Water treatment plant operation management support system and operation management support method
The system uses vibration monitoring and analysis to detect aeration device damage in water treatment plants, addressing visibility issues and preventing system downtime by identifying abnormalities in real-time.
Patent Information
- Application Number
- JP2021201449
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2021-12-13
- Publication Date
- 2025-11-26
- Estimated Expiration
- 2041-12-13
Smart Images

Figure 0007775686000001 
Figure 0007775686000002 
Figure 0007775686000003
Abstract
Description
[Technical Field]
[0001] The present invention relates to an operation management support system and an operation management support method for a water treatment plant. [Background technology]
[0002] Conventionally, sewage or wastewater treatment facilities (hereinafter referred to as "sewage or wastewater treatment plants" or simply "water treatment plants") have widely used the activated sludge method, which purifies water using aerobic microorganisms in a treatment tank (aeration tank). In water treatment plants using the activated sludge method, an aeration device is generally used to dissolve air and oxygen into the water, which are necessary for the respiration of aerobic microorganisms that decompose organic matter (see, for example, Patent Document 1). [Prior art documents] [Patent documents]
[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2011-230068 Summary of the Invention [Problem to be solved by the invention]
[0004] In water treatment plants, air and oxygen are supplied to an aeration device by a blower or the like, and the air and oxygen released from the aeration device into the water turns into bubbles and diffuses into the water. The smaller the diameter of the bubbles diffused into the water, the larger the specific surface area of the bubbles, allowing more oxygen to dissolve in the water. Furthermore, by dispersing the diffused bubbles evenly within the aeration tank, the residence time within the tank increases, allowing more oxygen to dissolve in the water.
[0005] However, if the diffuser is damaged due to deterioration or external forces, a large amount of large-diameter bubbles (coarse bubbles) may be generated from a part of the diffuser. When coarse bubbles are generated, the bubble diameter increases, reducing the oxygen dissolution efficiency. In addition, since a large amount of bubbles are released into the tank from a single part, the bubble distribution within the tank becomes uneven, shortening the bubble residence time and significantly reducing the oxygen dissolution efficiency.
[0006] In addition, the operation (air supply) of an aeration system may be temporarily suspended for performance maintenance, inspection, etc. If the supply of air or oxygen is stopped when the aeration system is deteriorated or damaged, wastewater containing solids may flow back from the damaged area through the aeration system or air supply pipe, and may even reach a working aeration system. If wastewater repeatedly enters the aeration system, it may become clogged from the inside by solids and be unable to generate bubbles properly. In this case, the purification of the sewage or wastewater will be insufficient, and the aeration system that has been contaminated by wastewater will need to be replaced.
[0007] Furthermore, sewage or wastewater treatment plants (water treatment plants) use aeration tanks with numerous aeration devices, and routine inspections of aeration tanks typically involve visually checking for coarse bubbles. However, in order to suppress odors emanating from the sewage or wastewater and to prevent objects from falling into the tank, the top of the aeration tank is often covered with a concrete slab or cover, and the opening is often small. This makes it difficult to see the entire aeration water surface during routine inspections, and it is even difficult to visually check for abnormal bubbles at the water level during operation. For this reason, it is difficult to visually detect damage to the aeration devices during routine inspections, and damage to the aeration devices often goes unnoticed.
[0008] For this reason, only when there was concern about damage due to an increase in the air volume blown by the blower or a deterioration in the quality of the treated water, did wastewater have to be drawn out of the aeration tank to lower the water level to near the top of the aeration device, and work was carried out to visually check for abnormal foaming such as coarse bubbles, damage to the aeration device, etc. In such cases, the aeration device was operated for a long period of time while damaged, and since wastewater had often already infiltrated the other working aeration devices in the same section through the damaged area, some of the aeration devices were clogged with mud and deemed unusable, and the aeration device had to be replaced.
[0009] In addition, lowering the water level required stopping the wastewater treatment system in question, which placed a heavy burden on operational management.Furthermore, it was difficult for the general facility manager (user) to determine whether coarse bubbles were occurring (abnormal foaming) or whether the aeration device was damaged, and an engineer from the aeration device supplier had to go to the site to check.
[0010] Therefore, an object of the present disclosure is to provide a technology that allows even a general facility manager to detect an abnormality in an aeration device during operation of a water treatment plant. [Means for solving the problem]
[0011] According to one aspect, an operation management support system for a water treatment plant includes an air supply pipe through which air or oxygen is supplied from a blower, an aeration device connected to the air supply pipe and through which air or oxygen is supplied from the air supply pipe, a treatment water tank in which the aeration device is disposed in water to be treated and from which the air or oxygen is released into the water, a vibration information collecting device that collects vibration information emitted by bubbles formed in the water in the treatment water tank by the air or oxygen released from the aeration device, and an information processing device that performs predetermined information processing on the vibration information collected by the vibration information collecting device, wherein the information processing device includes a vibration information acquiring unit that acquires vibration information from the vibration information collecting device and performs predetermined conversion processing on the acquired vibration information, a judgment algorithm executing unit that executes processing based on a predetermined judgment algorithm on the vibration information that has been subjected to the predetermined conversion processing by the vibration information acquiring unit, a vibration information determining unit that determines whether there is an abnormality in the aeration device based on the output result output from the judgment algorithm executing unit, and a judgment result notifying unit that notifies the result determined by the vibration information determining unit. The predetermined transformation process is a short-time Fourier transform performed on the vibration information acquired by the vibration information acquisition unit, the vibration information being cut out at a predetermined time interval. It is characterized by:
[0012] In the water treatment plant operation management support system according to one aspect, the vibration information collecting device may be disposed in a space above the treated water in the treatment water tank.
[0013] In addition, in the water treatment plant operation management support system according to one aspect, the vibration information collecting device may be placed underwater in the treatment water tank.
[0014] In addition, in the water treatment plant operation management support system according to one aspect, the vibration information collecting device may be disposed in proximity to or in contact with the air supply pipe.
[0015] In addition, in one embodiment of the water treatment plant operation management support system, the vibration information is acoustic information of the sounds of bubbles rising underwater, breaking, flowing water, splashing waves, or a combination of these sounds emitted by the bubbles, the vibration information collection device is a microphone that collects the acoustic information, and the vibration information acquisition unit may perform a predetermined conversion process on the acoustic information collected by the microphone.
[0016] In addition, in one embodiment of an operation management support system for a water treatment plant, the vibration information is vibration information of vibrations caused by bubbles rising underwater, bubble breaking vibrations, water flow vibrations, wave splash vibrations, or combinations of these vibrations, the vibration information collection device is a vibration sensor that collects the vibration information, and the vibration information acquisition unit may perform a predetermined conversion process on the vibration information collected by the vibration sensor.
[0018] In addition, in one embodiment of the water treatment plant operation management support system, the judgment algorithm execution unit calculates the average, standard deviation, or coefficient of variation of the time change in vibration intensity for each frequency for the vibration information that has been subjected to a predetermined conversion process by the vibration information acquisition unit, and outputs the result of accumulating the standard deviation or coefficient of variation for a predetermined range of frequencies or for all recorded frequencies, and the vibration information judgment unit may judge that there is an abnormality in the air diffuser when the numerical value for the time change in vibration intensity output from the judgment algorithm execution unit is greater than a predetermined threshold value.
