Dosing control system for sewage treatment

Abstract

The invention discloses a chemical adding control system for sewage treatment, which comprises a data acquisition module used for acquiring sewage inlet water quality data and inlet water flow, water quality data of each sewage treatment process and discharge water quality data in real time; the data acquisition module is used for acquiring dosing data and detection data; the data preprocessing module is used for assigning the partial water quality data Em based on a preset judgment threshold value; the dosing prediction module is used for inputting the real-time data into the self-attention dosing prediction model and outputting to obtain a next dosing strategy; the self-attention dosing prediction model corrects the attention score matrix based on the chemical reaction relation matrix E; and the dosing control module is used for dosing according to the output dosing strategy. The method can adapt to sewage component and concentration changes, and the dosing strategy of each process of sewage treatment is accurately predicted and obtained.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent control of sewage treatment, and more specifically, to a dosing control system for sewage treatment. Background Art

[0002] Traditional dosing control systems may not be able to accurately control the dosage of chemicals, resulting in unstable treatment effects or waste of chemicals. For example, due to the subjective factors of operators, excessive dosage of chemicals may occur, which in turn affects the treatment effect. For example, the "re-stabilization" phenomenon causes the total phosphorus in the effluent to exceed the standard. In addition, the composition and concentration of sewage faced by sewage treatment plants may change due to factors such as seasons, weather, and industrial activities. The intelligent dosing system needs to be able to adapt to these environmental changes to maintain the best treatment effect. In order to overcome these challenges, it is necessary to continuously strengthen technological research and development and improve the stability, accuracy and reliability of intelligent control systems. Summary of the invention

[0003] An object of the present invention is to solve at least the above problems and to provide at least the advantages which will be described hereinafter.

[0004] In order to achieve these purposes and other advantages according to the present invention, a dosing control system for sewage treatment is provided, comprising: Data collection module, which is used to collect real-time sewage inlet water quality data and inlet flow, water quality data of each sewage treatment process, and discharge water quality data; Data acquisition module, which is used to obtain the most recent dosing time t i , the type of dosing M, the amount of dosing m, and the multiple groups of influent water quality data E collected by the data acquisition module after the dosing treatment s and water inflow q, water quality data E of multiple groups of sewage treatment processes m , multiple sets of discharge water quality data E z , water temperature T; The data preprocessing module is used to preprocess the water quality data E of each process of multiple sewage treatment processes based on a preset judgment threshold. m Assignment, where water quality data E m When the measured value meets the judgment threshold, it is assigned a value of g, otherwise, it is assigned a value of p, where g and p are both real numbers; The dosing prediction module is used to input real-time data into the self-attention dosing prediction model and output the next dosing time t i+1 , dosing type M and dosing amount m, discharge water quality data E z ; Wherein, the real-time data includes the data acquired by the data acquisition module and the data preprocessed by the data preprocessing module; and, The self-attention dosing prediction model is based on a chemical reaction relationship matrix E of water quality data and discharged water quality data in each sewage treatment process caused by a series of chemical changes caused by two adjacent dosing amounts and dosing times. The attention score matrix is ​​multiplied by the chemical reaction relationship matrix E to obtain a corrected attention score matrix, and then the self-attention neural network model constructed based on the corrected attention score matrix is ​​trained; The dosing control module is used to add medicine according to the output dosing time, dosing type and dosing amount.

[0005] Preferably, the self-attention dosing prediction model further comprises: determining the current pre-output t i+1 Discharge water quality data at the moment E z Whether it meets the preset discharge water quality standard, if not, then the query is fed back forward according to the preset time gradient value until the discharge water quality data E is queried z When the discharge water quality standard is met, the output (t i+1 -Δt) time, discharged water quality data E z , and the initial pre-output t i+1 The drug type M and drug amount m corresponding to the moment, where Δt represents the sum of the time gradient values ​​of the forward feedback query.

[0006] Preferably, the inlet water quality data E s The water quality indicators include: pH value, chemical oxygen demand COD, five-day biochemical oxygen demand BOD5, suspended solids SS, total phosphorus TP, total nitrogen TN, ammonia nitrogen NH3-N, oil, heavy metal ions, sulfide, residual chlorine, turbidity, color, odor and taste, total bacteria count, coliform count, conductivity, solid matter, water temperature T and other combinations; The discharge water quality data E z The water quality indicators include: pH value, chemical oxygen demand COD, five-day biochemical oxygen demand BOD5, suspended solids SS, total phosphorus TP, total nitrogen TN, ammonia nitrogen NH3-N, oil, heavy metal ions, sulfide, residual chlorine, turbidity, chromaticity, total bacteria count, coliform count, and a variety of combinations of anionic surfactants.

