using the one obtained by usingFigure 9 In the case of a determination model of machine learning for the data set of (c), the control signal table can be omitted, and the output unit 14 outputs a control signal as the determination result obtained by the determination unit 13 to the processing component 10.

[0129] The control mode indicates which one of the filtration component 20, the cleaning component 30, and the replacement component 40 the filtration system 1 is processing. The control mode has a filtration mode related to the control of the filtration component 20, a filter cake cleaning mode related to the control of the filter cake cleaning in the cleaning component 30, a replacement mode related to the control of the replacement component 40, and a device cleaning mode related to the control of the device cleaning in the cleaning component 30.

[0130] The filtration mode can be further classified into a filtration mode and a filtrate dewatering mode. The filter cake cleaning mode can also be further classified into a filter cake cleaning mode and a cleaning liquid dewatering mode. The replacement mode can also be further classified into a filter element replacement mode and a filter element cleaning mode.

[0131] The control signal represents a signal for controlling the processing component 10 corresponding to the control mode. Multiple control signals can be set for one determination result. For example, the control signals related to the operation of the filtration component 20 corresponding to the determination result of filtration completion may include a control signal for controlling the pressure inside the filtration chamber generated by the filtration chamber pressure control unit and a control signal for controlling the opening and closing times of the supply valves of the slurry liquid and the filter aid liquid.

[0132] The control signal of the filtration mode corresponds to the determination result of the processing state of the filtration component 20, and has control signals related to the operation / stop of the filtration component 20, the increase / decrease of the slurry liquid supply amount, the increase / decrease of the filter aid liquid supply amount, the operation / stop of the filter cake dewatering process, etc. When the control signal output from the output unit 14 is obtained, the filtration component 20 controls the supply valves of the filtration chamber pressure control unit, the slurry liquid supply tank, and the filter aid liquid supply tank, etc.

[0133] Specifically, the filtration chamber pressure control unit controls the operation / stop of the filtration component 20 and the operation / stop of the filter cake dewatering process by pressurizing or depressurizing the pressure inside the filtration chamber on at least either the supply path side or the discharge path side. In addition, the supply valves of the slurry liquid supply tank or the filter aid liquid supply tank control the supply amounts of the slurry liquid or the filter aid liquid by changing the opening and closing times of the supply valves.

[0134] The control signal of the filter cake cleaning mode has signals for controlling the operation, stop, increase / decrease of the cleaning liquid supply amount, operation / stop of the filter cake dewatering process, etc. of the cleaning component 30 corresponding to the processing state of the cleaning component 30. When the control signal output from the output unit 14 is obtained, the cleaning component 30 controls the supply valves of the filtration chamber pressure control unit and the cleaning liquid supply tank, etc.

[0135] Specifically, the filtration chamber pressure control unit controls the operation / stop of the cleaning assembly 30 and the operation / stop of the filter cake dewatering process by pressurizing or depressurizing the internal pressure of the filtration chamber on at least one of the supply path side and the discharge path side. In addition, 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.

[0136] The control signal for the replacement mode has a signal for controlling the operation and stop of the replacement assembly 40 corresponding to the deterioration state of the filter element. When the control signal output from the output unit 14 is obtained, the replacement assembly 40 controls the filter element cleaning mechanism, the filter element supply mechanism, and the filter element discharge mechanism. The control signal for the replacement mode can be set as a control signal for different mechanisms according to the degree of the deterioration state of the filter element. It should be noted that the deterioration state of the filter element can also be determined in the filtration mode or the filter cake cleaning mode, and the replacement mode can be switched according to the determination result.

[0137] The control signal for the device cleaning mode has a signal for controlling the continued operation, stop, increase / decrease of the cleaning liquid supply amount, etc. of the cleaning assembly 30 corresponding to the processing state of the cleaning assembly 30. The control signal for the device cleaning in the cleaning assembly 30 that starts the device cleaning mode is output from the output unit 14 by receiving a device cleaning instruction operation from the control panel 60. When the control signal output from the output unit 14 is obtained, the cleaning assembly 30 controls the filtration chamber pressure control unit, the supply valve of the cleaning liquid supply tank, etc. The device cleaning instruction operation is performed by the operator when replacing an object such as the slurry liquid filtered by the filtration system 1.

[0138] Based on the determination result of the state of the processing assembly 10 obtained by the determination unit 13, the output unit 14 outputs a warning signal to the warning notification assembly 50. Figure 11 (b) of shows an example of the data structure of the warning signal table. The warning signal table has a control mode, a determination result, and a warning signal. The control mode of the warning signal table is one of the filtration mode and the filter cake cleaning mode, and warning signals corresponding to each control mode are set.

