Intelligent treatment flow control method based on sewage chemical feature recognition

Abstract

The invention relates to the technical field of sewage treatment, and discloses a sewage chemical characteristic identification-based treatment flow intelligent control method, which comprises the following steps of: acquiring inlet water chemical fingerprint spectrum data through a spectrum sensor deployed in a sewage pipe network, and predicting the time when the inlet water reaches a treatment plant; before inflow water arrives, inputting the fingerprint data into a biochemical reaction potential analysis model to obtain a biochemical reaction potential vector, and calculating a hydraulic affinity score of each parallel biochemical treatment unit according to the biochemical reaction potential vector; in combination with sludge function portrait data acquired from the plant, a targeted gain affinity score for deploying a specific return sludge source to a specific biochemical treatment unit is calculated. According to the method, based on the two scores, the optimal flow distribution and sludge allocation strategy is cooperatively determined through optimization solution, so that water quality impact is effectively dealt with, stable operation of the system is ensured, and the impact resistance and stability of the system are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of sewage treatment, and in particular to an intelligent control method for a treatment process based on sewage chemical feature identification. Background Art

[0002] Wastewater treatment is a critical component of environmental protection, and ensuring the stable operation of its treatment processes and the consistent compliance of effluent quality standards is crucial. However, modern wastewater treatment plants, particularly large centralized treatment facilities, have extensive service areas and complex upstream pipe networks, resulting in significant dynamic fluctuations and unpredictability in their influent quality. These plants are particularly susceptible to shock loads caused by events such as sudden industrial discharges.

[0003] Existing control methods typically rely on online monitoring instruments installed at the sewage treatment plant entrance. This approach suffers from a fundamental limitation: the control system can only respond passively once sewage with specific water quality characteristics has reached the plant. Faced with high-flow shock loads, this delayed emergency control often fails to provide sufficient buffer time, potentially leading to treatment unit overload, microbial system disruption, and even temporary effluent quality violations.

[0004] Furthermore, when distributing loads across multiple biochemical treatment units operating in parallel, traditional control strategies often employ a homogenized distribution principle, such as distributing the average flow rate based on the volume of each unit. This approach fails to fully consider the actual differences in the microbial communities within different biochemical treatment units, nor does it assess the compatibility of specific influent water quality with the biochemical reactions of different treatment units. Consequently, when faced with influents with specific chemical characteristics (such as high ammonia nitrogen or recalcitrant organic matter), this "one-size-fits-all" distribution approach fails to precisely direct the load to the unit with the greatest treatment capacity, resulting in a waste of treatment resources and a decrease in overall treatment efficiency.

[0005] Furthermore, existing control logic often treats the allocation of hydraulic load and the allocation of functional microorganisms (such as return sludge) as two independent decision-making processes. Operators may first adjust the flow rate and then, based on experience, determine whether to adjust the sludge return flow. This lacks a mechanism for coordinated optimization. This step-by-step, decoupled decision-making approach can easily lead to suboptimal results. In other words, the optimal solution for flow allocation may not be optimal for sludge resource utilization, and vice versa, thus failing to maximize the overall operational efficiency of the system. Summary of the Invention

[0006] Existing wastewater treatment control methods typically respond to water quality monitoring data within the treatment plant or at the inlet. This control approach suffers from time lags and is difficult to respond to rapidly changing water quality shocks within the sewage network. Furthermore, when a wastewater treatment plant contains multiple parallel treatment units, there is a lack of a technology that can proactively and collaboratively allocate hydraulic loads and schedule bio-augmentation based on the compatibility of the influent chemical characteristics with the various treatment units and microbial populations. This limits the overall stability and efficiency of the treatment system.

[0007] To solve the above technical problems, the present invention provides an intelligent control method for a treatment process based on sewage chemical feature recognition, the method comprising the following steps: S1. Obtain influent chemical fingerprint data collected in real time by a spectral sensor deployed in the sewage pipe network system; the data collected by the spectral sensor can reflect the comprehensive chemical composition of various pollutants such as organic matter, nitrogen, and phosphorus in the influent.

[0008] S2. Based on the influent chemical fingerprint data and a preset pipe network hydraulic transport model, predict the time when the influent carrying the corresponding chemical characteristics arrives at the sewage treatment plant.

[0009] S3. Before the influent reaches the sewage treatment plant, the influent chemical fingerprint data is input into a preset biochemical reaction potential analytical model to obtain a biochemical reaction potential vector (BRP, Biochemical Reaction Potential Vector) that characterizes the potential impact of the influent on subsequent biochemical treatment processes.

[0010] S4. Calculate a hydraulic affinity score (HAS) for each biochemical treatment unit based on the biochemical reaction potential vector and in combination with historical performance data of each biochemical treatment unit operating in parallel in the sewage treatment plant.

[0011] S5. Based on the hydraulic affinity score, determine a distribution ratio for flow distribution of the influent, and control the flow distribution equipment of the sewage treatment plant according to the determined distribution ratio based on the time predicted in step S2.

[0012] In one embodiment, in step S3, the biochemical reaction potential vector BRP includes at least one of the following potential energy components: Aerobic oxygen consumption rate potential (AOUR_P), nitrification inhibition potential (NIP), sludge production potential (SYP) or toxic impact potential (TIP). The biochemical reaction potential analytical model outputs quantified potential energy component values ​​by matching and calculating the influent chemical fingerprint data with a pre-established database.

[0013] Furthermore, the method further comprises: After step S3 and before step S4, sludge functional profile data collected by biological monitoring equipment deployed at each return sludge source in the sewage treatment plant is obtained; The sludge functional portrait data is input into a preset sludge functional pedigree analysis model to obtain a sludge functional pedigree vector (SFP, Sludge Functional Pedigree Vector) representing the functional characteristics of the return sludge source.

[0014] Preferably, the sludge functional profile vector SFP includes at least one of the following functional components: general biological activity (GBA), tolerance to specific toxicity (STR) or specific degradation effect (SDR) on specific refractory organic matter.

[0015] In a specific embodiment, the method further includes, after step S4 and before step S5: Based on the biochemical reaction potential vector (BRP) and the sludge functional spectrum vector (SFP) obtained in step S3, a targeted gain affinity score (TGAS) is calculated for each combination of the biochemical treatment unit and each return sludge source. The targeted gain affinity score is used to quantitatively represent the expected gain effect when sludge from a specific return sludge source is allocated to a specific biochemical treatment unit to treat a specific influent.

