Intelligent chemical adding control method and system for industrial sewage plant based on multi-parameter self-adaption

Abstract

The invention discloses an industrial sewage plant intelligent dosing control method and system based on multi-parameter self-adaption, and relates to the technical field of industrial sewage control. Through a water quality monitoring device arranged in a sewage treatment multi-parameter acquisition position group Pos, an original multi-parameter feature vector group Vec is constructed; and on this basis, a regional drug effect incomplete factor matrix Mat is obtained through calculation, so that quantitative identification of a drug response failure region in an industrial sewage treatment space structure is realized, a reaction failure region proportion Rtd is calculated in combination with a preset response failure judgment threshold Tau, a quantitative index is provided for identification of a local drug adding invalid or low-efficiency region, and the quantitative identification of the drug response failure region in the industrial sewage treatment space structure is realized. According to the method, the regional efficacy incomplete factor matrix Mat is utilized to perform differential correction on the original dosing plan dose vector Cref, and a corrected dose vector Adj is generated, so that spatial differential dosing regulation and control in a real sense are realized, and resource waste and treatment efficiency reduction caused by non-uniform dosing diffusion in a traditional scheme are effectively avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial wastewater control, and in particular to an intelligent dosing control method and system for an industrial wastewater plant based on multi-parameter self-adaptation. Background Art

[0002] In modern industrial systems, water treatment technology, as a key branch of environmental engineering, serves as a bridge between industrial production and ecological and environmental protection. Industrial wastewater treatment, as a core area of ​​water treatment, has become a key area of ​​research for process optimization and the application of intelligent technologies due to its complex pollutant composition, drastic fluctuations, and difficulty in treatment. In specific industrial wastewater treatment scenarios, dosing control systems are often deployed in sedimentation tanks, coagulation tanks, or neutralization reaction units, fulfilling key tasks such as pH adjustment, heavy metal removal, and suspended particle removal.

[0003] In the traditional dosing control process, it is often assumed that the reagent can quickly and evenly diffuse into the water body after addition, participating in subsequent reactions such as neutralization, coagulation, and sedimentation. However, in actual operation, due to the existence of complex hydraulic factors such as uneven water velocity distribution, local turbulent structure, dead corners in the pool body, or asymmetric inlet and outlet water, the reagent added in some areas cannot be fully mixed or does not react with the pollutants at all, forming a "reaction failure area". In these areas, even if the sensor detection shows that the reagent has been added and the set dosage has been reached, the actual reaction efficiency may be far lower than expected due to insufficient diffusion or the existence of hydraulic blind spots.

[0004] At the same time, traditional control strategies often lack the ability to perceive and provide feedback on this spatial distribution non-uniformity. In such systems, parameters such as pH, conductivity, and ORP are usually collected at a single measurement point, making it difficult to reflect the spatial distribution of the reagent throughout the tank. More critically, the existence of these "failure zones" not only leads to insufficient drug efficacy, but also triggers a chain of control misjudgments, which in turn may cause the system to determine that the drug efficacy does not meet the standard based on global feedback, thereby further increasing the dosage, ultimately leading to increased energy consumption and operating costs, while also causing problems such as side reactions or excessive treatment. Summary of the Invention

[0005] In view of the deficiencies in the prior art, the present invention provides an intelligent dosing control method and system for an industrial sewage treatment plant based on multi-parameter self-adaptation, which solves the problems mentioned in the background technology.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: an intelligent dosing control method for industrial sewage treatment plants based on multi-parameter self-adaptation, comprising the following steps:

[0007] S1. By deploying water quality monitoring devices in the sewage treatment multi-parameter collection position group Pos, collecting the original environmental response parameters of each sewage treatment multi-parameter collection position group Pos, and constructing the original multi-parameter feature vector group Vec of the industrial sewage treatment system;

[0008] S2. Based on the constructed original multi-parameter feature vector group Vec, construct the regional efficacy incomplete factor matrix Mat of the sewage treatment multi-parameter collection position group Pos;

[0009] S3. Calculate the response failure region ratio Rtd based on the regional efficacy defect factor matrix Mat and the preset response failure judgment threshold Tau, so as to identify the region with low efficacy utilization in the sewage treatment multi-parameter collection position group Pos;

[0010] S4. Based on the regional efficacy defect factor matrix Mat and the original dosing plan dose vector Cref, a modified dose vector Adj is generated to form a spatially differentiated dosing strategy for the sewage treatment multi-parameter collection position group Pos;

[0011] S5. Calculate the comprehensive dosing response effectiveness index Eff using the corrected dose vector Adj and the drug response efficiency function Eta to determine the overall effectiveness of the current dosing strategy.

[0012] Preferably, said S1 includes S11 and S12;

[0013] S11. At each sampling point Pnt∈Pos in a plurality of preset sewage treatment multi-parameter collection position groups Pos, a water quality monitoring device is deployed to collect the original environmental response parameters of each sewage treatment multi-parameter collection position group Pos in real time;

[0014] The water quality monitoring device includes a water flow velocity sensor, a pH value sensor and a conductivity sensor;

[0015] The original environmental response parameters include the water flow velocity Vel(Pnt,t) at the sampling point Pnt at time t, the pH value Phv(Pnt,t) at the sampling point Pnt at time t, and the conductivity Con(Pnt,t) at the sampling point Pnt at time t.