[0019] In addition, in one embodiment of an operation management support system for a water treatment plant, the vibration information acquisition unit converts vibration information that has undergone a predetermined conversion process into three-dimensional information of time, frequency, and vibration intensity, the judgment algorithm execution unit calculates and outputs the degree of deviation between the three-dimensional information converted by the vibration information acquisition unit and three-dimensional information based on vibration information under normal conditions that has been learned in advance, and the vibration information judgment unit may judge that there is an abnormality in the air diffuser when the numerical value for the time change in vibration intensity output from the judgment algorithm execution unit is greater than a predetermined threshold value.
[0020] In addition, in the water treatment plant operation management support system according to one aspect, the predetermined threshold value may be different for each season, each predetermined time period, each aeration tank, or each section of the aeration tank.
[0021] In addition, in one embodiment of an operation management support system for a water treatment plant, the vibration information acquisition unit converts vibration information that has undergone a predetermined conversion process into three-dimensional information of time, frequency, and vibration intensity, the judgment algorithm execution unit classifies the three-dimensional information converted by the vibration information acquisition unit based on three-dimensional information based on vibration information that has been learned in advance, and the vibration information judgment unit may judge that there is an abnormality in the air diffuser when the result of the classification of the time change in vibration intensity by the judgment algorithm execution unit is a classification that indicates an abnormality in the air diffuser.
[0022] In addition, the water treatment plant operation management support system according to one embodiment may further include a judgment algorithm generation unit that generates a predetermined judgment algorithm based on the vibration information that has been subjected to a predetermined conversion process by the vibration information acquisition unit.
[0023] In the water treatment plant operation management support system according to one aspect, the determination algorithm generation unit may update the predetermined determination algorithm for each predetermined period or each predetermined time period.
[0024] In addition, in the water treatment plant operation management support system according to one aspect, the determination result notification unit may notify the result determined by the vibration information determination unit to an outside of the information processing device via a display or an information processing terminal.
[0025] According to one aspect, a method for supporting operation management of a water treatment plant includes an air supply pipe through which air or oxygen is supplied from a blower, an aeration device connected to the air supply pipe and through which air or oxygen is supplied from the air supply pipe, a treatment water tank in which the aeration device is disposed in water to be treated and the air or oxygen is released into the water from the aeration device, and a vibration information collecting device that collects vibration information emitted by bubbles formed in the water in the treatment water tank by the air or oxygen released from the aeration device, the method comprising the steps of: a vibration information acquiring step of acquiring vibration information from the vibration information collecting device and performing a predetermined conversion process on the acquired vibration information; a determination algorithm executing step of executing a process based on a predetermined determination algorithm on the vibration information that has been subjected to the predetermined conversion process; a vibration information determining step of determining an abnormality in the aeration device based on an output result output by the determination algorithm executing step; and a determination result notifying step of notifying the result determined by the vibration information determining step. The predetermined transformation process is a short-time Fourier transform performed on the vibration information acquired by the vibration information acquisition step, which is cut out at a predetermined time interval. It is characterized by: [Effects of the Invention]
[0026] According to the present disclosure, even an ordinary facility manager can detect an abnormality in an aeration device during operation of a water treatment plant. [Brief explanation of the drawings]
[0027] [Figure 1] 1 is a diagram illustrating an embodiment of an operation management support system and an operation management support method for a water treatment plant. [Figure 2] 2 is a diagram showing an example of a method for arranging vibration information collecting devices in the operation management support system for the water treatment plant shown in FIG. 1. FIG. [Figure 3] 1. FIG. 4 is a diagram showing another example of a method for arranging vibration information collecting devices in the operation management support system for the water treatment plant shown in FIG. [Figure 4] 1. FIG. 4 is a diagram showing another example of a method for arranging vibration information collecting devices in the operation management support system for the water treatment plant shown in FIG. [Figure 5]FIG. 5 is a diagram illustrating an example of the configuration of the information processing device illustrated in FIGS. 2 to 4. [Figure 6] 5 is a diagram showing a flow of vibration information processing by the information processing device shown in FIGS. 2 to 4 and classification of processing methods. FIG. [Figure 7] FIG. 10 is a diagram showing vibration information cut out at a certain time interval. [Figure 8] FIG. 10 is a diagram showing a state in which the result of a short-time Fourier transform is converted into a grayscale image. [Figure 9] FIG. 9 is a diagram clearly showing the vertical streaks in the image shown in FIG. 8(b). [Figure 10] FIG. 8(b) is a diagram schematically showing the results of short-time Fourier transform during normal bubbling shown in FIG. 8(a). [Figure 11] FIG. 10 is a diagram schematically illustrating the results of short-time Fourier transform during the abnormal bubbling shown in FIGS. 8(b) and 9. [Figure 12] 6 is a flowchart showing an example of a processing flow when processing without machine learning is performed by the information processing device shown in FIGS. 2 to 5. [Figure 13] 6 is a flowchart showing an example of a processing flow when processing by unsupervised machine learning is performed by the information processing device shown in FIGS. 2 to 5. [Figure 14] 6 is a flowchart showing an example of a processing flow when processing by supervised machine learning is performed by the information processing device shown in FIGS. 2 to 5. [Figure 15] 15 is a conceptual diagram showing an example of the hardware configuration of a processing circuit included in the information processing device according to the embodiment shown in FIGS. 1 to 14. FIG. DETAILED DESCRIPTION OF THE INVENTION
[0028] Hereinafter, an embodiment of the operation management support system 1 and operation management support method for a water treatment plant 10 disclosed herein will be described with reference to the drawings.
[0029] <Configuration example of one embodiment> Fig. 1 is a diagram showing an embodiment of an operation management support system 1 and an operation management support method for a water treatment plant 10. Fig. 1 schematically shows an example of a water treatment plant 10 in the operation management support system 1 for the water treatment plant 10.
[0030] 1, in the operation management support system 1 for a water treatment plant 10, the water treatment plant 10 includes a blower 11, an air supply pipe 12, a treated water tank 20, an aeration device 30, and a vibration information collection device 40. The water treatment plant 10 is, for example, a sewage or wastewater treatment facility or a sewage or wastewater treatment plant that uses an activated sludge method for purifying water using aerobic microorganisms.
[0031] The blower 11 is connected to a plurality of air diffusers 30 via an air supply pipe 12. The blower 11 blows (supplies) air or oxygen to the plurality of air diffusers 30 via the air supply pipe 12.
[0032] One end of the air supply pipe 12 is connected to the blower 11, and branches into multiple parts along the way, with the other ends of the multiple branches each connected to an air diffuser 30. The air supply pipe 12 supplies air or oxygen supplied from the blower 11 to each of the multiple air diffusers 30 connected at the other end.
