JP2674226B2 - Flock formation control device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION A. Field of Industrial Application The present invention relates to a floc formation controller for aggregating suspended matter in treated water, and more particularly to a paddle for stirring treated water by a feedback control system. .

B. Summary of the Invention In the present invention, a treated water photographing device is installed in the rear part of a flock formation pond, and based on image information from the treated water photographing device, the average particle size of the flocs in the treated water and the fine floc per unit volume By obtaining the amount, obtaining the maximum flock particle size from the rotational strength of the paddle, and controlling the rotational strength of the paddle based on these values, the control accuracy is improved and the paddle control delay time is shortened.

C. Conventional technology Generally, in water purification plants, raw water taken from rivers and lakes is used in floc formation ponds by flocculants (polycalcium chloride (PAC), vantosulfate, etc.) and turbid components (clay, algae, etc.) Are flocculated and removed in a sedimentation basin.

Conventionally, in controlling floc formation, at new water treatment plants, flock formation was performed by a jar test (indoor experiment).
The control relational expression between the raw water turbidity and the coagulant injection rate or the ALT ratio (AL ³⁺ / turbidity) was obtained, and it was used on a real scale from parameters such as the GCT value. On the other hand, in existing water purification plants, floc formation control is performed using the accumulated data.

In such a flock formation control, in order to improve efficiency, a correction term (water temperature of raw water, PH, etc.) is provided in the injection formula or the like to perform control. Further, since this control is a feed-forward control system, it cannot cope with a sudden change in the turbidity of the raw water, and there is a risk that the subsequent filtration basin will be overloaded. Therefore, an aspect in which a feedback control system is incorporated has also been constructed.

FIG. 9 shows an example of paddle type floc formation control adopted in a water purification plant of medium scale or larger.

First, raw water is taken into the landing well 1 and the coagulant is added to the mixing pond 2. Then, in the floc formation pond 3, the paddle 4 is rotated by the rotation speed controller 10 to agitate the treated water to agglomerate the suspended matter components (form flocs).
Then, flocs are settled in the settling tank 5.

Turbidimeters 6 and 7 are attached to the landing well 1 and the sedimentation basin 5, respectively. The coagulant injection controller 8 calculates the coagulant injection amount based on the measured turbidity of the raw water and controls the coagulant injection into the mixing basin 2.

There is also a mode in which the rotational force of each paddle is controlled to a ratio of, for example, 5: 3: 1 to prevent the flocs from being destroyed by the rotational force of the paddle in the subsequent stage (see JP-A-61-111110).

D. Problems to be Solved by the Invention However, in the conventional flock formation control, the turbidity meter 7
Since the feedback control is performed based on the turbidity measured at the outlet of the sedimentation basin 5, the delay time is about 2 to 3 hours, so that there is a problem that the control accuracy deteriorates.

Since the turbidimeters 6 and 7 are light application devices such as a scattered light type and a near infrared light type, they measure all substances that scatter and absorb light such as chromaticity and algae. For this reason, the control system using the turbidimeter has a problem that the error fluctuation is large and the reliability is poor.

In view of these problems, it is an object of the present invention to improve a feedback control system and improve control accuracy in flock formation control.

E. Means and Actions for Solving the Problem In order to achieve the above-mentioned object, the present invention is to add a coagulant to a treated water and mix it, and to mix the treated water from the mixing pond with a paddle. A floc formation pond for forming flocs and a sedimentation pond for precipitating the flocs in the treated water from the floc formation pond, and controlling the turbidity of the treated water flowing out from the sedimentation pond to a set outflow turbidity The controller constitutes the following feedback control system.

That is, the maximum floc particle size calculation unit calculates the maximum flock particle size from the rotational strength of the paddle.

Also, the treated water was taken by a treated water camera installed at the rear of the flock formation pond, the flocs were identified from this image information, and statistical processing was performed by the statistical processing unit to determine the average particle size of the flocs and the fineness per unit volume. Find the amount of flock. Further, the predicted outflow turbidity calculation unit calculates the predicted outflow turbidity from the fine floc amount per unit volume. The legitimacy of these values can be confirmed as described below.

The particle size of flocs is related to turbidity as described later. As a general rule, if the particle size of the flock is large, the turbidity is small, but if there are many flock in the range that can be photographed by the treated water camera, the flock will be overlapped and photographed. There are circumstances such as getting bigger.

