Information processing device, information processing method and program, and method for controlling water treatment

Abstract

To provide an information processing device capable quickly and easily discriminating an addition amount of flocculant added to form a good quality flock.SOLUTION: An information processing device 10 includes a CPU 11 and a SSD 14, flocks grow in a flocculation basin in which many flocks are formed by injecting flocculant into water to be treated and coagulating suspended matters contained in the water to be treated, and is used for tap water treatment in which grown flocks sedimentate in a sedimentation tank to be removed from the water to be treated, and the CPU 11 trims an image of each flock from image data showing a state of the flocculation basin to generate a trimming image of each flock, calculates a size of each flock and a fractal dimension of the trimming image of each flock on the basis of the trimming image of each flock, and also calculates a sedimentation velocity of each block on the basis of the size of each block and each fractal dimension.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to an information processing device, an information processing method and a program used in water treatment, and a method for controlling water treatment. [Background technology]

[0002] Conventionally, water supply treatment for converting raw water such as river water, dam water, or lake water (hereinafter simply referred to as "raw water") into drinking water has been known. Water supply treatment includes a flocculation step in which a coagulant is injected into the raw water and the raw water is stirred to form flocs by coagulating turbid matter contained in the raw water, and a settling step in which the flocs are settled. In the settling step, flocs with good settling properties settle easily, and flocs with poor settling properties do not settle easily. Therefore, flocs with good settling properties are suitable for water supply treatment, and when the settling properties of flocs are poor, a manager who manages the water supply treatment adds a coagulant to the raw water.

[0003] If a manager managing water supply treatment has a high level of skill in water supply treatment, he or she will observe the formation of flocs and determine the amount of coagulant to be added to the raw water, but if the manager does not have a high level of skill in water supply treatment, the manager will not be able to quickly determine the amount of coagulant to be added to the raw water.In response to this, an information processing device is known that performs deep learning in advance on the relationship between image data showing flocs acquired within a specified time after the coagulant is injected and a category indicating whether or not the flocs have an adverse effect on water supply treatment, so that a manager who does not have a high level of skill in water supply treatment can quickly determine the amount of coagulant to be added to the raw water (see, for example, Patent Document 1).

[0004] When image data showing flocs obtained within a predetermined time after a coagulant is injected is newly input, the information processing device of Patent Document 1 outputs a predicted category indicating whether or not the flocs shown by the input image data will have an adverse effect on water treatment, based on the relationship between image data and categories that have been previously deep-learned. Specifically, the information processing device of Patent Document 1 outputs the feature amount of image data by a CNN (Convolutional Neural Network) model during deep learning or when recognizing newly input image data to output a predicted category. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2020-134284 A Summary of the Invention [Problem to be solved by the invention]

[0006] However, when the feature values ​​of image data are output, the points of interest in the image data are not unified. For example, FIG. 7 is an image diagram visualizing the points of interest in image data when a conventional information processing device outputs the feature values ​​of image data based on CNN, which is deep learning. In some cases, CNN focuses on the entire image data (FIG. 7(a)), and in other cases, CNN focuses on specific flocs reflected in the image data (FIG. 7(b)). The right diagrams in FIG. 7(a) and (b) show the parts captured by CNN as features, with red representing the parts with the highest weight and blue representing the parts with the lowest weight. Therefore, since the basis for the feature values ​​output by CNN is not unified, it is not possible to accurately learn the relationship between image data showing flocs acquired within a specified time after the injection of a flocculant and the category of whether or not the flocs have an adverse effect on water treatment, and therefore, when image data showing flocs is input, there is a risk that it is not possible to accurately output a predicted category of whether or not the flocs have an adverse effect on water treatment based on the image data. In other words, there is a problem in that it is not possible to accurately determine whether or not flocs with good settling properties have been formed in order to perform appropriate water treatment.

