A sewage treatment operation method and system based on AI algorithm

Abstract

The present invention provides a sewage treatment operation method and system based on an AI algorithm, which relates to the technical field of sewage treatment, including obtaining first information and second information, the first information including sewage image information and sewage index data information monitored by sewage real-time monitoring equipment, and the second information including historical sewage reuse strategy information; sending the first information to a preset feature extraction module to obtain sewage image feature information and sewage index data feature information; sending the sewage image feature information and sewage index data feature information to a trained sewage judgment model for judgment processing to obtain a judgment result of a cleanliness level corresponding to the sewage feature; sending the judgment result and the second information to a preset sewage operation model for analysis to obtain sewage reuse strategy information; and performing sewage operation according to the sewage reuse strategy. The present invention intelligently treats sewage, reduces sewage treatment costs, and reduces environmental impact.

Description

Technical Field

[0001] The present invention relates to the technical field of sewage treatment, and in particular to a sewage treatment operation method and system based on AI algorithm. Background Art

[0002] With the acceleration of urbanization and the improvement of industrialization, sewage treatment has become an important issue for environmental protection and sustainable development. At present, traditional sewage treatment methods have problems such as inaccurate monitoring methods and difficulty in formulating fine reuse strategies. Especially in terms of sewage reuse, due to the lack of intelligent decision support system, the reuse strategies formulated for sewage with different cleanliness levels lack scientificity and accuracy. Traditional sewage treatment operation methods are usually based on experience and rules, and it is difficult to fully explore the information in a large amount of real-time and historical data, resulting in high sewage treatment costs. In actual operation, due to the limited monitoring methods of sewage images and indicator data, it is difficult to accurately obtain sewage quality information, which affects the judgment of cleanliness level. In addition, the existing sewage treatment system fails to make full use of historical data to provide reference for decision-making. Therefore, there is an urgent need for an intelligent sewage treatment operation method and system to improve resource utilization. Summary of the invention

[0003] The purpose of the present invention is to provide a sewage treatment operation method and system based on AI algorithm to improve the above problems. In order to achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0004] On the one hand, the present application provides a sewage treatment operation method based on AI algorithm, including:

[0005] Acquire first information and second information, wherein the first information includes sewage image information and sewage indicator data information monitored by sewage real-time monitoring equipment, and the second information includes historical sewage reuse strategy information, wherein the historical sewage reuse strategy information includes a strategy for classified utilization according to the cleanliness level of the corresponding sewage;

[0006] Sending the first information to a preset feature extraction module to obtain sewage image feature information and sewage index data feature information;

[0007] The sewage image feature information and the sewage index data feature information are sent to the trained sewage judgment model for judgment processing to obtain a judgment result of the cleanliness level corresponding to the sewage feature;

[0008] Sending the judgment result and the second information to a preset sewage operation model for analysis to obtain sewage reuse strategy information;

[0009] A control command is sent, where the control command is a command for performing sewage operation according to the sewage reuse strategy.

[0010] On the other hand, the present application also provides a sewage treatment operation system based on AI algorithm, including:

[0011] an acquisition unit, configured to acquire first information and second information, wherein the first information includes sewage image information and sewage index data information monitored by the real-time sewage monitoring device, and the second information includes historical sewage reuse strategy information, wherein the historical sewage reuse strategy information includes a strategy for classified utilization according to the cleanliness level of the corresponding sewage;

[0012] A processing unit, used for sending the first information to a preset feature extraction module to obtain sewage image feature information and sewage index data feature information;

[0013] A judgment unit, used to send the sewage image feature information and the sewage index data feature information to the trained sewage judgment model for judgment processing, and obtain a judgment result of the cleanliness level corresponding to the sewage feature;

[0014] An analysis unit, used for sending the judgment result and the second information to a preset sewage operation model for analysis to obtain sewage reuse strategy information;

[0015] The sending unit is used to send a control command, where the control command is a command for performing sewage operation according to the sewage reuse strategy.