[0033] The treated water tank 20 is a tank in which treated water W, such as sewage, sewage, or wastewater, is stored, and a plurality of aeration devices 30 are disposed in the treated water W stored in the treated water tank 20. The treated water tank 20 is divided, for example, into a plurality of tanks (compartments), and a plurality of aeration devices 30 are disposed in each of the tanks (compartments). In the treated water tank 20, water quality purification of the treated water W is performed, for example, by an activated sludge method using aerobic microorganisms that decompose organic matter. In this specification, the treated water tank 20 is also referred to as an "aeration tank 20."
[0034] The air diffusers 30 are connected to the blower 11 via air supply pipes 12 and are disposed in multiple locations in the treated water W of one aeration tank 20. Each air diffuser 30 includes an air diffuser, such as a rubber or resin sheet or membrane or a porous body made of sintered ceramic powder, a holder for holding the air diffuser, and a gasket for maintaining airtightness between the air diffuser and the holder and preventing air or oxygen leakage. The air diffuser has multiple air diffusion holes that communicate between the inside and outside of the diffuser. Air or oxygen supplied from the air supply pipes 12 passes through the multiple air diffusion holes to become fine bubbles A, which are then supplied (foamed) into the treated water W in the aeration tank 20. The air diffuser 30 may be a pipe with an internal space, and multiple air diffusion holes communicating between the inside and outside of the pipe may be provided on the side of the pipe.
[0035] The aeration device 30 brings air bubbles A discharged from the air diffusion holes into contact with the treated water W to dissolve oxygen into the water, and the movement of the air bubbles A stirs the treated water W stored in the aeration tank 20, supplying the dissolved oxygen throughout the aeration tank 20. In this way, the aeration device 30 uniformly diffuses and dissolves the oxygen necessary for the respiration of the aerobic microorganisms in the aeration tank 20 into the treated water W, activating the aerobic microorganisms in the aeration tank 20 and purifying the treated water W in the aeration tank 20.
[0036] The vibration information collecting device 40 is, for example, a microphone or a vibration sensor, and periodically or in real time collects vibration information such as sound (acoustic, voice) or air vibrations generated by bubbles A bubbled in the treated water W from the air diffuser 30. For example, if the vibration information collecting device 40 is a microphone, the microphone periodically or in real time collects acoustic information of the sound (acoustic) generated by bubbles A bubbled in the treated water W from the air diffuser 30. The vibration information collecting device 40 is connected to an information processing device 50 (see FIGS. 2 to 4), which will be described later. The microphone and vibration sensor are examples of the vibration information collecting device 40, and other devices capable of collecting acoustic information and vibration information may be used. Sound (acoustic, voice) is air vibration detected by sound pressure level. In a broad sense, sound (acoustic, voice) is also air vibration. Therefore, in this specification, acoustic information will also be described as vibration information.
[0037] In this specification, the vibrations (sounds) emitted by bubbles include vibrations (sounds) emitted when bubbles rise in water, vibrations (sounds) generated when bubbles burst, vibrations (sounds) emitted by water currents induced by the rising bubbles, and vibrations (sounds) emitted by waves caused by water currents induced by the rising bubbles. Furthermore, the vibrations (sounds) emitted by bubbles may be a combination of any of these vibrations (sounds).
[0038] Fig. 2 is a diagram showing an example of a method for arranging vibration information collecting devices 40 in the operation management support system 1 of the water treatment plant 10 shown in Fig. 1. Fig. 2 shows a schematic example of a side view of one aeration tank 20 in the water treatment plant 10. Note that the air supply pipe 12 (see Fig. 1) is not shown.
[0039] The aeration tank 20 is provided with an air supply pipe 12 (see FIG. 1), a plurality of aeration devices 30, and a vibration information collecting device 40, all of which are not shown. The vibration information collecting device 40 is connected to an information processing device 50 outside the aeration tank 20 by wire or wirelessly. Treated water W is stored in the aeration tank 20. The top of the aeration tank 20 is covered with a cover (not shown).
[0040] The plurality of air diffusers 30 foams bubbles A made of air or oxygen supplied from the air supply pipe 12 through a large number of air diffusion holes. For example, if the air diffuser or air diffusion holes of the air diffuser 30 are torn or damaged, the bubbles A will foam as coarse bubbles B with large diameters (abnormal foaming). In this specification, the coarse bubbles B are also referred to as "abnormal bubbles B."
[0041] The vibration information collecting device (microphone, vibration sensor) 40 collects sounds (vibrations) emitted from bubbles A and coarse bubbles B bubbled in the treatment water W. If the vibration information collecting device 40 is a microphone, the vibration information collecting device 40 may collect and record the sounds emitted from the bubbles A and coarse bubbles B. If the vibration information collecting device 40 is a vibration sensor, it may collect (and record) the vibrations emitted from the bubbles A and coarse bubbles B bubbled in the treatment water W. The vibration information collecting device 40 outputs the recorded (collected) sounds (vibration information) to the information processing device 50.
[0042] 2, the vibration information collecting device 40 is placed in the air near the water surface of the aeration tank 20. For example, one vibration information collecting device 40 is placed for one aeration tank 20. Note that multiple vibration information collecting devices 40 may be placed for each section of one aeration tank 20, or one vibration information collecting device 40 may be placed for one aeration device 30.
[0043] The information processing device 50 is connected to the vibration information collecting device 40 and acquires vibration information output from the vibration information collecting device 40. Note that sound and vibration may be recorded or recorded by the information processing device 50 rather than by a microphone, a vibration sensor, or the like in the vibration information collecting device 40. Details of the information processing device 50 will be described later.
[0044] Fig. 3 is a diagram showing another example of a method for arranging vibration information collecting devices 40 in the operation management support system 1 of the water treatment plant 10 shown in Fig. 1. Like Fig. 2, Fig. 3 schematically shows an example of a side view of one aeration tank 20 in the water treatment plant 10. Note that in Fig. 3, the same or similar components as those in Fig. 2 are denoted by the same reference numerals, and detailed explanations thereof will be omitted or simplified.
[0045] In Fig. 3, a vibration information collecting device (microphone, vibration sensor) 40 is placed in the treatment water W. Even when it is underwater, the vibration information collecting device 40 can collect sounds (vibrations) emitted from bubbles A and coarse bubbles B. Note that other configurations are the same as those in Fig. 2, and therefore description thereof will be omitted.
[0046] Fig. 4 is a diagram showing another example of a method for arranging the vibration information collecting device 40 in the operation management support system 1 of the water treatment plant 10 shown in Fig. 1. Like Fig. 2 and Fig. 3, Fig. 4 schematically shows an example of a side view of one aeration tank 20 in the water treatment plant 10. Note that in Fig. 4, the same or similar components as those in Fig. 2 and Fig. 3 are denoted by the same reference numerals, and detailed description thereof will be omitted or simplified.
[0047] In Fig. 4, the vibration information collecting device (microphone, vibration sensor) 40 is arranged so as to be close to or in contact with the air supply pipe 12. For example, when coarse bubbles B are bubbled from the air diffuser 30, the vibration information collecting device 40 can collect the sound (vibration) of the bubble formation of the coarse bubbles B as well as the sound (vibration) of the air supplied from the blower 11 through the air supply pipe 12. The vibration information collecting device 40 may be arranged so as to be close to or in contact with each of a plurality of air supply pipes 12, which are not shown. The other configurations are the same as those in Figs. 2 and 3, and therefore description thereof will be omitted.
