Method for generating characteristic inversion heavy metal content model after industrial sewage treatment
By clustering and reverse deriving the time series data of the sewage treatment system, an inversion model of heavy metal content after industrial sewage treatment was generated, solving the problems of high monitoring costs, time-consuming and insufficient robustness in the existing technology, and achieving fast and accurate prediction of heavy metal content.
Patent Information
- Application Number
- CN202510545611.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-28
- Publication Date
- 2025-07-22
AI Technical Summary
In the prior art, the monitoring method of heavy metal content after industrial wastewater treatment depends on laboratory chemical analysis, which is costly, time-consuming, incapable of real-time monitoring and is susceptible to human errors. The machine learning-based model is not robust enough, making it difficult to achieve fast and accurate prediction of heavy metal content.
By obtaining the time series data of the same type of historical sewage in the sewage treatment system, performing clustering analysis and feature extraction, using the K-mean clustering algorithm to determine the dynamic data of important parameters, establishing a target calculation model, and generating a final inversion model through reverse derivation and least squares method optimization, realizing accurate calculation of heavy metal content.
It realizes rapid and accurate prediction of heavy metal content after industrial wastewater treatment, improves the reduction degree of heavy metal content historical data and the robustness of model, and meets the needs of on-site real-time monitoring.
Smart Images
Figure CN120356567A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of water treatment monitoring, and specifically to a method for generating a model for inverting heavy metal content based on the characteristics of industrial sewage after treatment. Background Art
[0002] Industrial sewage treatment is an important link in the field of environmental protection, and heavy metal pollution is one of the most intractable problems in industrial sewage. Even at low concentrations, heavy metal elements (such as lead, cadmium, mercury, arsenic, etc.) in industrial sewage can cause serious harm to the environment and human health. Therefore, accurate monitoring and prediction of heavy metal content after industrial sewage treatment are particularly important.
[0003] Traditional heavy metal monitoring methods mainly rely on laboratory chemical analysis techniques, such as atomic absorption spectrometry (AAS), inductively coupled plasma mass spectrometry (ICP-MS), etc. However, these methods have the following significant drawbacks: First, laboratory testing requires professional equipment and technical personnel, which is costly and time-consuming; second, laboratory testing cannot achieve on-site real-time monitoring and is difficult to meet the rapid detection requirements during industrial sewage treatment; finally, laboratory testing methods have high requirements for sample pretreatment, are complex to operate, and are easily affected by human errors.
[0004] In recent years, mathematical models based on machine learning have gradually been applied to heavy metal content prediction, but still face challenges such as insufficient data and insufficient model robustness. Therefore, there is an urgent need for a new method that can quickly and accurately predict the heavy metal content after industrial sewage treatment. Summary of the Invention
[0005] Technical Problems to be Solved In view of the deficiencies of the prior art, the present invention provides a method for generating a model for inverting heavy metal content based on the characteristics of industrial sewage after treatment to solve the problems raised in the above background art. Technical Solutions
[0006] To achieve the above objectives, the present invention is realized through the following technical solutions: A method for generating a model for inverting heavy metal content based on the characteristics of industrial sewage after treatment, comprising the following steps: Obtain a set of historical sewage treatment data in time series that belongs to the same type as the target sewage in the sewage treatment system, where the treatment data includes the dynamic parameter data of historical sewage at each time point within the time series; Cluster the dynamic parameter data to obtain the important dynamic parameter data of historical sewage; Based on the process parameters of the sewage treatment system, determine a time calculation model for the treatment data of sewage in the sewage treatment system, and use the important dynamic parameter data to optimize the time calculation model to obtain a target calculation model; Calculate the predicted sewage data of the second node through the target calculation model and the actual sewage data of the first node, and determine the inversion model of the sewage data of the first node based on the predicted sewage data of the second node and the actual data of the second node; Modify and optimize the inversion model of the sewage data of the first node based on the dynamic data of important parameters to obtain the final inversion model.
[0007] As a preference of this embodiment, the processed data reflects the water quality parameters and heavy metal content of the historical sewage in the sewage treatment system, and the parameter dynamic data of the historical sewage at each time point includes the parameter data after treatment at each process node of the sewage in the sewage treatment system.
