Water treatment dosing method and dosing system based on alumen ustum detection

Abstract

The invention discloses a water treatment dosing method and dosing system based on alumen ustum detection, relates to the technical field of water treatment, and aims at monitoring equivalent parameters of alumen ustum forms in real time, enabling the alumen ustum forms to be equivalent to spherical particles and ignoring the influence of Brownian motion and fluid shearing on alumen ustum mechanics. Van der Waals force among alumen ustum particles is deduced according to equivalent parameters of alumen ustum forms, and then Zeta potential is deduced. The current metal particle concentration is determined according to the Zeta potential, and the steady-state dosage is adjusted according to the current metal particle concentration. According to the method, the DLVO mechanical field and the electrochemical potential distribution and control theory are fused, a multi-physics field coupling model is constructed, real-time updating of force field parameters under spatio-temporal data driving is achieved, a sensing-decision-execution closed-loop control chain is formed, and the system response speed is remarkably increased.

Description

Technical Field

[0001] The present invention belongs to the technical field of environmental protection, and particularly relates to a water treatment chemical dosing method and a chemical dosing system based on floc detection. Background Art

[0002] In the coagulation sedimentation process, the key to water quality purification lies in the destabilization of colloids and the formation of flocs. Colloidal particles repel each other due to the negatively charged surface, forming a stable suspension system. By adding coagulants (such as Al³⁺, Fe³⁺), the surface charge can be neutralized, causing the colloids to destabilize and aggregate. Coagulation sedimentation is divided into three stages: coagulation, flocculation, and sedimentation. In the coagulation stage, the coagulant neutralizes the surface charge of the colloid, reducing the Zeta potential and destabilizing it; in the flocculation stage, the destabilized colloidal particles collide and combine to form visible flocs, which is affected by factors such as hydraulic conditions, bridging action, and sweep flocculation; in the sedimentation stage, the gravity sedimentation of the flocs and the solid-liquid separation are achieved.

[0003] Flocs are composed of destabilized colloids, hydrolysis products of coagulants, etc., and have a porous network structure, which can further adsorb pollutants. It is necessary to optimize by adjusting to the optimal range or selecting a suitable chemical combination. In addition, monitoring and evaluation methods such as Zeta potential, turbidimeter, floc imaging analysis, and sedimentation rate test are also needed to optimize the operation plan in real time to achieve efficient and low-cost solid-liquid separation. In summary, the destabilization of colloids and the formation of flocs are the core of coagulation sedimentation, and it is necessary to comprehensively control water quality conditions, chemical selection, and hydraulic parameters.

[0004] The coagulation effect has a decisive impact on the subsequent process control. Achieving effective coagulant dosing control can minimize chemical consumption and improve the effluent water quality on the basis of ensuring the optimal dosing amount. However, as the core link of water treatment, coagulant dosing has always been a major challenge in the water production field due to its inherent large lag and nonlinear characteristics. Traditional control methods, such as mathematical model method, streaming current method, simulated filter bed method, etc., have not been widely used due to their respective defects and insufficient reliability. For example, most water plants still adopt a control strategy based on the influent water flow, combined with manual observation of the floc morphology and the turbidity at the outlet of the sedimentation tank to evaluate the dosing effect, mainly relying on the flow ratio to control the dosage of the coagulant. However, this method cannot achieve precise control and real-time monitoring of water quality. The fixed-programmed flow ratio automatic control system, whose test results only reflect the water quality conditions at the sampling moment, has discontinuity and lag in the determined coagulant dosage, making it difficult to achieve optimization and precise control.

[0005] With the rapid development of artificial intelligence technology, especially the wide application of deep learning, underwater particulate matter imaging technology and computer artificial intelligence control provide new ideas for the intelligent control of coagulant dosing. The application of deep convolutional neural network models in fields such as image recognition and object monitoring has been quite mature. It can extract the texture features of floc images for recognition, and through steps such as threshold segmentation and morphological processing, screen out the main image features for standardization processing, and then combine machine learning algorithms such as SVM, BP neural network, and GRNN to judge the alum dosage. However, this method highly depends on high-quality images, is vulnerable to external factors such as light and noise interference, and the feature extraction may not be comprehensive enough. The generalization ability of the machine learning model is also limited, and the response speed of real-time monitoring is relatively slow. Summary of the Invention

[0006] The purpose of the present invention is to provide a water treatment dosing method and dosing system based on floc detection, aiming to solve the above problems. The present invention integrates the DLVO mechanical field, electrochemical potential distribution and control theory, constructs a multi-physical field coupling model, realizes the real-time update of force field parameters driven by spatio-temporal data, forms a closed-loop control chain of perception - decision - execution, and significantly improves the system response speed.

[0007] The present invention is mainly realized through the following technical solutions: A water treatment dosing method based on floc detection includes the following steps: Step S1: Real-time monitor the equivalent parameters of floc morphology; Step S2: Equivalent the floc morphology to spherical particles, and ignore the influence of Brownian motion and fluid shear on floc mechanics. According to the equivalent parameters of floc morphology, deduce the van der Waals force between floc particles F vdW ; Step S3: Deduce the Zeta potential according to the van der Waals force ; Step S4: Determine the current metal particle concentration according to the Zeta potential C Al ; Step S5: Adjust the steady-state dosing amount according to the current metal particle concentration.