[0016] Furthermore, the method further comprises: A targeted return flow strategy is determined based on the targeted gain affinity score; and according to the predicted time obtained in step S2, the return sludge mixing equipment of the sewage treatment plant is controlled according to the targeted return flow strategy to mix the sludge from the designated return sludge source to the designated biochemical treatment unit.

[0017] In a preferred embodiment, the steps of determining the allocation ratio and the targeted reflux strategy are implemented through a unified optimization process, which is specifically as follows: First, an optimization problem is constructed, in which the flow distribution ratio vector of each biochemical treatment unit is and the allocation strategy matrix of each return sludge source to each biochemical treatment unit is the decision variable, where is the number of biochemical treatment units, is the return sludge source index, Index to the biochemical processing unit.

[0018] Then, the objective function of the optimization problem is established The objective function is the weighted sum of the hydraulic affinity score and the targeted gain affinity score, and its mathematical expression is: Maximize ; Subject to the following conditions: ,in, For the The hydraulic affinity score of each biochemical treatment unit, For the The return sludge source and the The target gain affinity score of the biochemical processing unit combination, and is the preset weight coefficient, and + =1.

[0019] Finally, the objective function is maximized by solving the optimization problem , thus satisfying the flow balance constraint ( ) and sludge allocation constraints, and simultaneously obtain the optimal allocation ratio vector and the targeted reflow strategy matrix .

[0020] In one embodiment, the method further comprises: After step S5, for each biochemical treatment unit that receives the allocated flow, the operating parameters of the biochemical treatment unit are dynamically adjusted according to the biochemical reaction potential vector obtained in step S3 and combined with the targeted reflux strategy formulated for the biochemical treatment unit.

[0021] Preferably, the operating parameters include aeration volume, internal reflux ratio or dosage of reagent.

[0022] In a specific embodiment, in step S1, the spectral sensor is an ultraviolet-visible light full spectrum analyzer or a three-dimensional fluorescence spectrum analyzer.

[0023] The present invention provides an intelligent control method for wastewater treatment processes based on chemical feature identification. It has the following beneficial effects: 1. This invention deploys spectral sensors upstream of the sewage pipeline network and combines them with a hydraulic transport model to predict the arrival time of influent water. This advances the time window for control decisions. This overcomes the inherent response lag of traditional control methods that rely on monitoring at the treatment plant inlet. It enables the control system to calculate and deploy response strategies in advance of the arrival of influent water with specific chemical characteristics, thus achieving proactive control of the treatment process.

[0024] 2. This invention achieves non-balanced flow distribution based on the compatibility of water quality and treatment units by analyzing the biochemical reaction potential vector of the influent and calculating hydraulic affinity scores for each parallel treatment unit. This approach precisely directs influent with specific chemical characteristics to the unit with the highest treatment efficiency, improving the utilization of treatment resources and the system's ability to respond to complex water quality changes.

[0025] 3. This invention integrates decision-making at the hydraulic load scheduling and bioresource efficiency levels by constructing an optimization problem with flow distribution ratio and targeted return flow strategy as unified decision variables, and using hydraulic affinity and targeted gain affinity scores as collaborative optimization objectives. This collaborative optimization mechanism generates a globally coordinated control strategy, avoiding suboptimal and conflicting decisions that can result from single, step-by-step decision-making, and improving the overall stability and effectiveness of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 A schematic diagram of an execution environment of a treatment process intelligent control method based on sewage chemical feature recognition according to an embodiment of the present invention; Figure 2 A schematic diagram of a collaborative optimization problem solving process according to an embodiment of the present invention; Figure 3 Schematic diagram showing the comparison of the effects of the method of the present invention on dealing with impact loads. DETAILED DESCRIPTION

[0027] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the present specification. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0028] Please refer to the attached Figure 1 , Figure 1 The figure is a schematic diagram of the execution environment of an intelligent control method for a treatment process based on wastewater chemical signature recognition according to one embodiment of the present invention. The present invention provides an intelligent control method for a treatment process. In a specific embodiment, the method is executed in a system comprising an upstream monitoring unit 100, an in-plant monitoring and execution unit 200, and a central control and computing unit 300. The method specifically includes the following steps: Step 1: Real-time collection of multi-source heterogeneous data. This step aims to obtain comprehensive, real-time operating condition information for subsequent optimization decisions. This step is implemented in the following ways: Collecting influent chemical fingerprint data: Using at least one spectral sensor 110 (such as an ultraviolet-visible full-spectrum analyzer or a three-dimensional fluorescence spectrometer) in the upstream monitoring unit 100 deployed at the upstream pipeline node of the sewage treatment plant, continuous or high-frequency online spectral scanning is performed on the sewage about to enter the plant area, thereby obtaining influent chemical fingerprint data representing its water quality chemical component information.

[0029] Collecting sludge functional profile data: Using biological monitoring equipment 210 deployed at each return sludge source within the plant, the biological activity status of each return sludge is monitored, thereby obtaining sludge functional profile data that characterizes the strength of its specific degradation functions (such as nitrification, denitrification, anti-toxicity, etc.).

[0030] The two types of data collected are sent to the central control and computing unit 300 for processing in real time through their respective data transmission modules.

[0031] Step 2: Collaborative optimization decision-making based on data analysis. This step is the core of the method of the present invention and is completed by the processor in the central control and computing unit 300 executing a preset program. This step receives the above-mentioned collected data and performs a series of calculations and decisions, specifically including: Analyze affinity score: call the biochemical reaction potential analysis model to analyze the influent chemical fingerprint data into a quantitative hydraulic affinity score HAS; at the same time, call the sludge functional spectrum analysis model to combine the sludge functional portrait data with the influent characteristics to analyze it into a quantitative targeted gain affinity score TGAS.

[0032] Construct and solve the optimization problem: Based on the parsed affinity score, execute a decision optimization module. The core task of this module is to construct and solve a mixed integer linear programming (MILP) problem, that is, to find a set of parameters that can maximize the total objective function under the premise of satisfying a series of physical and process constraints. The optimal decision variable to maximize (flow distribution ratio vector) and (Sludge allocation strategy matrix).

