JPH0640928B2 - Coagulant injection controller for water purification plants - Google Patents

Description

The present invention relates to a coagulant injection control device for a water purification plant, which is configured to inject an appropriate amount of the coagulant to be injected in order to favorably form a block in the water purification plant. .

[Conventional technology]

In a water purification plant, a flocculant is added to the raw water taken up to flocculate suspended substances to form flocculates (hereinafter referred to as flocs), and the flocs are sedimented and removed. Specifically, after injecting the coagulant in the rapid mixing pond, it is introduced into the flocks forming pond and the flocs are formed while gently stirring. Raw water flowing out from the floc formation pond is guided to the sedimentation pond, and sediments the flocs to remove suspended substances. Fine particles that have not settled in the settling tank are removed in the filter tank.

When the water treatment is carried out in this way, if the flocks are not formed in the flocks, the clogging of the filtration pond will be accelerated. It is known to control the injection amount of the flocculant in order to form the block well. Conventionally, for example,
As described in Japanese Patent Laid-Open No. 54-298281, the coagulant injection amount is controlled on the basis of the turbidity of raw water, the particle size and surface area of suspended matter.

On the other hand, as described in, for example, Japanese Patent Publication No. 54-143296, there has been proposed a method of monitoring the shape and size of a block by image processing. Specifically, from a block image taken with an industrial camera or the like, a portion (pixel) brighter than a predetermined brightness (threshold value) is set as a "1" level, and this is recognized as a block, and conversely a predetermined value is detected. A portion (pixel) darker than the value is set as a "0" level and recognized as other than the block. In this way, the block image is binarized and image processing is performed to monitor the state of block formation.

[Problems to be solved by the invention]

Only by controlling the injection amount of the coagulant based on the turbidity of the raw water, the particle size and surface area of the turbidity, the floc formation is affected by the temperature, turbidity, particle size, pH and alkalinity, etc. I cannot guarantee that you can. In other words, since the flocculation state is not directly measured and the coagulant is not injected, the flocculation cannot always be maintained well. On the other hand, although the idea of image recognition of the flocks is well known, it is not known at all how to evaluate the quality of the flocs formation from the recognized images and control the coagulant injection for the flocs formation. For this reason, it is difficult to satisfactorily control the formation of the block based on the image-recognized block image.

An object of the present invention is to provide a coagulant injection control device capable of favorably forming flocs.

[Means for solving problems]

The sedimentation velocity of the flocs was calculated from the floc image information in the sedimentation basin, and the floc was calculated from the total volume of the flocs and the measured turbidity, which was calculated from the floc image information in the floc formation pond, and the floc was injected using the flocculant injection mass. Calculate the density. The floc formation situation is grasped by using the calculated values representing the flocculation characteristics of the flocc density and flocc sedimentation velocity as indexes, and the flocculant is injected.

[Action]

Since the raw water turbidity correlates with the mass of raw water turbidity, the mass of flocs formed in the flocks forming pond can be obtained from the turbidity and the coagulant injection rate. The floc density can be obtained from the mass of this floc. Since the floc density, floc sedimentation rate, and average particle size are related to the flocculation property, a good floc can be formed by grasping these factors and injecting the flocculant.

〔Example〕

 FIG. 1 shows an embodiment of the present invention.

In Fig. 1, raw water taken from rivers etc. is led to a landing well. The turbidity meter 11 measures the turbidity T _u of the raw water flowing out from the landing well 10. The turbidity measurement value T _u is input to the turbidity mass computing device 230. The suspended matter mass calculation device 230 calculates the mass MW of the inflowing suspended matter from the measured turbidity value T _u . The suspended matter mass signal MW is input to the block density calculation device 220.
The rapid mixing pond 20 is provided with a stirrer 21. Further, the quick mixing pond 20 is injected with a coagulant from a coagulant injector 22. The block forming pond 30 is provided with three stirring paddles 31A, 31B, 31C, and these stirring paddles 31A, 31B, 31C are rotationally driven by stirring motors 32A, 32B, 32C, respectively. A straightening wall 33 is provided between the stirring paddles 31A, 31B and 31C.
It is divided by A and 33B. The small flock groups that have flowed into the flock formation pond 30 from the rapid mixing pond 20 sequentially flow down in the flock formation pond 30 and are stirred by paddles 31A, 31.
B and 31C are sequentially stirred. The micro-flock grows into a large flock 34 by colliding and coalescing with each other while flowing down in the flock forming pond 30. in this way,
The particle size of the flock gradually increases while passing through the flock formation pond 30. The flock formation pond 30 has a flock 34
The image pickup apparatus 100 that converts the image of FIG. The imaging device 100 is an industrial television camera (ITV)
Is included. The image processing apparatus 110 performs image processing of the block 34 based on the brightness information obtained by the image pickup apparatus 100, calculates the particle size and particle size distribution of the block, calculates the volume assuming that the block is a sphere, and calculates the volume. The concentration distribution V is calculated. The volume / particle diameter calculation device 210 calculates the total volume V _{t of the} flocks and the average particle diameter from the flocc volume concentration distribution V calculated by the image processing device 110. The total volume V _t of the flocks calculated by the volume / particle size computing device 210 is input to the flot density computing device 220. On the other hand, the coagulant injection rate signal A output from the control computer 200 is input to the coagulant weight calculation device 400. Flocculant weight calculator 4
00 calculates the coagulant injection mass MA and outputs it to the block density calculator 220. Float density calculation device 220
Calculates the floc density ρ from the mass MW of the inflowing turbidity, the total volume V _{t of the} flocs, and the coagulant injection mass MA by the following formula.

