Operation control method and system for sewage treatment denitrification device

Abstract

The invention relates to the technical field of sewage treatment denitrification, and discloses an operation control method and system for a sewage treatment denitrification device, and the method comprises the following steps: obtaining inflow water quality monitoring data of a sewage treatment plant in real time; the method comprises the following steps: preprocessing inflow water quality monitoring data, and inputting the data into a denitrification device operation control parameter prediction model trained based on ensemble learning to obtain predicted operation control parameters; and controlling the sewage treatment denitrification device to operate based on the predicted operation control parameters. According to the embodiment of the invention, accurate operation control parameters can be generated based on real-time inflow water quality monitoring data, it is ensured that the denitrification device operates in an optimal state, the situations of energy waste and the like caused by unreasonable parameters are avoided, the operation cost of sewage treatment can be reduced, reasonable utilization of resources and effective saving of cost are realized, and the method is suitable for popularization and application. The economical efficiency of sewage treatment is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of sewage treatment and denitrification, and in particular to an operation control method and system of a sewage treatment and denitrification device. Background Art

[0002] Excessive application of nitrogen fertilizers and improper discharge of domestic sewage and industrial wastewater have caused a large amount of nitrogen to flow into aquatic ecosystems, triggering eutrophication of water bodies, which in turn has significantly reduced the quality of the water environment and posed a potential threat to human health.

[0003] During the denitrification process, sewage treatment plants generally rely on manual experience to set the operating parameters of the denitrification device. It is difficult to accurately control the operation mode of the denitrification device according to the real-time influent water quality and treatment requirements. This leads to incomplete denitrification reaction and low total nitrogen removal rate when the carbon source is insufficient; or excessive carbon source not only increases the treatment cost, but may also cause problems such as excessive effluent COD. Summary of the invention

[0004] In view of this, the present invention provides an operation control method and system for a sewage treatment denitrification device to solve the problem of how to accurately control the operation mode of the sewage treatment denitrification device and reduce management costs while meeting sewage discharge standards.

[0005] In a first aspect, the present invention provides an operation control method for a sewage treatment denitrification device, the method comprising:

[0006] Obtain real-time monitoring data on the influent water quality of sewage treatment plants;

[0007] After preprocessing the influent water quality monitoring data, the data is input into the denitrification device operation control parameter prediction model trained based on ensemble learning to obtain the predicted operation control parameters;

[0008] The operation of the sewage treatment denitrification device is controlled based on the predicted operation control parameters.

[0009] The operation control method of the sewage treatment denitrification device provided in the embodiment of the present invention can generate accurate operation control parameters based on real-time influent water quality monitoring data, ensure that the denitrification device operates in an optimal state, avoid energy waste due to unreasonable parameters, etc., help to reduce the operating cost of sewage treatment, achieve rational utilization of resources and effective cost savings, improve the economy of sewage treatment, and meet the development trend of efficient, stable and intelligent operation of modern sewage treatment plants.

[0010] In an optional embodiment, the process of training a denitrification device operation control parameter prediction model based on ensemble learning includes:

[0011] Obtaining historical operation data of the sewage treatment plant and preprocessing the data, the historical operation data including: inlet water quality parameters and corresponding operation modes and operation parameters, outlet water quality parameters, and consumables prices;

[0012] Use historical running data as sample data, and divide the sample data into a training set and a test set;

[0013] The test set is divided into multiple test data subsets and input into multiple learners respectively. The training is carried out with the constraints of meeting the effluent water quality parameters and minimizing the cost of consumables. The trained model is evaluated based on preset evaluation indicators, and the final learner or learner combination is determined according to the evaluation value as the final denitrification device operation control parameter prediction model.

[0014] The embodiment of the present invention obtains comprehensive historical operation data of the sewage treatment plant, including information such as inlet water quality parameters, operation mode, operation parameters, effluent water quality parameters, and consumables prices, so that the integrated learning model can learn the operation rules of the denitrification device under different working conditions. At the same time, the model is trained with the effluent water quality parameters meeting the standards and the minimum consumables price cost as constraints, so that the model closely revolves around the actual operation goals of the sewage treatment plant. This training method does not simply pursue prediction accuracy, but comprehensively considers the two key factors of economic cost and water quality compliance. In the actual sewage treatment process, substandard effluent water quality will lead to environmental problems and regulatory risks, while excessively high consumables prices will increase the operational burden. Therefore, the model can provide sewage treatment plants with more economical and environmentally beneficial operation control parameter predictions.