[0007] Preferably, the sewage treatment process includes a biochemical treatment process in a biochemical reaction tank, a chemical precipitation treatment process, a coagulation and flocculation treatment process, and a disinfection treatment process; and, The water quality indicators of the biochemical treatment process include: pH value, chemical oxygen demand COD, biochemical oxygen demand BOD, suspended solids SS, ammonia nitrogen NH3-N, nitrate nitrogen NO3-N, sludge concentration MLSS, redox potential ORP, water temperature T, and various combinations of dissolved oxygen DO in anaerobic zone, anoxic zone and aerobic zone; The water quality indicators of the chemical precipitation treatment process include: pH value, chemical oxygen demand COD, suspended solids SS, heavy metal ions, phosphorus PO4-P, water temperature T and other combinations; The water quality indicators of the coagulation and flocculation treatment process include: pH value, chemical oxygen demand COD, suspended solids SS, total phosphorus TP, total nitrogen TN, ammonia nitrogen NH3-N, oil, turbidity, chromaticity, and various combinations of water temperature T; The water quality indicators of the disinfection treatment process include: pH value, chemical oxygen demand COD, suspended solids SS, ammonia nitrogen NH3-N, heavy metal ions, residual chlorine, turbidity, color, odor, fecal coliform bacteria, total coliform bacteria, and various combinations of viruses.

[0008] Preferably, the types of drugs added in the biochemical treatment process include: a combination of carbon source, nitrogen source, phosphorus source, flocculant, coagulant aid, conditioner, pH adjuster, redox agent, disinfectant, microbial preparation, sludge improver, and phosphorus removal agent; The types of drugs added in the chemical precipitation treatment process include: hydroxide precipitants, sulfide precipitants, barium salt precipitants, carbonate precipitants, halide precipitants, reduction precipitants, organic reagent precipitants, magnesium ammonium phosphate precipitants, flocculants, coagulants, and various combinations of conditioning agents; The types of drugs added in the coagulation and flocculation treatment process include: polyaluminium chloride PAC, polyacrylamide PAM, aluminium sulfate, ferric chloride, coagulant aid, pH adjuster, and various combinations of oxidants; The types of drugs added in the disinfection treatment process include: chlorine-containing disinfectants, chlorine dioxide disinfectants, ozone, peroxide disinfectants, iodine-containing disinfectants, bromine-containing disinfectants, phenol disinfectants, and various combinations of quaternary ammonium salt disinfectants.

[0009] Preferably, the data preprocessing module further includes: a module for using a linear interpolation method to collect the influent water quality data E acquired by the data acquisition module. s and the inflow flow rate q, water quality data E of each sewage treatment process m , Discharge water quality data E z Perform interpolation processing, as shown in formula 1: Formula 1, Among them, A represents the detection time of the missing data in the detection data, B represents the missing data in the detection data, B0 and B1 represent the measured data before and after B respectively, and A0 and A1 represent the detection time corresponding to B0 and B1 respectively.

[0010] Preferably, the chemical reaction relationship matrix E is used to represent the subordinate relationship between the process itself and the test data after the process after the chemical addition treatment of multiple groups of sewage treatment processes. The test data subordinate relationship is expressed asa ×(X a -X a0 )+ b ×(X b -X b0 )+ c ×(X c -X c0 )∈E, where a Indicates the weight of the relationship between the current dosing process and the water quality index, b Indicates the relationship weight between the next sewage treatment process adjacent to the current dosing process and the water quality index, c represents the weight of the relationship between the next sewage treatment process adjacent to the current dosing process and the water quality index, X a0 and X a Respectively represent the test data of water quality indicators during and after the current dosing process, X b0 and X b They represent the test data of water quality index of the next sewage treatment process during and after the current dosing process, respectively. c0 and X c They respectively represent the test data of water quality indicators of the adjacent second sewage treatment process during and after the dosing of the current dosing process, and so on.

[0011] Preferably, the self-attention dosing prediction model is obtained based on the convergence of the consumption rate of the agent per ton of sewage within a period of time being lower than a set threshold.