[0016] Preferably, S12, based on the obtained original environmental response parameters, calculating the response change rate per unit time for each sampling point Pnt, specifically by calculating the change rate between time t and time t-Δt, where Δt represents the time interval, to form a derivative parameter reflecting the transient disturbance characteristics;

[0017] The derivative parameters include the pH value change Dph (Pnt, t) at the sampling point Pnt at time t and the conductivity change rate Dct (Pnt, t) at the sampling point Pnt at time t;

[0018] By integrating the derivative parameters and the original environmental response parameters, the original multi-parameter feature vector group Vec(Pnt, t) = {Dph(Pnt, t), Dct(Pnt, t), Vel(Pnt, t), Phv(Pnt, t), Con(Pnt, t)} of the sampling point Pnt at time t in the industrial wastewater treatment system is constructed.

[0019] Preferably, said S2 includes S21 and S22;

[0020] S21. Based on the water flow velocity Vel(Pnt,t), pH value change Dph(Pnt,t) and conductivity change rate Dct(Pnt,t) in the original multi-parameter feature vector group Vec(Pnt,t) at the sampling point Pnt at time t, construct the agent reaction efficiency function Eta(Pnt,t) at the sampling point Pnt at time t.

[0021] Preferably, S22, the drug reaction efficiency function Eta (Pnt, t) of all sampling points Pnt at time point t is normalized, and the drug reaction efficiency function Eta (Pnt, t) of each sampling point Pnt at time point t is compared after extracting the maximum value to obtain the regional drug efficacy defect factor Mat (pnt) of each sampling point Pnt, and the regional drug efficacy defect factor Mat (pnt) of all sampling points Pnt is integrated to construct a matrix representing the degree of regional drug efficacy attenuation of each sampling point, and obtain the regional drug efficacy defect factor matrix Mat.

[0022] Preferably, said S3 includes S31 and S32;

[0023] S31. Based on the regional efficacy defect factor matrix Mat and the preset response failure judgment threshold Tau, determine whether each sampling point Pnt meets the response failure condition, and obtain the response failure status label Flg of each sampling point Pnt;

[0024] The response failure status tag Flg of the sampling point Pnt is obtained by the following response failure conditions:

[0025] When the regional efficacy defect factor Mat(pnt) of the sampling point Pnt is ≥ the response failure judgment threshold Tau, the response failure status label Flg(Pnt) of the sampling point Pnt is obtained as 1, indicating that the sampling point Pnt is an area with unqualified efficacy utilization rate;

[0026] When the regional efficacy defect factor Mat(pnt) of the sampling point Pnt is less than the response failure judgment threshold Tau, the response failure state label Flg(Pnt) of the sampling point Pnt is obtained as 0, indicating that the sampling point Pnt is a region with qualified efficacy utilization rate;

[0027] S32. By extracting the number of sampling points Pnt with the response failure state label Flg=1 of all marked sampling points Pnt in the regional efficacy defect factor matrix Mat, and then calculating the ratio between them and the sewage treatment multi-parameter collection position group Pos, the response failure area ratio Rtd is obtained.

[0028] Preferably, said S4 includes S41;

[0029] S41. Based on the regional efficacy defect factor matrix Mat and the original dosing plan dose vector Cref, a modified dose vector Adj(Pnt) is generated for each sampling point Pnt, which is used to modify the dosing strategy of the sampling point Pnt, thereby forming a spatially differentiated dosing strategy for the wastewater treatment multi-parameter collection position group Pos;

[0030] Among them, the original dosing plan dosage vector Cref represents the preset dosing dosage for industrial wastewater;

[0031] The corrected dose vector Adj(Pnt) of the sampling point Pnt is obtained by the following calculation formula:

[0032] ;

[0033] Where Cref(Pnt) represents the original planned dosing dose vector at the sampling point Pnt, and Alp represents the dose correction adjustment coefficient, which is used to adjust the sensitivity of the defect level to the dose correction. The specific value is set by the user.

[0034] Preferably, said S5 includes S51 and S52;

[0035] S51, calculating the comprehensive dosing response effectiveness index Eff based on the obtained corrected dose vector Adj and the drug response efficiency function Eta;

[0036] The comprehensive dosing response effectiveness index Eff is obtained by the following calculation formula:

[0037] ;

[0038] Where Eta(Pnt, t) represents the drug reaction efficiency function of the sampling point Pnt at time point t, and Mat(pnt) represents the regional efficacy defect factor of the sampling point Pnt.

[0039] Preferably, S52, compare the comprehensive dosing response effectiveness index Eff with the preset system response evaluation threshold Thr, and trigger the execution of the dosing control strategy according to the comparison result to determine the overall effectiveness of the current dosing strategy

[0040] The execution of the dosing control strategy is triggered by the following comparison mechanism:

[0041] When the comprehensive dosing response effectiveness index Eff is less than the system response evaluation threshold Thr, the current dosing response effect is determined to be unqualified, indicating that the current original dosing plan dose vector Cref reflects an unqualified efficiency, and steps S1 to S4 are executed again to generate a new corrected dose vector Adj, and a second comparison is performed;

[0042] When the comprehensive dosing response effectiveness index Eff ≥ the system response evaluation threshold Thr, the current dosing response effect is determined to be qualified, and the executed dosage is the generated corrected dose vector Adj. At the same time, the corrected dose vector Adj can be used as a reference for adaptive updating of the original dosing plan dose vector Cref to form a new version of the original dosing plan dose vector, and the original dosing plan dose vector Cref continues to be used.