ρ = (MW + MA) / V _t (1) The block 34 formed in the block forming pond 30 flows into the settling tank 40 and is settled and removed. In the sedimentation basin 40, an image pickup device 300 for converting the sinking block image into luminance information is installed. The image pickup device 300 has the same structure as the image pickup device 100 in the block formation pond. Imaging device 3
00 has an ITV built-in. Image processing device 110
Calculates the flocculation velocity U based on the luminance signal sent from the image pickup device 300. Flocc settling velocity U is
It is input to the control computer 200 via the particle size calculation device 210. The supernatant water from which the flocs 34 have been removed in the settling tank 40 flows into the filter tank 50. In the filter basin 50, the remaining fine flocks that were not removed in the settling basin 40 are filtered and removed. The water flowing out from the filtration basin 50 is supplied to the customers via a distribution basin (not shown) and a reservoir (not shown).

The control computer 200 calculates the coagulant injection rate A on the basis of the floc density ρ, the floc average particle size and the flocculation velocity U. The coagulant injection machine 22 has a coagulant injection rate A
The coagulant injection amount is controlled by receiving the signal of.

FIG. 2 shows a detailed view of an example of the image pickup apparatus 100 and the image processing apparatus 110.

In FIG. 2, the ITV fixed in the airtight container 120
Reference numeral 130 magnifies and recognizes the image of the block 34 in the block formation pond 30 through the observation window 121 made of a transparent material such as glass by the close-up lens 131. The wiper 122 driven by the wiper driving device 123 is the observation window 12
1 and the back screen 124 are regularly operated to remove dirt from the surface. The back screen 124 is provided for accurately recognizing the block group with high contrast, and is installed in front of the observation window 121 by a back screen fixing device 125 fixed to the airtight container 120. The back screen 124 is preferably a dark color system in order to accurately recognize a white block with high contrast. There are multiple lamps 140 installed,
Irradiate the block 34 in multiple directions. In order to maintain the illuminance of the lamp 140 as a constant condition, the illuminance controller 141 controls the illuminance to be constant at a suitable time according to a change in ambient illuminance. As the lamp 140, an instant light emission type stroboscope or the like can be used in order to accurately recognize the block regardless of the function of the block.

The block brightness information captured by the ITV 130 is sent to the ITV controller 132. ITV controller 132
Also has a function of giving the ITV 130 a timing signal for determining the timing for extracting the luminance information. The ITV controller 130 and the illuminance controller 141 are command-controlled by the imaging device controller 143. The image capturing device controller 143 receives the image processing device 1 via the selector 144.
10 transmits and receives signals. The image control device 160 controls the number of times of block recognition by the image recognition device 150, the interval, and the like. The image processing apparatus 110 is the selector 1
Signals can be transmitted / received to / from a plurality of imaging devices via 44. In the present embodiment, two of the image pickup device 100 in the block formation pond and the image pickup device 300 in the sedimentation pond are connected to the image processing device 110.

FIG. 3 shows details of the image recognition device 150. Selector 1
The ITV luminance signal sent via 44 is processed by the processing selector 15
Input to 1. The selector 151 is the block formation pond 30.
And the brightness signal from the sedimentation basin 40 are branched to a floc particle size distribution calculation processing system and a flot sedimentation velocity calculation processing system, respectively. First, the floating particle size distribution calculation system will be described. The luminance signal is converted into a digital signal by the A / D converter 152A and stored in the image memory 153A. The image memory 153 has, for example, 256 (pixels) × 24.
Image memory 15 with a capacity of 0 (pixels) x 8 bits
The block grayscale image stored in 3A is a binarization circuit 15.
4A, the luminance of the block portion is binarized to "1" level and the background luminance is binarized to "0" level. As I explain in detail later,
The distribution calculation circuit 155A calculates the number of particles, the particle size, and the volume from one block binary image of the image memory 153A, and further calculates the volume concentration distribution V. The above processing is repeatedly executed the number of times set in the processing determination circuit 156A.