[0008] As a preference of this embodiment, clustering the parameter dynamic data, and the obtained important parameter dynamic data of the historical sewage includes: Extract features from the parameter dynamic data to obtain multiple time series parameter features; Vectorize the multiple time series parameter features to obtain parameter feature vectors; Use the K-means clustering algorithm to cluster the parameter feature vectors to obtain the clustering centers of the parameter feature vectors; Analyze the clustering centers to obtain the important parameter dynamic data of the parameter dynamic data.
[0009] As a preference of this embodiment, the process of determining the inversion model of the sewage data of the first node based on the predicted sewage data of the second node and the actual data of the second node includes: Modify the predicted sewage data of the second node based on the actual data of the second node to obtain the modified data; Determine the inversion model of the sewage data of the first node according to the modified data.
[0010] As a preference of this embodiment, the process of modifying and optimizing the inversion model of the sewage data of the first node based on the dynamic data of important parameters to obtain the final inversion model includes: Determine the full-process parameter data of each node corresponding to the sewage of the first node based on the dynamic data of important parameters, and obtain all node parameter data before the first node; Modify the inversion data of each node through all the node parameter data before the first node, and perform overall modification based on the correction results corresponding to each node and the inversion model of the sewage data of the first node to obtain the final inversion model.
[0011] As a preference of this embodiment, after obtaining the final inversion model by performing overall modification based on the correction results corresponding to each node and the inversion model of the sewage data of the first node, it further includes: Optimize the final inversion model based on the least squares method to obtain an optimized inversion model; Among them, the formula for optimizing the final inversion model based on the least squares method is: ; Among them, y′ is the output of the corrected model, ŷ is the predicted value of the sewage parameters, λ is the correction factor, and y is the actual value of the sewage parameters; The calculation formula of the correction factor is: ; Among them, N represents the number of process nodes for obtaining treatment data of the sewage in the sewage treatment system.
[0012] As a preference of this embodiment, a system for generating a heavy metal content model for the characteristics of industrial sewage after treatment includes: An acquisition processing module, configured to acquire a set of time-series treatment data of historical sewage of the same type as the target sewage in the sewage treatment system, where the treatment data includes parameter dynamic data of the historical sewage at each time point within the time series, and perform clustering on the parameter dynamic data to obtain important parameter dynamic data of the historical sewage; Among them, the treatment data reflects the water quality parameters and heavy metal content of the historical sewage in the sewage treatment system, and the parameter dynamic data of the historical sewage at each time point includes the parameter data after each process node of the sewage in the sewage treatment system; A target calculation model establishment module, configured to determine a time calculation model of the treatment data of the sewage in the sewage treatment system based on the process parameters of the sewage treatment system, and optimize the time calculation model by using the important parameter dynamic data to obtain a target calculation model; An inversion model establishment module, calculates the predicted sewage data of the second node through the target calculation model and the actual sewage data of the first node, determines the inversion model of the sewage data of the first node based on the predicted sewage data of the second node and the actual data of the second node, and also corrects and optimizes the inversion model of the sewage data of the first node based on the important parameter dynamic data to obtain a final inversion model.
[0013] As a preference of this embodiment, performing clustering on the parameter dynamic data to obtain important parameter dynamic data of the historical sewage includes: Extract features from the parameter dynamic data to obtain multiple time-series parameter features; Vectorize the multiple time-series parameter features to obtain parameter feature vectors; Use the K-means clustering algorithm to cluster the parameter feature vectors to obtain the clustering centers of the parameter feature vectors; Analyze the clustering centers to obtain the important parameter dynamic data of the parameter dynamic data.
[0014] Preferably in this embodiment, the process of the inversion model for determining the sewage data of the first node based on the sewage prediction data of the second node and the actual data of the second node includes: Correct the sewage prediction data of the second node based on the actual data of the second node to obtain the corrected data; Determine the inversion model of the sewage data of the first node according to the corrected data.