Under the circumstances described above, the present invention compares the maximum flock particle size with the average particle size of the flocs, compares the set outflow turbidity with the predicted outflow turbidity, and based on these comparison results, the paddle It enables feedback control of the rotational force of.

G. Examples Examples of the present invention will be described below.

G.1 Outline of Flock Measurement System According to Example FIG. 1 shows an example of paddle type flock formation control adopted in a water purification plant of medium scale or larger.

1 is a landing well, 2 is a mixing pond for adding coagulant to treated water,
Reference numeral 3 is a floc formation pond for forming flocs, and 5 is a sedimentation pond for precipitating flocs.

The landing well 1 is provided with a turbidity meter 6 for measuring the turbidity of raw water. The sedimentation tank 5 has a turbidity meter 7 for measuring runoff turbidity.
Is attached. The coagulant injection controller 8 controls coagulant injection.

A plurality of paddles 4 that stir the treated water are installed in the flock formation pond 3. Reference numeral 9 is a motor for rotating the paddle, and the rotation speed controller 10 controls the motor 9.

At the last stage of the flock formation pond 3, an underwater camera 11 with a cleaning mechanism for photographing treated water is attached. In order to obtain a flowing floc as a completely still image, a fully interlaced camera in an electronic shutter mode is used as the underwater camera 11.

The flock measurement device 12 processes the image from the underwater camera 11 to identify the flock, performs statistical processing to create various data about the flock, and the rotation speed controller 10
To give an operation instruction.

G2. Working principle of the embodiment The following formula is given as a coagulant injection formula capable of feedback control.

D = A · TB ⁿ + B (1) where D is the coagulant injection rate, TB is the raw water turbidity, B is the feedback correction value, and A and n are coefficients.

In addition to equation (1), GC is the most important factor in determining the maximum growth floc particle size in floc formation.
There is a T value. The GCT value is the stirring intensity (G value), the concentration of suspended solids in the raw water (C value), and the stirring time (T value). These factors are also design guidelines for the floc pond.

G.2.1 Outline of GCT value field experiment This section describes the field experiment that was performed to verify the validity of this GCT value.

This experiment was performed using the flock measurement system shown in FIG.

In the floc measuring device 12, the still image from the underwater camera 11 is processed to extract the feature amount, the feature amount is statistically processed, and the flock geometric average particle size (the average particle size in the histogram of the floc volume value and the average particle size by group) is calculated. When is expressed as a logarithmic value, a value that gives a normal distribution), standard deviation, total flock volume ratio per unit volume (FV value), etc. were measured.

And these screen factors (average particle size, number of flocs / screen, FV value) and injection factors (turbidity, ALT ratio), hydraulic factors (stirring strength, stirring time), water quality factors (PH, water temperature, conductivity) ) Etc., the consistency was analyzed by stepwise regression analysis.

That is, the explanatory variable X _1-
The relationship of X _n was extracted and deleted (variable increase / decrease, increase method), applied to the equation (2) as a linear combination equation, and the matching standard partial regression coefficients β _{1 to} β _n were obtained by least-squares approximation.

Y = β ₀ + β ₁ X ₁ + β ₂ X ₂ + ... + β _n X _n + ε (2) In this equation (2), the objective variable Y is the floc mean particle size, and the explanatory variables β _{1 to} β _n are The injection factor, hydraulic factor and water quality factor were taken and analyzed. As a result,
Equation (3) was obtained.

(Floc average particle size) = β ₀ + β ₁ × (number of flocs / screen) + β ₂ × (FV value) + β ₃ × (water intake amount) + β ₄ × (GT value) + β ₅ × (water temperature) × β ₀ (ALT Ratio) (3) As a result of multiple regression analysis assuming the average amount of flocs, which is an image measurement factor, as the maximum growth floc diameter, it was found that all variables were functions of GCT values except for water temperature. From this, it was found that the average particle size of flocs obtained by image measurement is a floc forming factor.

G.2.2 Optimal Flock Formation Control Conditions in Flock Formation The following three points are listed as optimum flock formation control conditions.

 Keep runoff turbidity in sedimentation basin stable and low.

 Reduce coagulant injection volume.