Next, the flotation sedimentation velocity calculation system will be described. The block image luminance signal is converted into a digital signal by the A / D converter 152B and stored in the image memory 153B.
The image memory 153B is, for example, 256 × 240 × 8 bit ×
It has a capacity for multiple screens such as 10 screens. The storage of the block image in the image memory 153B is performed at intervals t by the number N of times set in the initial setting circuit 154B of the screen loading process. The N-screen block image stored in the image memory 153B is added to the one-screen image memory 155B after the background brightness is removed. This screen stores the locus of the flocks for N screens. The interval t set in the initial setting circuit 154B is, for example, 0.1 second. The interval t and the ITV observation space are set such that the distance at which the sedimentation tank block sinks during the interval t is about several pixels at the maximum on the image memory. The binarization circuit 156B is the image memory 1
The 55B grayscale image is binarized. That is, the brightness "1" level is given to the block part and the brightness of the background part is set to "0" level to extract the block group. As will be described in detail later, the binarized block image is expanded / reduced by the expansion / contraction circuit 15.
6B, the expansion process is performed a plurality of times, and then the reduction process is performed the same number of times. These processes can connect the loci of one block. In one screen, a plurality of (for example, FN) block loci can be obtained. The thinning circuit 157B thins the floc locus into a line having a width of 1 pixel. One-point noise that does not correspond to the locus is removed by a known image processing technique.
The averaging circuit 158B counts the number of pixels in the pixel row representing the FN block loci, and calculates the average value of the lengths of the block loci from the number of pixels. Since the average value of the lengths of the loci represents the amount of flot sedimentation per initial set time, the sedimentation velocity U can be calculated.

By the way, Fig. 4 shows 2 of the flock image in the flock formation pond.
This is for explaining the function of the digitization circuit 154A.
FIG. 4A shows a block group image recognized by the image pickup apparatus 100. Since the flock group is white, the brightness level is high. On the other hand, since the background black screen 124 is black, the brightness level of water is low. Since the flock group is a grayscale image, the boundary between the flock 34 and water is not clear in reality, but FIG. 4 (a) shows only the outline of the flock group for simplicity. FIG. 4 (b) shows the distribution of the brightness levels scanned by the AA 'line on the screen of FIG. 4 (a). The brightness level is displayed in 128 steps, for example, and the brightness is higher in the vertical and horizontal directions and lower in the lower direction. Since the block 34 is white, the brightness is high. The brightness of the flock group differs depending on the degree of growth of the flock, and the brightness of the grown flock is high and the brightness of the minute flock is low. When the block is binarized and extracted, a fixed threshold value (for example, L _t )
When digitized, growth flock can be recognized, but minute flock cannot be recognized. FIG. 4C shows a luminance distribution when the spatial grayscale original image is subjected to the spatial filtering process for enhancing the luminance difference. As a space filter, for example, 6 lines 6
A matrix of rows is used. After emphasizing the brightness difference, binarization is performed with a constant threshold value L _t to extract the block. For the binarization method, other known image processing techniques such as floating binarization can be used. FIG. 5 shows FIG. 4 (a).
FIG. 3 is a diagram binarized by the line AA ′ in FIG.

Next, the number of pixels of each block (N _i , i = 1, ...
n) to the particle size F _i of each block (i = 1, 2 ... n)
Is calculated by the following formula.

A volume concentration distribution is calculated as a particle size distribution showing the state of flock formation. The volume concentration is a ratio of a volume occupied by a block having a certain particle diameter D _{i to} a unit volume. For example, if the flocs are classified into 50 divisions with a width of 0.1 mm up to 5.0 mm and 51 divisions by adding 5.0 mm or more, the classification is expressed by the following equation.

When i = 1 to 50 0.1 × (i−1) mm <D _i ≦ 0.1 × i mm When i = 51 D ₅₁ > 5.0 mm (4) Individual floc grain size F _i It is determined which classification D _i it belongs to, whether it belongs to each classification D _i, and the volume Vo (D _i ) of the block in each classification D _i is accumulated to calculate the volume concentration distribution of the block. The volume concentration V is calculated by the following formula.

V (D _i ) = Vo (D _i ) / V _t o (5) Here, V _t o is the total volume of the observation space.