[0015] In an optional embodiment, the influent water quality parameters include: influent COD cr , influent total nitrogen, influent flow rate; and according to influent COD cr Determine the carbon-nitrogen ratio of water quality based on the total nitrogen in the influent, and determine the water volume and water quality stability based on the changes in the influent flow rate and influent water quality within a preset unit time;

[0016] When the total nitrogen in the inlet water meets the preset standard, the water quality type is up-to-standard water quality, and the operation mode corresponding to the up-to-standard water quality is the bypass override mode;

[0017] When the total nitrogen in the inlet water does not meet the preset standards, the water quality carbon-nitrogen ratio and water quality stability are further judged, and the water quality is divided into different sewage types according to the numerical range of the water quality carbon-nitrogen ratio, water volume, and whether the water quality is stable, and different operating modes and operating parameters are set for different sewage types.

[0018] The embodiment of the present invention first performs preliminary classification based on whether the total nitrogen in the influent meets the preset standard, and adopts the bypass override mode for water quality that meets the standard. This is a way to efficiently utilize resources. When the total nitrogen in the influent does not meet the standard, the sewage type is further subdivided based on the carbon-nitrogen ratio, water volume, and water quality stability, and different operating modes and parameters are set. This detailed classification and targeted treatment method can enable the denitrification device to operate in the most appropriate manner when facing sewage of different properties. This refined treatment method helps to control operating costs.

[0019] In an optional embodiment, the water volume and water quality are divided into different sewage types according to the numerical range of the water quality carbon-nitrogen ratio and whether the water quality is stable, and different operating modes and operating parameters are set for different sewage types, including:

[0020] When the total nitrogen in the influent does not meet the preset standard, the carbon-nitrogen ratio and water quality stability of the water quality are further judged;

[0021] When the influent carbon-nitrogen ratio is greater than the first preset threshold value and the influent flow rate and water quality change within the preset unit time are less than the preset change threshold range, the influent carbon-nitrogen ratio is high and the water quality is stable, and the corresponding operation mode is the heterotrophic denitrification mode, and its control parameters include: central pool liquid level, central pool carbon source dosage, and central pool influent flow rate;

[0022] When the inlet carbon-nitrogen ratio is greater than the second preset threshold and less than the first preset threshold, and the inlet flow rate and water quality change within the preset unit time are less than the preset change threshold range, the inlet carbon-nitrogen ratio is low and the water quality is stable, and the corresponding operation mode is the autotrophic denitrification mode, and its control parameters include: peripheral pool liquid level, peripheral pool filler filling ratio, and peripheral pool inlet flow rate;

[0023] When the inlet carbon-nitrogen ratio is less than the second preset threshold value and the inlet flow rate and water quality changes within the preset unit time are greater than the preset change threshold range, the inlet carbon-nitrogen ratio is extremely low and the water quality is unstable. The corresponding operation mode is the autotrophic-heterotrophic coordinated denitrification mode, and its control parameters include: central tank liquid level, peripheral tank liquid level, and central tank and peripheral tank inlet ratio.

[0024] The embodiment of the present invention divides the sewage types according to different carbon-nitrogen ratio ranges and water quality and water quantity stability conditions and sets corresponding operation modes, so that the sewage treatment system can flexibly respond to various changes in the influent water quality. Regardless of whether the sewage has a high and stable carbon-nitrogen ratio or a low and unstable carbon-nitrogen ratio, there is a matching treatment method, which reduces the risk of the denitrification device being unable to operate effectively due to sudden changes in water quality, and ensures the continuity and stability of the entire sewage treatment process. From a maintenance perspective, the key maintenance parts and maintenance cycles of equipment under different operating modes can also be determined more specifically. In the heterotrophic denitrification mode, more attention needs to be paid to the maintenance of the carbon source addition equipment; in the autotrophic denitrification mode, the maintenance of equipment related to the peripheral pool filler is more critical, which helps to reasonably arrange maintenance resources, improve the service life and operational reliability of the equipment, and ensure the long-term efficient and stable operation of the entire sewage treatment system.