[0008] In order to better implement the present invention, further, the step S1 includes the following steps: Step S11: Construct a floc absorbance detection model for a monochromatic light source; Step S12: Construct a floc spatial concentration field model based on three-dimensional space scanning; Step S13: Determine the light source wavelength for multi-band compensation measurement; Step S14: Based on Step S12 and Step S13, construct a multi-wavelength compensated equivalent concentration field model of flocs: , where: C floc ( x , y , z ) is the multi-wavelength compensated equivalent concentration of flocs; λ is the light source wavelength; C λ ( x , y , z ) is the equivalent concentration of flocs determined based on the spatial concentration field model of flocs at the light source wavelength λ ; ξ λ is the equivalent concentration calculation weight of flocs corresponding to the light source wavelength λ ; Step S15: Extract the boundary of the floc geometry based on the concentration threshold τ to obtain the equivalent parameters of the floc morphology.

[0009] To better implement the present invention, further, in the Step S11, the absorbance detection model of the monochromatic light source for flocs is: , where: A sum is the total absorbance; C org is the organic matter concentration of the sample; ε org is the absorbance coefficient of the organic matter in the sample; C floc is the equivalent concentration of flocs; ε floc is the equivalent molar absorptivity of flocs; L floc is the equivalent optical path of flocs; L is the total optical path of the device sample cell.

[0010] To better implement the present invention, further, the Step S12 includes the following steps: Step A1: Data acquisition: Scan in the xoz and yoz planes respectively to measure the absorbance matrices in the x, y, and z directions; Step A2: Projection alignment: Unify the spatial reference system of the projection data through coordinate transformation; Step A3: Fourier reconstruction: Perform a three-dimensional Fourier transform on the projection data, fill the frequency domain space, and then perform an inverse transform to obtain the spatial concentration field model of the floc: , where: C ( x , y , z ) is the equivalent concentration of the floc in the spatial distribution; ε floc is the equivalent molar extinction coefficient of the floc; k x , k y , k z are the frequency domain variables of the Fourier transform of the process data in the x, y, and z directions respectively; C f ( k x , k y , k z ) is the frequency domain of the three-dimensional Fourier transform of the three-dimensional concentration distribution; i is the imaginary unit; e is the natural logarithm.

[0011] To better implement the present invention, further, in the step S13, the floc structure includes a core layer, a hydrophilic colloid coating layer, a microbial aggregation layer, a suspension adsorption layer, and an epitaxial extension layer arranged in sequence from the inside to the outside. Specifically, in the step S13, the light source wavelengths for determining the multi-band compensation measurement include 254 nm, 290 nm, 400 nm, 485 nm, and 890 nm.

[0012] To better implement the present invention, further, the step S15 includes the following steps: Step B1: Define the set of the floc geometric body boundaries through the concentration threshold τ: , where: is the scanned three-dimensional space; ∣ is the screening condition in the set; Step B2: Use the three-dimensional connected component labeling method for the binary isopotential surface S ( x , y ,z ) Perform segmentation and label each connected region as an independent floc individual; Step B3: The dataset of flocs is: , where: p q ( Am s ) is the spatial position information of each geometric body; Am s is the set S ( x , y , z ) the number of 26-connected geometric bodies; q is the counted number of flocs; Step B4: Obtain the equivalent parameters of the floc morphology.

[0013] To better implement the present invention, further, in the step S15, if the floc is a sphere model, the equivalent diameter of the floc is: , if the floc is an ellipsoid model, the equivalent diameter of the floc is: , where: Lx ( i ) is the individual length of the floc in the x direction; Ly ( i ) is the individual length of the floc in the y direction; Lz ( i ) is the individual length of the floc in the z direction.

[0014] To better implement the present invention, further, in the step S5, the steady-state chemical dosage is: , , where: k in is the chemical addition input rate constant; k out is the consumption rate constant; k 0 is the reaction activity coefficient; k ads is the metal ion adsorption equilibrium constant; ρ( t ) is the number density of flocs; D b ( t ) is the equivalent diameter of flocs; K H⁺ is the H⁺ competitive adsorption constant; C H⁺ is the H⁺ concentration; is an intermediate variable: , where: ε is the vacuum permittivity; k D is the length of the Debye reciprocal; F is the Faraday constant; Γ max is the maximum adsorption capacity.

[0015] Preferably, in the step S2, the van der Waals force F vdW is: , , where: r i and r j are the equivalent radii of the floc particles i and the floc particles j respectively; h is the equivalent distance between the surfaces of adjacent floc particles; A is the Hamaker constant; A 0 is the Hamaker constant in pure solvent; α is the ionic shielding coefficient; I ( t ) is the ionic strength; In the step S3, the Zeta potential is: , where: ε is the vacuum permittivity; kD The length of the Debye reciprocal.

[0016] The present invention is mainly implemented through the following technical solutions: A water treatment chemical dosing system based on floc detection, which is used to implement the above-mentioned water treatment chemical dosing method based on floc detection, includes a floc morphology monitoring module, a van der Waals force and Zeta potential calculation module, and a chemical dosing control module; The floc morphology monitoring module is used to monitor the equivalent parameters of the floc morphology in water; The van der Waals force and Zeta potential calculation module is used to calculate the van der Waals force and Zeta potential between floc particles in sequence based on the equivalent parameters of the floc morphology; The chemical dosing control module is used to calculate the current metal particle concentration according to the Zeta potential and adjust the steady-state chemical dosing amount according to the current metal ion concentration.

