A method for calculating the dosage of water treatment system

Abstract

The present invention discloses a method for calculating the dosage of a water treatment system, and the method includes the following steps: S1: taking raw water for beaker experiment, obtaining multiple groups of optimal dosage data under different water turbidity, PK value and temperature; S2: establishing a dosage initial model of quadratic polynomial; S3: given the initial value of the model parameter, using the Gauss-Newton method to iterate and calculate the calculation model parameter; S4: when the model parameter value meets the error accuracy requirement, the model parameter is output, and if it does not meet the requirement, the iteration is continued until the number of iterations is limited; S5: according to the dosage calculation model, the turbidity, temperature and pH value of the current water quality are input, and the dosage required per unit volume can be obtained; S6: according to the concentration of the liquid in the current medicine pool and the water flow in the pipeline, the dosage of the metering pump can be obtained. The present invention can accurately calculate the optimal value of adding coagulant according to the current water quality parameters, which can reduce the water production cost and improve the effluent quality, which is of great significance to the safe and efficient operation of the water plant.

Description

Technical Field

[0001] The invention relates to a method for calculating the dosage of a water treatment system, and belongs to the field of water treatment calculation. Background Art

[0002] The water source of the water plant contains various active colloids and coarse dispersed substances. The large suspended solids and colloidal particles in the water body can generally sink by themselves with the help of gravity and be removed from the water body. However, the suspended solids and colloidal particles with smaller particle sizes in the water body have a certain stability in the water body under the action of Brownian motion, electrostatic repulsion and hydration of the particle surface. Therefore, a certain amount of coagulant must be added to destroy the stability of the fine particles in order to remove them from the water body. The purpose of the coagulation treatment with added drugs is to condense the pollutants into high-quality alum flowers through the action of coagulants, and the alum flowers are separated from the water under the action of gravity. The important issue in the coagulation treatment link is how to timely and accurately determine the current amount of coagulant dosage, so as to put the coagulant required in the coagulation process into the sedimentation tank.

[0003] Dosing is an important part of the water treatment plant to control the turbidity index of the water quality out of the factory. The coagulation effect is related to many factors, such as treatment process, influent turbidity, pH value, temperature, treated water volume, type of coagulant, mixing method, dosage, etc. The coagulation effect is generally adjusted by adjusting the dosage on site to ensure the quality of the effluent water. There is an optimal value for the dosage of coagulant. If it is too little, the coagulation effect is not good, the colloid in the water is not completely destabilized, and the water quality cannot be guaranteed. If it is too much, re-stabilization will occur, which not only deteriorates the effluent effect, but also wastes coagulants. Therefore, finding the optimal value of the coagulant dosage can not only reduce the cost of water production, but also improve the effluent quality, which is of great significance to the safe and efficient operation of the water plant.

[0004] The dosing system is a complex nonlinear system with the characteristics of large hysteresis, strong time-varying and many uncertain factors. In addition, the sampling response and anti-interference capabilities of various automation equipment and instruments in the water treatment system are different. Therefore, it is very difficult to achieve precise dosing control by conventional means. Summary of the invention

[0005] In order to overcome the shortcomings of the above-mentioned prior art, the present invention provides a multi-parameter dosage calculation model, which calculates the accurate dosage of the coagulant based on the flow rate, turbidity, water temperature and pH value, and the water quality parameters of the raw water. The dosing system can output a signal to the frequency converter of the metering pump based on the model calculation result, thereby changing the dosage of the coagulant; this method can achieve precise control of the dosage, thereby reducing the cost of dosing and improving the water quality of the effluent.

[0006] The present invention is realized by the following technical solutions: 1. A method for calculating the dosage of a water treatment system, characterized in that: the specific method steps are as follows:

[0007] Step 1) Beaker experiment

[0008] Take the raw water and measure the water quality parameters of the raw water; then put the raw water into 5 1L beakers respectively, add the coagulant into each beaker according to a certain dosage gradient, stir the beakers fully, let them stand for more than 15 minutes, and measure the water quality parameters of the clarified water in each beaker; select the beaker with the dosage closest to the water quality parameters required for the factory water as the optimal dosage Y;

[0009] If all the water quality parameters of the clarified water do not meet the water quality parameters required for the factory water, repeat the above steps until the best dosage Y is obtained, and record its data as valid data; repeat the above steps again to measure n groups of valid data;