[0016] The beneficial effects of the present invention are:

[0017] The present invention makes full use of the images, index data and historical reuse strategy information obtained by the real-time sewage monitoring equipment, and realizes the intelligent judgment of the sewage cleanliness level and the scientific formulation of the reuse strategy through the comprehensive application of feature extraction, AI judgment model and operation model. The present invention realizes the efficient extraction of real-time monitoring information through the feature extraction of sewage image information and the processing of sewage index data by the Bert model. At the same time, the purpose, amount and price characteristics in the second information are matched with the judgment results through the KM matching algorithm, which provides a scientific basis for the formulation of sewage reuse strategies. Finally, all sewage reuse strategies are processed by the SAGA optimization model, and the optimal solution with the highest profit margin is obtained, which minimizes the operating cost and maximizes the economic benefit.

[0018] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or be understood by implementing the embodiments of the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0020] Figure 1 A schematic diagram of the process of sewage treatment operation method based on AI algorithm described in an embodiment of the present invention;

[0021] Figure 2 A schematic diagram of the topological structure of a sewage treatment operation system based on an AI algorithm described in an embodiment of the present invention;

[0022] Figure 3 A frame selection diagram of feature information of a sewage image of a sewage treatment operation system based on an AI algorithm described in an embodiment of the present invention;

[0023] Figure 4 Schematic diagram of sewage control and discharge of a sewage treatment operation system based on an AI algorithm described in an embodiment of the present invention.

[0024] In the figure: 701, acquisition unit; 702, processing unit; 703, judgment unit; 704, analysis unit; 705, sending unit. DETAILED DESCRIPTION

[0025] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments in 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.

[0026] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings. At the same time, in the description of the present invention, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0027] Embodiment 1:

[0028] This embodiment provides a sewage treatment operation method based on AI algorithm, see Figure 1 , Figure 3 and Figure 4 , the figure shows that the method includes step S1, step S2, step S3, step S4 and step S5.

[0029] Step S1, obtaining first information and second information, wherein the first information includes sewage image information and sewage indicator data information monitored by the sewage real-time monitoring equipment, and the second information includes historical sewage reuse strategy information, and the historical sewage reuse strategy information includes a strategy for classified utilization according to the cleanliness level of the corresponding sewage;

[0030] In this step, the sewage can be monitored in real time by real-time monitoring equipment, which includes sensors, cameras, etc., and can obtain image information and index data information of the sewage. The image information can show the actual situation of the sewage, and the index data includes various key parameters, such as pH value, chemical oxygen demand, ammonia nitrogen, etc. Figure 4 As shown, the present invention can obtain the previously adopted sewage reuse strategy information by accessing the historical database or log, and send the information in the database and the information collected by the sensor to the processing center for processing, wherein the processing center includes a processing unit, a judgment unit and an analysis unit, and can then quickly process and analyze the data to determine the sewage discharge level and the corresponding discharge strategy.

[0031] Step S2, sending the first information to a preset feature extraction module to obtain sewage image feature information and sewage index data feature information;

[0032] The feature extraction module can efficiently extract meaningful feature information from the original monitoring data. This information will play a key role in the subsequent sewage cleanliness level judgment and reuse strategy formulation, improving the data expression ability and processing efficiency.

[0033] Step S3, sending the sewage image feature information and the sewage index data feature information to the trained sewage judgment model for judgment processing, and obtaining a judgment result of the cleanliness level corresponding to the sewage feature;

[0034] By using the trained sewage judgment model, the system can accurately judge the cleanliness level of sewage based on the input image and indicator data feature information, improve the automatic judgment capability of sewage characteristics, and provide key input information for subsequent operation models.

[0035] Further, the step S3 includes step S31, step S32, step S33 and step S34:

[0036] Step S31, obtaining historical sewage image feature information and historical sewage index data feature information pre-stored in the database;

[0037] Therefore, by obtaining historical sewage image feature information and historical sewage indicator data feature information, the system can use these rich historical data for model training and comparative analysis, thereby improving the accuracy and robustness of the judgment model.