[0048] Fig. 5 is a diagram showing an example of the configuration of the information processing device 50 shown in Fig. 2 to Fig. 4. The information processing device 50 is installed, for example, in the water treatment plant 10 or in a location remote from the water treatment plant 10, and is connected to the vibration information collecting device 40 by wire or wirelessly, although wiring and the like are omitted in the figure. The information processing device 50 may also include a communication unit, an operation unit, a display unit, a storage unit, a control unit, and the like, which are not shown.
[0049] The information processing device 50 has a processor 91 (see FIG. 15) such as a CPU (Central Processing Unit), MPU (Micro Processing Unit), or GPU (Graphics Processing Unit) that operates by executing a program, which will be described later. The information processing device 50 also has a memory 92 (see FIG. 15) that will be described later, and executes a predetermined program stored in the memory 92 to operate the processor 91 and process vibration information.
[0050] The information processing device 50 functions as the following units by executing, for example, a predetermined program stored in a memory 92 (see FIG. 15 ), which will be described later. The information processing device 50 functions as a vibration information acquisition unit 51, a determination algorithm execution unit 52, a vibration information determination unit 53, a determination result notification unit 54, and a determination algorithm generation unit 55. Note that each of the above functions may be realized by a program executed by an arithmetic processing unit (not shown) included in the information processing device 50, or may be realized by hardware. The vibration information acquisition unit 51, the determination algorithm execution unit 52, the vibration information determination unit 53, the determination result notification unit 54, and the determination algorithm generation unit 55 execute the predetermined program to perform the following processes.
[0051] The vibration information acquisition unit 51 acquires vibration information from the vibration information collecting device 40 and performs a predetermined conversion process on the acquired vibration information. Details of the process performed by the vibration information acquisition unit 51 will be described later.
[0052] The determination algorithm execution unit 52 performs processing based on a predetermined determination algorithm on the vibration information that has been subjected to a predetermined conversion process by the vibration information acquisition unit 51. Details of the processing by the determination algorithm execution unit 52 will be described later.
[0053] The vibration information determination unit 53 determines whether there is an abnormality in the air diffuser 30 based on the output result output from the determination algorithm execution unit 52. Details of the processing by the vibration information determination unit 53 will be described later.
[0054] The determination result notification unit 54 notifies the result determined by the vibration information determination unit 53. The determination result notification unit 54 notifies the result by, for example, displaying it on a display device through contact output, current value output, or voltage value output. Alternatively, the determination result notification unit 54 notifies the result by, for example, displaying it on an information processing terminal such as a personal computer or a smartphone through email or the like. Alternatively, the determination result notification unit 54 notifies the result by, for example, displaying it on a display device connected to the information processing device 50. Note that the notification method is not limited to these. The determination result notification unit 54 may also notify the result to an external device of the information processing device 50 through the above-mentioned display device, information processing terminal, or the like.
[0055] The judgment algorithm generation unit 55 generates a predetermined judgment algorithm based on the vibration information that has been subjected to a predetermined conversion process by the vibration information acquisition unit 51. The judgment algorithm generation unit 55 may update the predetermined judgment algorithm at predetermined intervals, predetermined time periods, or in real time. This is because seasonal changes in water temperature and changes in the water quality of sewage and wastewater can change the water quality of the treated water W in the aeration tank 20, changing physical properties such as viscosity, and changing the frequency of the popping sound of bubbles on the water surface and the characteristics of vibration intensity.
[0056] 5, the process flow when generating a determination algorithm is indicated by the white arrows, and the process flow when determining vibration information is indicated by the gray arrows. When generating a determination algorithm, the vibration information sent to the information processing device 50 is sent to the determination algorithm generation unit 55 as indicated by the white arrows, and a determination algorithm is generated.
[0057] On the other hand, when the vibration information is judged (when the judgment algorithm is executed), the vibration information sent to the information processing device 50 is sent to the vibration information judgment unit 53 through processing by the judgment algorithm execution unit 52, as indicated by the gray arrow. The result of the judgment of the soundness of the air diffuser 30 by the vibration information judgment unit 53 is sent to the judgment result notification unit 54, which then notifies the outside of the information processing device 50.
[0058] Note that since the determination algorithm execution unit 52 processes information based on the determination algorithm generated by the determination algorithm generation unit 55, the processing by the determination algorithm execution unit 52 and the processing by the determination algorithm generation unit 55 are usually not performed simultaneously. However, the processing by the determination algorithm execution unit 52 and the processing by the determination algorithm generation unit 55 may be performed simultaneously.
[0059] The determination algorithm may be one in which vibration information (acoustic information) is decomposed into frequencies by a short-time Fourier transform, or may be one that uses machine learning including deep learning.
[0060] <Processing example of one embodiment> The following describes the flow of information processing performed by the information processing device 50. First, the information processing device 50 converts the vibration information (acoustic information) recorded (recorded) in the aeration tank 20 into a state that can be handled as digital data, and performs vibration analysis (acoustic analysis) using Fourier transform and machine learning (deep learning). Then, the information processing device 50 determines whether the foaming state is normal or abnormal based on the results of the vibration analysis (acoustic analysis), and if the foaming state is abnormal, notifies the user that this is an abnormality.
[0061] Fig. 6 is a diagram showing the flow of vibration information processing and classification of processing methods by the information processing device 50 shown in Fig. 2 to Fig. 4. Fig. 6 shows three examples of processing methods for information processing performed by the information processing device 50. That is, Fig. 6 shows three examples of processing methods: processing without machine learning, processing by unsupervised machine learning, and processing by supervised machine learning.
[0062] In Figure 6, in processing without machine learning, processing is performed in the order of (1) → (2-1) → (3) surrounded by solid lines. In processing with unsupervised machine learning, processing is performed in the order of (1) → (2-2) → (2-2-1) → (3) surrounded by dashed lines. In processing with supervision, processing is performed in the order of (1) → (2-2) → (2-2-2) surrounded by dashed lines.
[0063] 6(1), for example, the vibration information acquisition unit 51 of the information processing device 50 cuts out the recorded vibration information for a predetermined time (for example, 10 seconds). Then, the vibration information acquisition unit 51 performs a short-time Fourier transform over the entire recording time, for example, for a time period about 50 to 5000 times the sampling rate.
[0064] FIG. 7 is a diagram showing vibration information cut over a certain period of time. Note that FIG. 7 is a diagram created based on an experiment conducted by the applicant using a microphone as the vibration information collecting device 40. In FIG. 7, the horizontal axis represents time, and the vertical axis represents sound pressure level (vibration level). FIG. 7 shows raw acoustic data (vibration data) recorded (collected) by the vibration information collecting device 40, cut over a 10-second period by the vibration information acquiring unit 51. Since the raw acoustic information (vibration information) shown in FIG. 7 does not allow for distinguishing between normal and abnormal foaming, the vibration information acquiring unit 51 performs a short-time Fourier transform on the raw acoustic information (vibration information) shown in FIG. Note that in this specification, the vibration level (sound pressure level) is also referred to as vibration intensity (sound pressure intensity).