[0139] The warning signal has a signal for controlling the warning notification of the warning notification assembly 50 corresponding to the deterioration state or abnormal state of the filter element. When the warning signal output from the output unit 14 is obtained, the warning notification assembly 50 controls the display of the control panel 60, the voice or alarm sound of the alarm, the lighting of the lamp, etc.

[0140] Specifically, when a warning signal of a filter component abnormality is obtained, the control panel 60 displays a notification related to the abnormality. In addition, when a warning signal of filter component deterioration is obtained, the control panel 60 displays a notification related to the deterioration. The alarm and the lamp can also give warnings related to the abnormality and deterioration through voices and lighting patterns corresponding to the warning signals.

[0141] The output unit 14 can output a control signal for stopping the filter assembly 20 or the cleaning assembly 30 corresponding to the determination result of the abnormal state of the filter component. Thereby, the emergency stop action for abnormalities such as breakage of the filter component can be automated.

[0142] The first control mode switching unit 15 switches the control mode shown by the processing component 10 that is the control object in the filtration system 1. The first control mode switching unit 15 determines which control mode to switch to, the filtration mode, the filter cake cleaning mode, or the replacement mode, based on the determination result of the state of the processing component 10 in the determination unit 13 or the output of the control signal in the output unit 14. The first control mode switching unit 15 usually sequentially switches the filtration mode, the filter cake cleaning mode, and the replacement mode. However, for example, when the determination result of the deterioration state or abnormal state of the filter component is obtained during filtration, it can also switch from the filtration mode to the replacement mode.

[0143] The first control mode switching unit 15 switches to the next control mode according to the determination result of the states of filtration completion, cleaning completion, and filter component replacement completion. In addition, the first control mode switching unit 15 switches to the next control mode based on the output of a specified control signal such as the operation stop of the filter assembly 20, the operation stop of the cleaning assembly 30, and the operation stop of the replacement assembly 40.

[0144] The first control mode switching unit 15 can also switch the control mode to the warning notification mode based on the determination result of the abnormal state of the filter component. The first control mode switching unit 15 switches the warning notification mode to the original control mode by eliminating the abnormal state.

[0145] The first control mode switching unit 15 switches the control mode to the device cleaning mode by receiving a device cleaning instruction operation from the control panel 60. The first control mode switching unit 15 switches to the original control mode based on the determination result of cleaning completion or the output of the operation stop control signal.

[0146] Figure 12 A flowchart related to a series of controls performed by the filtration system 1 is shown. The filtration system 1 receives the input of basic settings such as the type, total amount, or operation time of the slurry liquid to be filtered and the type of filter component used via the control panel 60 and starts processing.

[0147] <Filtration step>

[0148] Step S21: The filtration component 20 performs a filtration process by controlling the filtration 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.

[0149] Step S22: The imaging unit 12 acquires image data of the slurry liquid and the filtrate respectively. The determination unit 13 inputs the image data into a determination model to determine the state of the processing component 10 including the processing state of the filtration component 20, the deterioration state of the filter element, and the abnormal state of the filter element. The output unit 14 outputs a control signal for controlling their supply amounts to the filtration component 20 based on the determination result related to the supply amounts of the slurry liquid and the filtrate, and continues the filtration process.

[0150] Step S23: When the determination unit 13 determines that the filtration is completed as the filtration processing state (in S23, it is "yes"), the output unit 14 outputs a control signal for stopping the operation related to the filtration process to the filtration component 20. The first control mode switching unit 15 switches the control mode to the filtrate dewatering mode. When the determination unit 13 determines that the filtration is in progress as the filtration processing state (in S23, it is "no"), the first determination process in step S22 is repeatedly executed until it is determined that the filtration is completed.

[0151] Step S24: When the determination unit 13 determines that the dewatering is completed as the dewatering processing state (in S24, it is "yes"), the output unit 14 outputs a control signal for stopping the operation related to the dewatering process to the filtration component 20. The first control mode switching unit 15 switches the control mode to the filter cake cleaning mode. When the determination unit 13 determines that the dewatering is in progress as the dewatering processing state (in S24, it is "no"), the first determination process in step S22 is repeatedly executed until it is determined that the dewatering is completed.