[0038] Step S32: clustering the historical sewage image feature information and the historical sewage index data feature information based on the K-means clustering algorithm, and calculating the threshold range corresponding to each cluster obtained by clustering based on the Laida criterion to obtain the level threshold of the historical sewage image feature information and the historical sewage index data feature information;

[0039] This step uses K-means clustering and the Laida criterion to automatically determine the threshold range corresponding to each cluster according to the distribution of historical data, thereby providing an accurate reference for subsequent grade judgment. This step infers the range of sewage images and indicator data of different grades based on the clustering characteristics of historical data, thereby improving the accuracy of grade judgment.

[0040] Step S33: non-dimensionalize all the level thresholds, and process all the non-dimensionalized level thresholds based on a preset level threshold adjustment formula to obtain the level thresholds after weight adjustment;

[0041] This step can better adapt to different data distributions and model preferences by making the level threshold dimensionless and adjusting the weight, thereby improving the accuracy and robustness of the sewage level. This makes the level threshold more universal and applicable to sewage treatment scenarios under different conditions. The preset level threshold adjustment formula is as follows:

[0042] ;

[0043] in, is the dimensionless covariance matrix of the historical sewage image feature information, is the dimensionless covariance matrix of the characteristic information of historical sewage index data, is the dimensionless total covariance matrix of sewage image feature information and sewage index data feature information, X is the dimensionless feature vector of historical sewage image feature information, Y is the dimensionless feature vector of sewage index data feature information, and The default initial weights.

[0044] Step S34: Compare the sewage image feature information and the sewage index data feature information with the weighted level thresholds respectively, wherein the cleaning level corresponding to the sewage feature is obtained based on the weighted level threshold interval to which the sewage image feature information and the sewage index data feature information belong.

[0045] Therefore, by comparing the sewage image feature information and sewage index data feature information with the level threshold after adjusting the weight, the cleanliness level of the sewage can be determined more accurately, which improves the accuracy of the cleanliness level determination and provides key input information for the subsequent operation model. Among them, primary sewage is sewage after primary treatment, usually including water bodies after coarse screening, grilles, grit chambers and other process treatments. Primary sewage mainly removes large particles of debris, suspended matter and some organic matter, but the water quality still contains a high concentration of pollutants and is not suitable for direct discharge or reuse, and requires further treatment. Secondary sewage is sewage that further removes organic matter, nitrogen, phosphorus and other pollutants in the water. Tertiary treated sewage is sewage that further removes trace organic matter, heavy metals, bacteria and other pollutants in the water, and can be directly used for specific reuse or discharge requirements.

[0046] Step S4, sending the judgment result and the second information to a preset sewage operation model for analysis to obtain sewage reuse strategy information;

[0047] In this step, through the analysis of the sewage operation model, the system can provide scientific and reasonable sewage reuse strategy information based on actual conditions and historical data, realize the optimization and efficient operation of the sewage treatment process, and improve the intelligence and adaptability of the sewage treatment system.

[0048] Furthermore, in order to ensure that the impact of the selected strategy on the environment is within an acceptable range, step S4 in the method may be followed by step S5, step S6 and step S7.

[0049] Step S5: performing correlation analysis based on the obtained sewage reuse strategy information and the historical environmental impact data to obtain a correlation value between the sewage reuse strategy information and the historical environmental impact data;

[0050] Among them, by calculating the correlation value, we can more accurately understand the relationship between the wastewater reuse strategy and the historical environmental impact data. This helps to optimize the wastewater reuse strategy to reduce the negative impact on the environment and thus improve overall sustainability. This step considers the environmental impact of the wastewater reuse strategy more comprehensively by quantifying the correlation value.