[0043] An intelligent dosing control system for industrial sewage treatment plants based on multi-parameter self-adaptation, comprising a water quality data acquisition module, a regional drug efficacy detection module, a regional drug efficacy response determination module, a correction module, and a decision-making module;

[0044] The water quality data acquisition module collects the original environmental response parameters of each sewage treatment multi-parameter acquisition position group Pos by deploying water quality monitoring devices in the sewage treatment multi-parameter acquisition position group Pos, and constructs the original multi-parameter feature vector group Vec of the industrial sewage treatment system;

[0045] The regional efficacy detection module constructs the regional efficacy incomplete factor matrix Mat of the sewage treatment multi-parameter collection position group Pos based on the constructed original multi-parameter feature vector group Vec;

[0046] The regional efficacy response determination module calculates the response failure region ratio Rtd based on the regional efficacy defect factor matrix Mat and the preset response failure determination threshold Tau, which is used to identify areas with low efficacy utilization in the sewage treatment multi-parameter collection position group Pos;

[0047] The correction module generates a correction dose vector Adj based on the regional efficacy defect factor matrix Mat and the original dosing plan dose vector Cref, forming a spatially differentiated dosing strategy for the sewage treatment multi-parameter collection position group Pos;

[0048] The decision module uses the modified dose vector Adj and the drug response efficiency function Eta to calculate the comprehensive drug dosing response effectiveness index Eff and determine the overall effectiveness of the current drug dosing strategy.

[0049] The present invention provides an intelligent dosing control method and system for industrial sewage treatment plants based on multi-parameter self-adaptation, which has the following beneficial effects:

[0050] (1) The water quality monitoring devices arranged in the multi-parameter collection position group Pos of the sewage treatment are used to construct the original multi-parameter feature vector group Vec, and on this basis, the regional efficacy defect factor matrix Mat is calculated, thereby realizing the quantitative identification of the "drug response failure area" in the spatial structure of industrial sewage treatment. Subsequently, combined with the preset response failure judgment threshold Tau, the response failure area ratio Rtd is calculated, which provides a quantitative indicator for identifying the local ineffective or inefficient areas of drug addition. The regional efficacy defect factor matrix Mat is used to perform differential correction on the original drug addition plan dose vector Cref to generate the corrected dose vector Adj, realizing the true sense of "spatial differentiated drug addition regulation", effectively avoiding the waste of resources and decreased treatment efficiency caused by uneven drug diffusion in traditional schemes. Finally, by constructing the comprehensive drug addition response effectiveness index Eff and combining the comparison results of this index with the system response evaluation threshold Thr, whether to execute the corrected dose vector Adj is dynamically adjusted, thereby forming a closed-loop control mechanism with response evaluation, strategy judgment, and adaptive correction.

[0051] (2) Based on the combined judgment mechanism of the regional efficacy defect factor matrix Mat and the response failure judgment threshold Tau, the response failure region ratio Rtd obtained based on the response failure status label Flg(Pnt) of all sampling points Pnt not only quantifies the proportion of inefficient regions in the system as a whole, but also provides a macro judgment threshold for whether to trigger the correction strategy. When the judgment is established, the correction dose vector Adj constructed by combining the regional efficacy defect factor matrix Mat with the original dosing plan dose vector Cref enables the system to accurately adjust the dosage according to the degree of defect of each sampling point Pnt, forming a spatially structured dosing control scheme. This mechanism significantly improves the control system's ability to handle "regional response failure", allowing the dosing strategy to evolve from "uniform dose" to "dynamic adjustment according to response performance", effectively avoiding problems such as local overdosing and insufficient dose in the boundary area.

[0052] (3) By constructing and calculating the comprehensive dosing response effectiveness index Eff, a global and quantitative evaluation of the current dosing response status of the entire sewage treatment system is achieved, and the overall effectiveness of the current dosing strategy can be independently identified. When the dosing response does not meet the standard, the system automatically backtracks to execute steps S1 to S4, regenerates the corrected dose vector Adj(Pnt), and verifies its response performance again; when the dosing response meets the expected standard, the system not only actually executes the generated corrected dose vector Adj(Pnt), but also uses it as a feedback basis to dynamically optimize the original dosing plan dose vector Cref(Pnt), achieving continuous evolution of control experience. This breaks the limitations of traditional industrial dosing systems that rely on fixed dose rules and lack data feedback judgment, and constructs an intelligent closed-loop control framework that drives strategy evolution with response performance and quantifies the advantages and disadvantages of strategies with system-level indicators. This greatly improves the adaptability, self-learning ability and stability of the control strategy, and is particularly suitable for dynamic treatment scenarios with frequent sewage quality fluctuations and complex pollutant structures. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 This is a schematic diagram of the steps of an intelligent dosing control method for an industrial sewage treatment plant based on multi-parameter self-adaptation according to the present invention;

[0054] Figure 2 This is a schematic diagram of a block diagram of an intelligent dosing control system for an industrial sewage plant based on multi-parameter self-adaptation according to the present invention;

[0055] Figure 3 Schematic diagram of the regional efficacy defect factor matrix Mat. DETAILED DESCRIPTION

[0056] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0057] Example 1

[0058] The present invention provides an intelligent dosing control method for industrial sewage treatment plants based on multi-parameter self-adaptation. Figure 1 , including the following steps:

[0059] S1. By deploying water quality monitoring devices in the sewage treatment multi-parameter collection position group Pos, collecting the original environmental response parameters of each sewage treatment multi-parameter collection position group Pos, and constructing the original multi-parameter feature vector group Vec of the industrial sewage treatment system;

[0060] S2. Based on the constructed original multi-parameter feature vector group Vec, construct the regional efficacy incomplete factor matrix Mat of the sewage treatment multi-parameter collection position group Pos;

[0061] S3. Calculate the response failure region ratio Rtd based on the regional efficacy defect factor matrix Mat and the preset response failure judgment threshold Tau, so as to identify the region with low efficacy utilization in the sewage treatment multi-parameter collection position group Pos;

[0062] S4. Based on the regional efficacy defect factor matrix Mat and the original dosing plan dose vector Cref, a modified dose vector Adj is generated to form a spatially differentiated dosing strategy for the sewage treatment multi-parameter collection position group Pos;

[0063] S5. Calculate the comprehensive dosing response effectiveness index Eff using the corrected dose vector Adj and the drug response efficiency function Eta to determine the overall effectiveness of the current dosing strategy.