[0017] In order to better implement the present invention, further, the floc morphology monitoring module includes an absorbance scanning module, a floc spatial concentration processing module, a light source wavelength determination module, a floc equivalent concentration processing module, and a floc morphology parameter extraction module; The absorbance scanning module is used to obtain the absorbance matrix of the floc through three-dimensional scanning; The floc spatial concentration processing module is used to perform Fourier transform processing based on the absorbance matrix to obtain a floc spatial concentration field model; The light source wavelength determination module is used to determine the light source wavelength for multi-band compensation measurement; The floc equivalent concentration processing module is used to obtain the floc equivalent concentration based on the light source wavelength and the floc spatial concentration field model; The floc morphology parameter extraction module is used to extract the boundary of the floc geometric body according to the floc equivalent concentration and obtain the equivalent parameters of the floc morphology.

[0018] The beneficial effects of the present invention are as follows: (1) The present invention uses light sources of different wavelengths to accurately monitor flocs in three-dimensional space, which not only gets rid of the bondage of image quality, but also realizes the comprehensive monitoring of multi-dimensional water quality parameters such as the size, dimension, density, turbidity, and UVCOD of flocs. It has the characteristics of high precision, strong real-time monitoring ability, strong adaptability, intelligence, and automation, and shows significant advantages in accurate monitoring, on-line monitoring, and refined analysis, providing more comprehensive, accurate, and scientific water quality management information support for the accurate and on-line monitoring of flocs in the water treatment process of water plants.

[0019] (2) The present invention has achieved a subversive breakthrough in the modeling and control technology in the field of water treatment, showing three-dimensional innovations: ①Innovatively integrating optical multi-wavelength sensing, DLVO mechanical field, electrochemical potential distribution and control theory, a multi-physical field coupling model is constructed, completely breaking the traditional single-dimensional analysis paradigm; ②Through a dynamic derivation algorithm, real-time update of force field parameters driven by spatio-temporal data is achieved, forming a closed-loop control chain of perception - decision - execution, enabling the system response speed to leap by an order of magnitude of 10 compared with traditional offline optimization; ③At the level of morphological parameter analysis, a fractal dimension intelligent matching wavelength selection mechanism is introduced, and an adaptive optical detection model is established in combination with scattering characteristics, significantly improving the morphological resolution; moreover, in the field of process control, the limitation of PID control is broken through, and the non-linear model predictive control (NMPC) technology is innovatively introduced to precisely regulate the complex non-linear dynamics of the dosing process, locking the dosing accuracy at an ultra-high level of ±5%, and the timing control reaching a millisecond-level response; This series of innovations form a full-chain technology upgrade from microscopic mechanism analysis to macroscopic process control, constituting the key technical support for the intelligent revolution in water treatment. Description of the Drawings

[0020] Figure 1 It is a schematic diagram of the principle of absorbance measurement for multi-band compensation of flocs; Figure 2 It is a flowchart of the water treatment dosing method based on floc detection of the present invention; Figure 3 It is a flowchart of the equivalent parameters for monitoring the morphology of flocs; Figure 4 It is a flowchart of constructing a spatial concentration field model of flocs based on three-dimensional space scanning; Figure 5 It is a flowchart of extracting the boundary of the floc geometry based on the concentration threshold τ; Figure 6 It is a schematic diagram of the principle of the water treatment dosing system based on floc detection of the present invention; Figure 7 It is an architecture diagram of the water treatment dosing system based on floc detection of the present invention. Detailed Description of the Invention

[0021] Example 1: A water treatment dosing method based on floc detection, as Figure 2 shown, includes the following steps: (1) Real-time monitoring of the equivalent parameters of the floc morphology; (2) Deriving the van der Waals force between floc particles according to the equivalent parameters of the floc morphology; (3) Deriving the Zeta potential according to the van der Waals force; (4) Determining the current relative concentration of metal ions according to the Zeta potential; (5) Optimize the amount of chemical addition according to the relative concentration of current metal ions.

[0022] Preferably, as Figure 3 shown, the steps for real-time monitoring of the equivalent parameters of floc morphology include the following: 1) Construct a basic floc measurement model - a floc absorbance detection model with a monochromatic light source; 2) Obtain three-dimensional space scan data and construct a floc spatial concentration field model; 3) Construct the floc structure and adapt the light source wavelength for multi-band measurement according to the floc structure; 4) Construct a floc equivalent concentration field model according to the floc equivalent concentration corresponding to different light source wavelengths; 5) Extract the boundary of the floc geometry based on the concentration threshold τ and determine the equivalent parameters of the floc morphology.

[0023] Preferably, the derivation of the van der Waals force between floc particles specifically includes the following steps: Basic theory: The stability theory of lyophobic colloids is generally known as the DLVO theory. Colloidal particles are attracted to each other by the van der Waals force. When they approach, the double electric layers overlap and generate repulsion. The stability of the colloid depends on the relative magnitudes of the two. This theory calculates the variation of the attractive energy and repulsive energy between particles of different shapes with distance. If the repulsive energy is greater than the attractive energy at a certain distance, the colloid is stable; if the attraction is always greater than the repulsion, the colloid coagulates. When the ion concentration or the valence number of counterions in the solution increases, the van der Waals force remains basically unchanged, the repulsive energy of the double electric layer decreases, and the stability of the colloid decreases until coagulation. According to the dynamic DLVO model theory, the colloidal force field is derived from the dynamic morphology data. Therefore, it is set that: 1. Equivalent the morphology of the floc to spherical particles; 2. Neglect the influence of Brownian motion and fluid shear on the mechanics of the floc in the derivation model; 3. The Hamaker constant is affected by the ionic strength of the solution; Then the van der Waals force is derived according to the equivalent parameters of the floc morphology as:

[0024] Where: F vdW : The van der Waals force acting between floc particles; r i and r j are the equivalent radii of floc particle i and floc particle j respectively; h : The equivalent distance between the surfaces of adjacent floc particles, obtained from the spatial positions of adjacent flocs; A: Hamaker constant:

[0025] Wherein: A 0 : The Hamaker constant in the pure solvent is obtained experimentally; α : Ion shielding coefficient, which can be an empirical coefficient, can be obtained from a table, or can be calibrated experimentally; I ( t ): Ionic strength (unit: mol / m³), which is obtained by collecting and converting with an on-line ion selective electrode; it can also be obtained by an accumulation method.