[0051] Step S6: multiply the correlation value as the environmental impact weight with the corresponding score of the preset historical environmental impact data to obtain the environmental impact score corresponding to each sewage reuse strategy information;

[0052] This step obtains a more comprehensive environmental impact assessment by considering the relevance value as a weight and combining it with the scoring of historical environmental impact data. More accurate quantification of the actual environmental impact of each wastewater reuse strategy will help make decisions that are more in line with sustainable development goals.

[0053] Step S7: compare the environmental impact score with a preset threshold value. If the environmental impact score is less than the preset threshold value, the sewage reuse strategy information corresponding to the environmental impact score is used as the final sewage reuse strategy information.

[0054] Therefore, in this step, by comparing with the preset threshold, the system can screen out the sewage reuse strategies with environmental impact scores lower than the set standard, thereby ensuring that the impact of the selected strategy on the environment is within an acceptable range, optimizing the selection of sewage reuse strategies to ensure environmental sustainability.

[0055] Step S8: sending a control command, wherein the control command is a command for performing sewage operation according to the sewage reuse strategy.

[0056] Embodiment 2:

[0057] The difference between this embodiment and embodiment 1 is that step S2 includes step S21, step S22, step S23 and step S24:

[0058] Step S21, extracting features from the sewage image information in the first information, wherein the sewage image information is subjected to target detection, the sewage image information is compared with preset clean water image information, all difference area image information is obtained, and the difference area image information is used as sewage image feature information;

[0059] Through target detection and comparison of difference areas, the system can effectively extract key features in sewage images, help accurately judge the condition of sewage, improve the ability to perceive the actual situation of sewage, and provide more detailed and accurate information for subsequent sewage grade judgment.

[0060] Step S22, sending the sewage index data information in the first information to the Bert model for pre-training, and determining the characteristic data information contained in the first information, wherein the characteristic data information is the index data information that is greater than or less than a preset threshold;

[0061] Through the pre-training of the Bert model, the system can convert sewage indicator data information into semantic representation, improving the abstract ability of indicator data. At the same time, by setting thresholds, it can more specifically screen out qualified feature data information for subsequent hierarchical analysis and feature weight calculation. The technical effect of this step is to enhance the understanding and mining ability of indicator data information.

[0062] Step S23, performing hierarchical analysis on the characteristic data information to determine the relative weight of each characteristic data information;

[0063] It can be understood that this step uses the hierarchical analysis method to construct a judgment matrix to compare the relative importance of each feature data information. Then, the relative importance of each feature data information can be quantitatively evaluated to ensure that the weight allocation of different features is more accurate and reasonable. The technical effect of this step is to improve the accuracy of the analysis of feature data information and the determination of weights, providing a more reliable basis for subsequent comprehensive judgment.

[0064] Step S24, sort the relative weights of all feature data information from large to small, and select the first N (for example, the first ten or the first twenty) feature data information as the sewage index data feature information in the order of arrangement, thereby extracting the most important and significant features through sorting, reducing information redundancy, while retaining important information on sewage indicators, and further optimizing the selection of feature data information, making the subsequent analysis process more efficient and the analysis results more accurate.

[0065] Furthermore, the step S21 includes step S211, step S212 and step S213:

[0066] Step S211, sending the sewage image information and the preset clean water image information to a preset three-dimensional rectangular coordinate system for coordinate transformation, to obtain the coordinate value and pixel value of each pixel point in the sewage image information and the preset clean water image information;

[0067] By transforming the three-dimensional space coordinates, the system can more accurately represent the image information, so that each pixel in the image has a clearly defined position and attributes. This helps in subsequent image comparison and more accurate extraction of difference areas, and improves the accuracy and interpretability of image information.

[0068] Step S212: performing a pairwise comparison between the coordinate value and pixel value of each pixel point in the sewage image information and the preset clean water image information to obtain distinguishing characteristic pixel points;

[0069] like Figure 3As shown, in this step, the characteristic information of the sewage image is framed, wherein the characteristic information is the characteristic pixel points that differ between the sewage image and the clean water image, and then the images are compared pairwise, thereby improving the sensitivity of the present invention to image differences and providing an effective basis for subsequent extraction of image difference areas.

