JPS63218217A - Floculator control device - Google Patents

Abstract

PURPOSE:To appropriately form flocs with high reliability, by controlling the number of revolutions of a flocculator based on the settling condition of flocs collectively obtained from the floc diameter distribution obtained at plural sites in a settling basin by floc image information processing. CONSTITUTION:In the control of the number of revolutions of the flocculators 31A, 31B, and 31C in a water purifying plant consisting of a floc settling basin 40 and a flocculating basin 30, plural floc image pickup devices 100 and 200 are provided in the settling basin 40, the floc image obtained at each site is processed 400, and the diameter distribution of the extracted flocs is calculated. The floc settling condition is detected based on the distributions obtained at the plural sites in the settling basin, and the number of revolutions of the flocculator is controlled from the condition. By this method, the formation of the flocs easy to settle can be directly detected, the flocculator can be appropriately controlled at all times, the load on filtration can be kept low, energy is saved, and reliability is improved.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a flocculator control device in a flocculation pond of a water purification plant.

[Conventional technology]

At water treatment plants, the suspended particles in raw water are small in size, so the process is to agglomerate them to form flocs, which are then allowed to settle.

In the rapid mixing pond, when a flocculant, which is a positively charged colloid, is injected into raw water suspended particles, which are negatively charged colloids, the flocculant separates the suspended particles from each other. The knot can be tied into a micro flock. Next, at the floc formation pond.

The micro flocs flow and collide with each other due to the rotation of the flocculator, forming large flocs.

It is known that if the rotation of the flocculator is too strong, the flocs will not grow in size because they will be destroyed by the shear force of the turbulent flow.

On the other hand, if the rotation of the flocculator is too weak, the flocs will not be destroyed by the shear force of the turbulent flow, but since the frequency of collision between flocs will be lower, all the microfloc particles will be destroyed for a limited time. In other words, if the agitation is too weak, even if large flocs can be formed in one part, there will be a large number of microfluorocarbon particles that will not form flocs. Therefore, these micro flocs do not settle in the sedimentation tank but flow into the appropriate area.
This will block suitable land.

Therefore, it is important to maintain proper rotation of the flocculator. However, floc formation is complicated, as it is affected by temperature sPH, etc., and there are cases where flocs cannot be formed properly due to unknown reasons.

Even if the particle size of flocs is large, if the density is low, sedimentation will be slow, and it is necessary to detect the sedimentation status to determine whether flocs are properly formed.

In addition, as described in Japanese Patent Application Laid-open No. 61-64307 as a related technology, the flocs in the sedimentation tank are image recognized, the size of the flocs in the sedimentation tank, the number 9 distribution, and other formation conditions are detected, and the amount of treated water, Methods of controlling flocculant injection are known.

[Problem that the invention seeks to solve]

JP-A No. 61-64307 measures the number and size of flocs at a point where a camera is installed in the sedimentation tank to measure the floc formation status, but does not describe how to understand the floc settling condition. Furthermore, the sedimentation behavior within the sedimentation basin cannot be understood by simply measuring the floc formation status at one point. There is no known method for controlling floc formation by evaluating the quality of floc sedimentation from floc image recognition information.

An object of the present invention is to provide a flocculator control device for forming flocs with good sedimentation properties.

[Means for solving problems]

The above purpose is to provide a means for controlling the rotational speed of a flocculator, a sedimentation basin for settling flocs, a means for converting the flocs in the sedimentation basin into brightness information, an image recognition means for extracting flocs from the brightness information, and an extraction means in a water purification plant. means for calculating the particle size distribution of the flocs, means for detecting the floc settling state based on the floc particle size distribution at a plurality of points in the sedimentation tank, and means for controlling the flocculator rotation speed operating means based on the settling state. This is achieved by being equipped with the following.

[Effect]

The state of the flocs in the sedimentation basin is imaged with an industrial television camera, and the flocs are extracted by image processing the floc image information and the particle size distribution is calculated. The particle size distribution, that is, the floc state, at multiple points in the sedimentation tank is combined to determine the floc settling behavior.