[0026] Specifically, the method of collecting and converting with an on-line ion selective electrode:

[0027] Wherein: β : Chemical dosing - ionic strength conversion coefficient, obtained from a table or calibrated experimentally; Q ( t ): Real-time chemical dosing amount; I 0 : The previous moment, or the initial ionic strength.

[0028] Preferably, the derivation of the Zeta potential specifically includes the following steps: 1) Set: The double-layer effect can be described by the Debye-Hückel approximation (under the condition of low surface potential). Based on the DLVO theory, the double-layer force formula is:

[0029]

[0030] Wherein: F electric : Double-layer force potential acting force; h : Equivalent distance between the surfaces of adjacent floc particles; k : Equivalent conversion empirical constant; ε : Vacuum permittivity; k D : Length of the reciprocal of Debye (unit: m⁻¹); : zeta potential; e v : Elementary charge; N A : Avogadro constant; k B : Boltzmann constant; T : Absolute temperature (unit: K).

[0031] Specifically, the conversion method of ionic strength is as follows:

[0032] Wherein: C i : Cation detected by the electrode i Molar concentration; Z i : Cation i Corresponding charge number.

[0033] 2) Derive the surface potential through the force balance relationship:

[0034] 3) Substitute into the double-layer potential formula to obtain:

[0035] Solve to get: .

[0036] Preferably, establish the quantitative relationship between the Zeta potential and the concentration of metal ions (such as aluminum salts, iron salts, etc.) , which specifically includes the following steps: Basic model: Stern double-layer model: Divide the double layer into a compact Stern layer and a diffusion layer, and metal ions affect the surface charge through adsorption. Adsorption equilibrium: The adsorption of metal ions in the Stern layer follows the Langmuir isotherm.

[0037] Step 1), According to the charge balance equation of the Stern layer, the surface charge density (Stern layer) can be obtained:

[0038] Wherein: σ Stern : The surface charge density can be obtained from the charge balance equation of the Stern layer; F: Faraday constant; Γ max : Maximum adsorption capacity (mol / m²), obtained from experimental data, empirical data or by looking up tables; kads : The metal ion adsorption equilibrium constant (m³ / mol), obtained from experimental data, empirical data, or by looking up tables; : The metal ion concentration (mol / m³); K H⁺ : The H⁺ competitive adsorption constant (m³ / mol), obtained from experimental data, empirical data, or by looking up tables; C H⁺ : The H⁺ concentration (mol / m³), obtained from an on-line pH sensor; the conversion relationship is: ; Where: PH is the detected value of the on-line pH sensor; Step 2), The correlation between Zeta potential and surface charge, the potential approximation formula:

[0039] Step 3): Substitute the Stern layer charge density:

[0040] Let the intermediate variable be:

[0041] Then:

[0042] Transform to obtain the metal particle concentration:

[0043] Specifically, traverse all the calculated zeta potentials and calculate the average metal particle concentration :

[0044]

[0045] Where: q : The counted number of flocs; : The zeta potentials of two adjacent floc individuals during traversal.

[0046] Preferably, optimizing the amount of chemical addition according to the current relative metal ion concentration specifically includes the following steps: Step 1): Assume that the metal ion concentration is jointly determined by the chemical addition input and flocculation consumption, and obtain from the particle mass balance equation:

[0047] Wherein: k in : Chemical dosing input rate constant (m³ / (mol·s)); Q ( t ): Adjust the chemical dosing amount; k out : Consumption rate constant (s -1 ), related to the floc formation rate;

[0048] k 0 : Reaction activity coefficient (m⁻²·s⁻¹); ρ ( t ): Floc number density (pieces / m³), , where q ( t ) is the number of flocs in the periodic scan result at time t; Dt x : Total length of the x-axis; Dt y : Total length of the y-axis; Dt z : Total length of the z-axis; D b ( t ): Equivalent diameter.

[0049] Step 2), Steady-state chemical dosing calculation When the system is stable then

[0050] Combined with the zeta potential, it is deformed to obtain:

[0051] Wherein: Q steady : Steady-state chemical dosing amount.

[0052] The present invention realizes the multi-dimensional coupling of optical full-spectrum perception, DLVO microscopic mechanical field, electrochemical interface characteristics and cybernetics, constructs a cross-scale digital twin model of multi-physical fields, and completely breaks through the single-dimensional shackles of traditional analysis methods. Secondly, an intelligent decision-making engine driven by spatio-temporal data is constructed through a dynamic derivation algorithm, forming a real-time closed-loop system of "perception - calculation - regulation", which enables the response speed to achieve an order-of-magnitude leap, accelerating more than ten times compared with traditional offline optimization. In the dimension of morphological analysis, a wavelength adaptive selection mechanism driven by fractal dimension is innovatively introduced, and an intelligent optical detection model is constructed by combining dynamic scattering characteristics, resulting in a significant improvement in the resolution of morphological parameters; moreover, a paradigm breakthrough is achieved in the field of process control, replacing traditional PID control with non-linear model predictive control (NMPC) technology to accurately regulate the complex non-linear dynamics of the dosing process, locking the dosing accuracy at a super-high level of ±5% and achieving a millisecond-level response for timing control. This systematic innovation constructs a full-chain intelligent upgrade system from microscopic mechanism understanding to macroscopic process regulation, provides a key technical paradigm for the intelligent transformation of the water treatment industry, and marks the official entry of water treatment technology into a new era of multi-field coupling and intelligent decision-making.