[0007] An object of the present invention is to provide an information processing device, an information processing method and program, and a method for controlling water treatment, which are capable of accurately determining whether or not flocs with good settling properties have been formed. [Means for solving the problem]

[0008] In order to achieve the above-mentioned object, an information processing device of the present invention is an information processing device used for drinking water treatment in which flocs grow in a flocculation basin in which turbid matters contained in water to be treated are coagulated to form flocs, and the grown flocs settle in a settling tank and are removed from the water to be treated, and is characterized in comprising: a binarization means for binarizing image data showing a state of the flocculation basin; a generation means for generating a cropped image of each of the flocs by cropping an image of each of the flocs from the binarized image data; and a calculation means for calculating a size of each of the flocs and a fractal dimension of the cropped image of each of the flocs based on the cropped image of each of the flocs, and for calculating the settling velocity of each of the flocs based on the size and the fractal dimension of each of the flocs.

[0009] In order to achieve the above object, an information processing method of the present invention is an information processing method used in drinking water treatment, in which flocs grow in a flocculation basin in which turbid matters contained in the water to be treated are coagulated to form flocs, and the grown flocs settle in a settling tank and are removed from the water to be treated, the information processing method comprising: a binarization step of binarizing image data showing a state of the flocculation basin; a generation step of generating a trimmed image of each of the flocs by cropping an image of each of the flocs from the binarized image data; and a calculation step of calculating a size of each of the flocs and a fractal dimension of the trimmed image of each of the flocs based on the trimmed image of each of the flocs, and calculating a settling velocity of each of the flocs based on the size and the fractal dimension of each of the flocs.

[0010] In order to achieve the above object, the program of the present invention is a program for causing a computer to execute an information processing method used in drinking water treatment, in which flocs are formed in a flocculation basin in which turbid matters contained in the water to be treated are coagulated, and the grown flocs are settled in a settling tank and removed from the water to be treated, the information processing method comprising: a binarization step of binarizing image data showing the state of the flocculation basin; a generation step of generating trimmed images of each of the flocs by cropping images of each of the flocs from the binarized image data; and a calculation step of calculating a size of each of the flocs and a fractal dimension of the trimmed images of each of the flocs based on the trimmed images of each of the flocs, and calculating the settling velocity of each of the flocs based on the size and the fractal dimension of each of the flocs.

[0011] In order to achieve the above object, a control method for water treatment of the present invention is a method for controlling water treatment used in drinking water treatment, in which flocs grow in a flocculation basin in which turbid matters contained in the water to be treated are coagulated to form flocs, and the grown flocs settle in a settling tank and are removed from the water to be treated, the method comprising the steps of: binarizing image data showing a state of the flocculation basin; generating a cropped image of each of the flocs by cropping the image of each of the flocs from the binarized image data; and calculating a size of each of the flocs and a fractal dimension of the cropped image of each of the flocs based on the cropped image of each of the flocs, and calculating a settling velocity of each of the flocs based on the size and the fractal dimension of each of the flocs. Effect of the Invention

[0012] According to the present invention, it is possible to accurately determine whether or not flocs with good settling properties have been formed. [Brief description of the drawings]

[0013] [Figure 1]1 is a block diagram illustrating a schematic configuration of an information processing device according to an embodiment of the present invention. [Diagram 2] 2 is a block diagram illustrating a schematic view of a water treatment system in which image data to be input to a data input unit in FIG. 1 is acquired. [Diagram 3] 2 is a flowchart showing the procedure of a calculation process for calculating the settling velocity of each floc from image data, which is executed by a CPU appearing in FIG. 1. [Figure 4] 4(a) is image data showing a state where a large number of flocs are crowded together when the flocculation pond in FIG. 2 is imaged; FIG. 4(b) is image data obtained by subjecting the image data in FIG. 4(a) to binarization processing; and FIG. 4(c) is image data obtained by trimming a predetermined floc from the image data in FIG. 4(a) based on the position coordinates determined in the image data in FIG. 4(b). [Diagram 5] 10 is a flow chart showing the procedure of a discrimination process for discriminating whether or not the settling properties of flocs in a settling basin are good based on the settling velocities vs of all flocs calculated in S38 of FIG. 3. [Figure 6] 6A and 6B are diagrams used to explain the settling velocity of each floc before and after rainfall calculated by the calculation process of FIG. 3, where FIG. 6(a) shows the settling velocity under normal conditions, and FIG. 6(b) shows the settling velocity under conditions of high turbidity. [Figure 7] These are image diagrams that visualize the focal points of image data when a conventional information processing device outputs the features of the image data based on CNN, which is deep learning. FIG. 7(a) is an image diagram that focuses on the entire image data, and FIG. 7(b) is an image diagram that focuses on a specific flock that appears in the image data. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0014] Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings.