[0015] Preferably in this embodiment, the process of correcting and optimizing the inversion model of the sewage data of the first node based on the dynamic data of important parameters to obtain the final inversion model includes: Determine the full-process parameter data of each node in the sewage treatment system corresponding to the sewage of the first node based on the dynamic data of important parameters, and obtain all the node parameter data before the first node; Correct the inversion data of each node through all the node parameter data before the first node, and perform overall correction based on the correction results corresponding to each node and the inversion model of the sewage data of the first node to obtain the final inversion model; Among them, after obtaining the final inversion model by performing overall correction based on the correction results corresponding to each node and the inversion model of the sewage data of the first node, it further includes: Optimize the final inversion model based on the least squares method to obtain the optimized inversion model; Among them, the formula for optimizing the final inversion model based on the least squares method is: ; Among them, y′ is the output of the corrected model, ŷ is the predicted value of the sewage parameter, λ is the correction factor, and y is the actual value of the sewage parameter; The calculation formula of the correction factor is: ; Among them, N represents the number of process nodes for obtaining treatment data of sewage in the sewage treatment system.
[0016] Beneficial effects The present invention provides a method for generating a model for inverting heavy metal content based on the characteristics of industrial sewage after treatment, which has the following beneficial effects: By obtaining a set of processing data in time series of historical sewage of the same type as the target sewage in the sewage treatment system, the processing data includes the dynamic parameter data of historical sewage at each time point within the time series; clustering the dynamic parameter data to obtain the important dynamic parameter data of historical sewage; determining a time calculation model for the processing data of sewage in the sewage treatment system based on the process parameters of the sewage treatment system, and optimizing the time calculation model using the important dynamic parameter data to obtain a target calculation model; calculating the predicted sewage data of the second node through the target calculation model and the actual sewage data of the first node, and determining an inversion model for the sewage data of the first node based on the predicted sewage data of the second node and the actual data of the second node; correcting and optimizing the inversion model for the sewage data of the first node based on the important dynamic parameter data to obtain a final inversion model. The present application extracts the heavy metal data from the whole-process data of sewage treatment, clusters and analyzes the dynamic data of each node of sewage in the sewage treatment system, uses the change relationship of the dynamic data of sewage at each node for forward deduction, then through reverse deduction, inverses each node separately, and then through the comprehensive evolution of the whole-process inversion data, realizes the accurate calculation of the heavy metal content in sewage treatment, and further optimizes the final inversion model to improve the reduction degree of the historical data of the heavy metal content after sewage treatment. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] Figure 1 It is a flowchart of the method for generating a model for inverting heavy metal content based on the characteristics of industrial sewage after treatment according to the present invention; Figure 2 It is a block diagram of the system for generating a model for inverting heavy metal content based on the characteristics of industrial sewage after treatment according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS
[0018] The following describes in detail the embodiments of the present invention. The examples of the embodiments are shown in the drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions from beginning to end. The embodiments described below by referring to the drawings are exemplary only for explaining the present invention and should not be construed as limiting the present invention.
[0019] The following disclosure provides many different embodiments or examples for implementing different structures of the present invention. To simplify the disclosure of the present invention, the components and settings of specific examples are described below. Of course, they are only examples and are not intended to limit the present invention. In addition, the present invention may repeat reference numerals and / or reference letters in different examples. This repetition is for the purpose of simplification and clarity, and does not itself indicate the relationship between the various embodiments and / or settings discussed. In addition, the present invention provides examples of various specific processes and materials, but those of ordinary skill in the art can be aware of the application of other processes and / or the use of other materials.
[0020] As Figure 1 shown, an embodiment of the present invention provides a method for generating a heavy metal content model by inverting the characteristics after industrial sewage treatment, including the following steps: Step 1: Obtain a set of processing data of historical sewage of the same type as the target sewage in the sewage treatment system in a time series. The processing data includes the dynamic parameter data of the historical sewage at each time point within the time series; Among them, the processing data reflects the water quality parameters and heavy metal content of the historical sewage in the sewage treatment system. The dynamic parameter data of the historical sewage at each time point includes the parameter data after the treatment of each process node of the sewage in the sewage treatment system.
[0021] It can be understood that the water quality parameters and heavy metal content of the historical sewage reflected by the processing data are the data of the historical sewage processed within a period or multiple periods.