[0033] Step 3: Execution of coordinated control instructions based on the optimal decision. This step aims to transform the abstract mathematical solution calculated in step 2 into precise control of physical devices. The central control and computing unit 300 converts the optimal solution into and The data is converted into specific equipment control instructions and sent to the factory monitoring and execution unit 200. The unit performs the following coordinated actions according to the instructions: Execute flow distribution: drive the flow distribution device 220 (such as a PLC-controlled adjustable valve) installed on the water inlet pipeline of each parallel biochemical pool to accurately adjust its opening so that the actual water inlet flow ratio of each unit strictly follows the optimal solution requirements.

[0034] Execute targeted sludge mixing: drive the return sludge mixing equipment 230 composed of sludge pump and pipe valve system, according to the optimal solution The instructions corresponding to the elements with a median value of 1 accurately add a specific return sludge source to the designated biochemical treatment unit.

[0035] Execute dynamic adjustment of operating parameters: drive the operating parameter adjustment equipment 240 (such as blower, internal reflux pump, etc.) associated with each biochemical pool, according to the optimal solution and The contained operating load and treatment objectives adjust key process parameters such as dissolved oxygen and internal reflux ratio in a feedforward manner to create the optimal local biochemical environment for the assigned tasks.

[0036] By cyclically executing the above steps, the method of the present invention can realize intelligent and coordinated closed-loop control of the entire sewage treatment process based on real-time perception of water quality characteristics.

[0037] In a specific embodiment of the present invention, the process of executing step S1 to obtain the chemical fingerprint data of the influent water is completed by the spectral sensor 110 deployed in the pipe network system.

[0038] In one embodiment, spectral sensor 110 is a three-dimensional fluorescence spectrometer. The analyzer is connected to a sewage pipeline via a flow cell, allowing sewage in the pipeline network to continuously flow through the analyzer's measurement area. The analyzer's light source emits excitation light at a series of preset wavelengths, sequentially illuminating the sewage sample flowing through the flow cell. For each excitation wavelength, the analyzer's photodetector synchronously records the emission spectral intensity across a complete wavelength range in a direction perpendicular to the excitation wavelength, thereby forming a raw fluorescence data matrix.

[0039] The raw fluorescence data matrix undergoes a series of correction and preprocessing steps to eliminate instrument background noise and physical interference, thereby forming an influent chemical fingerprint that accurately reflects the chemical composition of the water sample. This preprocessing process includes: first, measuring and subtracting a blank matrix generated by ultrapure water (Milli-Q water) under the same test conditions to eliminate water Raman scattering and instrument background fluorescence.

[0040] Next, the blank-subtracted data matrix is ​​corrected for the inner filter effect. The inner filter effect is a nonlinear attenuation of fluorescence intensity caused by absorption of excitation or emission light by high-concentration substances in the sample. The correction process can be performed using the following formula: ,in, is the corrected fluorescence intensity, is the fluorescence intensity measured by the instrument, is the absorbance of the sample at the excitation wavelength, The absorbance of the sample at the emission wavelength can be measured simultaneously using the same instrument or an auxiliary UV-visible absorption spectrometer.

[0041] After the above preprocessing steps, a two-dimensional excitation-emission matrix (EEM) is obtained. This matrix is ​​the chemical fingerprint data of the influent at a specific moment. The X-axis of the matrix is ​​the emission wavelength, the Y-axis is the excitation wavelength, and each element in the matrix The value is at the excitation wavelength At the emission wavelength The fluorescence intensity was measured and corrected.

[0042] Influent chemical fingerprint data The data is digitized and sent to the central control and computing unit 300 via the data transmission module of the upstream monitoring unit 100 for subsequent calculations in steps S2 and S3. This process is repeated at a preset frequency (e.g., every 5 minutes), forming a time series dataset that continuously reflects the changes in the chemical characteristics of the upstream water quality.

[0043] After obtaining the influent chemical fingerprint data and before the influent carrying the corresponding chemical characteristics reaches the sewage treatment plant, the central control and computing unit 300 executes step S3 to input the influent chemical fingerprint data into a preset biochemical reaction potential analytical model to calculate the biochemical reaction potential vector (BRP).

[0044] In a specific embodiment, the biochemical reaction potential analysis model is a pre-trained multiple linear regression model, such as a partial least squares regression (PLSR) model. The model construction process is completed before the system is deployed. The process includes the following steps: First, a large number of historical sewage samples are collected. For each sample, the chemical fingerprint of the influent is measured. , and also through laboratory-scale batch reaction experiments, a set of biochemical reaction parameters such as the corresponding aerobic oxygen consumption rate, nitrification inhibition rate, sludge yield and toxicity inhibition level to specific microorganisms were determined.

[0045] Secondly, the collected data is processed. Each two-dimensional water chemical fingerprint data matrix , is unfolded into a one-dimensional row vector The biochemical reaction parameter values ​​measured by multiple laboratories corresponding to the sample constitute a response vector . Combine the data of all historical samples to form an independent variable matrix and a dependent variable matrix.

[0046] Finally, the partial least squares regression algorithm is applied to analyze the statistical relationship between the independent variable matrix and the dependent variable matrix, and a method is established to transform the input fingerprint vector into Response vector mapped to output A stable mathematical relationship, which is represented by a regression coefficient matrix Defined. Regression coefficient matrix It is solidified and stored in the biochemical reaction potential analytical model of the central control and computing unit 300.

[0047] During the real-time operation of the system, when a new influent chemical fingerprint data After being acquired and transmitted to the central control and computing unit 300, the biochemical reaction potential analytical model performs the following calculations: First, the two-dimensional matrix Expand into a one-dimensional row vector in the same way as during training .

[0048] Then, the biochemical reaction potential vector is calculated by the following matrix multiplication operation: : ,in: is the output biochemical reaction potential vector, which is in the form of ; is the potential component of aerobic oxygen consumption rate; is the nitrification inhibition potential component; is the potential component of sludge yield; is the toxic impact potential component; is the one-dimensional row vector obtained by expanding the current influent chemical fingerprint data; is the regression coefficient matrix stored in the pre-trained PLSR model.