[0015] FIG. 1 is a block diagram showing a schematic configuration of an information processing device 10 according to an embodiment of the present invention.

[0016] The information processing device 10 in Fig. 1 includes a CPU 11 (binarization means, generation means, calculation means, discrimination means), a RAM 12, a ROM 13, an SSD 14, and a data input unit 15, which are connected to each other via an internal bus 16. An external device 17 such as an imaging device or a recording medium is also connected to the data input unit 15, and image data stored in the external device 17 or image data acquired by the external device 17 is input from the data input unit 15 to the information processing device 10. The image data input to the information processing device 10 is stored in the SSD 14. The CPU 11 loads a program stored in the ROM 13 or SSD 14 into the RAM 12, which is the work memory of the CPU 11, and executes it.

[0017] FIG. 2 is a block diagram that shows a schematic diagram of a water treatment system 20 from which image data to be input to the data input unit 15 in FIG. 1 is acquired.

[0018] The water treatment system 20 in FIG. 2 includes a rapid mixing basin 21 in which a coagulant, for example, polyaluminum chloride, is injected into raw water and stirred rapidly to form microflocs, which are nuclei for coagulating turbidity in the raw water, a plurality of flocculation basins 22 in which turbidity is coagulated into microflocs by slowly stirring, and a settling basin 23 (settling tank) in which the flocs settle. The flocculation basin 22 has flocculation basins 22a, 22b, and 22c in this order along the direction from the rapid mixing basin 21 to the settling basin 23. The water to be treated can be transported between the rapid mixing basin 21, the flocculation basins 22a, 22b, and 22c, and the settling basin 23, for example, by using piping. The size of the microflocs, which are nuclei for coagulating turbidity in the raw water and are formed in the rapid mixing basin 21, is usually about 10 μm. Two to five flocculation basins can be installed, and preferably three to four flocculation basins are installed. The flocs in the flocculation basin 22 grow (coarsen) by passing through the flocculation basins 22a, 22b, and 22c in sequence. The flocculation basin 22c is adjacent to the settling basin 23, and supplies the treated water containing the grown flocs to the settling basin 23. In this embodiment, an imaging device 17a is disposed in the flocculation basin 22c as the external device 17, and captures an image of a large number of flocs crowded together in the treated water in the flocculation basin 22c. The location of the imaging device 17a in the flocculation basin 22c is not particularly limited. The amount and rate of coagulant injected into the raw water are recorded in the SSD 14.

[0019] The rapid mixing basin 21 has an agitator 24 which agitates the water in the rapid mixing basin 21, and the flow rate of the water in the rapid mixing basin 21 is controlled to 1.5 m / s or more around the agitator 24. The flocculation basin 22a has an agitator 25 which agitates the water in the flocculation basin 22a at a rotation speed of, for example, 30 rpm, the flocculation basin 22b has an agitator 26 which agitates the water in the flocculation basin 22b at a rotation speed of, for example, 20 rpm, and the flocculation basin 22c has an agitator (not shown) which controls the flow rate of the water in the flocculation basin 22c. The flow rate of the water in the flocculation basin 22c can be controlled by a bypass type or pipe type agitation method which utilizes the force of the water flow without using an agitator. The flow velocities of the flocculation basins 22a, 22b, and 22c are controlled to 0.1-0.8 m / s and are successively decreased. For example, the flow velocity of the water to be treated in the flocculation basin 22a is controlled to 0.6-0.8 m / s around the agitator 25, the flow velocity of the water to be treated in the flocculation basin 22b is controlled to 0.2-0.6 m / s around the agitator 26, and the flow velocity of the water to be treated in the flocculation basin 22c is controlled to 0.1-0.2 m / s. The flow velocity of the water to be treated controlled in each flocculation basin can be determined, for example, from the diameter and rotation speed of the agitator blade.