[0022] Step 2: Cluster the dynamic parameter data to obtain the important dynamic parameter data of the historical sewage; Among them, after receiving the dynamic parameter data, the computer device can analyze the linear or non-linear relationship between the heavy metal content and each parameter in the process of sewage treatment.
[0023] It can be understood that different nodes in the sewage treatment system will cause corresponding changes in the parameter data of the sewage itself. During the treatment of the sewage within a period, every time the sewage passes through a node within this period, the heavy metal content, impurities, ammonia nitrogen content, and other substances to be removed in the sewage will change. By analyzing the changes of these substances to be removed during the entire process of sewage treatment and clustering these data, the data changes that occur at each node of the sewage can be obtained, as well as the specific change trends and data.
[0024] In this embodiment, clustering the dynamic parameter data to obtain the important dynamic parameter data of the historical sewage includes: Extract features from the dynamic parameter data to obtain multiple time series parameter features; Vectorize multiple time series parameter features to obtain parameter feature vectors; Use the K-means clustering algorithm to cluster the parameter feature vectors to obtain the cluster centers of the parameter feature vectors; Analyze the cluster centers to obtain the important parameter dynamic data of the parameter dynamic data.
[0025] It can be understood that the important parameter dynamic data of historical sewage is specifically the parameters related to the heavy metal content during the sewage treatment process.
[0026] Specifically, during the sewage treatment process, sewage will enter different nodes of the sewage treatment system due to time changes. When obtaining the important parameter dynamic data of historical sewage, the features of the parameter dynamic data can be extracted, and after vectorizing these extracted features, parameter feature vectors can be obtained. Then, use the K-means clustering algorithm to cluster the parameter feature vectors to obtain the cluster centers of the parameter feature vectors, so as to obtain the important parameter dynamic data of the parameter dynamic data.
[0027] It should be noted that the parameter dynamic data includes the data of sewage at each node in the sewage treatment system during the treatment process. These data have similar time differences, thus ensuring the consistency between the data of each node and further ensuring the reliability of subsequent analysis.
[0028] Step 3: Determine the time calculation model for the sewage treatment data in the sewage treatment system based on the process parameters of the sewage treatment system, and optimize the time calculation model using the important parameter dynamic data to obtain the target calculation model; In this embodiment, based on the parameter dynamic data of historical sewage at each time point within the collected time series, the data of sewage at each node in the sewage treatment system is determined. The change (fluctuation) data of the substances to be removed along the time line during the sewage treatment process can be determined, so as to determine the time calculation model. Use the dynamic data of important parameters to optimize the time calculation model to obtain the calculation model along the sewage treatment process (i.e., the target calculation model).
[0029] Step 4: Calculate the predicted sewage data of the second node through the target calculation model and the actual sewage data of the first node, and determine the inversion model of the sewage data of the first node based on the predicted sewage data of the second node and the actual data of the second node; In this embodiment, based on the target calculation model determined in step three, that is, based on the correlation of sewage data in the sewage treatment process, the problem can be deduced backward and still holds. During the backward deduction calculation process, substitute the parameters of the sewage at node T (any node in sewage treatment) into the backward deduction calculation formula (target calculation model) to obtain the sewage parameters at time T-1, and combine with the actual parameters of the sewage at time T-1 for model optimization (correction), then the inversion model at time T can be obtained.
[0030] Specifically, the process of determining the inversion model of the sewage data at the first node based on the predicted sewage data at the second node and the actual data at the second node includes: Correct the predicted sewage data at the second node based on the actual data at the second node to obtain the corrected data; Determine the inversion model of the sewage data at the first node according to the corrected data.
[0031] Step five: Correct and optimize the inversion model of the sewage data at the first node based on the dynamic data of important parameters to obtain the final inversion model.
[0032] In this embodiment, the inversion model obtained in step four is only for time T. There are several nodes in the sewage treatment process. Rely on the inversion model from time T to T-1 to the inversion model from time T-1 to T-2, and combine with the sewage parameters at time T-2 for correction, gradually deduce the inversion model of the initial node, and then analyze the inversion model of the entire sewage treatment process to obtain the final inversion model.