The flocculator rotation speed is controlled so that the floc settling state is within the specified range (normal).

〔Example〕

FIG. 1 is a diagram showing an embodiment of the present invention. In FIG. 1, taken raw water is led to a rapid mixing pond 20 via a landing well 10. The rapid mixing pond 20 is provided with an agitator 21, which agitates the raw water into which the flocculant has been injected from the flocculant injector 22. In the rapid mixing pond 20, suspended fine particles become micro flocs by the action of a flocculant. The flocculators 31A, 31B, 31 for stirring are provided in the flocculation pond 30.
These stirring flocculators are rotated by stirring motors 32A, 32B, and 32G. The stirring paddles 31A, 31B, and 31C are partitioned by rectifying walls 33A, 33B having a large number of holes. The micro flocs flowing from the rapid mixing pond 20 are placed in the floc formation pond.

By being sequentially stirred by the flocculators 31A, 31B, and 31C, the minute flocs collide and coalesce,
It aggregates into flocs 34. The particle size of the flocs 34 gradually increases while passing through the floc formation pond 30. The flocs 34 in the floc formation basin 30 flow into the settling basin 40 and are settled and removed. The supernatant water from which the flocs 34 have been removed flows into the filter basin 50. In the filtration tank 5o, the settling tank 4
The remaining fine flocs that were not removed in step 0 are removed by filtration. The water that has passed through the filtration site 50 is supplied through a drainage pond (not shown), a reservoir (not shown), and the like.

The imaging devices 100 and 200 installed in the sedimentation tank 40 convert images of flocs in the sedimentation tank into brightness information using an industrial television camera (not shown). Controller 110, 210
are respectively imaging devices 10o.

200 and transmits brightness information to the selector 500. The selector selects one of the luminance signals of the imaging devices 100 and 200 and transmits it to the image processing device 400. or,
The selector selects the installation position of the selected imaging device in the sedimentation basin.
Arithmetic device i! Send to 300. The selector first selects the imaging device 100 and transmits the luminance signal to the image processing device 400 and the position information to the arithmetic device 300.
00 monitors the transmitted brightness information (not shown)
At the same time, image processing calculations are performed based on the luminance information to calculate the particle size distribution and average diameter of the flocs, and these calculated values are transmitted to the calculation device 300. The arithmetic device 300 stores the particle size distribution, and at the same time, the selector 50
0 to switch the imaging device from 100 to 200. The selector 500 inputs the signal from the arithmetic device 300, and sends the luminance signal from the imaging device 200 to the image processing bag @400.
The image processing device that transmits the installation position of the imaging device 200 in the sedimentation tank to the calculation device 300 is the input imaging device 200.
Imaging device based on the luminance signal of! The particle size distribution and average diameter of the flocs at the 2oO position are calculated and sent to the processing unit WL300. The calculation device 1300 determines the current sedimentation state of the floc from the particle size distribution of the floc at the position of the imaging device 100, 200.

The flocculant injection device 22 receives the sedimentation state signal calculated by the calculation device 300 and controls the amount of flocculant injection.

Similarly, the flocculator control device 34 is also connected to the arithmetic device 300.
The rotational speed of each of the flocculator stirring motors 32A, 32B, and 32C is calculated in response to the sedimentation state signal from the flocculator stirring motor 32A, 32B, and 32C. Flocculator stirring motor 32A, 32B, 32C
receives the calculated rotational speed signal and operates the flocculator 3.
1A, 31B.

Manipulate the rotation speed of 31C.

FIG. 2 shows a detailed configuration of the imaging device 100.

An industrial television camera (
CCTV) 130 magnifies an image of the state of flocs in the sedimentation basin through an observation window 123 made of a transparent material such as glass. A back screen 121 is provided to accurately recognize flock groups with high contrast.