[0012] An electronic device is provided, comprising: one or more processors; and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor performs the processing of the dosing control system for sewage treatment.

[0013] A computer-readable storage medium is provided, on which a computer program is stored. When the program is executed by a processor, the processing of the sewage treatment dosing control system is realized.

[0014] The present invention has at least the following beneficial effects: First, the present invention updates and corrects the attention score matrix through the chemical reaction relationship matrix E, so as to more accurately reflect the impact of each dosing, optimize the output dosing strategy, significantly improve the dosing accuracy, and effectively improve the stability of the sewage treatment dosing control system under different working conditions, ensuring the continuity of the sewage treatment process and the stability of the effluent water quality.

[0015] Second, in order to avoid the discharge water quality data not meeting the standards, that is, to ensure that the discharge water quality data meets the discharge water quality standards, the present invention sets a strategy of forward feedback query according to the time gradient value, and adjusts the dosing time to maintain the best sewage treatment effect to adapt to the changes in sewage composition and concentration due to factors such as seasons, weather and industrial activities.

[0016] Third, under the premise of ensuring the quality of sewage treatment, the self-attention neural network model is optimized and adjusted by superimposing the consumption rate of chemicals per ton of sewage over a period of time, which can reduce the amount of chemicals added and achieve energy-saving effects.

[0017] Other advantages, objectives and features of the present invention will be embodied in part through the following description, and in part will be understood by those skilled in the art through study and practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a flow chart of one of the technical solutions of the present invention. DETAILED DESCRIPTION

[0019] The present invention is further described in detail below in conjunction with the accompanying drawings so that those skilled in the art can implement the invention with reference to the description.

[0020] It should be noted that the experimental methods described in the following embodiments are conventional methods unless otherwise specified, and the reagents and materials are commercially available unless otherwise specified; in the description of the present invention, the orientation or positional relationship indicated by the terms is based on the orientation or positional relationship shown in the accompanying drawings, which is only for the convenience of describing the present invention and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention.

[0021] like Figure 1 As shown, the present invention provides a dosing control system for sewage treatment, comprising: The data acquisition module is used to collect sewage inlet water quality data and inlet flow, water quality data of each sewage treatment process, and discharge water quality data in real time according to a preset frequency; for example, the inlet flow can be measured by a metering pump or detected by a flow sensor, and the water quality data can be detected in real time by an optical sensor, an electrochemical sensor, a temperature sensor, and a pressure sensor, and the data collected by the data acquisition module is stored.

[0022] Data acquisition module, which is used to obtain the most recent dosing time t i , the type of dosing M, the amount of dosing m, and the multiple groups of influent water quality data E collected by the data acquisition module after the dosing treatment sand water inflow q, water quality data E of multiple groups of sewage treatment processes m , multiple sets of discharge water quality data E z , water temperature T; dosing time t i , dosing type M, dosing amount m, etc. can be obtained from historical dosing records.

[0023] The data preprocessing module is used to preprocess a plurality of water quality data E of each process of sewage treatment based on a preset judgment threshold. m Assignment, where water quality data E m When the measured value meets the judgment threshold, it is assigned a value of g, otherwise, it is assigned a value of p, where g and p are both real numbers; usually g and p can be set to 0 and 1, or -1 and 1, respectively. In order to reduce the data processing difficulty of the self-attention dosing prediction model, several water quality data E can be selected according to the prior value. m Assigning values ​​to judge whether these water quality indicators meet the judgment threshold in the corresponding sewage treatment process stage is not an important influencing factor. The actual detection value will not affect the final discharge water quality, nor will it affect the overall total dosage, thereby simplifying the processing of such water quality indicator data.

[0024] The dosing prediction module is used to input real-time data into the self-attention dosing prediction model and output the next dosing time t i+1 , dosing type M and dosing amount m, discharge water quality data E z ; Wherein, the real-time data includes the data acquired by the data acquisition module and the data preprocessed by the data preprocessing module; based on the dosing prediction module, the dosing strategy can be accurately and real-time output, that is, the dosing type, dosing time, dosing amount, etc. of the sewage treatment process.