[0070] Step S213: connect the distinguishing characteristic pixel points with lines, and interpolate the lines using a linear interpolation method to obtain a difference area image between the sewage image information and the preset clean water image information.

[0071] This step fills the area between the lines by performing linear interpolation on the obtained lines. Linear interpolation can estimate the value of a new point between two known points, making the contour smoother, so that the system can generate a clearer, continuous and smoother difference area image, improving the accuracy and visualization of the system for the difference area, and providing more detailed information for subsequent judgment and analysis.

[0072] Embodiment 3:

[0073] The difference between this embodiment and embodiment 1 or 2 is that in this step, step S4 includes step S41, step S42 and step S43.

[0074] Step S41: Perform feature analysis on the data in the second information to obtain the use feature, usage feature and price feature in the second information, wherein the use feature is the use of sewage reuse, the usage feature is the water volume of sewage for all uses, and the price feature is the cost price and selling price of sewage for all uses;

[0075] Among them, the use characteristics include uses for irrigation water sources, industrial water, landscape water, etc., the usage characteristics include the usage of irrigation water sources, industrial water usage or landscape water usage, and the price characteristics include the price of reused water resources for each purpose. The cost price of sewage is the cost of sewage in the reuse process, and the selling price is the selling price of the reused sewage. By using text mining or keyword extraction technology, the key information describing the use of sewage reuse in the second information is identified. Obtaining key information such as the specific use of sewage reuse, water usage, cost price and selling price provides basic data for subsequent decision-making, and increases the system's understanding and comprehensive consideration of sewage reuse strategies.

[0076] Step S42: Based on the KM matching algorithm, the usage characteristics, the usage characteristics and the price characteristics in the second information are matched with the judgment result respectively to obtain at least two sewage reuse strategy information, wherein the sewage reuse strategy includes the usage, the usage and the price corresponding to each clean level of sewage;

[0077] By using the KM matching algorithm, the usage characteristics, usage characteristics and price characteristics are clustered separately to form several categories, and then their respective cluster centers are formed. Then, for each cleaning level, the cluster center closest to the judgment result is selected to determine the usage, usage and price. Then, the usage, usage and price information represented by the matching cluster centers are combined to obtain the strategy information.

[0078] Therefore, through the KM matching algorithm, the system can match the usage characteristics, usage characteristics and price characteristics in the second information with the actual judgment results, generate at least two sewage reuse strategies, and provide more specific operation plans for subsequent optimization models based on sewage of different cleanliness levels, thereby improving the system's adaptability and flexibility to sewage reuse strategies.

[0079] Step S43: Send all sewage reuse strategy information to the trained SAGA optimization model for processing to obtain the sewage reuse strategy information with the highest profit margin.

[0080] Through the processing of the SAGA optimization model, the scheme with the highest profit margin is screened out from all sewage reuse strategies, which maximizes the economic benefits of sewage reuse, improves the system's ability to optimize economic benefits, and makes the sewage treatment system more intelligent and economical.

[0081] Further, the step S43 includes step S431, step S432, step S433, step S434 and step S435:

[0082] Step S431: non-dimensionalize all sewage reuse strategy information, and input the dimensionless sewage reuse strategy information into a preset SAGA optimization model as an input parameter, and set the maximum optimization iteration number, crossover probability, mutation probability and population size of the SAGA optimization model;

[0083] By non-dimensionalizing the wastewater reuse strategy information and setting it according to the preset SAGA optimization model parameters, it can better adapt to parameters of different scales and improve the stability and convergence speed of the optimization model. The technical effect of this step is to optimize the parameter adjustment of the model so that it can search for the optimal solution more effectively.