 Maintain maximum growth floc diameter.

Conditions are especially important in the water purification process. This is because if the runoff turbidity becomes high, the burden will increase in the subsequent filter basin.

When flock formation is performed with a stable injection rate formula,
The occurrence of fine flocs is considered as a factor that increases the runoff turbidity of the sedimentation pond. The increase in fine flocs can be confirmed by measuring the average particle size distribution of the above flocs.

FIG. 2 shows the average particle size distribution of flocs. In the figure, X ₁ shows a normal average particle size, X ₂ shows a geometric average particle size, (a) shows a normal probability density distribution, and (b) shows a log-normal probability density distribution.
The normal mean particle size is 1.5488 and the standard deviation is 0.771291 for the normal probability density distribution, and the geometric mean particle size is 1.35858 and the standard deviation is 0.232319 for the log normal probability density distribution. The threshold is 34, TF is 8549, TV is 2.25643, F
The V value is 0.334305 and the number of measurement screens is 11.

As can be seen from this figure, the particle size distribution of flocs is always dispersed, and if the average particle size of flocs is large, the proportion of fine flocs is inevitably reduced. Therefore, the average particle diameter of flocs can be used as the representative diameter of flocs.

G.2.3 Effective density of flocs Figure 3 shows the relationship between the effective density of flocs and the geometric mean particle size of flocs.

The floc effective density and the floc geometric mean particle size have a correlation as shown in FIG. Since the adequacy of the floc geometric average particle diameter has been clarified as described above, the floc effective density relational expression based on the floc geometric average particle diameter can be derived. The flock effective density relational expression serves as a criterion for determining the flock formation state.

A general flock effective density relational expression is shown in Expression (4).
Here, ρ _e is the effective floc density (g / cm ³ ), ds is the diameter of the clock equivalent volume (cm), and a and n are constants determined from the flocculation floc formation conditions.

In image measurement, the total volume of floc (cm ³ /) per unit volume and raw water turbidity (g /
), The effective floc density is calculated. In this case, the flock effective density relational expression is as shown in expression (5). However,
ρ _E is the effective density of flocs (g / cm), X is the volume of flocs by series (cm ³ ), GC is the number of measurement screens, GS is the screen viewing range capacity (), and TB is raw water turbidity (g /). is there.

A feedback experiment was conducted to verify the validity of this equation (5). In this field experiment, in the equation (4), the floc geometric mean particle diameter was used as the representative diameter of the flocs.
HD was replaced with the diameter ds of the floc equivalent volume, and the relationship with the floc effective density ρ _E calculated by the equation (5) was obtained.

 As a result, the relationship shown in the following equation (6) was confirmed.

Generally, the constant a in the equation (4) is about 3 × 10 ⁻⁴ , and the constant n is about 1.3 to 1.4.
0) Tanbo et al., "Study on Aluminum Floc Density (I)"). Therefore, from the result shown in equation (6),
The validity of the relational expression of equation (5) was confirmed.

G.2.5 Calculation of Floc Settling Velocity from Flock Effective Density Next, the calculation of flock settling velocity from flock effective density will be explained. The floc sedimentation velocity is the main cause of the outflow turbidity that corresponds to the above-mentioned floc formation control conditions.

Generally, if the floc diameter, density, shape, and water temperature are known, the floc sedimentation velocity can be calculated using the sedimentation velocity formula of the following equation (7) (see the above-mentioned document). Where W is the flocculation rate (cm / sec), μ is the viscosity coefficient of water (g / cm · sec), g is the acceleration of gravity (cm · sec ² ), K is the shape factor, and 1/1 ⁿ is the dimension. It is a constant (1 cm) ⁿ for correction.

Substituting equation (4) into equation (7) yields the following.

However, there has been no reliable measurement example of the density of individual floc particles. This was the main factor that impeded the rational design of sedimentation ponds.

In this example, in the formula (8), the floc geometric mean particle size HD is used as the diameter ds of the flock equivalent volume, and the floc effective density ρ _e is the density ρ _{E2 of the} formula (6).
The relational expression is derived by using. As a result, the following equation (9) is determined. However, W _G is the floc sedimentation velocity (cm / sec).