[0025] Among them, the self-attention dosing prediction model is based on the chemical reaction relationship matrix E of water quality data and discharge water quality data in each sewage treatment process caused by a series of chemical changes caused by two adjacent dosing amounts and dosing times. The attention score matrix is ​​multiplied by the chemical reaction relationship matrix E to obtain a revised attention score matrix, and then the self-attention neural network model constructed based on the revised attention score matrix is ​​trained; based on the characteristics of the sewage treatment process, that is, the previous sewage treatment process will affect the next sewage treatment process, and even the operating status of multiple sewage treatment processes downstream, therefore, it is necessary to specially consider the correlation between the added agent and the water quality data, that is, the chemical reaction relationship, and setting the chemical reaction relationship matrix E to update the revised attention score matrix can more accurately reflect the impact of each dosing, thereby further optimizing the output dosing strategy.

[0026] The dosing control module is used to add medicine according to the output dosing time, dosing type and dosing amount. Usually, a controller is used to control the dosing device to execute the dosing strategy to avoid interference from human dosing. For example, a PLC programmable logic controller can adjust the dosing amount of each metering pump through PLC control.

[0027] In the above technical solution, the data acquisition module can be used to collect the status data of each stage of the current sewage treatment in real time, reflect the current sewage water quality status in real time, and provide an accurate data basis for prediction; the data preprocessing module can reduce the redundancy of some water quality data, reduce the occupation of computing resources by the dosing prediction module, and improve the prediction efficiency; the attention score matrix is ​​updated and corrected through the chemical reaction relationship matrix E, so as to more accurately reflect the impact of each dosing, further optimize the output dosing strategy, and improve the accuracy of dosing. Effectively improve the stability of the dosing control system for sewage treatment under different working conditions, ensure the continuity of the sewage treatment process and the stability of the effluent water quality.

[0028] The Self-Attention Neural Network Models used in this application mainly include the following key structures: Self-Attention Mechanism: The self-attention mechanism allows the model to dynamically assign different attention weights between different positions in the sequence, thereby capturing long-distance dependencies within the sequence. This mechanism is not limited by the length of the sequence and can effectively process sequence data of any length.

[0029] Positional Encoding: Since the self-attention mechanism itself does not contain position information, positional encoding needs to be added to provide the position information of the elements in the sequence. Positional encodings can be fixed or learnable, and they are usually added to the embedding representation of the input sequence.

[0030] Feed-Forward Networks: After the self-attention layer, the output of each position passes through a feed-forward network, which performs the same operation on the input of each position independently. The feed-forward network consists of two linear transformations and an activation function.

[0031] Residual Connection: After the self-attention layer and the feed-forward network layer, there is a residual connection that adds the input directly to the output. This design helps avoid information loss and gradient vanishing problems that occur in deep networks.

[0032] Layer Normalization: After each sublayer (self-attention layer and feed-forward network layer), layer normalization is applied. This helps stabilize the training process, speed up convergence, and improve the performance of the model.

[0033] Encoder-Decoder Architecture: The self-attention model uses an encoder-decoder architecture. The encoder processes the input sequence and the decoder generates the output sequence, and the two interact through the self-attention mechanism.

[0034] These structures together form a self-attention neural network model, which enables it to effectively process sequence data.

[0035] In the present invention, Q=Q′×X, K=K′×X, V=V′×X, X represents input data, Q′, K′, V′ respectively represent learnable parameters, Q, K, V respectively represent scores under the attention mechanism, and a softmax activation function is used to process the attention score matrix, and then the chemical reaction relationship matrix E is multiplied element by element with the attention score matrix to obtain a modified attention score matrix.

[0036] In another technical solution, the self-attention dosing prediction model further includes: determining the current pre-output t i+1 Discharge water quality data at the moment E z Whether it meets the preset discharge water quality standard, if not, then the query is fed back forward according to the preset time gradient value until the discharge water quality data E is queried z When the discharge water quality standard is met, the output (t i+1 -Δt) time, discharged water quality data E z , and the initial pre-output t i+1 The drug type M and drug amount m corresponding to the moment, where Δt represents the sum of the time gradient values ​​of the forward feedback query.

[0037] Since the sewage composition and concentration faced by sewage treatment plants will change due to factors such as seasons, weather and industrial activities, in the above technical solution, in order to avoid the discharge water quality data not meeting the standards, that is, to ensure that the discharge water quality data meets the discharge water quality standards, a strategy of forward feedback query based on the time gradient value is set up to adjust the dosing time to maintain the best sewage treatment effect.