[0084] Step S432, initializing all sewage reuse strategy information, wherein each sewage reuse strategy information is taken as a population to obtain an initial population;

[0085] In this step, the initial population is formed by random generation, which provides a starting point for the SAGA optimization model, helps to accelerate the convergence process of the model and improve the optimization efficiency. The technical effect of this step is to provide a reasonable initial population for the SAGA optimization model, which helps to find the global optimal solution more quickly.

[0086] Step S433: Perform a crossover mutation operation on all initial populations, and recombine the uses, amounts, and prices in all initial populations to obtain at least one offspring;

[0087] By selecting two individuals from the initial population, randomly selecting a crossover point, and exchanging the information of the two individuals after this point, the diversity of the population is increased, which helps to escape from the local optimal solution and find the global optimal solution. By introducing randomness, the diversity of the population is increased, which helps to search the solution space more comprehensively.

[0088] Step S434, construct an objective function for calculating the profit rate, and judge the profit rate of all offspring and the initial population. If the profit rate of the offspring is less than a preset threshold or greater than 1, the offspring is deleted to obtain the profit rate of the population with the abnormal offspring deleted and the profit rate of the initial population;

[0089] Therefore, by constructing the objective function and judging the profit rate, the wastewater reuse strategies that do not meet expectations can be filtered out, thereby improving the final optimization results. In addition, this step retains individuals with better profit rates in the population by deleting abnormal offspring, which helps the SAGA optimization model converge to the optimal solution faster.

[0090] The objective function is as follows:

[0091] ;

[0092] Where A represents the profit margin, n represents the total number of wastewater reuse strategies, i represents the i-th wastewater reuse strategy, represents the rate of return corresponding to the usage characteristics in the i-th sewage reuse strategy, represents the usage characteristics in the i-th sewage reuse strategy, represents the selling price in the i-th sewage reuse strategy, Represents the cost price in the i-th sewage reuse strategy.

[0093] And step S435, based on the tournament selection method, select a population with a higher profit rate, and repeat the mutation operation and the profit rate judgment operation until the current number of iterations reaches the preset maximum number of optimization iterations, and use the population with the highest profit rate finally obtained as the sewage reuse strategy information.

[0094] Tournament selection is a method of selecting individuals, which randomly groups individuals in a population and then selects individuals with the highest fitness from each group. Through tournament selection and iterative optimization, the sewage reuse strategy with the highest profit margin is finally obtained, which helps to improve the economic benefits of the sewage treatment system and obtain a better sewage reuse strategy.

[0095] Embodiment 4:

[0096] like Figure 2 As shown, this embodiment provides a sewage treatment operation system based on AI algorithm, see Figure 2 The system includes an acquisition unit 701 , a processing unit 702 , a judgment unit 703 , an analysis unit 704 and a sending unit 705 .

[0097] The acquisition unit 701 is used to acquire first information and second information, wherein the first information includes sewage image information and sewage index data information monitored by the sewage real-time monitoring device, and the second information includes historical sewage reuse strategy information, and the historical sewage reuse strategy information includes a strategy for classified utilization according to the cleanliness level of the corresponding sewage, and the process is the same as the above step S1;

[0098] The processing unit 702 is used to send the first information to a preset feature extraction module to obtain sewage image feature information and sewage index data feature information, and the process is the same as the above step S2;

[0099] The judgment unit 703 is used to send the sewage image feature information and the sewage index data feature information to the trained sewage judgment model for judgment processing to obtain the judgment result of the cleanliness level corresponding to the sewage feature. The process is the same as the above step S3;

[0100] An analysis unit 704 is used to send the judgment result and the second information to a preset sewage operation model for analysis to obtain sewage reuse strategy information. The process is the same as the above step S4;

[0101] The sending unit 705 is used to send a control command, where the control command is a command for performing sewage operation according to the sewage reuse strategy. The process is the same as the above step S5.