[0020] The water treatment system 20 has a turbidity meter (not shown) which measures the turbidity of the raw water flowing into the rapid mixing basin 21. The flocculation basin 22c has a water thermometer (not shown) which measures the water temperature in the flocculation basin 22c. The water thermometer (not shown) does not necessarily have to be disposed in the flocculation basin 22c, but may be disposed within the water treatment system 20 and measure the water temperature at the location where it is disposed. Information on the measured water temperature is used to calculate the viscosity coefficient of water, which will be described later.

[0021] The water supply treatment system 20 forms microflocs in the rapid mixing basin 21, forms flocs in the flocculation basin 22 and grows the flocs, and causes the flocs to settle under static conditions in the settling basin 23. Of the flocs formed in the water supply treatment system 20, those that settle and are quickly removed in the settling basin 23, i.e., flocs with good settling properties, are desired. Settling velocity is an index indicating flocs with good settling properties, and if the settling velocity of each floc in the flocculation basin 22c is known, it can be accurately determined based on this whether or not flocs with good settling properties have been formed in the settling basin 23. Therefore, the settling velocity v of each floc reflected in the image data is s The process of directly calculating from image data will be described with reference to FIGS.

[0022] FIG. 3 shows the settling velocity v of each floc calculated from the image data executed by the CPU 11 in FIG. s 10 is a flowchart showing a procedure for a calculation process for calculating

[0023] In the calculation process (information processing method, water treatment control method) of FIG. 3, first, the imaging device 17a captures an image of a large number of flocs crowded in the water to be treated in the flocculation pond 22c to generate image data I (FIG. 4(a)) (S31). The size of the image data I is set according to the resolution of the camera (imaging device), and in this embodiment, the resolution of the camera (imaging device) used is full high-definition (1,920×1,080). The generated image data I is stored in the SSD 14 via the data input unit 15. Next, the CPU 11 performs binarization processing on the image data I (S32). The CPU 11 generates a luminance histogram based on the luminance of each pixel constituting the image data I, and sets a threshold value that is a reference for binarization while referring to the luminance histogram. The CPU 11 colors pixels having a luminance greater than the set threshold value as white and pixels having a luminance less than the set threshold value as black, and generates image data J by performing binarization processing on the image data I, for example, binarization processing using Otsu (FIG. 4(b)). The size of the image data J is the same as the size of the image data I.

[0024] The CPU 11 executes edge detection processing to detect edges from the binarized image data J, for example, using the cv2.findcontours function (S33). The detected edges indicate the edges (contours) of the flocks, and the CPU 11 outputs position coordinates indicating the edges of the flocks. Next, the CPU 11 generates a cropped image K ( FIG. 4(c) ) by cropping a square including the entire white portion surrounded by the edges of the flock and consisting of a minimum amount of black portion, for each flock, based on the output position coordinates (S34, generation step), and crops each of all flocks included in the image data J. Here, in the case where a plurality of white portions surrounded by edges are joined, all the joined white portions are cropped as one flock. The number of flocks included in the image data J may vary depending on the aggregation conditions, but is usually 20 to 100, and the settling velocity v of the flocks is determined based on the output position coordinates. s In order to accurately calculate the settling velocity, the number of flocs included in the image data J is preferably 30 to 50. In order to accurately calculate the settling velocity, the number of flocs included in the image data J is preferably a number such that the flocs do not overlap each other. In the following S35 to S37, a process of calculating the settling velocity of a given floc using a trimmed image K of the floc will be described.

[0025] Since the cropped image K cropped from the image data J is a square, in this embodiment, the size of the block is the length d of one side of the cropped image K, and this length d is calculated by the following formula (1) based on the number of pixels that make up one side of the cropped image K and the resolution of the cropped image K (S35).

[0026]

number

[0027] The size d of the flocs calculated in this embodiment is not particularly limited as long as the flocs can be confirmed in an image, but in the case of grown flocs, it is usually 0.5 to 3.5 mm.