[0038] In another technical solution, the inlet water quality data E sThe water quality indicators include: pH value, chemical oxygen demand COD, five-day biochemical oxygen demand BOD5, suspended solids SS, total phosphorus TP, total nitrogen TN, ammonia nitrogen NH3-N, oil, heavy metal ions, sulfide, residual chlorine, turbidity, color, odor and taste, total bacteria count, coliform count, conductivity, solid matter, water temperature T and other combinations; The discharge water quality data E z The water quality indicators include: pH value, chemical oxygen demand COD, five-day biochemical oxygen demand BOD5, suspended solids SS, total phosphorus TP, total nitrogen TN, ammonia nitrogen NH3-N, oil, heavy metal ions, sulfide, residual chlorine, turbidity, color, total bacteria count, coliform count, and a variety of combinations of anionic surfactants; The sewage treatment process includes a biochemical treatment process, a chemical precipitation treatment process, a coagulation and flocculation treatment process, and a disinfection treatment process; and, The water quality indicators of the biochemical treatment process include: pH value, chemical oxygen demand COD, biochemical oxygen demand BOD, suspended solids SS, ammonia nitrogen NH3-N, nitrate nitrogen NO3-N, sludge concentration MLSS, redox potential ORP, water temperature T, and various combinations of dissolved oxygen DO in anaerobic zone, anoxic zone and aerobic zone; The water quality indicators of the chemical precipitation treatment process include: pH value, chemical oxygen demand COD, suspended solids SS, heavy metal ions, phosphorus PO4-P, water temperature T and other combinations; The water quality indicators of the coagulation and flocculation treatment process include: pH value, chemical oxygen demand COD, suspended solids SS, total phosphorus TP, total nitrogen TN, ammonia nitrogen NH3-N, oil, turbidity, chromaticity, and various combinations of water temperature T; The water quality indicators of the disinfection treatment process include: pH value, chemical oxygen demand COD, suspended solids SS, ammonia nitrogen NH3-N, heavy metal ions, residual chlorine, turbidity, color, odor, fecal coliform bacteria, total coliform bacteria, and various combinations of viruses.

[0039] In the above technical scheme, a combination of water quality indicators is selected based on specific sewage treatment standards, processes, and a priori conclusions. Although the sewage treatment process also includes pretreatment and primary treatment before the biochemical treatment process, including screens, grit chambers, etc., to protect the facilities of the biological treatment process from damage and reduce the treatment load; and including sedimentation tanks, sedimentation or flotation methods to remove suspended matter and colloidal particles or low-concentration organic matter, providing better conditions for biological treatment; in order to reduce the data redundancy of the self-attention neural network model, the detection data of the above two processes are not selected for the establishment and training of the self-attention neural network model. The actual results also show that the self-attention neural network model of this scheme is sufficient to meet the accurate prediction of the dosing strategy.

[0040] In another technical solution, the types of drugs added in the biochemical treatment process include: a combination of carbon source, nitrogen source, phosphorus source, flocculant, coagulant aid, conditioner, pH adjuster, redox agent, disinfectant, microbial preparation, sludge improver, and phosphorus removal agent; The types of drugs added in the chemical precipitation treatment process include: hydroxide precipitants, sulfide precipitants, barium salt precipitants, carbonate precipitants, halide precipitants, reduction precipitants, organic reagent precipitants, magnesium ammonium phosphate precipitants, flocculants, coagulants, and various combinations of conditioning agents; The types of drugs added in the coagulation and flocculation treatment process include: polyaluminium chloride PAC, polyacrylamide PAM, aluminium sulfate, ferric chloride, coagulant aid, pH adjuster, and various combinations of oxidants; The types of drugs added in the disinfection treatment process include: chlorine-containing disinfectants, chlorine dioxide disinfectants, ozone, peroxide disinfectants, iodine-containing disinfectants, bromine-containing disinfectants, phenol disinfectants, and various combinations of quaternary ammonium salt disinfectants.

[0041] In the above technical solution, each sewage treatment process selects a combination of dosing types based on specific sewage treatment standards, processes, a priori conclusions and actual conditions of existing sewage treatment.