The screen 121 is installed in front of an observation window 123 via a back screen fixing fitting 122 fixed to an airtight case. The back screen 121 is preferably a dark color in order to accurately recognize white flocks with high contrast.Although not shown in the figure, Ii! ! The airtight case 120 is equipped with a wiper or a cleaning water spray device for cleaning the surfaces of the inspection window 123 and the back screen 121 from dirt.

A plurality of illumination devices 140 are installed to irradiate the flock from multiple directions, giving uniform illuminance to the flock. The controller 110 includes lighting 140 and an industrial television camera 13.
0 control and transmits the brightness signal from the industrial television camera 130 to the selector 500.

FIG. 3 shows the configuration of the image processing bag [400].

Flock luminance information from selector 500 is input to A/D converter 401. When the A/D converter 401 receives a trigger from the timing circuit 408, it converts the luminance signal into, for example, an 8-bit digital signal. The converted digital signal is stored in a multi-level memory 402 having a capacity of, for example, 256 x 256 x 8 bits. The flock image information in the multi-level memory 402 is input to the luminance difference emphasizing circuit 403 as pre-processing, and at the same time the luminance of the flock portion is emphasized.
Background high-frequency noise is cut. Luminance difference enhancement circuit 4
Based on the signal output from 03, the binarization circuit 404 binarizes and extracts the flock portion and stores it in a binary memory 405 having a capacity of 256 x 256 x 1 bits. That is,
The luminance signal of the pixel in row i and column j of the input image signal is expressed as G(
xtj), the binarization threshold is Lt, and the signal after binarization is B
(i.

j), the binarization circuit 404 executes the calculation of the following equation.

When G(IIj)≧Lt, B (i, j) = 1 (
1) When G (ipj)<Lt, B (i, j) = O(
2) As a result, pixels whose image signal a (i, j) is higher than the threshold Lt are recognized as pixels corresponding to flocks and set to the "1" level, and conversely, the parts whose luminance is lower than the threshold Lt are recognized as pixels other than flocks. and becomes the Pa0” level.II
1 A set of pixels represented by the jfi level is recognized as a flock. A size distribution calculation circuit 406 performs a labeling process using a known image processing technique on the floc binary image stored in the binary memory 405, and extracts the number of flocs and the area of each floc from the connectivity thereof. The area of each extracted floc is calculated by calculating the circular equivalent diameter assuming a circle with the same area as this, and by calculating the volume of this circular equivalent diameter, the floc volume distribution is extracted as a particle size distribution. , the zero particle size distribution memory stored in the internal particle size distribution memory is divided into, for example, the following 51 divisions.

D5: 0.0~0.1■ Dt: 0.1~0.2■ Da: 0.2~0.3sm DISO: 4.9~5.0III Dll: 5.0~ ■ Operation count counter 407 is incremented each time the particle size distribution calculation for one screen is completed, and it is determined whether or not the flock image processing is completed for the number of N[strokes. If the number of times is less than eight, a signal is sent to the timing circuit 408 to instruct generation of a trigger to start the A/D. At the end of N times, N screen size particle distribution integrated data is output to the calculation device 500 via the input/output device 409.

A/D converted flock image signal, multi-value memory 402
．． The binary memory 404 and particle size distribution are stored in the input/output device 409.
It can be displayed on the monitor via.

Figure 4 shows the calculation device! 300 configuration is shown.

The signal of the particle size distribution D1 from the image processing device [400 input to the arithmetic device 300 is stored in the particle size distribution memory 301.
The average value and standard deviation Dv of the particle size distribution of 1 are calculated and output to the determination circuit 303. Information on the installation position of the imaging device in the settling tank inputted from the selector 500 is inputted into the normal data memory 306. The normal data memory 306 receives the position information input from the selector 500 and stores the average value Dmo and standard deviation DvO of the normal particle size distribution at that position.
Extract to comparison memory 307. The determination circuit compares the average diameter calculated online with the normal value DIIo, and stores the deviation Eo in the deviation memory Eo.