[0025] In an optional embodiment, the effluent water quality parameters include: effluent COD cr , total nitrogen in effluent, effluent flow rate; the prices of consumables: including liquid carbon source price, solid carbon source price, and sulfur autotrophic composite consumables price.

[0026] In the embodiment of the present invention, the effluent COD cr , effluent total nitrogen and effluent flow rate into historical data, so that the model can establish a complete relationship between the inlet water quality parameters and the final effluent water quality. The model can learn how the denitrification device can operate under different inlet conditions to achieve the expected effluent water quality standards. The price information of various types of consumables allows the model to consider economic costs during the training process. The model can not only learn how to optimize the water quality treatment effect, but also consider the impact of consumable prices on operating costs.

[0027] In an optional embodiment, the method of determining the final learner or learner combination as the final denitrification device operation control parameter prediction model according to the evaluation value includes:

[0028] When a single evaluation index is used, the evaluation values ​​are sorted in order of size, and the final coagulant prediction model is determined based on the sorting results;

[0029] When multiple evaluation indicators are used, the average of the prediction results corresponding to the learners with evaluation values ​​greater than the preset threshold is taken as the final prediction result.

[0030] When a single evaluation index is used in the embodiment of the present invention, the final denitrification device operation control parameter prediction model is determined by sorting the evaluation values, which provides a clear comparison standard for different learners. The use of a single evaluation index allows the operator to focus directly on the key performance factors and select the learner with the best performance on the factor as the final model, so as to better achieve the operation goals. By combining multiple evaluation indicators, the final prediction result can meet certain standards in multiple important performance dimensions. Multiple comprehensive considerations can improve the reliability and practicality of the model in practical applications, making it more in line with the complex and changeable actual operation needs of sewage treatment plants.

[0031] In a second aspect, the present invention provides an operation control system for a sewage treatment denitrification device, comprising:

[0032] Real-time water quality parameter monitoring module, used to obtain real-time influent water quality monitoring data of sewage treatment plants;

[0033] The operation mode prediction module is used to pre-process the influent water quality monitoring data and input it into the denitrification device operation control parameter prediction model trained based on ensemble learning to obtain the predicted operation control parameters;

[0034] An operation control module is used to control the operation of the sewage treatment denitrification device based on the predicted operation control parameters.

[0035] In a third aspect, the present invention provides a computer device, comprising: a memory and a processor, the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the operation control method of the sewage treatment denitrification device of the first aspect or any corresponding embodiment thereof by executing the computer instructions.

[0036] In a fourth aspect, the present invention provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the operation control method for a sewage treatment denitrification device according to the first aspect or any corresponding embodiment thereof.

[0037] In a fifth aspect, the present invention provides a computer program product, comprising computer instructions for causing a computer to execute the operation control method of a sewage treatment denitrification device according to the first aspect or any corresponding embodiment thereof. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0039] Figure 1 A schematic diagram of a flow chart of an operation control method of a sewage treatment denitrification device according to an embodiment of the present invention;

[0040] Figure 2 This is the sewage treatment process flow chart;

[0041] Figure 3 A schematic diagram of a flow chart of a denitrification device operation control parameter prediction model based on ensemble learning training according to an embodiment of the present invention;

[0042] Figure 4 Schematic diagram of a sewage treatment denitrification device according to an embodiment of the present invention;

[0043] Figure 5 A structural block diagram of an operation control system of a sewage treatment and denitrification device according to an embodiment of the present invention;

[0044] Figure 6 It is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention.