[0102] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A sewage treatment operation method based on AI algorithm, characterized in that: include: Acquire first information and second information, wherein the first information includes sewage image information and sewage indicator data information monitored by sewage real-time monitoring equipment, and the second information includes historical sewage reuse strategy information, wherein the historical sewage reuse strategy information includes a strategy for classified utilization according to the cleanliness level of the corresponding sewage; Sending the first information to a preset feature extraction module to obtain sewage image feature information and sewage index data feature information; The sewage image feature information and the sewage index data feature information are sent to the trained sewage judgment model for judgment processing to obtain a judgment result of the cleanliness level corresponding to the sewage feature; Sending the judgment result and the second information to a preset sewage operation model for analysis to obtain sewage reuse strategy information; sending a control command, wherein the control command is a command for performing sewage operation according to the sewage reuse strategy; The first information is sent to a preset feature extraction module to obtain sewage image feature information and sewage index data feature information, including: Extracting features from the sewage image information in the first information, wherein the sewage image information is subjected to target detection, the sewage image information is compared with preset clean water image information to obtain all difference area image information, and the difference area image information is used as sewage image feature information; Sending the sewage index data information in the first information to the Bert model for pre-training to determine the characteristic data information contained in the first information, wherein the characteristic data information is the indicator data information that is greater than or less than a preset threshold; Performing hierarchical analysis on the characteristic data information to determine the relative weight of each characteristic data information; The relative weights of all characteristic data information are sorted from large to small, and the top 20 characteristic data information are selected as sewage indicator data characteristic information according to the order of arrangement: The sewage image information is subjected to target detection, and the sewage image information is compared with the preset clean water image information to obtain all the difference area image information, including: Sending the sewage image information and the preset clean water image information to a preset three-dimensional rectangular coordinate system for coordinate transformation to obtain the coordinate value and pixel value of each pixel point in the sewage image information and the preset clean water image information; Compare the coordinate values ​​and pixel values ​​of each pixel point in the sewage image information with those in the preset clean water image information in pairs to obtain distinguishing characteristic pixel points; Connecting the distinguishing characteristic pixel points, and interpolating the connecting lines by linear interpolation to obtain a difference area image between the sewage image information and the preset clean water image information; The judgment result and the second information are sent to a preset sewage operation model for analysis to obtain sewage reuse strategy information, including: Performing feature analysis on the data in the second information to obtain the use feature, usage feature and price feature in the second information, wherein the use feature is the use of sewage reuse, the usage feature is the water volume of sewage for all uses, and the price feature is the cost price and selling price of sewage for all uses; Based on the KM matching algorithm, the purpose characteristics, the quantity characteristics and the price characteristics in the second information are matched with the judgment result respectively to obtain at least two sewage reuse strategy information, wherein the sewage reuse strategy includes the purpose, quantity and price corresponding to each clean level of sewage; All wastewater reuse strategy information is sent to the trained SAGA optimization model for processing to obtain the wastewater reuse strategy information with the highest profit margin; Among them, all sewage reuse strategy information is sent to the trained SAGA optimization model for processing to obtain the sewage reuse strategy information with the highest profit margin, including: All sewage reuse strategy information is dimensionless processed, and the dimensionless sewage reuse strategy information is input into the preset SAGA optimization model as input parameters, and the maximum optimization iteration number, crossover probability, mutation probability and population size of the SAGA optimization model are set; Initializing all sewage reuse strategy information, wherein each sewage reuse strategy information is taken as a population to obtain an initial population; Perform crossover mutation operations on all initial populations, recombine the uses, quantities and prices within all initial populations, and obtain at least one offspring; Construct an objective function for calculating the profit rate, and judge the profit rate of all offspring and the initial population. If the profit rate of the offspring is less than a preset threshold or greater than 1, the offspring is deleted to obtain the profit rate of the population with the abnormal offspring deleted and the profit rate of the initial population; Based on the tournament selection method, a population with a higher profit rate is selected, and the mutation operation and the profit rate judgment operation are repeated until the current iteration number reaches the preset maximum optimization iteration number, and the population with the highest profit rate is used as the sewage reuse strategy information; The objective function is as follows: ； Where A represents the profit margin, n represents the total number of wastewater reuse strategies, i represents the i-th wastewater reuse strategy, represents the rate of return corresponding to the usage characteristics in the i-th sewage reuse strategy, represents the usage characteristics in the i-th sewage reuse strategy, represents the selling price in the i-th sewage reuse strategy, Represents the cost price in the i-th sewage reuse strategy.