[0065] 6, in (2-1), the vibration information acquisition unit 51 of the information processing device 50 calculates the average, standard deviation, coefficient of variation, etc. of the time change in the vibration level (sound pressure level) for each frequency. Then, the vibration information acquisition unit 51 adds up the standard deviation or coefficient of variation for a predetermined range of frequencies or for all recorded frequencies.
[0066] In addition, in (2-2), the vibration information acquisition unit 51 of the information processing device 50 converts the result of the short-time Fourier transform into three-dimensional information of time, frequency, and vibration level (sound pressure level).
[0067] 8A and 8B are diagrams showing the state in which the results of the short-time Fourier transform are converted into a grayscale image. Fig. 8A shows the state in which the results of the short-time Fourier transform during normal bubbling are converted into a grayscale image, and Fig. 8B shows the state in which the results of the short-time Fourier transform during abnormal bubbling are converted into a grayscale image. Note that Figs. 8A and 8B are diagrams created based on an experiment conducted by the applicant using a microphone as the vibration information collecting device 40.
[0068] In Figures 8(a) and (b), the horizontal axis represents time, and the example in Figure 8 shows the short-time Fourier transform of sound data (vibration data) recorded for 10 seconds. The vertical axis represents frequency, with the upper side of Figure 8 representing high-frequency sounds (vibrations) and the lower side of Figure 8 representing low-frequency sounds (vibrations). Different colors represent differences in sound pressure level (vibration level), with darker colors representing lower sound pressure level (vibration level) and lighter colors representing higher sound pressure level (vibration level).
[0069] For example, when foaming occurs in still water, i.e., clean water, the popping or fizzing sounds of the bubbles can be heard, but as shown in Figure 7, it is not possible to distinguish between normal foaming sounds (vibrations) and abnormal foaming sounds (vibrations) from the raw acoustic data (vibration data). For this reason, a short-time Fourier transform is performed, and the results of the short-time Fourier transform are displayed as a grayscale image as shown in Figure 8, and then analyzed.
[0070] For example, a well-known document (Donald E. Spiel "Acoustical Measurements of Air Bubbles Bursting at a Water Surface: Bursting Bubbles as Helmholtz Resonators", J. Geophysical Research, 97, 11443-11452, 1992) shows that the frequency of the sound of a bubble bursting varies depending on the radius of the bubble. The above paper shows that bubble bursting is based on Helmholtz resonance, and that the frequency of the sound when a bubble bursts can be calculated based on the principle of Helmholtz resonance.
[0071] In Figure 8, the horizontal stripes are almost identical between Figure 8(a), which shows the results of the short-time Fourier transform during normal foaming, and Figure 8(b), which shows the results of the short-time Fourier transform during abnormal foaming. However, unlike Figure 8(a), which shows the results of the short-time Fourier transform during normal foaming, Figure 8(b), which shows the results of the short-time Fourier transform during abnormal foaming, has vertical lines (vertical lines) resembling scratches scattered randomly. While the above paper only conducted experiments on small bubbles, for example, those with a diameter of 4 mm or less, the applicant's experiments have shown that the results shown in Figure 8(b) are obtained when coarse bubbles B with a diameter of 10 mm to several tens of mm are generated, as occurs when the air diffuser 30 is damaged.
[0072] FIG. 9 clearly shows the vertical streaks in the image shown in FIG. 8(b). FIG. 9 shows the results of the short-time Fourier transform during abnormal foaming in FIG. 8(b), with the vertical streaks surrounded by ellipses. As described above, FIG. 9 reveals that during abnormal foaming, vertical streaks (vertical lines) resembling scratches are randomly observed. That is, FIGS. 8(b) and 9 reveal a temporal change in sound pressure intensity (sound pressure level) that is not observed during normal foaming (FIG. 8(a)). This is because during abnormal foaming, coarse bubbles B are generated discontinuously (randomly), and therefore the sounds of coarse bubbles B are generated discontinuously (randomly).
[0073] Furthermore, as shown in Figures 8(b) and 9, when abnormal foaming such as coarse bubbles B occurs in the aeration tank 20, it can be seen that sound is heard (sound pressure is distributed) over a very wide frequency range from approximately 3,000 Hz to 20,000 Hz. In other words, if it is possible to detect changes in volume and sound pressure over time in a wide frequency band, it is possible to determine whether foaming is normal or abnormal. If it is possible to detect changes in sound pressure intensity (sound pressure level) over time at least in a predetermined frequency band, it is possible to determine whether foaming is normal or abnormal.
[0074] FIG. 10 is a diagram showing the results of a short-time Fourier transform during normal foaming as shown in FIG. 8(a). In FIG. 10, as in FIG. 8, the horizontal axis (X-axis direction) represents time, and the vertical axis (Y-axis direction) represents frequency. The height direction (Z-axis direction) represents sound pressure (vibration) intensity. The Z-axis direction represents height, whereas in FIG. 8 it was represented by grayscale brightness (shade). In other words, FIG. 10 shows a three-dimensional graph of time, frequency, and sound pressure (vibration) intensity.
[0075] Figure 10 shows that the horizontal stripes in Figure 8(a) change little over time and are almost constant. That is, Figure 10 shows that during normal foaming, the distribution of sound pressure (vibration) intensity changes little over time (temporal change) and is almost constant.
[0076] Figure 11 is a diagram showing the results of a short-time Fourier transform performed during the abnormal bubbling shown in Figures 8(b) and 9. Similar to Figure 10, Figure 11 shows a three-dimensional graph of time, frequency, and sound pressure (vibration) intensity.
[0077] Figure 11 shows that vertical lines (vertical lines) like scratches in Figure 8(b) are randomly scattered, as indicated by arrows, in the sound pressure distribution when coarse bubbles burst. In other words, Figure 11 shows that the distribution of sound pressure (vibration) intensity changes over time during abnormal bubble formation.
[0078] 8 to 11 show grayscale images or three-dimensional graphs, but these are actually numerical values of three-dimensional information consisting of time, frequency, and vibration intensity. Therefore, the information processing device 50 determines that abnormal foaming occurs if the change over time of sound pressure intensity in a predetermined frequency band (or a wide frequency band) is equal to or greater than a predetermined value (without machine learning). Note that the information processing device 50 may learn information about a normal foaming state from the three-dimensional information by machine learning, calculate the difference in feature amount (or change over time) from the abnormal foaming state, and determine that abnormal foaming occurs if the calculated difference is equal to or greater than a predetermined value (unsupervised machine learning). Alternatively, the information processing device 50 may learn three-dimensional information about a normal foaming state and three-dimensional information about an abnormal foaming state by machine learning, and determine the type of abnormal foaming state based on the feature amount (or change over time) of the three-dimensional information (supervised machine learning).
[0079] 6, in (2-2-1), the determination algorithm execution unit 52 of the information processing device 50 uses the vibration information, which is the three-dimensional information during normal foaming shown in Fig. 8(a), as learning data and learns it using a machine learning algorithm such as deep learning. Then, the determination algorithm execution unit 52 inputs the newly learned vibration information into the previously learned algorithm and calculates the degree of deviation from the learned image information.