[0033] It can be understood that the heavy metal content characteristics of the sewage at each node in the sewage treatment system can be deduced through the final inversion model.
[0034] Correcting and optimizing the inversion model of the sewage data at the first node based on the dynamic data of important parameters to obtain the final inversion model includes: Determine the full-process parameter data of the sewage corresponding to the first node at each node in the sewage treatment system based on the dynamic data of important parameters to obtain the parameter data of all nodes before the first node; Correct the inversion data of each node through the parameter data of all nodes before the first node, and perform overall correction based on the correction results corresponding to each node and the inversion model of the sewage data at the first node to obtain the final inversion model.
[0035] Furthermore, after obtaining the final inversion model by performing overall correction based on the correction results corresponding to each node and the inversion model of the sewage data at the first node, it further includes: Optimize the final inversion model based on the least squares method to obtain the optimized inversion model; Among them, the formula for optimizing the final inversion model based on the least squares method is as follows: ; Among them, y′ is the output of the corrected model, ŷ is the predicted value of the sewage parameters, λ is the correction factor, and y is the actual value of the sewage parameters; The calculation formula for the correction factor is: ; Among them, N represents the number of process nodes where the sewage obtains treatment data in the sewage treatment system.
[0036] The method for generating the heavy metal content inversion model for the characteristics after industrial sewage treatment provided in this embodiment obtains a set of time-series treatment data of historical sewage of the same type as the target sewage in the sewage treatment system. The treatment data includes the parameter dynamic data of the historical sewage at each time point within the time series; clusters the parameter dynamic data to obtain the important parameter dynamic data of the historical sewage; determines the time calculation model of the sewage treatment data in the sewage treatment system based on the process parameters of the sewage treatment system, and uses the important parameter dynamic data to optimize the time calculation model to obtain the target calculation model; calculates the predicted sewage data of the second node through the target calculation model and the actual sewage data of the first node, and determines the inversion model of the sewage data of the first node based on the predicted sewage data of the second node and the actual data of the second node; corrects and optimizes the inversion model of the sewage data of the first node based on the important parameter dynamic data to obtain the final inversion model. This application extracts the heavy metal data from the full-process data of sewage treatment, clusters and analyzes the dynamic data of each node of the sewage in the sewage treatment system, uses the change relationship of the dynamic data of each node of the sewage for forward deduction, then through reverse derivation, inverses each node separately, and then through the comprehensive evolution of the full-process inversion data, realizes the accurate calculation of the heavy metal content in sewage treatment, and further optimizes the final inversion model to improve the reduction degree of the historical data of the heavy metal content after sewage treatment.
[0037] This embodiment also provides a system for generating a heavy metal content inversion model for the characteristics after industrial sewage treatment, including: An acquisition processing module, configured to acquire a set of time-series treatment data of historical sewage of the same type as the target sewage in the sewage treatment system. The treatment data includes the parameter dynamic data of the historical sewage at each time point within the time series, and clusters the parameter dynamic data to obtain the important parameter dynamic data of the historical sewage; Among them, the treatment data reflects the water quality parameters and heavy metal content of the historical sewage in the sewage treatment system. The parameter dynamic data of the historical sewage at each time point includes the parameter data after each process node treatment of the sewage in the sewage treatment system; The target calculation model establishment module is used to determine the time calculation model for the sewage treatment data in the sewage treatment system based on the process parameters of the sewage treatment system, and optimize the time calculation model using the dynamic data of important parameters to obtain the target calculation model; The inversion model establishment module calculates the predicted sewage data of the second node through the target calculation model and the actual sewage data of the first node, determines the inversion model of the sewage data of the first node based on the predicted sewage data of the second node and the actual data of the second node, and also corrects and optimizes the inversion model of the sewage data of the first node based on the dynamic data of important parameters to obtain the final inversion model.
[0038] Further, clustering the parameter dynamic data to obtain the dynamic data of important parameters of historical sewage includes: Performing feature extraction on the parameter dynamic data to obtain multiple time series parameter features; Vectorizing the multiple time series parameter features to obtain parameter feature vectors; Using the K-means clustering algorithm to cluster the parameter feature vectors to obtain the clustering centers of the parameter feature vectors; Analyzing the clustering centers to obtain the dynamic data of important parameters of the parameter dynamic data.