[0042] In another technical solution, the data preprocessing module further includes: a module for using a linear interpolation method to collect the influent water quality data E acquired by the data acquisition module. s and the inflow flow rate q, water quality data E of each sewage treatment process m , Discharge water quality data E z Perform interpolation processing, as shown in formula 1: Formula 1, Among them, A represents the detection time of the missing data in the detection data, B represents the missing data in the detection data, B0 and B1 represent the measured data before and after B respectively, and A0 and A1 represent the detection time corresponding to B0 and B1 respectively.

[0043] In the above technical solution, when establishing, training and using the self-attention neural network model, historical data and one or more data of real-time detection may be missing. Considering that during the operation of sewage treatment, the water quality data of each sewage treatment process changes slowly, the linear interpolation method is used to interpolate the missing data, thereby ensuring the integrity of the input data without affecting the prediction accuracy of the dosing strategy.

[0044] In another technical solution, the chemical reaction relationship matrix E is used to represent the subordinate relationship between the process itself and the test data after the process after the chemical treatment of multiple groups of sewage treatment processes. The test data subordinate relationship is expressed asa ×(X a -X a0 )+ b ×(X b -X b0 )+ c ×(X c -X c0 )∈E, where a Indicates the weight of the relationship between the current dosing process and the water quality index, b Indicates the relationship weight between the next sewage treatment process adjacent to the current dosing process and the water quality index, c represents the weight of the relationship between the next sewage treatment process adjacent to the current dosing process and the water quality index, X a0 and X a Respectively represent the test data of water quality indicators during and after the current dosing process, X b0 and X b They represent the test data of water quality index of the next sewage treatment process during and after the current dosing process, respectively. c0 and X c They represent the water quality index test data of the next second sewage treatment process during and after the current dosing process, and so on. Taking the chemical oxygen demand COD water quality index as an example, X a0 and X a They represent the COD test data before and after the addition of chemicals in the biochemical treatment process, a is the preset weight, or it can be the updated weight after the training is completed, X b0 and X b They represent the COD detection data of the chemical precipitation treatment process before and after the addition of the biochemical treatment process, b is the preset weight, and can also be the updated weight after the training is completed, X c0 and X c They respectively represent the chemical oxygen demand (COD) detection data of the coagulation and flocculation treatment processes before and after the addition of chemicals in the biochemical treatment process, and c is a preset weight, which can also be an updated weight after the training is completed.

[0045] In the above technical scheme, based on the characteristics of the sewage treatment process, that is, the previous sewage treatment process will affect the next sewage treatment process immediately adjacent, or even the operating status of multiple sewage treatment processes downstream, but will not affect the water quality of the upstream sewage treatment process, therefore, a ternary expression is established according to the order of the biochemical treatment process, the chemical precipitation treatment process, the coagulation and flocculation treatment process, and the disinfection treatment process. The ternary expression is selected first to ensure the correlation and reduce data redundancy. For example, for the biochemical treatment process, a dosing treatment is performed in the biochemical treatment process, a represents the relationship weight of the biochemical treatment process of the dosing treatment, b represents the relationship weight of the chemical precipitation treatment process of the dosing treatment, and c represents the relationship weight of the coagulation and flocculation treatment process of the dosing treatment. a, b, and c are determined after the self-attention neural network model training is completed. Using the chemical reaction relationship matrix E to update the modified attention score matrix can more accurately reflect the impact of each dosing, thereby further optimizing the output dosing strategy.

[0046] In another technical solution, the self-attention dosing prediction model is obtained based on the convergence of the consumption rate of the agent per ton of sewage within a period of time being lower than the set threshold. Under the premise of ensuring the quality of sewage treatment, the self-attention neural network model is optimized and adjusted by superimposing the consumption rate of the agent per ton of sewage within a period of time, which can reduce the amount of dosing and achieve energy saving effect.

[0047] The division of modules in this application is schematic and is only a logical function division. There may be other division methods in actual implementation. In addition, each functional module in each embodiment of the present invention may be integrated into one processor, or may exist physically separately, or two or more modules may be integrated into one module. The above-mentioned integrated modules may be implemented in the form of hardware or in the form of software functional modules.

[0048] The present invention also provides an electronic device, comprising: one or more processors; and a memory connected to the at least one processor in communication, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor performs the processing of the sewage treatment dosing control system. The electronic device can be any terminal device including a mobile phone, a laptop computer, a desktop computer, a tablet computer, a PDA (Personal Digital Assistant), a POS (Point of Sales), a vehicle-mounted computer, etc.