[0010] The water quality parameters are specifically water temperature, turbidity, and pH value, wherein water temperature is denoted as T in degrees Celsius, turbidity is denoted as R in NTU, and pH value is denoted as P;

[0011] Step 2) Obtain the dosage model

[0012] The dosage of each dose is Y i , where i = 1, 2, 3…n, and the water quality parameter of each raw water is X i , where i = 1, 2, 3…n, X i =[R i , T i , P i ] T , based on the effective data of n measurements, the best approximation function Y = f (X) is obtained, which is the best dosage model under the raw water quality parameter X;

[0013] In the best approximation function f(x), X is a three-dimensional variable, the relationship between the dosage and the water quality parameters is nonlinear, and f(X) is a three-variable nonlinear regression model;

[0014] Since the relationship between the dosage Y and the water quality parameters T, R and P is nonlinear, in order to improve the calculation speed, the polynomial regression method is used to take the nonlinear function relationship between Y and T, R and P to the quadratic term, and the polynomial regression function expression of Y is set as:

[0015] Y=a1R+a2R 2 +a3T+a4T 2 +a5P+a6P 2 +a7RT+a8RP+a9TP+a 10 (1)

[0016] make

[0017]

[0018] n sets of sampling data can be expressed in matrix form as:

[0019] Y = AB (2);

[0020] Since the three variables T, R and P are uncorrelated, there is no multicollinearity in the regression model formula (1). The Gauss-Newton iteration method can be used to find the least squares fit and obtain the optimal regression coefficient a1-a 10 ; First construct the minimum square error function:

[0021]

[0022] Make it as small as possible;

[0023] Building the Jacobian Matrix

[0024]

[0025] According to the conventional experience of water treatment plants, the initial value A0 is given as:

[0026]

[0027] According to the following formula (4),

[0028] A k+1 =A k -(J T J) -1 J T e, (4)

[0029] Where k = 0, 1, 2, ..., M; M is the limited number of iterations;

[0030] Calculation can be obtained A k+1 The value of

[0031] Given the allowable error value ε, at the kth iteration,

[0032] Calculate according to the following formula (5)

[0033]

[0034] When satisfied hour,

[0035] If the accuracy requirement is met, the iteration is stopped. k+1 This is the output result.

[0036] After obtaining the regression coefficient A of formula (1), the dosage Y of 1L raw water under specific parameters can be calculated according to formula (1);

[0037] Step 3) Calculate the dosage

[0038] Assuming the raw water flow is Q and the liquid concentration is D, the dosage of the metering pump under the current water quality parameters is:

[0039]

[0040] As a preferred solution of the method for calculating the dosage of a water treatment system described in the present invention: the raw water refers to water that has not been artificially purified and is used as a water supply source.

[0041] Beneficial technical effects of the present invention:

[0042] The present invention provides a multi-parameter dosage calculation model, which calculates the accurate dosage of coagulant according to the combined judgment of the water quality parameters of raw water, such as flow rate, turbidity, water temperature and pH value. The dosing system can output a signal to the frequency converter of the metering pump according to the calculation result of the model, thereby changing the dosage of the coagulant. The method can realize the precise control of the dosage, achieve the purpose of reducing the dosing cost and improving the effluent water quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 It is a flow chart of calculating the dosage of the water treatment system of the present invention. DETAILED DESCRIPTION

[0044] The present invention will be further described below in conjunction with the embodiments and drawings.

[0045] Example 1

[0046] Take the raw water from Linhuan Water Co., Ltd. and conduct a beaker experiment. The beaker capacity is 1L. A total of 30 groups of valid data are obtained. The data are as follows:

[0047]

[0048]

[0049] According to formula (1), the Jacobian matrix is ​​established:

[0050]

[0051] Given the initial value of the coefficient;

[0052] A0=[30.02-0.001-0.00010.020.0050.0010.0010.0001-3.5] T

[0053] Allowable error ε=1*e -3 , the number of iterations is 30.