[0049] Calculated biochemical reaction potential vector , a digital vector containing multiple quantitative components. Each component of the vector represents the potential impact of the influent on a specific biochemical process in the subsequent biochemical treatment. This vector is then passed to the decision optimization module as the core data input for step S4 and subsequent steps.

[0050] In a specific embodiment, the method of the present invention further includes, after step S3 and before step S4, obtaining sludge functional portrait data from the biological monitoring equipment 210 deployed at each return sludge source, and inputting the data into a preset sludge functional pedigree analysis model to obtain a sludge functional pedigree vector (SFP) that characterizes the functional characteristics of the sludge source.

[0051] The biological monitoring device 210 can be an online respirometer. The online respirometer is connected to the return sludge pipeline and can automatically and periodically extract sludge samples into the reaction chamber inside it. In the reaction chamber, the online respirometer can accurately measure and record the oxygen consumption rate of the sludge under different conditions. These oxygen consumption rate data constitute the sludge functional profile data. The sludge functional spectrum analysis model is a set of preset calculation rules and formulas integrated into the central control and calculation unit 300, which is used to convert the sludge functional profile data into a standardized sludge functional spectrum vector. .

[0052] Sludge functional lineage vector One component of is the overall biological activity (GBA). The calculation process is as follows: the online respirometer first measures the endogenous oxygen consumption rate of the extracted sludge sample under the condition of no external carbon source supply ( The sludge functional spectrum analysis model calculates the GBA component using the following formula: : ,in, It is a preset baseline oxygen consumption rate value, which represents the average biological activity level of the activated sludge under normal operating conditions of the treatment plant.

[0053] Sludge functional lineage vector Another component of STR is specific toxicity tolerance (STR). Its calculation process is as follows: After measuring the endogenous oxygen consumption rate, the online respirometer automatically adds a standard concentration of a pre-selected representative toxicant (such as 3,5-dichlorophenol) to the same sludge sample in the reaction chamber. After a fixed contact time, the sludge oxygen consumption rate is measured again and recorded as the post-inhibition oxygen consumption rate ( ). STR component It is calculated by the following formula: , which directly quantifies the tolerance of sludge microorganisms to this specific toxic substance.

[0054] Sludge functional lineage vector Another component of the SDR is the specific degradation rate (SDR) for a specific refractory organic compound. The calculation process is similar to the STR determination process, except that the substances added to the reaction chamber are one or more pre-selected, representative refractory organic compounds. By measuring the increase in oxygen consumption rate after adding the substance, the specific oxygen consumption rate (SDR) for the specific substance can be calculated. ). SDR component It can be expressed as: ,in, It is the total oxygen consumption rate measured after adding specific refractory organic matter.

[0055] By performing the above series of measurements and calculations, the central control and calculation unit 300 generates a sludge functional spectrum vector containing multiple functional components for each return sludge source. The vector is stored digitally and passed to the decision optimization module for calculating the target gain affinity score and formulating the target refluxing strategy.

[0056] After the central control and computing unit 300 generates the biochemical reaction potential vector (BRP), step S4 is immediately executed to calculate a hydraulic affinity score (HAS) for each unit j (j=1, 2, ..., n) of the n biochemical treatment units operating in parallel in the sewage treatment plant.

[0057] In a storage module of the central control and computing unit 300, a historical performance portrait vector is pre-stored for each biochemical processing unit j . Historical Performance Portrait Vector It is composed of the statistical analysis results of the long-term operation monitoring data of the unit, which includes the biochemical reaction potential vector The performance indicators corresponding to each component. In a specific embodiment, the historical performance portrait vector The form is: , where each component is a value extracted from the historical database and standardized: is the historical average aerobic load treatment capacity of unit j, which represents its performance in coping with the impact of high organic matter concentration; is the historical nitrification stability index of unit j, which characterizes its performance in maintaining nitrification function in the face of nitrification inhibitors; is the inverse of the historical sludge production coefficient of unit j, which characterizes its low sludge production performance; is the historical toxic shock tolerance of unit j, which characterizes its performance in recovering biological activity after encountering toxic substances.

[0058] The calculation process of the hydraulic affinity score is to convert the biochemical reaction potential vector at the current moment and the historical performance portrait vector of each biochemical treatment unit j Perform mathematical operations to obtain a scalar value that quantifies the degree of adaptation of the evaluation unit j to the current influent chemical characteristics. .

[0059] In a specific embodiment, the hydrodynamic affinity score It is calculated through a linear weighted summation model. The mathematical expression of the model is: ; ,in, is the calculated hydraulic affinity score of biochemical treatment unit j; is the index of the component in the biochemical reaction potential vector and the historical performance profile vector, representing a specific biochemical reaction dimension (such as AOUR, NIP, SYP, TIP); For the The preset weight coefficients of the components are set to reflect the relative importance of different biochemical reaction dimensions in the overall goal of wastewater treatment; is the biochemical reaction potential vector No. The value of the component; is the historical performance portrait vector of biochemical treatment unit j No. The value of the component; , , , The weight coefficients corresponding to the four dimensions of "aerobic oxygen consumption rate", "nitrification inhibition", "sludge yield" and "toxic impact" respectively; They correspond to the biochemical reaction potential components of the incoming water in the four dimensions of "aerobic oxygen consumption rate", "nitrification inhibition", "sludge production rate" and "toxic impact" respectively; , , , They correspond to the historical performance portrait components of the j-th biochemical treatment unit in the four dimensions of "aerobic oxygen consumption rate", "nitrification inhibition", "sludge yield" and "toxic impact".

[0060] The structure of the calculation formula makes a high specific historical performance ( The biochemical treatment unit with a large value) will be When the water inlet is large (with a large value), a higher hydraulic affinity score can be obtained.

[0061] The central control and calculation unit 300 is a controller for all biochemical treatment units (j = 1 to ) calculate their respective Then, the obtained score vector The data is output to the decision optimization module as the core basis for determining the traffic allocation ratio.

[0062] In a specific embodiment, after step S4 and before step S5, the method further includes: , a target gain affinity score (TGAS) is calculated. Used to quantify the source of return sludge The sludge is allocated to the biochemical treatment unit j to cope with the expected gain effect when a specific influent is applied.