[0039] Further, the process of determining the inversion model of the sewage data of the first node based on the predicted sewage data of the second node and the actual data of the second node includes: Correcting the predicted sewage data of the second node based on the actual data of the second node to obtain the corrected data; Determining the inversion model of the sewage data of the first node according to the corrected data.
[0040] Further, the process of correcting and optimizing the inversion model of the sewage data of the first node based on the dynamic data of important parameters to obtain the final inversion model includes: Determining the full-process parameter data of each node in the sewage treatment system for the sewage corresponding to the first node based on the dynamic data of important parameters to obtain the parameter data of all nodes before the first node; Correcting the inversion data of each node through the parameter data of all nodes before the first node, and performing overall correction based on the correction results corresponding to each node and the inversion model of the sewage data of the first node to obtain the final inversion model; Among them, after obtaining the final inversion model by performing overall correction based on the correction results corresponding to each node and the inversion model of the sewage data of the first node, it further includes: Optimizing the final inversion model based on the least squares method to obtain the optimized inversion model; Among them, the formula for optimizing the final inversion model based on the least squares method is: ; Among them, y′ is the output of the corrected model, ŷ is the predicted value of the sewage parameter, λ is the correction factor, and y is the actual value of the sewage parameter; The calculation formula of the correction factor is: ; Among them, N represents the number of process nodes for the sewage to obtain treatment data in the sewage treatment system.
[0041] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.
Claims
1. A method for generating a model for inverting heavy metal content based on the characteristics of industrial sewage after treatment, characterized in that, It includes the following steps: Obtain a set of processing data in time series of historical sewage of the same type as the target sewage in the sewage treatment system, where the processing data includes the dynamic parameter data of the historical sewage at each time point within the time series; Cluster the parameter dynamic data to obtain the important parameter dynamic data of the historical sewage; Determine a time calculation model for the processing data of the sewage in the sewage treatment system based on the process parameters of the sewage treatment system, and optimize the time calculation model using the important parameter dynamic data to obtain a target calculation model; Calculate the predicted sewage data of the second node through the target calculation model and the actual sewage data of the first node, and determine an inversion model for the sewage data of the first node based on the predicted sewage data of the second node and the actual data of the second node; Revise and optimize the inversion model for the sewage data of the first node based on the important parameter dynamic data to obtain a final inversion model.
2. A method for generating a heavy metal content model by inverting the characteristics after industrial sewage treatment according to claim 1, characterized in that, The processing data reflects the water quality parameters and heavy metal content of the historical sewage in the sewage treatment system, and the parameter dynamic data of the historical sewage at each time point includes the parameter data after each process node of the sewage in the sewage treatment system.
3. A method for generating a heavy metal content model by inverting the characteristics of industrial sewage after treatment according to claim 1, characterized in that, Clustering the parameter dynamic data to obtain the important parameter dynamic data of the historical sewage includes: Extract features from the parameter dynamic data to obtain multiple time series parameter features; Vectorize the multiple time series parameter features to obtain parameter feature vectors; Use the K-means clustering algorithm to cluster the parameter feature vectors to obtain the clustering centers of the parameter feature vectors; Analyze the clustering centers to obtain the important parameter dynamic data of the parameter dynamic data.
4. A method for generating a heavy metal content model by inverting the characteristics after industrial sewage treatment according to claim 1, wherein, The process of determining the inversion model for the sewage data of the first node based on the predicted sewage data of the second node and the actual data of the second node includes: Correct the predicted sewage data of the second node based on the actual data of the second node to obtain corrected data; Determine the inversion model for the sewage data of the first node according to the corrected data.
5. A method for generating a heavy metal content model by inverting the characteristics after industrial sewage treatment according to claim 4, characterized in that, Revising and optimizing the inversion model for the sewage data of the first node based on the important parameter dynamic data to obtain a final inversion model includes: Determine the full-process parameter data of the sewage corresponding to the first node at each node in the sewage treatment system based on the important parameter dynamic data, and obtain the parameter data of all nodes before the first node; Correct the inversion data of each node through the parameter data of all nodes before the first node, and perform overall correction based on the correction results corresponding to each node and the inversion model of the sewage data of the first node to obtain the final inversion model.