[0064] In this embodiment, the original multi-parameter feature vector group Vec is constructed by using water quality monitoring devices deployed within the sewage treatment multi-parameter acquisition position group Pos. On this basis, the regional efficacy defect factor matrix Mat is calculated, thereby achieving quantitative identification of the "drug response failure area" in the spatial structure of industrial sewage treatment. Subsequently, combined with the preset response failure judgment threshold Tau, the response failure area ratio Rtd is calculated, providing a quantitative indicator for identifying local ineffective or inefficient drug dosing areas. The regional efficacy defect factor matrix Mat is used to perform differential correction on the original drug dosing plan dose vector Cref to generate a corrected dose vector Adj, realizing true "spatially differentiated drug dosing regulation" and effectively avoiding the waste of resources and decreased treatment efficiency caused by uneven drug diffusion in traditional schemes. Finally, by constructing a comprehensive drug dosing response effectiveness index Eff and combining the comparison results of this index with the system response evaluation threshold Thr, whether to execute the corrected dose vector Adj is dynamically adjusted, thereby forming a closed-loop control mechanism with response evaluation, strategy judgment, and adaptive correction. As a result, not only the spatial balance of the dosing reaction and the coordination of the system response are significantly improved overall, but also the key problems in the existing technology such as the difficulty in identifying the reaction dead zone, the difficulty in quantifying the dosing failure area, and the inability to reconstruct the control strategy for spatial heterogeneity are solved.

[0065] Example 2

[0066] Specifically: S1 includes S11 and S12;

[0067] S11. At each sampling point Pnt∈Pos in a plurality of preset sewage treatment multi-parameter collection position groups Pos, a water quality monitoring device is deployed to collect the original environmental response parameters of each sewage treatment multi-parameter collection position group Pos in real time;

[0068] The water quality monitoring device includes a water flow velocity sensor, a pH value sensor and a conductivity sensor;

[0069] The original environmental response parameters include the water flow velocity Vel(Pnt,t) at the sampling point Pnt at time t, the pH value Phv(Pnt,t) at the sampling point Pnt at time t, and the conductivity Con(Pnt,t) at the sampling point Pnt at time t.

[0070] S12. Based on the obtained original environmental response parameters, calculate the response change rate per unit time for each sampling point Pnt, specifically by calculating the change rate between time t and time t-Δt, where Δt represents the time interval, to form a derivative parameter reflecting the transient disturbance characteristics;

[0071] The derivative parameters include the pH value change Dph (Pnt, t) at the sampling point Pnt at time t and the conductivity change rate Dct (Pnt, t) at the sampling point Pnt at time t;

[0072] The pH value change Dph(Pnt, t) at the sampling point Pnt at time t is obtained by the following calculation formula:

[0073] ;

[0074] The conductivity change rate Dct(Pnt,t) of the sampling point at time t is obtained by the following calculation formula:

[0075] ;

[0076] By integrating the derivative parameters and the original environmental response parameters, the original multi-parameter feature vector group Vec(Pnt, t) = {Dph(Pnt, t), Dct(Pnt, t), Vel(Pnt, t), Phv(Pnt, t), Con(Pnt, t)} of the sampling point Pnt at time t in the industrial wastewater treatment system is constructed.

[0077] In this embodiment, real-time acquisition of raw environmental response parameters, including water velocity Vel(Pnt,t), pH value Phv(Pnt,t), and conductivity Con(Pnt,t), not only enables spatially distributed monitoring of the sewage system's operating status but also introduces dynamic response characteristics in the temporal dimension. By calculating the rate of change of the sampling point Pnt between time t and time t−Δt, derivative parameters such as the pH change Dph(Pnt,t) and conductivity change rate Dct(Pnt,t) are constructed. This "transient disturbance behavior" is integrated into the sewage treatment process control as diagnostic fundamental data. Based on the combined expression of these derivative parameters and the raw environmental response parameters, the constructed raw multi-parameter feature vector group Vec(Pnt,t) comprehensively characterizes the dynamic state of water quality at a sampling point at a specific moment, providing extremely high time sensitivity and parameter linkage for subsequent regional efficacy response evaluation and control strategy optimization. Compared with the existing technology that only relies on static concentration values ​​or single parameter triggering for dosing decision-making, this feature vector construction method significantly improves the system's ability to perceive changes in water quality caused by minor disturbances, and has stronger predictive and feedforward control potential. It is especially suitable for industrial wastewater treatment scenarios with narrow reaction windows and fast disturbance responses.

[0078] Example 3

[0079] See also Figure 1 and Figure 3 Specifically: S2 includes S21 and S22;

[0080] S21. Based on the water flow velocity Vel(Pnt,t), pH value change Dph(Pnt,t), and conductivity change rate Dct(Pnt,t) in the original multi-parameter feature vector group Vec(Pnt,t) at the sampling point Pnt at time t, construct the drug response efficiency function Eta(Pnt,t) at the sampling point Pnt at time t. This function reflects the response efficiency of the unit dose at the sampling point Pnt. A higher value indicates a more significant drug efficacy at the sampling point Pnt.