[0063] The calculation process is performed by the central control and calculation unit 300. The process is based on the biochemical reaction potential vector at the current moment , m sludge functional lineage vectors from m return sludge sources (i=1,2,…,m), and the historical performance portrait vector of each biochemical treatment unit as input.

[0064] Target gain affinity score The calculation aims to evaluate the functional characteristics of sludge source i (given by Characterization) and the performance shortcomings of biochemical treatment unit j (by characterization) in response to the biochemical challenges posed by influent water (by The degree of co-matching when characterizing.

[0065] In a specific embodiment, the score is calculated using the following mathematical expression: ,in, The target gain affinity score calculated for the combination of return sludge source i and biochemical treatment unit j; and are the preset weight coefficients for the toxicity tolerance dimension and the specific degradation dimension respectively; is the biochemical reaction potential vector The toxic impact potential component in is the sludge functional spectrum vector of the i-th return sludge source in, the tolerance component to specific toxicity; is the historical performance portrait vector of the jth biochemical treatment unit The historical toxic shock tolerance component after normalization (value between 0 and 1). It represents the unit's "performance shortcoming" or "gain requirement" in terms of toxicity tolerance; is the biochemical reaction potential vector In the middle, it characterizes the degradation demand potential component of the influent containing specific refractory organic matter; is the sludge functional spectrum vector of the i-th return sludge source Among them, the specific degradation effect component on specific difficult-to-degrade organic matter; is the historical performance portrait vector of the jth biochemical treatment unit In, one with The corresponding normalized historical substance-specific degradation performance component. It represents the “performance shortcoming” of the unit in this specific degradation function.

[0066] Through this revised calculation model, the score The level is determined by three parties: There are clear challenges in water inflow ( The sludge has the ability to cope (high c value), and the target unit has deficiencies in this aspect (high (1−h) value). This allows the score to truly and accurately reflect the logic of "targeted efficiency enhancement" and provides a solid data foundation for subsequent collaborative optimization decisions.

[0067] The central control and calculation unit 300 generates a set of sludge return sources for each possible sludge return source according to the above model. Combined with biochemical treatment unit j, the , thus forming an m×n-dimensional target gain affinity score matrix. This matrix is ​​then output to the decision optimization module as the core basis for determining the targeted reflow strategy.

[0068] After calculating the hydraulic affinity score and the target gain affinity score, the decision optimization module within the central control and computing unit 300 executes step S5 to generate a control decision by constructing and solving a collaborative optimization problem. The first step in this process is to accurately define the decision variables of the collaborative optimization problem.

[0069] In a specific embodiment of the present invention, the collaborative optimization problem includes two groups of decision variables: one group is used to determine the allocation of the influent flow, and the other group is used to determine the allocation of the return sludge.

[0070] The first set of decision variables is the flow distribution ratio vector This vector is used to determine the distribution ratio of the total influent flow among n biochemical treatment units running in parallel. Flow distribution ratio vector The mathematical form is a A one-dimensional vector of elements: , where each element is a continuous variable, and its physical meaning is assigned to the The proportion of the water flow of each biochemical treatment unit to the total water flow. Therefore, each element is a continuous variable, and its physical meaning is assigned to the The proportion of the water flow of each biochemical treatment unit to the total water flow. The value range of is limited to: , and the sum of all elements must be equal to 1 to satisfy the basic physical constraint of flow conservation: .

[0071] The second set of decision variables is the sludge allocation strategy matrix This matrix is ​​used to represent the specific plan of allocating m different return sludge sources to n biochemical treatment units. Sludge allocation strategy matrix The mathematical form of OK A 2D matrix of columns: , where each element is a binary integer variable whose value is defined as: ; when When The sludge from the return sludge source is transferred to the The operation of a biochemical treatment unit; , it means that the operation will not be performed.

[0072] To ensure the specificity and effectiveness of targeted gains, the sludge allocation strategy matrix The elements in must also satisfy the following constraints: The physical meaning of this constraint is that within a single control cycle, each specific return sludge source i (especially sludge sources with special functional characteristics) can only be deployed to one target biochemical treatment unit at most. , or not used.

[0073] Traffic distribution ratio vector and sludge allocation strategy matrix Together they form the complete set of decision variables for the collaborative optimization problem. The goal of the decision optimization module is to calculate the optimal values ​​of these two sets of variables.

[0074] In one embodiment of the present invention, the objective function of the collaborative optimization problem is defined as , and is set as a maximization problem. The objective function performs a weighted summation of the hydraulic affinity score and the targeted gain affinity score to form a comprehensive indicator that can reflect the overall operating efficiency of the system.

[0075] The mathematical expression of the objective function is: Maximize ,in, is the objective function to be maximized, and its calculation result is a scalar value representing the maximum value of the decision variable (flow distribution ratio vector) and The "overall operational affinity" or "comprehensive benefit score" that the entire sewage treatment system can achieve under the specific control strategy defined in the (Sludge Allocation Strategy Matrix).

[0076] Item 1 : Represents the total hydraulic affinity score. This item is intended to reward those who transfer a larger portion of the inflow (larger ) is assigned to the water that is more suitable for the current influent water quality (i.e., has a higher hydraulic affinity score It reflects the degree of optimization of the system at the macro traffic distribution level.

[0077] Item 2 : represents the total score of targeted gain. This item aims to quantify the total score of all bio-enhancement operations performed (i.e., all =1 deployment) brings about the cumulative benefits. It is a binary variable that acts as a “switch”: it is activated only when the decision model decides to implement a certain option from the sludge source. To processing unit The potential gain of the deployment Otherwise, its contribution will be zero.

[0078] + These two dimensionless parameters are hyperparameters that can be preset by the system operator or advanced control logic to make a trade-off and balance between the two optimization objectives. In a specific embodiment, these two weight coefficients must satisfy the constraints + =1 and , ≥0. By adjusting these two coefficients, the optimization goal can be more focused on improving the treatment efficiency and stability of conventional pollutants (for example, setting =0.8, = 0.2), or more focused on proactively and quickly responding to sudden toxic shocks or shocks of refractory substances (e.g., setting =0.3, =0.7).

[0079] The ultimate goal of the decision optimization module is to meet all the constraints defined above (such as and Under the premise of (etc.), a set of solutions that can make the objective function The optimal decision variable that achieves the maximum value and .