[0053] Example 2: This embodiment is optimized on the basis of Embodiment 1. In step (1), the specific content of constructing the absorbance detection model of the alum flocs for the monochromatic light source is as follows: 1. Monochromatic light measurement principle The interaction between light and matter is mainly manifested in three forms: absorption, reflection, and transmission. When the frequency of the incident light wave matches the natural vibration frequency of the atoms of the substance, energy absorption occurs. When the frequencies do not match, the unabsorbed light wave acts in the form of reflection (the energy of electron vibration is re-radiated on the surface) or transmission (the energy is radiated after passing through the material). This selective action mechanism determines the color characteristics of the substance. In the field of alum flocs detection, the turbidity index reflects the reflection characteristics of particulate matter to light, while the change in transmitted light intensity forms a functional relationship with the morphological characteristics of alum flocs (such as floc density, particle size distribution, and three-dimensional structure). By establishing a mathematical model between the transmission spectrum parameters and the physical properties parameters of the flocs, non-invasive dynamic monitoring of the alum flocs growth process can be achieved. This photo-physical action mechanism provides an important online analysis basis for optimizing the water treatment process.

[0054] 2. Equivalent measurement method During the flocculation sedimentation process, as aggregates of multi-component colloidal particles, the dynamic formation process of flocs is accompanied by complex optical effects. When incident light passes through a suspension system containing flocs, the change in the transmitted light intensity is composed of two parts: (1) the selective absorption of dissolved organic matter at specific wavelengths; (2) the scattering-absorption coupling effect of floc particles on light. Based on the equivalent absorbance theoretical framework of Lambert-Beer's law, the attenuation effect of floc aggregates on transmitted light can be equivalent to an absorption behavior, thereby establishing a synergistic action model for the absorbance of dissolved organic matter (Aorg) and the equivalent absorbance of flocs (Afloc). This model needs to consider the non-linear effects of multi-scale parameters such as floc particle size distribution, porosity, and refractive index difference on the equivalent absorbance.

[0055] Additivity of absorbance: When there are multiple light-absorbing components in a medium, if there is no interaction between the components, the total absorbance of the system is equal to the algebraic sum of the absorbances of each component at the same wavelength. As Figure 1 shown, when measuring absorbance, a light source can be emitted by a light source array, passing through a condenser lens, a flocculation tank, and a condenser lens in sequence and then reaching the detector array.

[0056] 3. Additivity of absorbance and equivalent absorbance model 3.1 Establishment of the total absorbance of monochromatic light Expression of the total absorbance:

[0057]

[0058] Simplify and transform the calculation formula of the total absorbance to obtain:

[0059]

[0060] Among them: A sum : Total absorbance; C org : Concentration of sample organic matter, obtained by three-dimensional space scanning measurement method; ε org : Absorption coefficient of sample organic matter; C floc : Equivalent concentration of flocs; ε floc : Equivalent molar absorption coefficient of flocs; L org : Equivalent optical path of organic matter; L floc : Equivalent optical path of floc.

[0061] L : Total optical path.

[0062] 3.2 Equivalent molar absorption coefficient of floc The equivalent molar absorption coefficient of floc is simplified according to the Mie scattering law:

[0063] Where: Q ext : Mie scattering coefficient; k : Equivalent conversion empirical constant; r : Average equivalent particle length of water body.

[0064] Other parts of this embodiment are the same as those of Embodiment 1, so they will not be described in detail.

[0065] Embodiment 3: This embodiment is optimized on the basis of Embodiment 1 or 2. In step (1), the specific content of constructing the spatial concentration field model of floc is as follows: 1. By the three-dimensional measurement method of scanning the Cartesian coordinate system 1.1 Scanning data to obtain the total absorbance of the monochromatic light source The floc scanning adopts step-by-step plane scanning. In the plane scanning process, the movement scanning process of the x-axis is carried out first, then the movement scanning process of the y-axis is carried out, then the z-axis is moved, and then the plane scanning is repeated. Of course, if conditions permit, a surface light source can be used for one-time transmission to obtain all three-dimensional transmission absorbances at the same time.

[0066] Obtaining the scanning data of the xoz coordinate plane through spatial scanning:

[0067] Obtaining the scanning data of the yoz coordinate plane through spatial scanning:

[0068] Where: l is the number of subdivision steps on the x-axis; n is z the number of subdivision steps on the m is y the number of subdivision steps on the

[0069] 1.2 Orthogonal plane projection constraint When scanning in the xoz and yoz planes, the absorbance distribution can be decomposed into:

[0070] 。

[0071] 2. Construct a spatial concentration field model of floc by means of a spatial field As Figure 4 shown, data acquisition is carried out in sequence: measuring the absorbance matrix in the x, y, and z directions; projection alignment: unifying the spatial reference system of the projection data through coordinate transformation; Fourier reconstruction: performing a three-dimensional Fourier transform on the projection data, filling the frequency domain space and then inverse-transforming to obtain the spatial concentration field model of floc. Specifically, it includes the following steps: 2.1 Fourier transform of single-wavelength projection data 2.1.1 Scan the xoz plane (integrate along the y direction), where the absorbance A x ( z , x ) is expressed as:

[0072] For A x ( z , x ), perform a two-dimensional Fourier transform (variables are x and z):

[0073] Substitute A x ( z , x ) into the formula to obtain:

[0074] It is observed that y is not contained in the integral kernel, and the integral can be split into:

[0075] Since the integral with respect to y is an impulse function δ ( k y ), finally obtain:

[0076] Where: F zx ( k z , k x ) is the two-dimensional Fourier transform of A x ( z , x ); k x , k y , k z : Process data Fourier transform in the frequency domain variables in the x, y, and z directions, where k y = 0; ε floc : Equivalent molar absorptivity of flocs; C ( x , y , z ): Equivalent concentration of flocs in spatial distribution; C f ( k x , k y , k z ): Three-dimensional concentration distribution C ( x , y , z ): Frequency domain of the three-dimensional Fourier transform of ( ; C f ( k x , 0, k z ): Frequency domain mapping of the two-dimensional Fourier transform in the xoz plane to the frequency domain of the three-dimensional Fourier transform; 2.1.2 Scan the yoz plane (integrate along the x direction), where the absorbance A y ( z , y ) is expressed as:

[0077] For A y ( z , y ) perform two-dimensional Fourier transform (variables are y, z):

[0078] Substitute A y ( z , y ) into the formula to get:

[0079] It is observed that y is not contained in the integral kernel, so the integral can be split into:

[0080] Since the integral with respect to x is an impulse function δ ( k x ) Finally, we get:

[0081] where: F zy ( k z , k y ): A y ( z , y ) two-dimensional Fourier transform; C f (0, k y , k z ): The frequency domain mapping of the two-dimensional Fourier transform in the yoz plane to the frequency domain of the three-dimensional Fourier transform; 2.1.3 Filling Rules for Three-Dimensional Fourier Space Through orthogonal plane scanning, the following two planes in the three-dimensional Fourier space C f ( k x , k y , k z ) are filled: ① From the xoz plane scan:

[0082] ② From the yoz plane scan:

[0083] ③ The frequency components in the remaining regions C f ( k x , k y , k z ) are not directly measured. Assuming that the components in the unmeasured regions are zero (i.e., high-frequency information is ignored), then:

[0084] 2.1.4 Three-dimensional inverse Fourier transform Perform an inverse transform on the filled three-dimensional Fourier space to obtain the concentration distribution:

[0085] The other parts of this embodiment are the same as those of Embodiment 1 or Embodiment 2, so they will not be elaborated here.

[0086] Embodiment 4: This embodiment is optimized based on any one of Embodiments 1 - 3. In step (1), the specific content of constructing the equivalent concentration field model of flocs and determining the equivalent parameters of floc morphology is as follows: 1. Determine the final floc morphology and quantity according to the floc data sets corresponding to different wavelengths 1.1 Floc quantity statistics (concentration field) Step a1: Obtain the equivalent concentration fields of different wavelengths through the spatial concentration field model. Of course, different engineers can adapt the concentration fields of different bands to establish an equivalent concentration field model of flocs:

[0087] C floc ( x , y , z ): Equivalent concentration of flocs with multi-wavelength compensation; λ : Wavelength of the light source; C λ ( x , y , z ): Equivalent concentration of flocs determined based on the spatial concentration field model of flocs at the light source wavelength λ ; Preferably, the detection wavelengths for determining the substance composition in the flocs are 254 nm, 290 nm, 400 nm, 485 nm, and 890 nm. Of course, engineers can adapt according to different water bodies. Specifically, obtain the equivalent concentrations at wavelengths of 254 nm, 290 nm, 400 nm, 485 nm, and 890 nm: C 254 ( x , y , z ), C 290 ( x , y , z ), C 400 ( x , y ,z ), C 485 ( x , y , z ), C 890 ( x , y , z ).

[0088] ξ λ : The contribution degree to the equivalent floc concentration at a wavelength of λ is obtained through empirical values or experimental methods: ① Take a certain volume (usually 1 L) of the floc solution and filter it with ultra-fine filter paper (0.015 μm - 3 μm); and weigh the mass as M; ② Dissolve m mass of the flocs in excessive hydrochloric acid, then add excessive caustic soda solution, filter and weigh the mass of aluminum hydroxide m , then the contribution degree of the hydrolysis product of the flocculant: 0 , where is the experimental error coefficient, defaulting to 1; ③ Estimate the mass of different layers in the structure of m mass of the flocs through different experimental methods , and its contribution degree is: . .

[0089] Step a2: As Figure 5 shown, extraction of the floc geometry boundary (isosurface condition): ① Define the set of geometry boundaries through the concentration threshold τ:

[0090] where: S ( x , y , z ): The set of floc geometric surfaces; : The three-dimensional space scanned by the instrument; ∣: The screening condition in the set.

[0091] ② Use the "3D connected component labeling algorithm": Segment the binary isosurface S ( x , y , z ) and label each connected region as an independent floc individual. Connectivity criterion: 26-connectivity (voxels are considered connected if they are diagonally adjacent in space). Count the set S ( x , y, z The number of 26-connected geometric bodies of ().