[0080] In (2-2-2), the determination algorithm execution unit 52 of the information processing device 50 performs learning using a machine learning algorithm such as deep learning, using the vibration information, which is the three-dimensional information during normal and abnormal foaming shown in FIGS. 8(a) and 8(b), as training data. The determination algorithm execution unit 52 then inputs the newly acquired vibration information into a previously learned algorithm. The vibration information determination unit 53 of the information processing device 50 then identifies the type of abnormality based on the results output by the determination algorithm execution unit 52. The results output by the determination algorithm execution unit 52 may be a table that is compiled by dividing the results into normal and abnormal cases.
[0081] In addition, in (3), the vibration information determination unit 53 of the information processing device 50 determines that there is an abnormality when the value obtained by (2-1) or (2-2-1) is greater than a preset threshold value (predetermined threshold value).
[0082] <Example of processing without machine learning> An example of processing by the information processing device 50 without machine learning will be described below.
[0083] Fig. 12 is a flowchart showing an example of the flow of processing when processing without machine learning is performed by the information processing device 50 shown in Fig. 2 to Fig. 5. The flowchart shown in Fig. 12 shows an example when the information processing device 50 performs processing in the order of (1) → (2-1) → (3) surrounded by solid lines in Fig. 6.
[0084] In step S11, the vibration information acquisition unit 51 receives the vibration information sent from the vibration information collection device 40 (microphone or vibration sensor).
[0085] In step S12, the vibration information acquisition unit 51 extracts the received vibration information (sound information) at a fixed time interval (see, for example, FIG. 7).
[0086] In step S13, the vibration information acquisition unit 51 performs a short-time Fourier transform on the cut vibration information (sound information). Note that a short time is, for example, about 2 milliseconds to 50 milliseconds.
[0087] In step S14, the determination algorithm execution unit 52 calculates the average, standard deviation, coefficient of variation, etc., for the time change of the vibration intensity (sound pressure intensity) for each frequency.
[0088] In step S15, the determination algorithm execution unit 52 adds up the standard deviation or coefficient of variation of a certain range of frequencies or of all the recorded frequencies.
[0089] In step S16, the vibration information determination unit 53 compares the value output by the determination algorithm from the determination algorithm execution unit 52 with a preset threshold value (predetermined threshold value). Note that the predetermined threshold value may be different for each season or each predetermined time period depending on seasonal changes in water temperature or changes in sewage and wastewater water quality, for each aeration tank 20, or for each section of the aeration tank 20.
[0090] In step S17, the vibration information determination unit 53 determines that an abnormality has occurred if the numerical value output by the determination algorithm is greater than a predetermined threshold value.
[0091] In step S18, the determination result notification unit 54 notifies the result output by the determination algorithm and the result determined by the vibration information determination unit 53. The determination result notification unit 54 may notify the result determined by the vibration information determination unit 53, or may notify the result only when the vibration information determination unit 53 determines that an abnormality has occurred.
[0092] <Example of unsupervised machine learning processing> Next, an example of unsupervised machine learning processing by the information processing device 50 will be described.
[0093] Fig. 13 is a flowchart showing an example of the flow of processing when processing by unsupervised machine learning is performed by the information processing device 50 shown in Fig. 2 to Fig. 5. The flowchart shown in Fig. 13 shows an example when the information processing device 50 performs processing in the order of (1) → (2-2) → (2-2-1) → (3) surrounded by dashed lines in Fig. 6.
[0094] In step S21, the vibration information acquisition unit 51 receives the vibration information sent from the vibration information collection device 40 (microphone or vibration sensor).
[0095] In step S22, the vibration information acquisition unit 51 extracts the received vibration information (sound information) at a fixed time interval (see, for example, FIG. 7).
[0096] In step S23, the vibration information acquisition unit 51 performs a short-time Fourier transform on the cut vibration information (sound information). Note that a short time is, for example, about 2 milliseconds to 50 milliseconds.
[0097] In step S24, the vibration information acquisition unit 51 outputs the result of the short-time Fourier transform as three-dimensional information of time, frequency, and vibration level (sound pressure level). Note that Fig. 8 shows the result converted into a grayscale image in which time is represented as width, frequency as height, and vibration level (sound pressure level) as brightness.
[0098] In step S25, the determination algorithm execution unit 52 inputs the three-dimensional information of time, frequency, and vibration intensity (sound pressure intensity) obtained by the short-time Fourier transform into the determination algorithm learned by machine learning. Note that the determination algorithm is updated by the determination algorithm generation unit 55, for example, at the above-mentioned predetermined intervals.
[0099] In step S26, the determination algorithm execution unit 52 compares the input three-dimensional information with the learned normal vibration information and outputs the magnitude of the error. In unsupervised machine learning, normal vibration information is learned by machine learning, so the degree to which the input three-dimensional information is close to the normal vibration information is quantified.
[0100] In step S27, the vibration information determination unit 53 compares the value output by the determination algorithm from the determination algorithm execution unit 52 with a preset threshold value (predetermined threshold value). Note that the predetermined threshold value may be different for each season or each predetermined time period depending on seasonal changes in water temperature or changes in sewage and wastewater water quality, for each aeration tank 20, or for each section of the aeration tank 20.
[0101] In step S28, the vibration information determination unit 53 determines that an abnormality has occurred if the numerical value output by the determination algorithm is greater than a predetermined threshold value.
[0102] In step S29, the determination result notification unit 54 notifies the result output by the determination algorithm and the result determined by the vibration information determination unit 53. Note that the determination result notification unit 54 may notify the result determined by the vibration information determination unit 53, or may notify the result only when the vibration information determination unit 53 determines that an abnormality has occurred.
[0103] <Example of supervised machine learning processing> Finally, an example of supervised machine learning processing by the information processing device 50 will be described.
[0104] Fig. 14 is a flowchart showing an example of the flow of processing when processing by supervised machine learning is performed by the information processing device 50 shown in Fig. 2 to Fig. 5. The flowchart shown in Fig. 13 shows an example when the information processing device 50 performs processing in the order of (1) → (2-2) → (2-2-2) surrounded by the dashed dotted line in Fig. 6.
[0105] In step S31, the vibration information acquisition unit 51 receives vibration information sent from the vibration information collection device 40 (microphone or vibration sensor).
[0106] In step S32, the vibration information acquisition unit 51 cuts out the received vibration information (sound information) at a fixed time interval (see, for example, FIG. 7).
[0107] In step S33, the vibration information acquisition unit 51 performs a short-time Fourier transform on the cut vibration information (sound information). Note that a short time is, for example, about 2 milliseconds to 50 milliseconds.
[0108] In step S34, the vibration information acquisition unit 51 outputs the result of the short-time Fourier transform as three-dimensional information of time, frequency, and vibration level (sound pressure level). Note that Fig. 8 shows the result converted into a grayscale image in which time is represented as width, frequency as height, and vibration level (sound pressure level) as brightness.
[0109] In step S35, the determination algorithm execution unit 52 inputs the three-dimensional information of time, frequency, and vibration intensity (sound pressure intensity) obtained by the short-time Fourier transform into an algorithm trained by machine learning. Note that the determination algorithm is updated by the determination algorithm generation unit 55, for example, at the above-mentioned predetermined intervals.