[0152] <Filter cake cleaning step>

[0153] Step S25: The cleaning component 30 performs a cleaning process on the filter cake by controlling the filtration chamber pressure control unit and the cleaning liquid supply valve.

[0154] Step S26: The imaging unit 12 acquires image data of the cleaning drain liquid. The determination unit 13 inputs the image data into a determination model to determine the state of the processing component 10 including the processing state of the cleaning component 30, the deterioration state of the filter element, and the abnormal state of the filter element. The output unit 14 outputs a control signal for controlling the supply amount to the cleaning component 30 based on the determination result related to the cleaning liquid supply amount, and continues the cleaning process.

[0155] Step S27: When the determination unit 13 determines that the cleaning is completed as the processing state of the cleaning component 30 (Yes in S27), the output unit 14 outputs a control signal for stopping the operation related to the cleaning process to the cleaning component 30. The first control mode switching unit 15 switches the control mode to the cleaning liquid draining mode. When the determination unit 13 determines that the cleaning is in progress as the processing state of the cleaning component 30 (No in S27), the second determination process in step S26 is repeatedly executed until it is determined that the cleaning is completed.

[0156] Step S28: When the determination unit 13 determines that the draining is completed as the processing state of the draining (Yes in S28), the output unit 14 outputs a control signal for stopping the operation related to the draining process to the cleaning component 30. The first control mode switching unit 15 switches the control mode to the replacement mode. When the determination unit 13 determines that the draining is in progress as the processing state of the draining (No in S28), the second determination process in step S26 is repeatedly executed until it is determined that the draining is completed.

[0157] <Filter element replacement step>

[0158] Step S29: In the filtering step or the filter cake cleaning step, the determination unit 13 determines whether there is deterioration of the filter element as the deterioration state of the filter element. When it is determined that there is no deterioration of the filter element (No in S29), the determination unit 13 sets that there is no need to replace the filter element, and the filter element replacement step is completed.

[0159] Step S30: When the determination unit 13 determines that there is deterioration of the filter element (Yes in S29), the output unit 14 outputs a control signal related to the operation of the replacement component 40 to the replacement component 40 to replace the filter element or clean the filter element. Whether to replace the filter element or clean the filter element is determined by the degree of the deterioration state of the filter element. The first control mode switching unit 15 switches the control mode to the filtering mode, and the process is completed. It should be noted that the determination unit 13 can also be configured to immediately execute step S30 when it is determined that there is deterioration of the filter element in the filtering step or the filter cake cleaning step.

[0160] Step S31: After the filter element replacement step, the filtration system 1 executes a filter cake discharging step for discharging the filter cake accumulated on the filter element. It should be noted that in the case of adopting a configuration in which the filter cake is discharged together with the discharge of the filter element in the filter element replacement step, the filter cake discharging step can also be omitted.

[0161] Step S32: When the filter cake discharge process is completed, the first control mode switching unit 15 switches the control mode to the filtration mode and returns to step S21 again, and the process can be repeatedly executed from the filtration step (Yes in S32). The filtration system 1 completes the process when an instruction input indicating the completion of filtration is received or when a prescribed filtration process is completed (No in S32).

[0162] As described above, according to the present disclosure, a series of processes for filtering a slurry liquid implemented by a filtering device can be automatically controlled.

[0163] The filtration system 1 can reduce the workload by automatically controlling a series of processes. In addition, safety risks caused by poor cleaning during the treatment of chemicals can be reduced. In addition, the product quality can be improved by preventing poor liquid separation, maintaining the clarity of the filtrate, and preventing poor filter cake cleaning. In addition, energy losses caused by preventing excessive liquid separation operation can be reduced. In addition, by preventing excessive filter cake cleaning and excessive device cleaning, resource usage can be reduced.

[0164] <Embodiment 2>

[0165] Hereinafter, a different embodiment 2 of the filtration system 1 will be described. It should be noted that the same components as those in Embodiment 1 are denoted by the same reference numerals and the description thereof is omitted.

[0166] Figure 13 It is a block diagram showing the filtration system 1 of Embodiment 2. In the second embodiment, the filtration system 1 includes a determination device 100 and a filtration device 200.

[0167] The determination device 100 includes a first control unit 110, a photographing unit 12, a determination unit 13, an output unit 14, and a storage unit DB as a database. Each component is connected to the first control unit 110 and is controlled. In the present embodiment, the output unit 14 is connected to the filtration device 200, and controls the operations in the filtration device 200 by outputting a control signal.

[0168] The photographing unit 12 is arranged on the path of the fluid in the filtration device 200 and acquires image data of the fluid.