[0049] The present invention also provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, the processing of the sewage treatment dosing control system is realized.

[0050] Through the description of the above implementation mode, the technicians in the relevant field can clearly understand that the present invention can be implemented by means of software plus necessary general hardware, and of course, it can also be implemented by special hardware including special integrated circuits, special CPUs, special memories, special components, etc. In general, all functions completed by computer programs can be easily implemented by corresponding hardware, and the specific hardware structure used to implement the same function can also be various, such as analog circuits, digital circuits or special circuits. However, for the present invention, software program implementation is a better implementation mode in more cases. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a readable storage medium, such as a computer floppy disk, a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, RandomAccess Memory), a disk or an optical disk, etc., including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0051] Although the embodiments of the present invention have been disclosed as above, they are not limited to the applications listed in the specification and the implementation modes, and they can be fully applied to various fields suitable for the present invention. For those familiar with the art, additional modifications can be easily implemented. Therefore, without departing from the general concept defined by the claims and the scope of equivalents, the present invention is not limited to the specific details and the illustrations shown and described herein.

Claims

1. A dosing control system for sewage treatment, characterized in that: include: Data collection module, which is used to collect real-time sewage inlet water quality data and inlet flow, water quality data of each sewage treatment process, and discharge water quality data; Data acquisition module, which is used to obtain the most recent dosing time t i , the type of dosing M, the amount of dosing m, and the multiple groups of influent water quality data E collected by the data acquisition module after the dosing treatment s and water inflow q, water quality data E of multiple groups of sewage treatment processes m , multiple sets of discharge water quality data E z , water temperature T; The data preprocessing module is used to preprocess the water quality data E of each process of multiple sewage treatment processes based on a preset judgment threshold. m Assignment, where water quality data E m When the measured value meets the judgment threshold, it is assigned a value of g, otherwise, it is assigned a value of p, where g and p are both real numbers; The dosing prediction module is used to input real-time data into the self-attention dosing prediction model and output the next dosing time t i+1 , dosing type M and dosing amount m, discharge water quality data E z ; Wherein, the real-time data includes the data acquired by the data acquisition module and the data preprocessed by the data preprocessing module; and, The self-attention dosing prediction model is based on a chemical reaction relationship matrix E of water quality data and discharged water quality data in each sewage treatment process caused by a series of chemical changes caused by two adjacent dosing amounts and dosing times. The attention score matrix is ​​multiplied by the chemical reaction relationship matrix E to obtain a corrected attention score matrix, and then the self-attention neural network model constructed based on the corrected attention score matrix is ​​trained; The dosing control module is used to add medicine according to the output dosing time, dosing type and dosing amount.

2. The sewage treatment dosing control system according to claim 1, characterized in that: The self-attention dosing prediction model also includes, before outputting, determining the current pre-output t i+1 Discharge water quality data at the moment E z Whether it meets the preset discharge water quality standard, if not, then the query is fed back forward according to the preset time gradient value until the discharge water quality data E is queried z When the discharge water quality standard is met, the output (t i+1 -Δt) time, discharged water quality data E z , and the initial pre-output t i+1 The drug type M and drug amount m corresponding to the moment, where Δt represents the sum of the time gradient values ​​of the forward feedback query.

3. The sewage treatment dosing control system according to claim 1, characterized in that: The inlet water quality data E s The water quality indicators include: pH value, chemical oxygen demand COD, five-day biochemical oxygen demand BOD5, suspended solids SS, total phosphorus TP, total nitrogen TN, ammonia nitrogen NH3-N, oil, heavy metal ions, sulfide, residual chlorine, turbidity, color, odor and taste, total bacteria count, coliform count, conductivity, solid matter, water temperature T and other combinations; The discharge water quality data E z The water quality indicators include: pH value, chemical oxygen demand COD, five-day biochemical oxygen demand BOD5, suspended solids SS, total phosphorus TP, total nitrogen TN, ammonia nitrogen NH3-N, oil, heavy metal ions, sulfide, residual chlorine, turbidity, chromaticity, total bacteria count, coliform count, and a variety of combinations of anionic surfactants.