[0110] In step S36, the determination algorithm execution unit 52 classifies the input vibration information (acoustic information) using a machine learning algorithm (determination algorithm). In supervised machine learning, not only normal vibration information but also various abnormal vibration information is learned by machine learning, so the input three-dimensional information is classified into one of the learned categories.
[0111] In step S37, if the result of classification by the determination algorithm is a classification indicating an abnormality, the vibration information determination unit 53 determines that there is an abnormality. That is, in the case of supervised machine learning, a numerical value is not output in step S36, but the type is classified as normal or abnormal, so in the case of an abnormality, the vibration information determination unit 53 makes a determination based on the classified output result as is (without comparing it with a predetermined threshold, etc.).
[0112] In step S38, the determination result notification unit 54 notifies the result output by the determination algorithm and the result determined by the vibration information determination unit 53. Note that the determination result notification unit 54 may notify the result determined by the vibration information determination unit 53, or may notify the result only when the vibration information determination unit 53 determines that an abnormality has occurred.
[0113] <Hardware configuration example> FIG. 15 is a conceptual diagram showing an example of the hardware configuration of a processing circuit included in the information processing device 50 according to the embodiment shown in FIGS. 1 to 14. The above-described functions are realized by the processing circuit. In one aspect, the processing circuit includes at least one processor 91 and at least one memory 92. In another aspect, the processing circuit includes at least one dedicated hardware 93.
[0114] When the processing circuit includes a processor 91 and a memory 92, each function is realized by software, firmware, or a combination of software and firmware. At least one of the software and firmware is written as a program. At least one of the software and firmware is stored in the memory 92. The processor 91 realizes each function by reading and executing the program stored in the memory 92.
[0115] If the processing circuitry comprises dedicated hardware 93, the processing circuitry may be, for example, a single circuit, multiple circuits, a programmed processor, or a combination thereof. Each function is implemented by the processing circuitry.
[0116] Each function of the information processing device 50 may be partially or entirely configured by hardware, or may be configured as a program executed by a processor. That is, the information processing device 50 can also be realized by a computer and a program, and the program can be stored in a storage medium or provided via a network.
[0117] <Effects of one embodiment> 1 to 15, in the treated water tank 20, abnormalities or damage to the aeration device 30 are detected based on vibration information (acoustic information) emitted by bubbles formed in the treated water W by air or oxygen released from the aeration device 30. This makes it possible to detect abnormal foaming, such as the foaming of coarse bubbles B, before the quality of the treated water W deteriorates, thereby enabling early detection of abnormalities or damage to the aeration device 30. This also makes it possible to detect abnormalities or damage to the aeration device 30 without draining the treated water W from the aeration tank 20 or stopping operation of the water treatment plant 10 for inspection.
[0118] 1 to 15, abnormalities or damage to the aeration device 30 can be detected early, thereby preventing deterioration in the quality of the treated water W in the treatment water tank 20. This makes it possible to prevent sludge from flowing back into other aeration devices 30 in the same system, and as a result, it is possible to prevent the abnormality or damage to the aeration device 30 from spreading.
[0119] 1 to 15, a predetermined transformation process such as a short-time Fourier transform is performed on the vibration information (acoustic information) acquired from the vibration information collecting device 40. This makes it possible to determine the change over time in vibration intensity (sound pressure intensity) for each frequency, which cannot be determined from the raw vibration information (acoustic information).
[0120] 1 to 15, vibration information (acoustic information) that has been subjected to a predetermined conversion process is processed based on a predetermined judgment algorithm, thereby detecting abnormal foaming and other abnormalities or damage to the air diffusion device 30. As a result, abnormalities or damage to the air diffusion device 30 can be detected by using a general-purpose computer while the air diffusion device 30 is in operation.
[0121] 1 to 15, a general-purpose computer detects an abnormality or damage to the air diffuser 30 and notifies the user of the abnormality or damage. This allows even a general facility manager, rather than a specialized engineer, to detect an abnormality or damage to the air diffuser 30 while the air diffuser 30 is in operation. This also allows the user to detect an abnormality or damage to the air diffuser 30 by referring to the notification, even if the target water treatment plant 10 is located in a remote location, without the need for a specialized engineer to visit the site.
[0122] 1 to 15, the judgment algorithm is updated for each predetermined period or time period by the judgment algorithm generation unit 55. As a result, even if the time of year or time period varies depending on seasonal changes in water temperature or changes in sewage or wastewater quality, or even if the aeration tank 20 or the compartments of the aeration tank 20 are different, it is possible to detect abnormalities or damage to the aeration device 30 based on an appropriate judgment algorithm that is suited to the target water treatment plant 10.
[0123] <Supplementary information on the implementation form> 1 to 15, the processing by the information processing device 50 has been described as being divided into a processing example without machine learning shown in Fig. 12, a processing example using unsupervised machine learning shown in Fig. 13, and a processing example using supervised machine learning shown in Fig. 14, but the present invention is not limited to this. Some or all of these may be combined in series or in parallel and processed by the information processing device 50. By combining these processes, the combined process can achieve the respective operational effects achieved by the individual processes before being combined.
[0124] 1 to 15, an operation management support system 1 for a water treatment plant 10 has been described as an example of one aspect of the present disclosure, but it can also be realized as an information processing device 50 in the operation management support system 1 for a water treatment plant 10. It can also be realized as an operation management support method for a water treatment plant 10 and an information processing method in the operation management support system 1 for a water treatment plant 10.
[0125] The present disclosure can also be realized as an operation management support program for the water treatment plant 10 that causes a computer to execute processing steps in an operation management support method for the water treatment plant 10. Alternatively, the present disclosure can also be realized as an information processing program that causes a computer to execute processing steps in an information processing method in the operation management support system 1 for the water treatment plant 10.
[0126] The present disclosure can also be realized as a storage medium (non-transitory computer-readable medium) on which the above-mentioned operation management support program or information processing program is stored. The operation management support program or information processing program can be stored and distributed on a removable disk such as a CD (Compact Disc), DVD (Digital Versatile Disc), or USB (Universal Serial Bus) memory. The operation management support program or information processing program may be uploaded to a network via a network interface (not shown) provided in the information processing device 50 or the like, or may be downloaded from the network and stored in the memory 92 or the like.