[0092] ③ Let the function p ( Am s ) be the spatial position information of each geometric body, and let the function of the position information be S ( x , y , z ) Then:[[]] p ( Am s ) = S ( x , y , z ) Where:[[]] p ( Am s ): The spatial geometric body that satisfies the equipotential surface, p ( Am s ) ∈ S ( x , y , z ); S ( x , y , z ): The spatial position function of the spatial geometric body that satisfies the equipotential surface; Am s is the number of 26-connected geometric bodies of the set S ( x , y , z ); ④ The floc dataset to be counted is:

[0093] Where:[[]] p q ( Am s ) is the spatial position information of each geometric body; q : The number of flocs counted.[[]]

[0094] 1.2 Calculation of equivalent parameters of floc morphology 1.2.1 Unit size calculation The step size of the unit subdivision steps on the x-axis: ; The step size of the unit subdivision steps on the y-axis: ; Step size of the z-axis unit subdivision steps: 。

[0095] Where: Δx: Step size of the x-axis unit subdivision step; Δy: Step size of the y-axis unit subdivision step; Δz: Step size of the z-axis unit subdivision step; Dt x : Total length of the x-axis; Dt y : Total length of the y-axis; Dt z : Total length of the z-axis; l : Total number of x-axis subdivision steps; m : Total number of y-axis subdivision steps; n : Total number of z-axis subdivision steps.

[0096] 1.2.2 Parameter calculation (1) The number of flocs corresponding at a certain moment is: q 。 Then the size calculation is: Floc individual length in the x direction: ; Floc individual length in the y direction: ; Floc individual length in the z direction: ; Where: x right is x the right critical value of the floc in the x left is x the left critical value of the floc in the y right is y the right critical value of the floc in the y left is y the left critical value of the floc in the z right is z the right critical value of the floc in the z left is z the left critical value of the floc in the

[0097] (2) The equivalent diameter is: 1) The floc is a spherical model, and the equivalent diameter is:

[0098] 2) The floc is an ellipsoidal model, and the equivalent diameter is:

[0099] (3) The floc density is: 1) The volume of a single floc. For the i-th floc, its volume is the number of voxels in the region multiplied by the voxel volume:

[0100] Where: i : The process data is meaningless; N voxel,i : The i position information function in the S ( x , y , z ) The number of voxels included. 2) The volume density: .

[0101] Other parts of this embodiment are the same as any one of Embodiments 1 - 3, so they will not be elaborated here.

[0102] Embodiment 5: A water treatment chemical dosing system based on floc detection, as Figure 7 shown, includes a main control unit, a photodetector, a light source, a signal acquisition unit, a signal processing unit, and a light source driving unit. The photodetector is connected to the signal acquisition unit, and the signal acquisition unit is connected to an isolation unit and an active unit through the signal processing unit. The main control unit is connected to the light source through the light source driving unit.

[0103] As Figure 6 shown, the main control unit includes a floc morphology monitoring module, a van der Waals force and Zeta potential calculation module, and a chemical dosing control module. The floc morphology monitoring module includes an absorbance scanning module, a floc spatial concentration processing module, a light source wavelength determination module, a floc equivalent concentration processing module, and a floc morphology parameter extraction module.

[0104] The van der Waals force and Zeta potential calculation module is used to calculate the van der Waals force and Zeta potential between floc particles in sequence based on the equivalent parameters of the floc morphology; The chemical dosing control module is used to calculate the current metal particle concentration according to the Zeta potential and adjust the steady-state chemical dosing amount according to the current metal ion concentration.

[0105] The absorbance scanning module is used to obtain the absorbance matrix of alum flocs through three-dimensional scanning; The alum floc spatial concentration processing module is used to perform Fourier transform processing based on the absorbance matrix to obtain the alum floc spatial concentration field model; The light source wavelength determination module is used to determine the light source wavelength for multi-band compensation measurement; The alum floc equivalent concentration processing module is used to obtain the alum floc equivalent concentration based on the light source wavelength and the alum floc spatial concentration field model; The alum floc morphological parameter extraction module is used to extract the boundary of the alum floc geometric body according to the alum floc equivalent concentration and obtain the equivalent parameters of the alum floc morphology.

[0106] Preferably, the system further includes a communication unit, a data storage unit, a motor drive unit, and a motor, all of which are connected to the main control unit. The motor drive unit is used to drive the motor to drive the photodetector to perform scanning measurement in three-dimensional space. The main control unit is connected to a communication device through the communication unit.

[0107] The present invention integrates the DLVO mechanical field, the electrochemical potential distribution and control theory, constructs a multi-physical field coupling model, realizes the real-time update of the force field parameters driven by spatio-temporal data, forms a closed-loop control chain of perception-decision-execution, and significantly improves the system response speed.

[0108] The above are only the preferred embodiments of the present invention, and do not impose any form of limitation on the present invention. Any simple modification or equivalent change made to the above embodiments based on the technical essence of the present invention falls within the protection scope of the present invention.

Claims

1. A water treatment dosing method based on alum flower detection, characterized in that: The following steps are involved: Step S1: real-time monitoring of equivalent parameters of alum flower morphology; Step S2: The morphology of the alum flower is equivalent to spherical particles, and the effects of Brownian motion and fluid shear on the mechanics of the alum flower are ignored. The van der Waals force between the alum flower particles is derived based on the equivalent parameters of the alum flower morphology. F vdW ; Step S3: Derivation of Zeta potential based on van der Waals forces ; Step S4: determining the current metal particle concentration according to the Zeta potential; Step S5: adjusting the steady-state dosage according to the current metal particle concentration.