2. The sewage treatment operation method according to claim 1, characterized in that , sending the sewage image feature information and sewage index data feature information to the trained sewage judgment model for judgment processing, and obtaining the judgment result of the cleanliness level corresponding to the sewage feature, including: Acquire historical sewage image feature information and historical sewage index data feature information pre-stored in the database; Based on the k-means clustering algorithm, the historical sewage image feature information and the historical sewage index data feature information are clustered, and the threshold range corresponding to each cluster obtained by clustering is calculated based on the Laida criterion to obtain the level threshold of the historical sewage image feature information and the historical sewage index data feature information; All the level thresholds are dimensionally processed, and all the dimensionally processed level thresholds are processed based on a preset level threshold adjustment formula to obtain the level thresholds after weight adjustment; The sewage image feature information and the sewage index data feature information are respectively compared with the weighted level thresholds, wherein the cleanliness level corresponding to the sewage feature is obtained based on the weighted level threshold interval to which the sewage image feature information and the sewage index data feature information belong.

3. A system for implementing the sewage treatment operation method based on AI algorithm as described in any one of claims 1-2, characterized in that: The system comprises: an acquisition unit, configured to acquire first information and second information, wherein the first information includes sewage image information and sewage index data information monitored by the real-time sewage monitoring device, and the second information includes historical sewage reuse strategy information, wherein the historical sewage reuse strategy information includes a strategy for classified utilization according to the cleanliness level of the corresponding sewage; A processing unit, used for sending the first information to a preset feature extraction module to obtain sewage image feature information and sewage index data feature information; A judgment unit, used to send the sewage image feature information and the sewage index data feature information to the trained sewage judgment model for judgment processing, and obtain a judgment result of the cleanliness level corresponding to the sewage feature; An analysis unit, used for sending the judgment result and the second information to a preset sewage operation model for analysis to obtain sewage reuse strategy information; A sending unit, used to send a control command, wherein the control command is a command for performing sewage operation according to the sewage reuse strategy; Wherein, the processing unit includes: a first processing subunit, configured to extract features from the sewage image information in the first information, wherein the sewage image information is subjected to target detection, the sewage image information is compared with preset clean water image information, all difference area image information is obtained, and the difference area image information is used as sewage image feature information; A second processing subunit is used to send the sewage index data information in the first information to the Bert model for pre-training, and determine the characteristic data information contained in the first information, where the characteristic data information is the index data information that is greater than or less than a preset threshold; A third processing subunit is used to perform hierarchical analysis on the characteristic data information to determine the relative weight of each characteristic data information; The fourth processing subunit is used to sort the relative weights of all the characteristic data information from large to small, and select the top 20 characteristic data information as the sewage index data characteristic information according to the arrangement order; Wherein, the first processing subunit includes: a fifth processing subunit, configured to send the sewage image information and the preset clean water image information to a preset three-dimensional rectangular coordinate system for coordinate transformation, and obtain the coordinate value and pixel value of each pixel point in the sewage image information and the preset clean water image information; The sixth processing subunit is used to compare the coordinate value and pixel value of each pixel point in the sewage image information with the preset clean water image information in pairs to obtain distinguishing characteristic pixel points; The seventh processing subunit is used to connect the distinguishing characteristic pixel points and interpolate the lines by linear interpolation to obtain a difference area image between the sewage image information and the preset clean water image information.

Images