[0028] Next, the CPU 11 calculates the fractal dimension D of the trimmed image K (S36). Here, the fractal dimension D is used for a fractal figure including a figure whose entire figure is similar to the whole figure, and the fractal density ρ f is a dimension related to, and is calculated by, for example, a scale transformation method, a cover method, a box counting method, a visual field expansion method, a rotation radius method, or a density correlation function method. In this embodiment, the fractal dimension D of the trimmed image K is calculated by a box counting method. Specifically, when a plurality of square boxes δ are formed by dividing the trimmed image K into a lattice shape, the fractal dimension D of the trimmed image K is calculated using the following formula (2) in which the number of pixels on one side of the square box δ is the size n of the box δ and the number of boxes δ containing the fractal figure of the flock is N(n).

[0029]

number

[0030] The fractal dimension D of the trimmed image K calculated in this embodiment is usually 1.2 to 2, taking into account good settling properties.

[0031] By the way, the settling velocity v of a particle when it settles in a fluid has been conventionally known as s The Stokes equation (Equation (3) below) based on Stokes' law is known as a formula for calculating the settling velocity v s (cm / s) is the floc diameter d (cm), the gravitational constant g (cm / s 2 ), drag coefficient Cd, floc density ρ f (g / cm 3 ) and the density of water ρ w (g / cm 3 ) parameters. It is known that the drag coefficient Cd satisfies the following formula (4) when the Reynolds number Re, which indicates the ratio of the viscous force and the inertial force of the fluid, is much smaller than 1. Furthermore, the Reynolds number Re is proportional to the settling velocity v sThe relationship between the flow rate (cm / s), floc diameter d (cm), and water viscosity coefficient μ (g / cm s) is given by the following equation (5).

[0032]

number

[0033]

number

[0034]

number

[0035] In addition, the floc density ρ f (g / cm 3 ), the density of water ρ w (g / cm 3 ), floc diameter d (cm), and fractal dimension D. The floc density function (Equation (6) below) derived based on actual measurements is known as a relational expression that shows the relationship between floc density, floc diameter d (cm), and fractal dimension D. Incidentally, Equation (6) includes the constant a (g / cm 3 ) is used. The constant a is 2.5×10 -4 <a<4×10 -4 It depends on the shape of the floc. When polyaluminum chloride is used as the flocculant, a is usually in the range of a = 3.0 × 10 -4 When using other flocculants, the constant a can be set by back-calculating the Stokes' law from the measured settling velocity and narrowing down the appropriate range.

[0036] In this embodiment, the settling velocity of the flocs v s The Reynolds number Re of the settling tank 23 is much smaller than 1, and the water density ρ w is close to 1, the following formula (7) is derived based on formulas (3) to (6). According to the following formula (7), the settling velocity v s (cm / s) are the floc diameter d (cm), the fractal dimension D, and the constant a (g / cm 3), gravitational constant g(cm / s 2 ) and the water viscosity coefficient μ (g / cm s), which is determined by the water temperature. The water viscosity coefficient μ can be set based on the water temperature, for example, by referring to the Science Annals 2023 (page 400).

[0037]

number

[0038]

number

[0039] Returning to FIG. 3, the CPU 11 calculates the floc settling velocity v of the floc in the trimmed image K using the floc diameter d calculated in the process of S35, the fractal dimension D calculated in the process of S36, and the above-mentioned formula (7). s The CPU 11 executes the processes of S35 to S37 for all the trimmed images trimmed from the image data J, and calculates the settling velocities v of all the flocs. s (S38, calculation step), and this process ends.

[0040] Floc settling velocity v s Specifically, for example, if the floc size (floc diameter d) is 1.19 × 10 -1 cm, the fractal dimension D of the cropped image K is 1.67, and the constant a when polyaluminum chloride is used as the flocculant is 3.0×10 -4 g / cm 3 , the gravitational constant g is 980 cm / s 2 When the water temperature is 24.56°C and the viscosity coefficient μ of water is 0.90g / cm s, the settling velocity of the flocs v s is 0.020 cm / s.