[0127] The features and advantages of the embodiments will be apparent from the above detailed description. It is intended that the claims encompass the features and advantages of the above-described embodiments without departing from the spirit and scope of the claims. Furthermore, any improvements and modifications will be readily apparent to those skilled in the art. Therefore, it is not intended that the scope of the inventive embodiments be limited to the above-described embodiments, and appropriate improvements and equivalents within the scope of the disclosed embodiments may be utilized. [Explanation of symbols]
[0128] 1...Operation management support system; 10...Water treatment plant (sewage treatment or wastewater treatment facility, sewage or wastewater treatment plant); 11...Blower; 12...Air supply pipe; 20...Treated water tank (aeration tank); 30...Aeration device; 40...Vibration information collection device (microphone, vibration sensor); 50...Information processing device; 51...Vibration information acquisition unit; 52...Judgment algorithm execution unit; 53...Vibration information judgment unit; 54...Judgment result notification unit; 55...Judgment algorithm generation unit; 91...Processor; 92...Memory; 93...Hardware; A...Bubbles; B...Coarse bubbles (abnormal bubbles); W...Treated water
Claims
1. an air supply pipe through which air or oxygen is supplied from a blower; an air diffuser connected to the air supply pipe and receiving the air or oxygen from the air supply pipe; a treatment water tank in which the air diffuser is disposed in the treatment water and the air or oxygen is released into the water from the air diffuser; a vibration information collecting device that collects vibration information emitted by bubbles formed in the water by the air or oxygen released from the air diffuser in the treatment water tank; an information processing device that performs predetermined information processing on the vibration information collected by the vibration information collecting device; An operation management support system for a water treatment plant comprising: The information processing device includes: a vibration information acquisition unit that acquires the vibration information from the vibration information collecting device and performs a predetermined conversion process on the acquired vibration information; a determination algorithm execution unit that executes processing based on a predetermined determination algorithm on the vibration information that has been subjected to the predetermined conversion processing by the vibration information acquisition unit; a vibration information determination unit that determines an abnormality in the air diffusion device based on the output result output from the determination algorithm execution unit; a determination result notification unit that notifies the result of the determination made by the vibration information determination unit; Equipped with The water treatment plant operation management support system is characterized in that the specified transformation processing is a short-time Fourier transform performed on the vibration information acquired by the vibration information acquisition unit, which is cut out at a specified time.
2. 2. The water treatment plant operation management support system according to claim 1, The vibration information collecting device is disposed in a space above the treated water in the treated water tank. The present invention relates to an operation management support system for a water treatment plant.
3. 2. The water treatment plant operation management support system according to claim 1, The vibration information collecting device is disposed in the treated water in the treated water tank. The present invention relates to an operation management support system for a water treatment plant.
4. 2. The water treatment plant operation management support system according to claim 1, The vibration information collecting device is disposed in proximity to or in contact with the air supply pipe. The present invention relates to an operation management support system for a water treatment plant.
5. 5. The water treatment plant operation management support system according to claim 1, the vibration information is acoustic information of the underwater rising sound of the bubbles, the breaking sound of the bubbles, the water flow sound, the splashing sound of the waves, or a combination thereof, which is generated by the bubbles; the vibration information collecting device is a microphone that collects the acoustic information, The vibration information acquisition unit performs the predetermined conversion process on the acoustic information collected by the microphone. The present invention relates to an operation management support system for a water treatment plant.
6. 5. The water treatment plant operation management support system according to claim 1, the vibration information is vibration information of vibrations caused by the bubbles rising underwater, bubble breaking vibrations, water current vibrations, wave splash vibrations, or a combination of these vibrations; the vibration information collecting device is a vibration sensor that collects the vibration information, The vibration information acquisition unit performs the predetermined conversion process on the vibration information collected by the vibration sensor. The present invention relates to an operation management support system for a water treatment plant.
7. In the operation management support system for a water treatment plant according to any one of claims 1 to 6, the determination algorithm execution unit calculates an average, standard deviation, or coefficient of variation of the time change of vibration intensity for each frequency for the vibration information that has been subjected to the predetermined conversion process by the vibration information acquisition unit, and outputs a result of integrating the standard deviation or coefficient of variation for a predetermined range of frequencies or for all recorded frequencies; The vibration information determination unit determines that there is an abnormality in the air diffuser when the value of the time change in vibration intensity output from the determination algorithm execution unit is greater than a predetermined threshold value. The present invention relates to an operation management support system for a water treatment plant.
8. 7. The water treatment plant operation management support system according to claim 1, the vibration information acquisition unit converts the vibration information that has been subjected to the predetermined conversion process into three-dimensional information of time, frequency, and vibration intensity; the determination algorithm execution unit calculates and outputs a degree of discrepancy between the three-dimensional information converted by the vibration information acquisition unit and the three-dimensional information based on vibration information in a normal state that has been learned in advance; and The vibration information determination unit determines that there is an abnormality in the air diffuser when the value of the time change in vibration intensity output from the determination algorithm execution unit is greater than a predetermined threshold value. The present invention relates to an operation management support system for a water treatment plant.
9. 9. The water treatment plant operation management support system according to claim 7 or 8, The predetermined threshold value may be different for each season, each aeration tank, or each section of the aeration tank. The present invention relates to an operation management support system for a water treatment plant.
10. 7. The water treatment plant operation management support system according to claim 1, the vibration information acquisition unit converts the vibration information that has been subjected to the predetermined conversion process into three-dimensional information of time, frequency, and vibration intensity; the determination algorithm execution unit classifies the three-dimensional information converted by the vibration information acquisition unit based on the three-dimensional information based on vibration information learned in advance; The vibration information determination unit determines that there is an abnormality in the air diffusion device when the result of the time change of vibration intensity classified by the determination algorithm execution unit is a classification indicating an abnormality in the air diffusion device. The present invention relates to an operation management support system for a water treatment plant.
11. The water treatment plant operation management support system according to any one of claims 1 to 10, The vibration information acquisition unit may further include a determination algorithm generation unit that generates the predetermined determination algorithm based on the vibration information that has been subjected to a predetermined conversion process. The present invention relates to an operation management support system for a water treatment plant.
12. The water treatment plant operation management support system according to claim 11, The determination algorithm generation unit updates the predetermined determination algorithm every predetermined period or every predetermined time period. The present invention relates to an operation management support system for a water treatment plant.
13. The water treatment plant operation management support system according to any one of claims 1 to 12, The determination result notification unit notifies the result determined by the vibration information determination unit to an outside of the information processing device by a display or an information processing terminal. The present invention relates to an operation management support system for a water treatment plant.
14. an air supply pipe through which air or oxygen is supplied from a blower; an air diffuser connected to the air supply pipe and receiving the air or oxygen from the air supply pipe; a treatment water tank in which the air diffuser is disposed in the treatment water and the air or oxygen is released into the water from the air diffuser; a vibration information collecting device that collects vibration information emitted by bubbles formed in the water by the air or oxygen released from the air diffuser in the treatment water tank; In a water treatment plant comprising: a vibration information acquisition step of acquiring the vibration information from the vibration information collecting device and performing a predetermined conversion process on the acquired vibration information; a determination algorithm execution step of executing a process based on a predetermined determination algorithm on the vibration information that has been subjected to the predetermined conversion process; a vibration information determination step of determining an abnormality in the air diffusion device based on the output result output by the determination algorithm execution step; a determination result notifying step of notifying the result determined by the vibration information determining step; Equipped with The method for supporting operation management of a water treatment plant is characterized in that the predetermined transformation process is a short-time Fourier transform performed on the vibration information acquired by the vibration information acquisition step, which is cut out at a predetermined time.
Citation Information
Patent Citations
Operation supporting system for plant
JP1995014090A
Liquid treatment method or sensing method
JP1998085719A
active acoustic spectroscopy
JP2004504600A
Cleaning apparatus
JP2009285571A
Air diffusing body
JP2011230068A