4. The sewage treatment dosing control system according to claim 1, characterized in that: The sewage treatment process includes a biochemical treatment process in a biochemical reaction tank, a chemical precipitation treatment process, a coagulation and flocculation treatment process, and a disinfection treatment process; as well as, The water quality indicators of the biochemical treatment process include: pH value, chemical oxygen demand COD, biochemical oxygen demand BOD, suspended solids SS, ammonia nitrogen NH3-N, nitrate nitrogen NO3-N, sludge concentration MLSS, redox potential ORP, water temperature T, and various combinations of dissolved oxygen DO in anaerobic zone, anoxic zone and aerobic zone; The water quality indicators of the chemical precipitation treatment process include: pH value, chemical oxygen demand COD, suspended solids SS, heavy metal ions, phosphorus PO4-P, water temperature T and other combinations; The water quality indicators of the coagulation and flocculation treatment process include: pH value, chemical oxygen demand COD, suspended solids SS, total phosphorus TP, total nitrogen TN, ammonia nitrogen NH3-N, oil, turbidity, chromaticity, and various combinations of water temperature T; The water quality indicators of the disinfection treatment process include: pH value, chemical oxygen demand COD, suspended solids SS, ammonia nitrogen NH3-N, heavy metal ions, residual chlorine, turbidity, color, odor, fecal coliform bacteria, total coliform bacteria, and various combinations of viruses.

5. The sewage treatment dosing control system according to claim 3, characterized in that: The types of drugs added in the biochemical treatment process include: a combination of carbon source, nitrogen source, phosphorus source, flocculant, coagulant aid, conditioner, pH adjuster, redox agent, disinfectant, microbial preparation, sludge improver, and phosphorus removal agent; The types of drugs added in the chemical precipitation treatment process include: hydroxide precipitants, sulfide precipitants, barium salt precipitants, carbonate precipitants, halide precipitants, reduction precipitants, organic reagent precipitants, magnesium ammonium phosphate precipitants, flocculants, coagulants, and various combinations of conditioning agents; The types of drugs added in the coagulation and flocculation treatment process include: polyaluminium chloride PAC, polyacrylamide PAM, aluminium sulfate, ferric chloride, coagulant aid, pH adjuster, and various combinations of oxidants; The types of drugs added in the disinfection treatment process include: chlorine-containing disinfectants, chlorine dioxide disinfectants, ozone, peroxide disinfectants, iodine-containing disinfectants, bromine-containing disinfectants, phenol disinfectants, and various combinations of quaternary ammonium salt disinfectants.

6. The sewage treatment dosing control system according to claim 1, characterized in that: The data preprocessing module also includes: a module for using a linear interpolation method to collect the influent water quality data E acquired by the data acquisition module; s and the inflow flow rate q, water quality data E of each sewage treatment process m , Discharge water quality data E z Perform interpolation processing, as shown in formula 1: B= Formula 1, Among them, A represents the detection time of the missing data in the detection data, B represents the missing data in the detection data, B0 and B1 represent the measured data before and after B respectively, and A0 and A1 represent the detection time corresponding to B0 and B1 respectively.

7. The dosing control system for sewage treatment according to claim 1, characterized in that: The chemical reaction relationship matrix E is used to represent the subordinate relationship between the process itself and the test data after the process after the chemical addition treatment of multiple groups of sewage treatment processes. The test data subordinate relationship is expressed as a ×(X a -X a0 )+ b ×(X b -X b0 )+ c ×(X c -X c0 )∈E, where a Indicates the weight of the relationship between the current dosing process and the water quality index, b Indicates the relationship weight between the next sewage treatment process adjacent to the current dosing process and the water quality index, c represents the weight of the relationship between the next sewage treatment process adjacent to the current dosing process and the water quality index, X a0 and X a Respectively represent the test data of water quality index during and after the current dosing process, X b0 and X b They represent the test data of water quality index of the next sewage treatment process during and after the current dosing process, respectively. c0 and X c They respectively represent the test data of water quality indicators of the adjacent second sewage treatment process during and after the current dosing process, and so on.

8. The sewage treatment dosing control system according to claim 1, characterized in that: The self-attention dosing prediction model is obtained based on the convergence of the consumption rate of the agent per ton of sewage within a period of time being lower than a set threshold.

9. An electronic device, characterized in that include: one or more processors; And a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor performs the processing of the dosing control system for sewage treatment according to any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that A computer program is stored on the storage medium, and when the program is executed by the processor, the processing of the sewage treatment dosing control system described in any one of claims 1 to 8 is implemented.

Images