W _G = 0.6165 · HD ^0.6502 (9) Generally, in the formula (8), the coefficient (2-n) is 0.6 to 0.7.
It is considered to be about (see the above-mentioned document). This also
The validity of equation (9) can be confirmed.

G.2.6 Relationship between average floc particle size and number of flocs / screen for runoff turbidity Figure 4 shows the relationship between geometrical mean floc particle size and runoff turbidity in a sedimentation basin. The sediment effluent turbidity treated here is that after 6 hours from photography. As can be seen from this figure, the outflow turbidity increases as the floc geometric mean particle size decreases, and the outflow turbidity decreases as the floc geometric mean particle size increases.

Fig. 5 shows the relationship between the number of flocs / screen and the turbidity of sedimentation basin runoff. As can be seen from this figure, when the number of flocks / screen increases, the turbidity of the outflow increases, when the number of flocks / screen decreases, the turbidity of the outflow decreases, and when the number of flocks / screen is 800 or less, the turbidity of the outflow changes almost do not do.

In this way, there is a strong alternating relationship between the average particle size of flock and the number of flock particles / screen, and it can be seen that the turbidity of outflow decreases when the average particle size of flock is large and the number of flock particles / screen is small. . Here, the white turbidity phenomenon of aluminum caused by excessive injection of the coagulant is not taken into consideration.

G.2.7 Prediction of runoff turbidity 1) Surface area load factor Since the relationship between the geometric mean particle size of flocs and the sedimentation basin has been clarified by the equation (9), the sedimentation pond of fine floc particles is based on this equation (9). The effect on runoff turbidity was evaluated.

In this evaluation, a commonly used surface area load factor will be used. The surface area load factor is an index for evaluating how many floc particles are removed when a certain size of floc particles are allowed to flow into a sedimentation tank under certain conditions. The surface area load factor W ₀ (cm / sec) is shown in equation (10). However, Q is the flow rate in the sedimentation pond (cm ³ / sec), A
Is the surface area (cm ² ) of the sedimentation pond.

W ₀ = Q / A (10) The surface area load factor W ₀ has the same dimension as the floc sedimentation velocity. That is, it is a basic numerical value that gives the removal rate of floc particles having a floc sedimentation velocity W smaller than the surface area load factor W ₀ . The criteria for this judgment are as follows.

When W <W ₀ ... Removal rate W / W _{0 When} W ≧ W ₀ ... Removal rate W / 100% Based on this judgment, the removal rate of fine floc particles is determined and shown in Table 1. In Table 1, the surface area load factor W ₀ = 0.05
(Cm / sec) In this table, when the removal rate is less than 100%, that is, when there is a possibility that fine floc particles may flow out, is marked with.

Generally, the flow velocity in the chemical sedimentation basin is 0.66 cm / sec or less, and the flow velocity in the inclined plate sedimentation basin is 1.0 cm / sec or less. In Table 1, it is based on the flow velocity of 0.05 cm / sec.
Although it is about 10 to 1/20, fine flock particles of about 150 μm may flow out.

2) Fine floc amount per unit volume in image measurement Fine floc amount Mg per unit volume in image measurement
(G /) can be calculated by the following equation (11).
Where X _n is the volume of flock volume (cm ³ ) by series of diameter n (50 to 300 μm), G _c is the number of image measurements, G _s is the screen field of view capacity (), ρ _E2 is the average fine flock density (150 μm). Density = 0.074 g / cm ³ ).

3) Predicted runoff turbidity of sedimentation basin obtained from screen measurement When the predicted runoff turbidity T _TB (mg /) of the sedimentation basin is calculated based on surface area load factor W ₀ and floc sedimentation velocity W _G , (12)
Use the formula. However, W G _AVE is the sedimentation velocity of the average fine floc diameter (150 μm sedimentation velocity = 0.04 cm / sec), and W ₀ is the surface load factor (cm / sec) at the time of image measurement.

T _TB = Mg {1- (WG _AVE / W ₀ )} × 1000 (12) In actual image measurement, for example, geometric mean particle size 1.
When Mg = 0.00457, WG _AVE = 0.04, W ₀ = 0.05 at 095 mm,
From equation (12), the predicted outflow turbidity T _TB is 0.914 mg /.