[0080] In one embodiment of the present invention, the collaborative optimization problem The solution must be performed under the following set of constraints: Intrinsic constraints on decision variables: This set of constraints defines the range and type of the decision variables themselves and is a formal component of building an optimization model: Traffic distribution ratio range constraints: ,This constraint ensures that the proportion of traffic allocated to any processing unit is a valid percentage.

[0081] Sludge allocation decision type constraints: ,This constraint clarifies that the sludge allocation decision is a binary switch of "execute" or "not execute".

[0082] Material conservation and logical constraints: This set of constraints is key to ensuring the correctness of the system's material balance and core control logic: Total flow conservation constraint: ,This equality constraint ensures that the sum of the total flow proportions received by all ,parallel biochemical treatment units is exactly equal to 100%, i.e. all the influent flows are distributed without omissions or ,generated out of thin air.

[0083] Sludge source uniqueness constraint: , this inequality constraint is the core logic for achieving “targeted” gain. It stipulates that each return sludge source (Especially those sludges identified as having specific functions, such as efficient nitrogen removal or toxicity resistance) can only be directed to one target biochemical treatment unit in a single control cycle. This avoids the ineffective dispersion of valuable special-effect microbial resources.

[0084] Operational process and physical limitations: This set of constraints further refines the technical details of the present invention and ensures the feasibility of the solution. It takes into account the upper limits of the physical and biological processing capacity of the treatment unit: Treatment unit hydraulic load constraints: ,in, is the total water flow at the current moment (m 3 / h), and and They are Each biochemical treatment unit is designed to allow a minimum and maximum operating water flow. This constraint prevents the flow from being unevenly distributed (excessive ) caused by hydraulic shock, short flow, or too small flow (too small ) causes problems such as excessive aeration and reduced microbial activity.

[0085] Treatment unit sludge acceptance constraints: , this constraint is limited from the perspective of the receiving end, stipulating that any biochemical processing unit , can only receive sludge from one source at most in a single control cycle This measure has important technological significance: it avoids the simultaneous addition of multiple sludges with different functional properties (which may be antagonistic to each other) into one reactor, thus simplifying process control and making it possible to evaluate and verify the specific bioaugmentation behavior.

[0086] All the above constraints together define a "feasible region" for the optimization problem. The solution process of the decision optimization module is essentially to find a solution that can make the objective function The only solution that reaches the maximum value process.

[0087] Refer to the attached Figure 2 , Figure 2 Figure 1 is a schematic diagram of the collaborative optimization problem-solving process according to one embodiment of the present invention. Once the objective function and all constraints are fully defined, they together form a formalized, structured mathematical optimization problem. This problem-solving process is automatically performed by the decision optimization module within the central control and computing unit 300, with the goal of efficiently and accurately calculating the optimal control decision.

[0088] In a specific embodiment of the present invention, by analyzing the structure of the optimization problem, it can be classified as a Mixed-Integer Linear Programming (MILP) problem. The basis for this qualitative analysis is: The objective function is linear: the objective function is the decision variable and The linear weighted sum of .

[0089] Constraints are linear: All constraints, including equality and inequality constraints, are linear expressions in terms of the decision variables.

[0090] The variable types are mixed: the decision variable set contains both continuous variables with values ​​in the interval [0,1] (traffic distribution ratio), also contains binary integer variables with values ​​​​{0,1} (Sludge blending switch).

[0091] The decision optimization module integrates one or more professional solvers optimized for this type of MILP problem. At the beginning of each control decision cycle, the module first calculates the hydraulic affinity score in real time. and target gain affinity scores As a constant parameter, it is substituted into the objective function and constraint expressions to generate a specific mathematical instance to be solved at that moment.

[0092] The module then calls the MILP solver to solve the instance. The solver internally uses a robust algorithm such as Branch and Bound combined with the Simplex Method. This algorithm can systematically search within the complex, multi-dimensional feasible region defined by all constraints to ensure that a solution that satisfies the objective function is found within a limited time. The optimal solution that reaches the global maximum.

[0093] The final output of the solution process is not just the maximum value of the objective function, but the value of a set of optimal decision variables corresponding to the maximum value, which is recorded as and .

[0094] Optimal flow distribution ratio vector , where each element It is a specific value, for example [0.55,0.45], which means that 55% of the total influent flow is allocated to biochemical treatment unit 1 and 45% is allocated to unit 2.

[0095] Optimal sludge allocation strategy matrix , which is an m-row, n-column matrix consisting of 0s and 1s, where the elements with value 1 (e.g. ) accurately instructs which return sludge source (source 2) should target the sludge to which biochemical treatment unit (unit 1).

[0096] The central control and computing unit 300 finally converts this set of optimal solutions ( These instructions are translated into specific physical equipment control commands. For example, precise flow distribution can be achieved by adjusting the valve openings on relevant pipelines, or specific sludge mixing tasks can be performed by controlling the start and stop of pumps and flow switching valves. This entire solution and execution process repeats itself at a preset interval (e.g., every 30 minutes), achieving continuous, dynamic, and intelligent coordinated optimization control of the wastewater treatment process.

[0097] The decision optimization module solves the optimal flow distribution ratio vector of the current control cycle and optimal sludge allocation strategy matrix Afterwards, these decision instructions are not issued instantaneously and synchronously. Considering that the sewage treatment plant is a large physical system with significant time lag effects, the present invention further discloses a collaborative control execution method based on time prediction to ensure the accuracy and timeliness of control actions.

[0098] The core idea of ​​this method is to transform the static optimal decision solution into a dynamic, precisely time-scheduled execution sequence. The control execution module inside the central control and computing unit 300 receives the optimal solution. and After that, the following steps are performed: The control execution module first retrieves and calculates a set of key time-delay parameters from the system configuration database or calculates them based on real-time operating conditions. These parameters quantify the time required from the issuance of a control instruction to the actual process effect.

[0099] Hydraulic transmission time lag ( ): refers to the point where the total inflow is distributed (for example, the flow distribution well) and the flow to the first In one embodiment, the time delay is not a fixed constant, but is determined by the pipe geometry (length , pipe diameter ) and the current flow rate allocated to the unit ( ) is dynamically calculated to reflect the flow velocity changes under different flow rates.