[0041] According to the calculation process of Fig. 3, image data I showing a state in which a large number of flocs are crowded in the water to be treated in the flocculation basin 22c is subjected to binarization and edge detection processes, and the image is cropped for each floc to form a cropped image K (S32 to S34). Based on the cropped image K, the floc diameter d (cm) and fractal dimension D of each floc are calculated (S35, S36). It is necessary to form flocs with good settling properties in the flocculation basin 22 and for the flocs to easily settle in the settling basin 23, and the settling velocity v is used as an index showing flocs with good settling properties. s However, the settling velocity of flocs v s floc density ρ f are different, so they are not necessarily proportional to the size of the flocs.

[0042] Therefore, conventionally, the settling velocity of flocs v s was calculated from the distance traveled by the flocs when they moved downward in the vertical direction and the time required for the movement, so the settling velocity v s could only be obtained from the displacement of flocs in the settling tank 23. Correspondingly, the floc density ρ f The fractal dimension D related to the settling velocity v of each floc is calculated based on the cropped image K. s is calculated using the fractal dimension D, so that the settling velocity v of each floc reflected in the image data I is calculated almost simultaneously with the formation of the image data I in the flocculation pond 22c. s can be obtained.

[0043] In addition, conventionally, the settling velocity of flocs v s was measured in the settling basin 23 under stationary conditions. In other words, the water to be treated in the settling basin 23 does not have fluidity. On the other hand, in this embodiment, the water to be treated in the flocculation basin 22c has fluidity due to the influence of being fed to the settling basin 23, so that the flocs in the flocculation basin 22c move in all directions, but the settling velocity v sis calculated based on image data showing the state in which a large number of flocs are crowded in the flocculation basin 22c, so the settling velocity v of the flocs can be calculated regardless of the direction of movement of the flocs. s Therefore, the settling property of flocs in the settling basin 23 can be predicted in advance.

[0044] Figure 5 shows the settling velocity v of all flocs calculated in S38 of Figure 3. s 10 is a flow chart showing the procedure of a discrimination process for discriminating whether the settling properties of flocs in the settling basin 23 are good or not.

[0045] In this embodiment, it is assumed that the surface loading rate V0 of the settling basin 23 is stored in advance in the SSD 14. The surface loading rate V0 is an index showing the settling property of flocs, and a settling velocity v smaller than the surface loading rate V0 is s The flocs having a settling velocity v greater than the surface loading rate V0 do not settle in the settling tank 23 but remain in the treated water. s This indicates that flocs having a surface area of ​​0.01 cm / s settle in the settling basin 23. Therefore, the smaller the surface loading rate V0, the better the settling property of the flocs. The surface loading rate V0 (cm / s) is expressed as 2 ) to the flow rate Q(cm 3 / s) (Equation (8) below). The surface loading rate V0 varies depending on the configuration of the sedimentation basin 23 and the turbidity distribution of the raw water, but is usually 4 to 30 mm / min.

[0046]

number

[0047] In FIG. 5, the CPU 11 calculates the sedimentation velocity v s The number of flocs N1 for which the flocs are calculated is counted (S51), and the settling velocity v sThe CPU 11 counts the number N2 of flocs having the above floc number N1 (S52). Next, the CPU 11 calculates the ratio R of the number N2 of flocs to the number N1 of flocs counted (S53). The calculated ratio R indicates the proportion of flocs that are difficult to settle, and indicates the proportion N2 / N1 of flocs that may flow out of the settling basin 23 (hereinafter referred to as the "flow out rate").

[0048] In order to efficiently remove flocs, the lower the outflow rate, the better, for example, 10% or less, 7% or less, and 5% or less. On the premise that the control standard of the outflow rate is stored in the SSD 14 in advance, the CPU 11 judges whether the actual outflow rate exceeds the control standard of the outflow rate (S54). The control standard of the outflow rate can be, for example, 10%, 8%, 6%, or 4% depending on the accuracy required for the water treatment. When the actual outflow rate exceeds the control standard of the outflow rate as a result of the judgment in S54, the CPU 11 displays on a display device (not shown) connected to the information processing device 10 that the settling property of the flocs is not good, or notifies by voice (S55), and ends this process. Since such a notification allows the user to quickly grasp that the settling property of the flocs is not good, the timing to add a flocculant can be quickly confirmed. On the other hand, when the judgment in S54 shows that the actual outflow rate does not exceed the control standard of the outflow rate, the CPU 11 skips S55 as the settling property of the flocs is good, and ends this process.