[0054] According to formula (4), we can calculate:

[0055] A19 =[3.02460.00243-0.00271-0.0004270.02490.0040320.0009630.000386-0.000301-3.3755] T

[0056] When the calculation error meets the requirements, the dosage model is recorded as formula (7) as follows:

[0057] Y=3.0246R+0.00243R 2 -0.00271T-0.000427T 2 +0.0249P+0.004032P 2 +0.000963RT+0.000386RP-0.000301TP-3.3755

[0058] The water plant needs to add chemicals at a certain moment. At this time, the raw water turbidity is 4.7, the water temperature is 20℃, and the pH value is 7.4.

[0059] According to the dosage model formula (7), the dosage of 1L raw water can be calculated as;

[0060] Y=10.77(mg)

[0061] At this time, the raw water flow rate is 60m 3 / min, the concentration of the liquid is 55000g / m 3 , then according to formula (6), the dosage of the metering pump is:

[0062] Z = 11.75 L / min.

Claims

1. A method for calculating the dosage of a water treatment system, characterized in that: The specific steps are as follows: Step 1) Beaker experiment Take the raw water and measure the water quality parameters of the raw water; then put the raw water into 5 1L beakers respectively, add the coagulant into each beaker according to a certain dosage gradient, stir the beakers fully, let them stand for more than 15 minutes, and measure the water quality parameters of the clarified water in each beaker; select the beaker with the dosage closest to the water quality parameters required for the factory water as the optimal dosage Y; If all the water quality parameters of the clarified water do not meet the water quality parameters required for the factory water, repeat the above steps until the best dosage Y is obtained, and record its data as valid data; repeat the above steps again to measure n groups of valid data; The water quality parameters are specifically water temperature, turbidity, and pH value, wherein water temperature is recorded as T in degrees Celsius, turbidity is recorded as R in NTU, and pH value is recorded as P; Step 2) Obtain the dosage model The dosage of each dose is Y i , where i = 1, 2, 3…n, and the water quality parameter of each raw water is X i , where i = 1, 2, 3…n, X i =[R i , T i , P i ] T , based on the effective data of n measurements, the best approximation function Y = f (X) is obtained, which is the best dosage model under the raw water quality parameter X; In the best approximation function f(X), X is a three-dimensional variable, the relationship between the dosage and the water quality parameters is nonlinear, and f(X) is a three-variable nonlinear regression model; Since the relationship between the dosage Y and the water quality parameters T, R and P is nonlinear, in order to improve the calculation speed, the polynomial regression method is used to take the nonlinear function relationship between Y and T, R and P to the quadratic term, and the polynomial regression function expression of Y is set as: <h2 style=";text-align:left;direction:ltr">Y = a1R + a2R<h2 style=";text-align:left;direction:ltr"> 2 <h2 style=";text-align:left;direction:ltr"> + a3 T + a4 T<h2 style=";text-align:left;direction:ltr"> 2 <h2 style=";text-align:left;direction:ltr"> + a5 P + a6 P<h2 style=";text-align:left;direction:ltr"> 2 <h2 style=";text-align:left;direction:ltr"> + a7 RT + a8 RP + a9 TP + a<h2 style=";text-align:left;direction:ltr"> 10 <h2 style=";text-align:left;direction:ltr"> (1) make n sets of sampling data can be expressed in matrix form as: Y = AB (2); Since the three variables T, R and P are uncorrelated, there is no multicollinearity in the regression model formula (1). The Gauss-Newton iteration method can be used to find the least squares fit and obtain the optimal regression coefficient a1-a 10 ; First construct the minimum square error function: Make it as small as possible; Building the Jacobian matrix According to the conventional experience of water treatment plants, the initial value A0 is given as: According to the following formula (4), HAS k+1 = A k -(J T J) -1 I T e, (4) Where k = 0, 1, 2, ..., M; M is the limited number of iterations; Calculation can be obtained A k+1 The value of Given the allowable error value ε, at the kth iteration, Calculate according to the following formula (5) When satisfied When the accuracy requirement is met, the iteration is stopped. k+1 That is the output result; After obtaining the regression coefficient A of formula (1), the dosage Y of 1L raw water under specific parameters can be calculated according to formula (1); Step 3) Calculate the dosage Assuming the raw water flow is Q and the liquid concentration is D, the dosage of the metering pump under the current water quality parameters is:

2. The method for calculating the dosage of a water treatment system according to claim 1, characterized in that: The raw water refers to water that has not been artificially purified and is used as a water supply source.

Images