Next, the method for calculating the amount of flotation in the sedimentation tank will be described with reference to FIG. FIG. 6A shows an N screen (N = 4) block locus grayscale image added to the image memory 155B. This grayscale image is binarized with a fixed threshold value, and blocks having a certain size or more are binarized and extracted. When the binary image of this block group is expanded and contracted a plurality of times by the expansion / reduction circuit 156B and then reduced, FN binary binary images having loci connected as shown in FIG. 6B are obtained. When this trajectory binary image is thinned by the supplementary line conversion circuit 157B, FN lines each having a width of 1 pixel are obtained as shown in FIG. 6 (c). These represent the length of the floc locus. From the number of pixels P _i (i = 1, 2, ..., FN) representing the locus of each block, the average value L of the length of the locus of the block is calculated by the following equation.

From this, the sedimentation velocity U is obtained by the following equation.

U = L / (N, t) (7) where N is the number of capture screens and t is the capture interval time.

The volume / particle size calculating device 210 is supplied with the flocc volume concentration distribution V and the flocc sedimentation velocity U calculated by the above method. The volume / particle size calculation device 210 has a volume concentration distribution V
Based on the sedimentation velocity U and the total flot volume V _t and the average floe particle diameter are calculated by the following equations.

The average particle size is a representative value that represents the floc volume concentration distribution, and the average particle size can represent the state of flock formation in the flock formation pond.

FIG. 7 shows the floc average particle diameter with respect to the coagulant injection rate A,
The relationship between the sedimentation velocity U and the density ρ is shown. The control computer 200 determines an optimum coagulant injection rate for the plant based on the information obtained by the online measurement. As one example, first, the coagulant injection rate point A at which the sedimentation rate is maximized
Measure _p in advance. Using this A _p as a target value, the flotation rate of the sedimentation basin is measured online and the coagulant injection rate is feed back controlled. Further to fed back control Furotsuku the average particle size is measured online image coagulant injection rate in Furotsuku forming pond Furotsuku average particle diameter D _p which corresponds to the coagulant injection point A _p As another example as the target value.

〔The invention's effect〕

In the present invention, the amount of flocculant injection is controlled by constantly grasping the condition of floc formation in the flocculation pond, raw water turbidity, and flocculation rate of the flocculation pond. In this way, the coagulant injection amount is controlled while directly judging the quality of the actual formation of the flocks, so that the optimal flocs can be formed. For this reason, the load on the sedimentation basin and the filtration basin can be kept low, which in turn makes it possible to save energy and labor for water purification plant maintenance and improve reliability.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention, FIG. 2 is a detailed block diagram showing an example of an image pickup apparatus as an image processing apparatus, and FIG. 3 is a detailed block diagram showing an example of an image recognition apparatus. Figure, fifth
FIGS. 6, 6 and 7 are views for explaining the operation of the present invention. 10 ... Water well, 11 ... Turbidimeter, 20 ... Rapid mixing pond, 22 ... Flocculant injection machine, 30 ... Flot formation pond,
40 ... Settling basin, 100, 300 ... Imaging device, 110
...... Image processing device, 220 …… Float density calculation device,
210 ... Arithmetic unit, 200 ... Control computer,
230: Float weight calculator.

Claims (2)

[Claims] 1. A mixing pond in which raw water is supplied and a flocculant is injected to form a suspended floc in the raw water, a flocculation pond in which the flocs are grown, and a sedimentation pond in which the grown flocs are sedimented and removed. In the coagulant injection control device of a water purification plant having: a first flock imaging means for photographing the state of the flock in the flock formation pond and converting the luminance information into an electric signal; and a luminance state for photographing the state of the flock in the sedimentation pond. To an electric signal, a turbidity measuring means for measuring the turbidity of the raw water, and a particle size and particle size distribution of the flock based on the flock brightness information obtained from the first flock imaging means. The floc volume concentration distribution is calculated to calculate the floc total volume and the floc average particle size, and at the same time, the floc obtained from the second flock imaging means. Image processing means for calculating the floc sedimentation rate based on the brightness information, density calculation means for determining the flock density based on the mass of the influent suspended solid calculated from the raw water turbidity measurement value and the total volume of the floc and the coagulant injection mass, and The flocculation rate is maximized by applying the relationship between the floc density and the coagulant injection rate and the relationship between the floc sedimentation rate and the coagulant injection rate, which are calculated in advance. A coagulant injection control device for a water purification plant, comprising: an injection control unit that determines an injection rate of the agent and controls an injection amount of the coagulant injected into the mixing pond. 2. A coagulant for a water purification plant according to claim 1, wherein a relationship between a floc average particle diameter and a coagulant injection rate is set in advance in the coagulant injection control means. Injection control device.