[0049] In the discrimination process of Figure 5, the outflow rate is calculated by the settling velocity v s The settling velocity of each floc was calculated using the surface loading rate V0 of the settling basin 23, and s For example, when the residence time of the flocs in the settling basin 23 is one hour, the movement distance of each floc is calculated based on the settling velocity v sand residence time. Of the calculated migration distances of each floc, the number N3 of flocs that have a migration distance shorter than the water depth distance (cm), which is the distance from the bottom of the settling basin 23 to the water surface, is counted. At this time, the ratio S of the number N3 of flocs to the number N1 of counted flocs may be used as the discharge rate. By using this ratio S as the discharge rate, the settling property of flocs can be determined more easily than when the ratio R calculated using the surface loading rate V0 is used.

[0050] According to the discrimination process of FIG. 5, the number N1 of all flocs is counted (S51), and the settling velocity v is smaller than the surface loading rate V0 of the settling basin 23. s The number N2 of flocs having the same value is counted (S52), and the outflow rate, which is the ratio R of the number N2 of flocs to the number N1 of flocs, is calculated (S53). It is then determined whether the calculated outflow rate exceeds the control standard for the outflow rate stored in the SSD 14 (S54). As a result, the settling velocity v of each floc in the flocculation basin 22c calculated by the calculation process of FIG. s Based on this, the outflow rate, which indicates the proportion of flocs that do not easily settle, can be grasped, so that it is possible to accurately determine whether flocs with good settling properties have been formed or not from image data showing a state in which a large number of flocs are crowded together when an image of the flocculation basin 22c is taken.

[0051] FIG. 6 is a diagram used to explain the settling velocity of each floc before and after rainfall calculated by the calculation process of FIG. 3. FIG. 6(a) shows the settling velocity at normal times (turbidity: 12.5 to 20 degrees) before rainfall, and FIG. 6(b) shows the settling velocity at high turbidity times (turbidity: 60 to 142 degrees) after rainfall.

[0052] Conventionally, whether or not flocs with good settling properties have been formed has been judged based on a test using a jar test or an empirical formula using turbidity and alkalinity. Therefore, when it rains and the turbidity of the raw water increases, a flocculant has been mechanically added. In the present embodiment, whether or not flocs with good settling properties have been formed is judged based on the settling velocity v s This is based on the sedimentation velocity v of each floc when rain falls and the turbidity of the raw water increases. s is the settling velocity of each floc before rain, v s In this case, the settling property of the flocs is determined based on the settling velocity of each floc, so that it is possible to determine that there is no need to add a flocculant, thereby preventing unnecessary addition of a flocculant.

[0053] Although the embodiments of the present invention have been described above, the present invention is not limited to these embodiments.

[0054] The present invention can also be realized by supplying a program that realizes one or more of the functions of the above-mentioned embodiments to a system or device via a network or storage medium, and having one or more processors in a computer of the system or device read and execute the program, or by a circuit (e.g., an ASIC) that realizes one or more functions. [Explanation of symbols]

[0055] 10. Information processing device 11 CPU 14 SSD 15 Data Entry Section 17a Imaging device 22 Flocculation pond 23 Sedimentation pond

Claims

1. 1. An information processing device for use in drinking water treatment, in which flocs are formed in a flocculation basin in which suspended solids contained in water to be treated are coagulated, the flocs grow, and the grown flocs are precipitated in a settling tank and removed from the water to be treated, a binarization means for binarizing image data showing the state of the flocculation pond; a generating means for generating a trimmed image of each of the flocks by trimming the image of each of the flocks from the binarized image data; and calculating means for calculating a size of each of the flocs and a fractal dimension of the trimmed image of each of the flocs based on the trimmed image of each of the flocs, and for calculating a settling velocity of each of the flocs based on the size of each of the flocs and the fractal dimension.

2. 2. The information processing device according to claim 1, wherein the calculation means further calculates a runoff rate, which is a ratio of flocs having a settling velocity smaller than a surface loading rate calculated from a flow rate of the water to be treated in the settling tank relative to a surface area of ​​the settling tank, among the flocs for which the settling velocity has been calculated.