The processing explained above is performed by an imaging device! By performing this for both Zoo and Zoo 200, the deviation memory includes the imaging device 110
The deviation EO1 of the particle size distribution at the 0 position from the normal value and the deviation El of the position of the imaging device 1200 are stored. These deviations Eo and Es are sent to the sedimentation state detection circuit 305, and the deviations Eo and Es are sent to the sedimentation state detection circuit 305.
The signal is divided into two signals and sent to the flocculant injection device 22 and flocculator control device 34.

FIG. 5 shows an explanation of the sedimentation state signal.

The flocs 43 formed in the floc formation basin 30 are sent to the settling basin 40 and settle therein. The shaded area 41 represents settled flocs. A normal floc sedimentation state means that the flocs are settling in a manner that matches the size of the sedimentation basin, that is,
By the time all the flocs reach the settling tank outlet, they are in a state where they have settled to a depth that does not flow out from the outlet. On the other hand, in the case of C2, which settles too quickly, the size of the settling basin cannot be utilized and the settling ends midway, so there is too much leeway with respect to the settling behavior of normal CO. To maximize the performance of the settling tank, the flocculant can be reduced to bring it closer to the normal state of CO. In addition, in the case of C1, which settles slowly, the flocs flow out, so the amount of flocculant is increased or the flocculator rotational speed is lowered to bring it closer to the normal state CO.

The imaging device can be moved by remote control in two dimensions, up, down, left and right, from the installation location, so by increasing the data at and around key points at each water treatment plant, more accurate sedimentation data can be obtained. can get.

FIG. 6 is a diagram showing the relationship between the flocculant injection amount P and the sedimentation state. When the injection device i22 receives a signal that the sedimentation state is slow, the injection amount P is increased, and on the other hand, when it receives a signal that the sedimentation state is fast, it decreases the injection amount P. The injection amount P is set to a maximum value P, JI, and a minimum value Psi to prevent abnormal injection.

FIG. 7 is a diagram showing the relationship between flocculator rotational speed N and sedimentation state. When the flocculator control device 34 receives a signal that the settling state is slow, the flocculator control device 34 turns on the stirring motors 32A, 32.
B, rotation speed N(A) of 32C.

Operate so that N(B) and N(C) become smaller. Conversely, when the sedimentation state is rapid, the stirring motors 32A, 32B, and 32C are operated to be large in order to reduce the flocs. The rotation speed of the stirring motor is as follows:
, a lower limit N111 is provided. If the flocculator is a Cheebird type, the rotation speed N (A),
Manipulate the ratio of N(B) and N(C).

The stirring motors 32A, 32B, and 32C operate the rotation speeds of the flocculators 31A, 31B, and 31C, and the stirring force in the flocculation pond is controlled so that stable flocs are always formed.

〔Effect of the invention〕

According to the present invention, the formation of flocs with good sedimentation properties, which is the purpose of flocculator control, can be directly detected, so the flocculator can always be appropriately controlled. For this reason,
The load on filtration can be kept low, making it possible to save energy and improve the reliability of water treatment plant maintenance.

[Brief explanation of the drawing]

Fig. 1 is a system diagram showing one embodiment of the present invention, Fig. 2 is a partial detailed view of the embodiment, Figs. 3 and 4 are block diagrams of other embodiments, and Fig. 5 is a block diagram of another embodiment. The explanatory drawings, FIGS. 6 and 7, are explanatory diagrams of the sedimentation state.

Claims (1)

[Claims] 1. In a water purification plant, a means for operating the rotational speed of a flocculator, a sedimentation basin for settling flocs, a means for converting flocs in the sedimentation basin into luminance information, an image recognition means for extracting flocs from the luminance information, and an extracted flocs The present invention is characterized by comprising means for calculating the particle size distribution of the sediment, means for detecting the state of floc sedimentation based on the floc particle size distribution at a plurality of points in the sedimentation tank, and means for controlling the flocculator rotation speed control means based on the sedimentation state. A flocculator control device.