[0081] The drug reaction efficiency function Eta(Pnt, t) of the sampling point Pnt at time point t is constructed in the following way:

[0082] ;

[0083] Where Eps is a minimum constant used to prevent division by zero, and Lam is a weight coefficient used to adjust the contribution ratio of the pH change Dph (Pnt, t) and the conductivity change rate Dct (Pnt, t).

[0084] S22, normalizing the drug response efficiency function Eta(Pnt, t) of all sampling points Pnt at time point t, extracting the maximum value and comparing the drug response efficiency function Eta(Pnt, t) of each sampling point Pnt at time point t, obtaining a regional drug efficacy defect factor Mat(pnt) of each sampling point Pnt, integrating the regional drug efficacy defect factors Mat(pnt) of all sampling points Pnt to construct a matrix representing the degree of regional drug efficacy attenuation of each sampling point, and obtaining a regional drug efficacy defect factor matrix Mat;

[0085] The regional efficacy defect factor matrix Mat is obtained by the following calculation formula:

[0086] ;

[0087] Wherein, Mat(Pnt) represents the regional efficacy defect factor of the sampling point Pnt relative to the maximum drug response efficiency function Eta(Pnt, t), and max represents the maximum value function, which is used to extract the maximum drug response efficiency function Eta(Pnt, t) at time t.

[0088] Example of constructing the drug reaction efficiency function Eta(Pnt, t) and the regional drug efficacy defect factor matrix Mat:

[0089] Taking the three sampling points Pnt=1, Pnt=2, and Pnt=3 set in an industrial wastewater treatment section as an example, the original multi-parameter feature vector group Vec(Pnt,t) at the same time point t includes the following parameter values:

[0090] Sampling point Pnt=1: water flow velocity Vel(Pnt,t)=0.85; pH value change Dph(Pnt,t)=0.18; conductivity change rate Dct(Pnt,t)=0.11;

[0091] Sampling point Pnt=2: water flow velocity Vel(Pnt,t)=0.42; pH value change Dph(Pnt,t)=0.05; conductivity change rate Dct(Pnt,t)=0.03;

[0092] Sampling point Pnt=3: water flow velocity Vel(Pnt,t)=0.73; pH value change Dph(Pnt,t)=0.12; conductivity change rate Dct(Pnt,t)=0.07;

[0093] The calculation formula of the drug reaction efficiency function Eta(Pnt,t) at the sampling point Pnt at time t is as follows:

[0094] Sampling point Pnt=1: Eta(Pnt=1,t)=0.235 / 0.851≈0.276;

[0095] Sampling point Pnt=2: Eta(Pnt=2,t)=0.065 / 0.421≈0.1544;

[0096] Sampling point Pnt=3: Eta(Pnt=3,t)=0.155 / 0.731≈0.212;

[0097] The Eta(Pnt,t) of the three sampling points are normalized to obtain the regional efficacy defect factor Mat(Pnt) of each sampling point.

[0098] Sampling point Pnt=1: Mat(Pnt=1,t)=1-0.276 / 0.276=0;

[0099] Sampling point Pnt=2: Mat(Pnt=2,t)=1-0.1544 / 0.2760≈0.44;

[0100] Sampling point Pnt=3: Mat(Pnt=3,t)=1-0.212 / 0.276≈0.23;

[0101] In this embodiment, by introducing a drug reaction efficiency function Eta(Pnt,t) at sampling point Pnt, constructed with water flow velocity Vel(Pnt,t), pH change Dph(Pnt,t), and conductivity change rate Dct(Pnt,t) as core parameters, real-time modeling of the reactivity of a unit dose of drug at different treatment locations is achieved. This function not only quantifies the strength of the reaction efficiency but also controls the relative contribution of the two perturbations, pH and conductivity, to drug efficacy using an adjustment coefficient Lam. Compared to existing technologies that determine drug efficacy solely based on changes in terminal water quality, this function possesses stronger local resolution and real-time response sensitivity. Furthermore, the drug reaction efficiency function Eta(Pnt,t) at all sampling points Pnt is differentially characterized using a maximum normalization method to construct a regional drug efficacy incompleteness factor Mat(Pnt), thereby achieving a quantitative expression of the spatial attenuation of drug efficacy within the system. The regional efficacy defect factor matrix Mat enables the system to not only identify the problem of "how much medicine to add", but also identify the problem of "which areas the medicine is ineffective or inefficient", greatly enhancing the system's ability to identify reaction dead zones and potential waste areas, filling the key defect of the lack of spatial response difference expression method in traditional dosing control, and also providing a data-driven response benchmark for subsequent dosing correction strategies.

[0102] Example 4

[0103] Specifically: S3 includes S31 and S32;

[0104] S31. Based on the regional efficacy defect factor matrix Mat and the preset response failure judgment threshold Tau, determine whether each sampling point Pnt meets the response failure condition, and obtain the response failure status label Flg of each sampling point Pnt;

[0105] The response failure status tag Flg of the sampling point Pnt is obtained by the following response failure conditions:

[0106] When the regional efficacy defect factor Mat(pnt) of the sampling point Pnt is ≥ the response failure judgment threshold Tau, the response failure status label Flg(Pnt) of the sampling point Pnt is obtained as 1, indicating that the sampling point Pnt is an area with unqualified efficacy utilization rate;

[0107] When the regional efficacy defect factor Mat(pnt) of the sampling point Pnt is less than the response failure judgment threshold Tau, the response failure state label Flg(Pnt) of the sampling point Pnt is obtained as 0, indicating that the sampling point Pnt is a region with qualified efficacy utilization rate;

[0108] S32. By extracting the number of sampling points Pnt with the response failure state label Flg=1 of all marked sampling points Pnt in the regional efficacy defect factor matrix Mat, and then calculating the ratio between them and the sewage treatment multi-parameter collection position group Pos, the response failure area ratio Rtd is obtained.