3. 2. The information processing apparatus according to claim 1, wherein the calculation means further calculates a runoff rate, which is a ratio of flocs whose settling distance per unit time is shorter than a predetermined distance, to the flocs whose settling velocity has been calculated.

4. 4. The information processing apparatus according to claim 3, wherein the predetermined distance is a vertical length of the settling tank or a depth of the water to be treated in the settling tank.

5. 5. The information processing apparatus according to claim 3, further comprising a determination means for determining whether the outflow rate is 10% or less.

6. 1. An information processing method used in drinking water treatment, in which flocs are formed in a flocculation basin where suspended solids contained in the water to be treated are coagulated, the flocs grow, and the grown flocs settle in a settling tank and are removed from the water to be treated, comprising: a binarization step of binarizing the image data showing the state of the flocculation pond; a generating step of generating a trimmed image of each of the flocks by trimming an image of each of the flocks from the binarized image data; and calculating a size of each of the flocs and a fractal dimension of the trimmed image of each of the flocs based on the trimmed image of each of the flocs, and calculating a settling velocity of each of the flocs based on the size of each of the flocs and the fractal dimension.

7. 7. The information processing method according to claim 6, further comprising the step of: calculating a runoff rate, which is the proportion of flocs having a settling velocity smaller than a surface loading rate calculated from the flow rate of the water to be treated in the settling tank relative to the surface area of ​​the settling tank, among the flocs whose settling velocity has been calculated.

8. 7. The information processing method according to claim 6, further comprising the step of calculating a runoff rate, which is a ratio of flocs whose settling distance per unit time is shorter than a predetermined distance, to the flocs whose settling velocity has been calculated.

9. A program for causing a computer to execute an information processing method used in drinking water treatment, in which flocs are formed in a flocculation basin where suspended solids contained in the water to be treated are coagulated, the flocs grow, and the grown flocs settle in a settling tank and are removed from the water to be treated, comprising: The information processing method includes: a binarization step of binarizing the image data showing the state of the flocculation pond; a generating step of generating a trimmed image of each of the flocks by trimming an image of each of the flocks from the binarized image data; a calculating step of calculating a size of each of the flocs and a fractal dimension of the trimmed image of each of the flocs based on the trimmed image of each of the flocs, and calculating a settling velocity of each of the flocs based on the size of each of the flocs and the fractal dimension.

10. 10. The program according to claim 9, further comprising: calculating, in the calculation step, a flow-out rate, which is the proportion of flocs having a settling velocity smaller than a surface loading rate calculated from a flow rate of the water to be treated in the settling tank relative to a surface area of ​​the settling tank, among the flocs whose settling velocity has been calculated.

11. 10. The program according to claim 9, further comprising the step of calculating a runoff rate, which is a ratio of flocs whose settling distance per unit time is shorter than a predetermined distance, to the flocs whose settling velocity has been calculated.

12. A method for controlling water treatment used in drinking water treatment, in which flocs are formed in a flocculation basin where suspended solids contained in the water to be treated are coagulated, the flocs grow, and the grown flocs settle in a settling tank and are removed from the water to be treated, comprising the steps of: a binarization step of binarizing the image data showing the state of the flocculation pond; a generating step of generating a trimmed image of each of the flocks by trimming an image of each of the flocks from the binarized image data; a calculating step of calculating a size of each of the flocs and a fractal dimension of the trimmed image of each of the flocs based on the trimmed image of each of the flocs, and calculating a settling velocity of each of the flocs based on the size of each of the flocs and the fractal dimension.

13. 13. The method for controlling water treatment according to claim 12, further comprising the step of calculating a flow-out rate, which is the proportion of flocs having a settling velocity smaller than a surface loading rate calculated from the flow rate of the water to be treated in the settling tank relative to the surface area of ​​the settling tank, among the flocs whose settling velocity has been calculated.

14. 13. The method for controlling water treatment according to claim 12, further comprising the step of calculating a runoff rate, which is a ratio of flocs whose settling distance per unit time is shorter than a predetermined distance, to the flocs whose settling velocity has been calculated.

Images