[0100] Sludge transport and mixing time lag ( ) refers to the total time required from the start of the pump or valve that transfers sludge from the i-th sludge source to the j-th biochemical treatment unit until the sludge is essentially evenly mixed into the reaction zone of the biochemical treatment unit and sufficiently mixed to begin to exert its bioaugmentation effect. This parameter combines the pumping time, pipeline transportation time, and mixing time within the reactor.

[0101] Determine the control synchronization target time based on the calculated optimal flow distribution decision , the module will determine a control synchronization target time for each biochemical treatment unit j receiving water ( The physical meaning of this time is the time when the "batch" of influent water currently being measured and analyzed at the flow distribution point is expected to arrive at biochemical treatment unit j. Its calculation formula is: ,in, The start time for optimization calculation for the current control cycle.

[0102] The key task of the control execution module is to ensure that the effects of the control actions that need to be coordinated (i.e. flow distribution and sludge targeted dosing) can be achieved at the target time. To achieve synchronization, it "backwards" calculates the specific startup time of each physical device and generates an asynchronous instruction sequence: Flow distribution instructions: The instructions to adjust the flow distribution valve can usually be issued immediately or with a very short system response delay. Its execution time Approximately .

[0103] Sludge mixing instruction: For any sludge mixing operation that is determined to be required to be executed (i.e. ), the corresponding pump and valve start instructions are not in Instead, it is scheduled at a calculated future time. Execution. The execution time is calculated as follows: This calculation ensures that although the sludge mixing action is started "in advance", the time when its biological enhancement effect takes effect coincides with the time when the target water body that needs to be treated arrives.

[0104] Through this mechanism, the present invention avoids control mismatch problems caused by physical time lags. For example, it prevents functional sludge from being added to a reactor prematurely, while the characteristic pollutants it needs to degrade are still far away in the incoming water pipes. Similarly, it avoids the ineffective efficiency gains caused by the functional sludge arriving only after the characteristic pollutants have already entered the reactor. This time-prediction-based collaborative execution strategy greatly improves the actual conversion efficiency of targeted gains, accurately translating the mathematical optimal solution into the optimal process control effect in the physical world.

[0105] The collaborative optimization control of the present invention is not limited to macroscopic flow distribution and sludge allocation, but further deepens to the refined and adaptive adjustment of the internal operating environment of each independent biochemical treatment unit. When the collaborative control instruction is issued, a dynamic parameter adjustment mechanism closely coupled with the decision is activated. Its core goal is to provide the assigned task (by and definition) to create optimal local biochemical reaction conditions.

[0106] In a specific embodiment, the central control and calculation unit 300 includes an operation parameter adjustment submodule. This module receives the optimal solution output by the decision optimization module. and as the primary input, and combined with the affinity score that leads to the decision and The working condition interpretation information behind it is for each biochemical treatment unit The key operating parameters can be adjusted dynamically in a feed-forward manner.

[0107] This dynamic adjustment is mainly reflected in the following aspects: The feedforward aeration rate adjustment based on hydraulic load will adjust the operating parameters according to the optimal flow ratio allocated to the jth treatment unit. and total water flow , and calculates in real time the organic loading rate (OLR) that the unit will be subjected to. Compared with the traditional feedback control based on effluent dissolved oxygen (DO), the present invention adds a feedforward control loop: When a processing unit j is assigned a higher flow ratio (i.e. When the load surge is large, the system predicts a significant increase in organic matter and ammonia nitrogen loads. To address this, the module proactively raises the dissolved oxygen (DO) setpoint in the aerobic zone (for example, from the standard 2.0 mg / L to 2.5 mg / L) and increases the blower output accordingly. This ensures that when the load surge occurs, microorganisms have sufficient oxygen for respiratory metabolism and nitrification, preventing a decrease in pollutant removal efficiency due to hypoxia.

[0108] One of the core innovations of the process parameter optimization and operation parameter adjustment submodule with targeted gain function is that it can “understand” sludge allocation decisions. The strategic intention behind it and collaboratively adjust process parameters to maximize the gain effect.

[0109] Example 1: Enhanced nitrification If the decision is to allocate a sludge source i rich in efficient nitrifying bacteria to treatment unit j (i.e. ,and If a high score indicates a need for enhanced denitrification, the module will not only ensure adequate DO as described above, but may also moderately reduce the Mixed Liquor Recirculation (MLR) ratio within that unit. This is done to extend the sludge retention time (SRT) in the aerobic zone, providing a more favorable breeding environment for slower-growing nitrifying bacteria and maximizing the targeted gain effect.

[0110] Example 2: Enhanced denitrification If the decision is to allocate a sludge source i with high denitrification capacity to treatment unit j, the module will take the opposite strategy. It will significantly increase the internal reflow ratio of the unit, and the more nitrate (NO3 −−N) is quickly returned to the anoxic zone (Anoxic Zone), providing sufficient electron acceptors for the newly added dominant denitrifying bacteria, thereby achieving efficient denitrification.

[0111] Achieve differentiated energy-saving control of system energy efficiency. This dynamic adjustment mechanism not only pursues processing effects, but also strives to optimize overall energy consumption.

[0112] For those units that were assigned a lower hydraulic load during the control period (i.e. smaller), and no targeted gain operation is performed (i.e., all ) processing unit , the operating parameter adjustment submodule identifies it as "baseload" operation. In this mode, the module lowers its DO setpoint to a low level (e.g., 1.5 mg / L) sufficient to meet basic processing requirements and reduces the frequency of the internal reflux pump. This differentiated energy-saving control precisely directs limited energy to the units responsible for primary processing, avoiding energy waste in low-load units.

[0113] In this way, the present invention organically combines macro-level allocation decisions with micro-level unit internal parameter adjustments, forming a multi-level, synergistic control system. This ensures that each control decision is not only optimal in terms of resource allocation, but also provides the most suitable biochemical environment for its execution.

[0114] In one embodiment, Figure 3 This diagram illustrates the changes in effluent quality recorded by the system described in this invention, in the specific scenario of a sudden high-concentration pollution shock load in the upstream pipeline network. This diagram is a two-dimensional graph, with the horizontal axis representing time and the vertical axis representing the concentration of a key pollutant (ammonia nitrogen) in the effluent from the sewage treatment plant.