2. A water treatment dosing method based on alum flower detection according to claim 1, characterized in that: The step S1 comprises the following steps: Step S11: constructing a monochromatic light source alum flower absorbance detection model; Step S12: constructing a model of alum flower spatial concentration field based on three-dimensional spatial scanning; Step S13: determining the wavelength of the light source for multi-band compensation measurement; Step S14: Based on step S12 and step S13, a multi-wavelength compensated alum equivalent concentration field model is constructed: ， in: C floc ( x , y , z ) is the equivalent concentration of alum flowers for multi-wavelength compensation; λ is the wavelength of the light source; C λ ( x , y , z ) is the wavelength of the light source λ The equivalent concentration of alum flowers determined based on the alum flower spatial concentration field model under ; ξ λ is the wavelength of the light source λ The corresponding alum equivalent concentration calculation weight; Step S15: Extract the geometric boundary of the alum flower based on the concentration threshold τ to obtain the equivalent parameters of the alum flower morphology.

3. A water treatment dosing method based on alum flower detection according to claim 2, characterized in that: In step S11, the alum flower absorbance detection model of the monochromatic light source is: ， in: A sum is the total absorbance; C org is the concentration of organic matter in the sample; ε org is the sample organic matter absorption coefficient; C floc is the equivalent concentration of alum flowers; ε floc is the equivalent molar absorption coefficient of alum flower; L floc is the equivalent optical path of the alum flower; L is the total optical path of the sample cell of the device.

4. A water treatment dosing method based on alum flower detection according to claim 3, characterized in that: The step S12 comprises the following steps: Step A1: Data collection: xoz and yoz Plane scanning, measurement x , y , z Absorbance matrix of directions; Step A2: Projection alignment: unify the spatial reference system of the projection data through coordinate transformation; Step A3: Fourier reconstruction: Perform three-dimensional Fourier transform on the projection data, fill the frequency domain space and then perform inverse transform to obtain the alum flower spatial concentration field model: ， in: C ( x , y , z ) is the equivalent concentration of alum flowers in spatial distribution; ε floc is the equivalent molar absorption coefficient of alum flower; k x , k y , k z are the frequency domain variables of process data Fourier transform in x, y and z directions respectively; C f ( k x , k y , k z ) is the frequency domain of the three-dimensional Fourier transform of the three-dimensional concentration distribution; i Is an imaginary unit.

5. A water treatment dosing method based on alum flower detection according to claim 2, characterized in that: In the step S13, it is determined that the wavelengths of the light source for the multi-band compensation measurement include 254 nm, 290 nm, 400 nm, 485 nm and 890 nm.

6. A water treatment dosing method based on alum flower detection according to claim 2, characterized in that: The step S15 comprises the following steps: Step B1: Define the set of alum flower geometry boundaries by the concentration threshold τ: ， in: The three-dimensional space to be scanned; ∣ is the filtering condition in the collection; Step B2: Binarized equipotential surfaces using the 3D connected component labeling method S ( x , y , z ) to segment and mark each connected area as an independent alum flower individual; Step B3: The data set for counting alum flowers is: ， in: p q ( Am s ) is the spatial position information of each geometric body; Am s For collection S ( x , y , z )’s 26-connected geometric bodies; q The number of alum flowers counted; Step B4: Obtain equivalent parameters of alum flower morphology.

7. A water treatment dosing method based on alum flower detection according to claim 6, characterized in that: In step S15, if the alum flower is a spherical model, the equivalent diameter of the alum flower is: ， If the alum flower is an ellipsoid model, the equivalent diameter of the alum flower is: ， in: Lx ( i ) is the individual length of the alum flower in the x direction; Ly ( i ) is the individual length of the alum flower in the y direction; Lz ( i ) is the individual length of the alum flower in the z direction.

8. A water treatment dosing method based on alum flower detection according to claim 1, characterized in that: In step S5, the steady-state dosage is: ， ， in: k in Enter the rate constant for dosing; k out is the consumption rate constant; k 0 is the reaction activity coefficient; k ads is the metal ion adsorption equilibrium constant; ρ ( t ) is the number density of alum flowers; D b ( t ) is the equivalent diameter of the alum flower; K H⁺ is the H⁺ competitive adsorption constant; C H⁺ is the H⁺ concentration; For the intermediate variable: ， in: ε is the dielectric constant of vacuum; k D The length of the Debye reciprocal; F is the Faraday constant; Γ max is the maximum adsorption capacity.

9. A water treatment dosing system based on alum floc detection, used to implement the water treatment dosing method based on alum floc detection according to any one of claims 1 to 8, characterized in that: It includes alum flower morphology monitoring module, van der Waals force and Zeta potential calculation module and dosing control module; The alum flower morphology monitoring module is used to monitor the equivalent parameters of the alum flower morphology in water; The van der Waals force and Zeta potential calculation module is used to sequentially calculate the van der Waals force and Zeta potential between alum flower particles based on equivalent parameters of alum flower morphology; The dosing control module is used to calculate the current metal particle concentration according to the Zeta potential, and adjust the steady-state dosing amount according to the current metal ion concentration.

10. A water treatment dosing system based on alum flower detection according to claim 9, characterized in that: The alum flower morphology monitoring module includes an absorbance scanning module, an alum flower space concentration processing module, a light source wavelength determination module, an alum flower equivalent concentration processing module and an alum flower morphology parameter extraction module; The absorbance scanning module is used for three-dimensional scanning to obtain the absorbance matrix of alum flowers; The alum flower spatial concentration processing module is used to perform Fourier transform processing based on the absorbance matrix to obtain an alum flower spatial concentration field model; The light source wavelength determination module is used to determine the light source wavelength for multi-band compensation measurement; The alum floc equivalent concentration processing module is used to obtain the alum floc equivalent concentration based on the light source wavelength and the alum floc spatial concentration field model; The alum flower morphological parameter extraction module is used to extract the alum flower geometric body boundary according to the alum flower equivalent concentration, and obtain the equivalent parameters of the alum flower morphology.

Images