6. A method for generating a heavy metal content model by inverting the characteristics after industrial sewage treatment according to claim 5, wherein: After obtaining the final inversion model by performing overall correction based on the correction results corresponding to each node and the inversion model of the sewage data of the first node, it further includes: Optimize the final inversion model based on the least squares method to obtain an optimized inversion model; Among them, the formula for optimizing the final inversion model based on the least squares method is: ； Among them, y′ is the output of the corrected model, ŷ is the predicted value of the sewage parameter, λ is the correction factor, and y is the actual value of the sewage parameter; The calculation formula for the correction factor is: ； Among them, N represents the number of process nodes for obtaining the processing data of the sewage in the sewage treatment system.
7. An industrial sewage treatment characteristic inversion heavy metal content model generation system, characterized in that, It includes: An acquisition processing module, configured to acquire a set of processing data in time series of historical sewage of the same type as the target sewage in the sewage treatment system, where the processing data includes the dynamic parameter data of the historical sewage at each time point within the time series, and perform clustering on the dynamic parameter data to obtain the important dynamic parameter data of the historical sewage; Wherein, the processing data reflects the water quality parameters and heavy metal content of the historical sewage in the sewage treatment system, and the dynamic parameter data of the historical sewage at each time point includes the parameter data after the treatment of each process node of the sewage in the sewage treatment system; A target calculation model establishment module, configured to determine a time calculation model of the processing data of the sewage in the sewage treatment system based on the process parameters of the sewage treatment system, and optimize the time calculation model by using the important dynamic parameter data to obtain a target calculation model; An inversion model establishment module, calculates the predicted sewage data of the second node through the target calculation model and the actual sewage data of the first node, determines an inversion model of the sewage data of the first node based on the predicted sewage data of the second node and the actual data of the second node, and also corrects and optimizes the inversion model of the sewage data of the first node based on the important dynamic parameter data to obtain a final inversion model.
8. The heavy metal content model generation system for industrial sewage treatment feature inversion according to claim 7, wherein Performing clustering on the dynamic parameter data to obtain the important dynamic parameter data of the historical sewage includes: Performing feature extraction on the dynamic parameter data to obtain multiple time series parameter features; Vectorizing the multiple time series parameter features to obtain parameter feature vectors; Using the K-means clustering algorithm to cluster the parameter feature vectors to obtain the clustering centers of the parameter feature vectors; Analyzing the clustering centers to obtain the important dynamic parameter data of the dynamic parameter data.
9. A method for generating a heavy metal content model by inverting the characteristics after industrial sewage treatment according to claim 8, characterized in that, The process of determining the inversion model of the sewage data of the first node based on the predicted sewage data of the second node and the actual data of the second node includes: Correcting the predicted sewage data of the second node based on the actual data of the second node to obtain corrected data; Determining the inversion model of the sewage data of the first node according to the corrected data.
10. A method for generating a heavy metal content model by inverting the characteristics after industrial sewage treatment according to claim 9, characterized in that, The process of correcting and optimizing the inversion model of the sewage data of the first node based on the important dynamic parameter data to obtain a final inversion model includes: Determining the full-process parameter data of the sewage corresponding to the first node at each node in the sewage treatment system based on the important dynamic parameter data, and obtaining all node parameter data before the first node; Correcting the inversion data of each node through all the node parameter data before the first node, and performing overall correction based on the correction results corresponding to each node and the inversion model of the sewage data of the first node to obtain a final inversion model; Wherein, after obtaining the final inversion model by performing overall correction based on the correction results corresponding to each node and the inversion model of the sewage data of the first node, it further includes: Optimizing the final inversion model based on the least squares method to obtain an optimized inversion model; Wherein, the formula for optimizing the final inversion model based on the least squares method is: ； Wherein, y′ is the output of the corrected model, ŷ is the predicted value of the sewage parameter, λ is the correction factor, and y is the actual value of the sewage parameter; The calculation formula of the correction factor is: ； Among them, N represents the number of process nodes at which the sewage obtains processing data in the sewage treatment system.