Similarly, with a geometric mean particle size of 1.795 mm, Mg = 0.0004, WG
_{When AVE} = 0.04, W ₀ = 0.05, the predicted runoff turbidity T _TB is 0.1 mg /
Becomes

As can be seen from this result, there is a relationship between the average particle size and the amount of fine flocs. That is, there is dispersion in the floc particle size distribution, and if the floc average particle size becomes large, the proportion of fine flocs inevitably decreases and the outflow turbidity also decreases.

4) Relationship between Paddle Speed and Flock Maximum Growth In the case of a paddle type floc pond, it is possible to control the stirring strength, which is a growth factor of the flocs, by controlling the paddle speed.

Generally, the maximum flock diameter is expressed by the following equation (13) (see the above-mentioned document).

As can be seen from equation (13), the maximum flock diameter is inversely proportional to the paddle rotation speed. In order to maintain stable growth of flocs, the rotation speed may be adjusted appropriately.

Substituting the slope n = 1.3498 (see the equation (6)) of the floc density function obtained from the geometric average particle diameter of the flocs into this equation (13) gives the following equation.

d _max ∝ Nr ^−1.0345 (14) Generally, kp = 1.2 to 1.5 is used.

As described above, it is possible to control the paddle rotation speed of the paddle type flock formation pond from the factors such as effective floc density, floc sedimentation velocity, predicted runoff turbidity, and maximum floc growth obtained by image measurement.

G.3 Paddle Rotational Speed Control According to this Example As described above, for the growth of flocs, the paddle stirring intensity (G value), stirring time (T value), and turbidity concentration (C) of raw water
Value) is relevant. FIG. 6 shows the effect of GCT value on floc growth. In the figure, the horizontal axis represents the raw water turbidity and the vertical axis represents the stirring time. GCT is constant (about 100000
0), for example, if the stirring strength G = 20 and the raw water turbidity C = 10, the time T = 5000 (seconds) required for the formation of flocs. Of these GCT values, only the stirring intensity (G value) of the paddle is suitable for control.

The paddle rotation speed control by image measurement is basically performed by sampling control (at intervals of 30 to 60 minutes).

The flock measuring device 12 performs image measurement at sampling time intervals. In this image measurement, an image is first captured from the underwater camera 11 and binarized. Next, the image is inverted, and processing such as area / volume conversion is performed to extract characteristic quantities such as the particle size and number of flocs.

The extraction of the characteristic amount is repeated a predetermined number of times, and the host computer 13 performs statistical processing.

Then, the maximum flock particle size d _max obtained from the rotational speed of the paddle using the equation (14) is compared with the floc geometric mean particle size HD _AVE obtained by the statistical processing. Furthermore, using the equation (12), the predicted runoff turbidity T _TB obtained from the floc particle size distribution,
The outflow turbidity set value (threshold value) set in advance is compared.

The rotation speed of the paddle is controlled based on these comparison results. In the present embodiment, rather than the determination result of the floc particle size,
The judgment result of runoff turbidity shall be given priority.

 Table 2 shows the criteria for controlling the rotation speed of the paddle.

When the conditions in the table are applicable, the predicted outflow turbidity is higher than the set outflow turbidity, even though the floc geometric mean particle size obtained by image measurement is larger than the maximum flock diameter d _max obtained from the current rotational speed. .

This phenomenon is not practical in view of the particle size distribution of flocs. It is considered that the cause of this phenomenon is that a large amount of flock exists in the visual field range in image measurement and the flock overlaps with each other, so that the flock particle size becomes apparently large.

Therefore, it is necessary to keep the stirring strength (paddle rotation speed) low.

When the condition is satisfied, the maximum flock diameter d _max is small and the predicted outflow turbidity is high. That is, it is considered that flocs have not grown because the current stirring strength is high. Therefore, the paddle rotation speed is reduced.

In addition, if the conditions are met, the predicted outflow turbidity is low, so there is basically no problem, but if the stirring intensity is too low,
Flock may settle in the floc pond.
Therefore, the rotation speed of the paddle is increased to prevent the flocs from settling.

Further, when the condition is satisfied, the particle size of floc is small, but the turbidity of runoff satisfies the condition. Since the turbidity of runoff is given priority, the rotation speed of the paddle will be maintained.

In the case of conditions (1) to (4), the condition may be further divided according to the difference between the threshold value and the control factor, and the amount of change in the paddle rotation speed may be determined stepwise according to the difference between the two. it can.