[0109] Said S4 includes S41;

[0110] S41. Based on the regional efficacy defect factor matrix Mat and the original dosing plan dose vector Cref, a modified dose vector Adj(Pnt) is generated for each sampling point Pnt, which is used to modify the dosing strategy of the sampling point Pnt, thereby forming a spatially differentiated dosing strategy for the wastewater treatment multi-parameter collection position group Pos;

[0111] Among them, the original dosing plan dosage vector Cref represents the preset dosing dosage for industrial wastewater;

[0112] The corrected dose vector Adj(Pnt) of the sampling point Pnt is obtained by the following calculation formula:

[0113] ;

[0114] Where Cref(Pnt) represents the original planned dosing dose vector at the sampling point Pnt, and Alp represents the dose correction adjustment coefficient, which is used to adjust the sensitivity of the defect level to the dose correction. The specific value is set by the user.

[0115] In this embodiment, based on a combined determination mechanism based on the regional efficacy deficiency factor matrix Mat and the response failure determination threshold Tau, the present invention introduces a response failure status tag Flg(Pnt) for the first time in industrial wastewater treatment dosing control. This tag is used to indicate whether a sampling point Pnt exhibits a low efficacy response. Compared to traditional approaches using "overall error scoring" or "global concentration deviation," the introduction of the status tag Flg(Pnt) enables the system to determine response point by point, effectively identifying local dead zones or areas of inefficient response. Furthermore, the response failure region ratio Rtd, calculated based on the response failure status tags Flg(Pnt) for all sampling points Pnt, not only quantifies the proportion of inefficient areas in the system as a whole but also provides a macro-level threshold for determining whether to trigger a correction strategy. If the determination is successful, the modified dose vector Adj, constructed by combining the regional efficacy deficiency factor matrix Mat with the original dosing plan dose vector Cref, enables the system to precisely adjust the dosage based on the degree of deficiency at each sampling point Pnt, forming a spatially differentiated dosing control scheme. This mechanism significantly improves the control system's ability to handle "regional ineffective response", allowing the dosing strategy to evolve from "uniform dosage" to "dynamic adjustment according to response performance", effectively avoiding problems such as local over-dosing and insufficient dosage in boundary areas, and solving the core shortcomings of traditional systems in the face of spatial efficacy heterogeneity, such as rough control methods and insensitive feedback.

[0116] Example 5

[0117] Specifically: the S5 includes S51 and S52;

[0118] S51, calculating the comprehensive dosing response effectiveness index Eff based on the obtained corrected dose vector Adj and the drug response efficiency function Eta;

[0119] The comprehensive dosing response effectiveness index Eff is obtained by the following calculation formula:

[0120] ;

[0121] Where Eta(Pnt, t) represents the drug reaction efficiency function of the sampling point Pnt at time point t, and Mat(pnt) represents the regional efficacy defect factor of the sampling point Pnt.

[0122] S52: Compare the comprehensive dosing response effectiveness index Eff with the preset system response evaluation threshold Thr, and trigger the execution of the dosing control strategy based on the comparison result to determine the overall effectiveness of the current dosing strategy.

[0123] The execution of the dosing control strategy is triggered by the following comparison mechanism:

[0124] When the comprehensive dosing response effectiveness index Eff is less than the system response evaluation threshold Thr, the current dosing response effect is determined to be unqualified, indicating that the current original dosing plan dose vector Cref reflects an unqualified efficiency, and steps S1 to S4 are executed again to generate a new corrected dose vector Adj, and a second comparison is performed;

[0125] When the comprehensive dosing response effectiveness index Eff ≥ the system response evaluation threshold Thr, the current dosing response effect is determined to be qualified, and the executed dosage is the generated corrected dose vector Adj. At the same time, the corrected dose vector Adj can be used as a reference for adaptive updating of the original dosing plan dose vector Cref to form a new version of the original dosing plan dose vector, and the original dosing plan dose vector Cref continues to be used.

[0126] In this embodiment, by constructing and calculating a comprehensive dosing response effectiveness index (Eff), a global, quantitative evaluation of the current dosing response status of the entire sewage treatment system is achieved. This not only considers the drug reaction efficiency function (Eta(Pnt,t)) at the sampling point (Pnt), but also incorporates the spatial failure characteristics of the regional efficacy incompleteness factor (Mat(Pnt)). This effectively avoids the blind spot in traditional control systems where excellent local response masks global failure. By comparing the comprehensive dosing response effectiveness index (Eff) with the system response evaluation threshold (Thr), the system can autonomously identify the overall effectiveness of the current dosing strategy. When the dosing response does not meet the standard, steps S1 to S4 are automatically backtracked to regenerate the corrected dose vector Adj(Pnt) and verify its response performance again; when the dosing response meets the expected standard, the system not only actually executes the generated corrected dose vector Adj(Pnt), but also uses it as feedback to dynamically optimize the original dosing plan dose vector Cref(Pnt), thereby achieving continuous evolution of control experience. This breaks the limitations of traditional industrial dosing systems that rely on fixed dose rules and lack data feedback judgment, and constructs an intelligent closed-loop control framework that drives strategy evolution with response performance and quantifies the pros and cons of strategies with system-level indicators. This greatly improves the adaptability, self-learning ability and stability of the control strategy, and is particularly suitable for dynamic treatment scenarios with frequent fluctuations in sewage quality and complex pollutant structures.