[0115] Reference Figure 3 , the diagram contains the following elements: A horizontal line, labeled the Discharge Standard, represents a pre-set compliance threshold for effluent pollutant concentration.

[0116] A blue dotted line is marked as the traditional control method. This curve shows the change process of the effluent pollutant concentration in a system that does not adopt the method described in this invention. Before the moment, the value of the blue dotted line is lower than the emission standard threshold. After the moment, its value rises rapidly and is higher than the threshold value specified by the emission standard for a period of time, and then slowly decreases and returns to below the threshold value.

[0117] A black line is marked as the method of the present invention. This curve shows the change process of the effluent pollutant concentration in the system using the method of the present invention. In the entire time axis range, the value of the black line is always lower than the threshold value defined by the emission standard. Near the moment, the value of the curve fluctuates in a small range and then returns to stability.

[0118] Figure 3 Two specific time points are also marked: Time point : indicates the moment when the spectral sensor 110 of the upstream monitoring unit 100 detects a significant change in the chemical fingerprint of the sewage in the pipe network (corresponding to high concentration pollution). After that moment, the data analysis and the calculation of collaborative optimization decisions will be started.

[0119] Time point :Indicates The time when the high-concentration polluted sewage detected at the time actually arrives at the sewage treatment plant after being transported through the pipeline.

[0120] The figure shows that when dealing with the technical problem of high-concentration pollution shock at the same upstream, the traditional control method shown in curve A lacks predictability and is ineffective after the shock load arrives ( In contrast, the method of the present invention shown by the black line, by Detection is performed at all times and predictive collaborative control is performed, so that the system can At that moment) and thereafter, the effluent water quality is always maintained below the threshold value specified by the discharge standard, thus solving the technical problem that the existing technology cannot effectively cope with sudden impact loads.

Claims

1. An intelligent control method for treatment processes based on sewage chemical feature identification, characterized in that: The following steps are involved: S1. Obtaining influent chemical fingerprint data collected in real time by spectral sensors deployed in the sewage pipe network system; S2. Based on the influent chemical fingerprint data and the pipe network hydraulic transport model, predicting the time when the influent carrying the corresponding chemical characteristics will arrive at the sewage treatment plant; S3. Before the influent reaches the sewage treatment plant, inputting the influent chemical fingerprint data into a preset biochemical reaction potential analytical model to obtain a biochemical reaction potential vector representing the potential impact of the influent on subsequent biochemical treatment processes; S4. Calculating a hydraulic affinity score for each biochemical treatment unit based on the biochemical reaction potential vector and in combination with historical performance data of each biochemical treatment unit operating in parallel in the sewage treatment plant; S5. Based on the hydraulic affinity score, determine the distribution ratio for flow distribution of the influent, and according to the predicted time when the influent carrying the corresponding chemical characteristics arrives at the sewage treatment plant, control the flow distribution equipment of the sewage treatment plant according to the determined distribution ratio.

2. The intelligent control method for treatment process based on sewage chemical feature identification according to claim 1 is characterized in that: In step S3, the biochemical reaction potential vector includes at least one of the following potential energy components: aerobic oxygen consumption rate potential, nitrification inhibition potential, sludge production potential or toxic shock potential.

3. The intelligent control method for treatment process based on sewage chemical feature identification according to claim 1 is characterized in that: Also includes: After step S3 and before step S4, sludge functional profile data collected by biological monitoring equipment deployed at each return sludge source in the sewage treatment plant is obtained; The sludge functional portrait data is input into a preset sludge functional pedigree analysis model to obtain a sludge functional pedigree vector that characterizes the functional characteristics of the return sludge source.

4. The intelligent control method for treatment process based on sewage chemical feature identification according to claim 3 is characterized in that: The sludge functional spectrum vector includes at least one of the following functional components: Overall biological activity, tolerance to toxicity or degradation specificity for refractory organic matter.

5. The intelligent control method for treatment process based on sewage chemical feature identification according to claim 3 is characterized in that: After step S4 and before step S5, the method further includes: Calculating a targeted gain affinity score for each combination of the biochemical treatment unit and each return sludge source according to the biochemical reaction potential vector and the sludge functional spectrum vector; The targeted gain affinity score is used to characterize the expected gain effect when the sludge from the return sludge source is deployed to the biochemical treatment unit to cope with the influent.

6. The intelligent control method for treatment process based on sewage chemical feature identification according to claim 5 is characterized in that: Also includes: determining a targeted refluxing strategy based on the targeted gain affinity score; And based on the predicted time when the influent carrying corresponding chemical characteristics arrives at the sewage treatment plant obtained in step S2, the return sludge mixing equipment of the sewage treatment plant is controlled according to the targeted return flow strategy to mix the sludge from the designated return sludge source to the designated biochemical treatment unit.

7. The intelligent control method for treatment process based on sewage chemical feature identification according to claim 6 is characterized in that: The steps of determining the distribution ratio of the flow distribution of the influent and determining the targeted reflux strategy are: Constructing an optimization problem, wherein the allocation ratio and the targeted reflux strategy are used as decision variables; Establishing an objective function of the optimization problem, wherein the objective function is a weighted sum of the hydraulic affinity score and the targeted gain affinity score; By solving the optimization problem to maximize the objective function, the optimal allocation ratio and the targeted reflux strategy are simultaneously obtained.

8. The intelligent control method for treatment process based on sewage chemical feature identification according to claim 6 is characterized in that: After step S5, for each biochemical treatment unit that receives the allocated flow, the operating parameters of the biochemical treatment unit are dynamically adjusted according to the biochemical reaction potential vector obtained in step S3 and combined with the targeted reflux strategy formulated for the biochemical treatment unit.

9. The intelligent control method for treatment process based on sewage chemical feature recognition according to claim 8 is characterized in that: The operating parameters include aeration volume, internal reflux ratio or reagent dosage.

10. The intelligent control method for treatment process based on sewage chemical feature recognition according to claim 1 is characterized in that: In step S1, the spectrum sensor is an ultraviolet-visible light full spectrum analyzer or a three-dimensional fluorescence spectrum analyzer.

Images