[0169] The determination unit 13 determines the state of the processing component 10 in the filtration device 200 based on the image data acquired by the photographing unit 12.

[0170] In Embodiment 2, the output unit 14 outputs a control signal for the filtration device 200 based on the determination result of the state of the processing component 10 obtained by the determination unit 13. The processing component 10 is controlled based on the control signal output by the output unit 14.

[0171] The determination device 100 can use a general-purpose computer. As a hardware component, the determination device 100 includes an arithmetic device based on a CPU or the like, a main storage device based on a RAM or the like, an auxiliary storage device, a communication device, an input / output device, and the like.

[0172] The filtration device 200 includes a control panel 60 and a processing component 10. In Embodiment 2, the filtration device 200 can use an existing filtration device. In addition, the control panel 60 can use a PLC (Programmable Logic Controller) or the like.

[0173] 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 the input of the 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 by the signal input unit 61. The signal output unit 63 is connected to the processing component 10, obtains the control signal as the processing result obtained by the second control unit 62, and outputs it to the processing component 10, thereby controlling the operation of the processing component 10.

[0174] The signal processing unit 64 obtains the control signal of the output unit 14 and performs signal processing. The signal processing includes processing for converting the control signal into a signal output to the processing component 10 and processing for determining the output target.

[0175] The second control mode switching unit 65 switches the control mode shown by the processing component 10 that is the control object in the filtration device 200. The second control mode switching unit 65 switches the control mode based on the determination result of the state of the processing component 10 obtained by the determination unit 13 or the output of the control signal by the signal output unit 63. The second control mode switching unit 65 can also output the switched control mode to the determination device 100.

[0176] The processing component 10 is the same as in Embodiment 1, and includes a filtration component 20 that controls the filtration step, a cleaning component 30 that controls the cleaning step, a replacement component 40 that controls the filtration element replacement step, and a warning notification component 50 that controls the notification of the deterioration state or abnormal state of the filtration element. The filtration component 20, the cleaning component 30, the replacement component 40, and the warning notification component 50 control their operations based on the control signal output from the signal output unit 63.

[0177] The filtration system 1 only needs to achieve the same effect as a whole, and is not limited to the above-described embodiments, and various configurations can be adopted.

[0178] Industrial Applicability

[0179] The present invention can be used in various filtration devices such as pressure filtration devices, vacuum filtration devices, centrifugal filtration devices, and gravity filtration devices.

[0180] Description of Reference Numerals

[0181] 1: Filtration system;

[0182] 7: Model generation device;

[0183] 11: Control unit;

[0184] 12: Photographing unit;

[0185] 13: Judgment unit;

[0186] 14: Output unit;

[0187] 15: First control mode switching unit;

[0188] 20: Filtration component;

[0189] 30: Cleaning component;

[0190] 40: Replacement component;

[0191] 50: Warning notification component;

[0192] 60: Control panel;

[0193] DB: Storage unit;

[0194] 100: Judgment device;

[0195] 110: First control unit;

[0196] 200: Filtration device;

[0197] 61: Signal input unit;

[0198] 62: Second control unit;

[0199] 63: Signal output unit;

[0200] 64: Signal processing unit;

[0201] 65: Second control mode switching unit;

[0202] F: Fluid;

[0203] F1: Slurry liquid;

[0204] F2: Filtrate;

[0205] F3: Filter aid liquid;

[0206] F4: Filtrate including filter aid liquid;

[0207] F5: Mixed liquid;

[0208] F6: Cleaning liquid;

[0209] F7: Drainage of cleaning liquid.

Claims

1. A filtering system that can automatically control a processing component for filtering a slurry liquid, wherein, the processing component includes a filtering component, the filtering system comprises: a photographing unit, disposed on a path of a fluid within the filtering component, for acquiring image data of the fluid; a determination unit, for determining the state of the processing component based on the image data; and an output unit, for outputting a control signal for the processing component based on a determination result of the state of the processing component, and, the state of the processing component includes a processing state of the filtering component, a deterioration state of the filter element, or an abnormal state of the filter element, the determination unit uses a determination model to determine the processing state of the filtering component, the deterioration state of the filter element, or the abnormal state of the filter element, wherein the determination model has performed machine learning with the image data as input data and the processing state of the filtering component, the deterioration state of the filter element, or the abnormal state of the filter element as output data.