[0127] Example 6

[0128] An intelligent dosing control system for industrial wastewater plants based on multi-parameter self-adaptation, please refer to Figure 2 ,Specifically: including water quality data acquisition module, regional efficacy detection module, regional efficacy response determination module, correction module and decision module;

[0129] The water quality data acquisition module collects the original environmental response parameters of each sewage treatment multi-parameter acquisition position group Pos by deploying water quality monitoring devices in the sewage treatment multi-parameter acquisition position group Pos, and constructs the original multi-parameter feature vector group Vec of the industrial sewage treatment system;

[0130] The regional efficacy detection module constructs the regional efficacy incomplete factor matrix Mat of the sewage treatment multi-parameter collection position group Pos based on the constructed original multi-parameter feature vector group Vec;

[0131] The regional efficacy response determination module calculates the response failure region ratio Rtd based on the regional efficacy defect factor matrix Mat and the preset response failure determination threshold Tau, which is used to identify areas with low efficacy utilization in the sewage treatment multi-parameter collection position group Pos;

[0132] The correction module generates a correction dose vector Adj based on the regional efficacy defect factor matrix Mat and the original dosing plan dose vector Cref, forming a spatially differentiated dosing strategy for the sewage treatment multi-parameter collection position group Pos;

[0133] The decision module uses the modified dose vector Adj and the drug response efficiency function Eta to calculate the comprehensive drug dosing response effectiveness index Eff and determine the overall effectiveness of the current drug dosing strategy.

[0134] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent dosing control method for industrial wastewater treatment plants based on multi-parameter self-adaptation, characterized by: The following steps are involved: S1. By deploying water quality monitoring devices in the sewage treatment multi-parameter collection position group Pos, collecting the original environmental response parameters of each sewage treatment multi-parameter collection position group Pos, and constructing the original multi-parameter feature vector group Vec of the industrial sewage treatment system; S2. Based on the constructed original multi-parameter feature vector group Vec, construct the regional efficacy incomplete factor matrix Mat of the sewage treatment multi-parameter collection position group Pos; S3. Calculate the response failure region ratio Rtd based on the regional efficacy defect factor matrix Mat and the preset response failure judgment threshold Tau, so as to identify the region with low efficacy utilization in the sewage treatment multi-parameter collection position group Pos; S4. Based on the regional efficacy defect factor matrix Mat and the original dosing plan dose vector Cref, a modified dose vector Adj is generated to form a spatially differentiated dosing strategy for the sewage treatment multi-parameter collection position group Pos; S5. Calculate the comprehensive dosing response effectiveness index Eff using the corrected dose vector Adj and the drug response efficiency function Eta to determine the overall effectiveness of the current dosing strategy.

2. The intelligent dosing control method for industrial wastewater treatment plants based on multi-parameter self-adaptation according to claim 1 is characterized in that: Said S1 includes S11 and S12; S11. At each sampling point Pnt∈Pos in a plurality of preset sewage treatment multi-parameter collection position groups Pos, a water quality monitoring device is deployed to collect the original environmental response parameters of each sewage treatment multi-parameter collection position group Pos in real time; The water quality monitoring device includes a water flow velocity sensor, a pH value sensor and a conductivity sensor; The original environmental response parameters include the water flow velocity Vel(Pnt,t) at the sampling point Pnt at time t, the pH value Phv(Pnt,t) at the sampling point Pnt at time t, and the conductivity Con(Pnt,t) at the sampling point Pnt at time t.

3. The intelligent dosing control method for industrial wastewater treatment plants based on multi-parameter self-adaptation according to claim 2 is characterized in that: S12. Based on the obtained original environmental response parameters, calculate the response change rate per unit time for each sampling point Pnt, specifically by calculating the change rate between time t and time t-Δt, where Δt represents the time interval, to form a derivative parameter reflecting the transient disturbance characteristics; The derivative parameters include the pH value change Dph (Pnt, t) at the sampling point Pnt at time t and the conductivity change rate Dct (Pnt, t) at the sampling point Pnt at time t; By integrating the derivative parameters and the original environmental response parameters, the original multi-parameter feature vector group Vec(Pnt, t) = {Dph(Pnt, t), Dct(Pnt, t), Vel(Pnt, t), Phv(Pnt, t), Con(Pnt, t)} of the sampling point Pnt at time t in the industrial wastewater treatment system is constructed.

4. The intelligent dosing control method for industrial wastewater treatment plants based on multi-parameter self-adaptation according to claim 3 is characterized in that: Said S2 includes S21 and S22; S21. Based on the water flow velocity Vel(Pnt,t), pH value change Dph(Pnt,t) and conductivity change rate Dct(Pnt,t) in the original multi-parameter feature vector group Vec(Pnt,t) at the sampling point Pnt at time t, construct the agent reaction efficiency function Eta(Pnt,t) at the sampling point Pnt at time t.

5. The intelligent dosing control method for industrial wastewater treatment plants based on multi-parameter self-adaptation according to claim 4 is characterized in that: S22. Normalize the drug reaction efficiency function Eta(Pnt, t) of all sampling points Pnt at time point t, and extract the maximum value and compare the drug reaction efficiency function Eta(Pnt, t) of each sampling point Pnt at time point t to obtain the regional drug efficacy defect factor Mat(pnt) of each sampling point Pnt. Integrate the regional drug efficacy defect factors Mat(pnt) of all sampling points Pnt to construct a matrix representing the degree of regional drug efficacy attenuation of each sampling point, and obtain the regional drug efficacy defect factor matrix Mat.