As mentioned above, there are multiple paddles 4 in the flock formation pond 3.
The agitator is equipped with a control system that keeps all the paddle rotation speeds constant, and a tapered flocculation control method that lowers the rotation speed toward the latter stage (see JP-A-61-111110). ) There is. The latter method maintains a high collision probability by a strong stirring force in the first stage, and weakens the stirring force in the latter stage to avoid re-dissociation and destruction of flocs.

FIG. 7 shows the control range of the paddle rotation speed in the former control method, and FIGS. 8 (a), (b) and (c) show the control range of the paddle rotation speed in the latter control method. In the figure, Ro is a steady rotation speed, R _min is a minimum rotation speed, and R _max is a maximum rotation speed. In the latter control method, for example, the ratio of Ro is first stage: middle stage: second stage = 5: 3: 1.

This embodiment is applicable to any control method,
Within the control range, the rotation speed of each paddle may be controlled based on the above-mentioned control reference.

H. Effect of the Invention As described above, in the flock formation control device according to the present invention, the fine floc amount per unit volume is obtained from the image information in the rear part of the flock formation pond, and the predicted outflow in the sedimentation pond is calculated from this fine floc amount. The turbidity is calculated, and the predicted outflow turbidity is used as a feedback control factor to control the paddle rotation speed, thereby performing the flock formation control.

Therefore, the delay time is about 30 to 60 minutes, and in the case of the feedback control based on the measured value of runoff turbidity in the conventional sedimentation basin, the delay time can be greatly shortened compared to the delay time of 2 to 3 hours. The advantage is that the control accuracy can be improved.

Further, by obtaining the average particle size of the flocs from the image information and using this as a feedback control factor, there is an advantage that the floc formation control can be performed in consideration of hydraulic factors.

Furthermore, the difference between the maximum particle size of flocs and the average particle size of flocs, and the difference between the set outflow turbidity and the estimated outflow turbidity can be obtained to further subdivide the judgment criteria, and more detailed control can be realized. There are advantages.

[Brief description of the drawings]

FIG. 1 is a block diagram showing paddle type flock formation control according to an embodiment of the present invention, FIG. 2 is a distribution diagram of the flock average particle size, and FIG. 3 is a relationship diagram of the flock effective density and the flock geometric average particle size. Fig. 4 shows the relationship between geometric mean particle size of flocs and runoff turbidity of sedimentation basin, Fig. 5 shows the relationship between number of flocs / screen and sediment turbidity of sedimentation basin, and Fig. 6 shows growth of flocs of GCT value. FIGS. 7 and 8 (a), (b) and (c) are explanatory views showing the control range of the rotational speed of the paddle, and FIG. 9 is a block showing conventional paddle type flock formation control. It is a figure. 3 ... Flock formation pond, 4 ... Paddle, 10 ... Rotation speed controller, 11 ... Underwater camera, 12 ... Flock measurement device.

 ─────────────────────────────────────────────────── ─── Continuation of front page (56) Reference JP 62-266107 (JP, A) JP 1-266813 (JP, A) JP 62-214331 (JP, A) JP 61- 111110 (JP, A) JP-A-2-284605 (JP, A)

Claims (1)

(57) [Claims] 1. A mixing basin for introducing and mixing a coagulant into the treated water, a floc forming pond for agitating the treated water from the mixing pond with a paddle to form flocs, and a treated water from the floc forming pond. A maximum floc particle size calculator that calculates the maximum flock particle size from the rotational strength of the paddle in the case of controlling the turbidity of the treated water flowing out from the settling tank to a set turbidity And the treated water camera installed in the rear part of the floc formation pond and the flocs are identified based on the image information from this treated water camera, and the particle size distribution of the flocs in the treated water is obtained. A statistical processing unit that calculates the average particle size of flocs and the amount of fine flocs per unit volume, a predicted outflow turbidity calculation unit that calculates the estimated outflow turbidity from the amount of fine flocs per unit volume, The average particle size and maximum flock particle size of the paddle are compared, and the set outflow turbidity and the predicted outflow turbidity are compared, and a paddle control unit that controls the rotation intensity of the paddle based on these comparison results is provided. A flock formation control device characterized by the above.