2. A filtering system that can automatically control a processing component for filtering a slurry liquid, wherein, the processing component includes a cleaning component, the filtering system comprises: a photographing unit, disposed on a path of a fluid within the cleaning component, for acquiring image data of the fluid; a determination unit, for determining the state of the processing component based on the image data; and an output unit, for outputting a control signal for the processing component based on a determination result of the state of the processing component, and, the state of the processing component includes a processing state of the cleaning component, a deterioration state of the filter element, or an abnormal state of the filter element, the determination unit uses a determination model to determine the processing state of the cleaning component, the deterioration state of the filter element, or the abnormal state of the filter element, wherein the determination model has performed machine learning with the image data as input data and the processing state of the cleaning component, the deterioration state of the filter element, or the abnormal state of the filter element as output data.

3. The filtering system according to claim 1, wherein, the processing component includes a filtering component, the fluid is filtrate, the determination unit determines the processing state of the filtering component, the output unit outputs a control signal for the filtering component based on a determination result of the processing state of the filtering component.

4. The filtering system according to claim 3, wherein, the control signal for the filtering component includes at least any one or more of a control signal related to the operation of the filtering component, a control signal related to the supply amount of the slurry liquid, a control signal related to the dewatering process, and a control signal related to the supply amount of the filter aid liquid.

5. The filtering system according to claim 3, wherein, the control signal for the filtering component is a control signal for pressurizing or depressurizing the internal pressure of the filter chamber on at least either the supply path side or the discharge path side.

6. The filtering system according to claim 3, wherein, the control signal for the filtering component is a control signal for changing the opening and closing time of the supply valve of the slurry liquid or the filter aid liquid.

7. The filtering system according to claim 2, wherein, the processing component includes a cleaning component, the fluid is cleaning drain liquid, The determination unit determines the processing state of the cleaning component. The output unit outputs a control signal for the cleaning component based on the determination result of the processing state of the cleaning component.

8. The filtration system according to claim 1, wherein, the processing component includes a filtration component, a cleaning component, a replacement component, or a warning notification component, the fluid is filtrate, the determination unit determines the deterioration state of the filter element, the output unit outputs at least any one or more of a control signal for the cleaning component, a control signal for the replacement component, and a warning signal for the warning notification component based on the determination result of the deterioration state of the filter element.

9. The filtration system according to claim 2, wherein, the processing component includes a cleaning component, a replacement component, or a warning notification component, the fluid is cleaning drainage, the determination unit determines the deterioration state of the filter element, the output unit outputs at least any one or more of a control signal for the cleaning component, a control signal for the replacement component, and a warning signal for the warning notification component based on the determination result of the deterioration state of the filter element.

10. The filtration system according to claim 1 or 2, wherein, the processing component includes a filtration component, a cleaning component, or a warning notification component, the fluid is filtrate or cleaning drainage, the determination unit determines the abnormal state of the filter element, the output unit outputs a control signal related to operation for the filtration component or a warning signal related to the abnormal state of the filter element for the warning notification component based on the determination result of the abnormal state of the filter element.

11. The filtration system according to claim 1, wherein, the processing component includes a filtration component, the fluid is the slurry liquid, the imaging unit is disposed on the supply path of the slurry liquid to acquire image data of the slurry liquid, the output unit outputs a control signal related to the supply amount of the filter aid liquid for the filtration component based on the determination result of the processing state of the filtration component.

12. The filtration system according to claim 1 or 2, wherein, the imaging unit is disposed on the path of the fluid at a field angle that is substantially horizontal with respect to the ground plane.

13. The filtration system according to claim 1 or 2, wherein, the filtration system further includes a control mode switching unit for switching the control target processing component, the processing component includes a filtration component, a cleaning component, and a replacement component, the determination unit determines the state of the processing component including the processing state of the filtration component, the processing state of the cleaning component, and the deterioration state of the filter element, the output unit outputs control signals for the filtration component, the cleaning component, and the replacement component based on the determination result of the state of the processing component, the control mode switching unit switches the control mode based on the determination result of the state of the processing component or the output of the control signal.

14. The filtration system according to claim 13, wherein, the processing component includes a warning notification component, the determination unit further determines the state of the processing component including the abnormal state of the filter element. The output unit outputs a warning signal corresponding to the control mode to the warning notification component based on the determination result of the deterioration state or the abnormal state of the filter element.

15. The filtration system according to claim 13, wherein, the filtrate discharge path of the filtration component also serves as the cleaning liquid discharge path of the cleaning component, the photographing unit is disposed on the filtrate discharge path.

Images