6. The intelligent dosing control method for industrial wastewater treatment plants based on multi-parameter self-adaptation according to claim 5, characterized in that: Said S3 includes S31 and S32; S31. Based on the regional efficacy defect factor matrix Mat and the preset response failure judgment threshold Tau, determine whether each sampling point Pnt meets the response failure condition, and obtain the response failure status label Flg of each sampling point Pnt; The response failure status tag Flg of the sampling point Pnt is obtained by the following response failure conditions: When the regional efficacy defect factor Mat(pnt) of the sampling point Pnt is ≥ the response failure judgment threshold Tau, the response failure status label Flg(Pnt) of the sampling point Pnt is obtained as 1, indicating that the sampling point Pnt is an area with unqualified efficacy utilization rate; When the regional efficacy defect factor Mat(pnt) of the sampling point Pnt is less than the response failure judgment threshold Tau, the response failure state label Flg(Pnt) of the sampling point Pnt is obtained as 0, indicating that the sampling point Pnt is a region with qualified efficacy utilization rate; S32. By extracting the number of sampling points Pnt with the response failure state label Flg=1 of all marked sampling points Pnt in the regional efficacy defect factor matrix Mat, and then calculating the ratio between them and the sewage treatment multi-parameter collection position group Pos, the response failure area ratio Rtd is obtained.

7. The intelligent dosing control method for industrial wastewater treatment plants based on multi-parameter self-adaptation according to claim 6, characterized in that: Said S4 includes S41; S41. Based on the regional efficacy defect factor matrix Mat and the original dosing plan dose vector Cref, a modified dose vector Adj(Pnt) is generated for each sampling point Pnt, which is used to modify the dosing strategy of the sampling point Pnt, thereby forming a spatially differentiated dosing strategy for the wastewater treatment multi-parameter collection position group Pos; Among them, the original dosing plan dosage vector Cref represents the preset dosing dosage for industrial wastewater; The corrected dose vector Adj(Pnt) of the sampling point Pnt is obtained by the following calculation formula: ； Where Cref(Pnt) represents the original planned dosing dose vector at the sampling point Pnt, and Alp represents the dose correction adjustment coefficient, which is used to adjust the sensitivity of the defect level to the dose correction. The specific value is set by the user.

8. The intelligent dosing control method for industrial wastewater treatment plants based on multi-parameter self-adaptation according to claim 7, characterized in that: Said S5 includes S51 and S52; S51, calculating the comprehensive dosing response effectiveness index Eff based on the obtained corrected dose vector Adj and the drug response efficiency function Eta; The comprehensive dosing response effectiveness index Eff is obtained by the following calculation formula: ； Where Eta(Pnt, t) represents the drug reaction efficiency function of the sampling point Pnt at time point t, and Mat(pnt) represents the regional efficacy defect factor of the sampling point Pnt.

9. The intelligent dosing control method for industrial wastewater treatment plants based on multi-parameter self-adaptation according to claim 8, characterized in that: S52: Compare the comprehensive dosing response effectiveness index Eff with the preset system response evaluation threshold Thr, and trigger the execution of the dosing control strategy based on the comparison result to determine the overall effectiveness of the current dosing strategy. The execution of the dosing control strategy is triggered by the following comparison mechanism: When the comprehensive dosing response effectiveness index Eff is less than the system response evaluation threshold Thr, the current dosing response effect is determined to be unqualified, indicating that the current original dosing plan dose vector Cref reflects an unqualified efficiency, and steps S1 to S4 are executed again to generate a new corrected dose vector Adj, and a second comparison is performed; When the comprehensive dosing response effectiveness index Eff ≥ the system response evaluation threshold Thr, the current dosing response effect is determined to be qualified, and the executed dosage is the generated corrected dose vector Adj. At the same time, the corrected dose vector Adj can be used as a reference for adaptive updating of the original dosing plan dose vector Cref to form a new version of the original dosing plan dose vector, and the original dosing plan dose vector Cref continues to be used.

10. An intelligent dosing control system for an industrial wastewater treatment plant based on multi-parameter self-adaptation, applied to the intelligent dosing control method for an industrial wastewater treatment plant based on multi-parameter self-adaptation according to any one of claims 1 to 9, characterized in that: It includes water quality data acquisition module, regional efficacy detection module, regional efficacy response determination module, correction module and decision module; The water quality data acquisition module collects the original environmental response parameters of each sewage treatment multi-parameter acquisition position group Pos by deploying water quality monitoring devices in the sewage treatment multi-parameter acquisition position group Pos, and constructs the original multi-parameter feature vector group Vec of the industrial sewage treatment system; The regional efficacy detection module constructs the regional efficacy incomplete factor matrix Mat of the sewage treatment multi-parameter collection position group Pos based on the constructed original multi-parameter feature vector group Vec; The regional efficacy response determination module calculates the response failure region ratio Rtd based on the regional efficacy defect factor matrix Mat and the preset response failure determination threshold Tau, which is used to identify areas with low efficacy utilization in the sewage treatment multi-parameter collection position group Pos; The correction module generates a correction dose vector Adj based on the regional efficacy defect factor matrix Mat and the original dosing plan dose vector Cref, forming a spatially differentiated dosing strategy for the sewage treatment multi-parameter collection position group Pos; The decision module uses the modified dose vector Adj and the drug response efficiency function Eta to calculate the comprehensive drug dosing response effectiveness index Eff and determine the overall effectiveness of the current drug dosing strategy.

Images