[0045] Reference numerals:

[0046] 1-Secondary sedimentation tank; 2-Water outlet of the second sedimentation tank; 3-Packed bed water inlet pump; 4-Air washing pump; 5-Central pool water inlet valve; 6-Peripheral pool water inlet valve; 7-Air washing main valve; 8-Peripheral pool water inlet; 9-First peripheral pool air washing valve; 10-Central pool air washing valve; 11-Second peripheral pool air washing valve; 12-Supporting layer; 13-Peripheral pool water washing valve; 14-Water washing pump; 15-Central pool water inlet; 16-Peripheral pool; 17-Central pool; 18-Bypass valve; 19-Bypass water inlet; 20-Filter material supplement Filling port; 21-consumables replenishing pipe; 22-autotrophic consumables replenishing pipe; 23-heterotrophic consumables replenishing pipe; 24-outlet; 25-outlet valve; 26-clear water tank; 27-central tank water washing valve; 28-influent COD monitoring instrument; 29-outlet flow meter; 30-liquid level meter; 31-outlet total nitrogen; 32-outlet COD; 33-control system; 34-inlet total nitrogen; 35-disinfection tank; 36-overflow weir; 37-water washing sewage outlet; 38-water washing sewage valve; 39-wastewater inlet. DETAILED DESCRIPTION

[0047] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution 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, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0048] In order to achieve an accurate, efficient and economical deep denitrification operation mode, an embodiment of the present invention provides an operation control method embodiment of a sewage treatment denitrification device. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that here.

[0049] In this embodiment, a method for controlling the operation of a sewage treatment denitrification device is provided. Figure 1 is a flow chart of an operation control method for a sewage treatment denitrification device according to an embodiment of the present invention. Figure 1 As shown, the process includes the following steps:

[0050] S101, real-time acquisition of influent water quality monitoring data of the sewage treatment plant.

[0051] Specifically, the influent water quality of a sewage treatment plant changes dynamically, and this change may be caused by factors such as the diversity of sewage sources, seasonal changes, or emergencies. By acquiring the influent water quality monitoring data of the sewage treatment plant in real time, it is possible to timely grasp the dynamic changes in the influent situation, avoiding the deviation caused by relying on fixed parameters or outdated data to control the operation of the device, so that the subsequent prediction of the operation control parameters of the denitrification device is more in line with the actual influent conditions, thereby ensuring more accurate device operation control. The embodiment of the present invention obtains the influent COD in real time. cr (chemical oxygen demand), influent total nitrogen, influent flow rate, and based on influent COD cr The carbon-nitrogen ratio of water quality is determined by the total nitrogen in the inlet water, and the water quality stability is determined according to the changes in the inlet water quality and flow rate within the preset unit time. The carbon-nitrogen ratio of the inlet water and the water quality and flow rate stability information are obtained, which can provide basic data basis for the accurate operation control parameters of the denitrification device.

[0052] S102, after pre-processing the influent water quality monitoring data, input it into the denitrification device operation control parameter prediction model trained based on ensemble learning to obtain predicted operation control parameters.

[0053] Specifically, the influent water quality monitoring data is normalized and other preprocessed before being input into the prediction model to predict the operation control parameters. The main purpose of the denitrification device is to remove nitrate nitrogen from sewage. By real-time monitoring of the influent water quality and using the model to predict the operation control parameters, the conditions for the denitrification reaction can be better maintained to ensure a stable denitrification effect. In the prediction of the operation control parameters of the sewage treatment denitrification device, there may be a complex relationship between the influent water quality monitoring data and the operation control parameters. For example, the relationship between parameters such as the water quality carbon-nitrogen ratio, stability data, and the carbon source dosage of the denitrification device may be affected by the interaction of multiple factors. Ensemble learning can combine the advantages of multiple learning algorithms or models, effectively reduce the problems of overfitting and underfitting that may exist in a single model, improve the accuracy of the prediction of the operation control parameters, and help to more accurately set the operation parameters of the denitrification device to ensure stable and efficient treatment effects.

[0054] S103, controlling the operation of the sewage treatment denitrification device based on the predicted operation control parameters.

[0055] Specifically, the following can be achieved through the operation of the sewage treatment denitrification device: Figure 2 The denitrification treatment process shown in the figure is used to obtain water that meets the standards. The embodiment of the present invention provides intelligent decision support for the operation and management of the sewage treatment plant through the operation control parameters predicted by the model. The control system can directly make operational adjustments based on the operation control parameters predicted by the model, reducing the subjectivity and errors of manual judgment. At the same time, with the continuous accumulation of data, the model can also be continuously optimized and updated to further improve the accuracy and reliability of the prediction, and promote the development of sewage treatment plants in the direction of intelligence and efficiency.

[0056] The operation control method of the sewage treatment denitrification device in the embodiment of the present invention can generate accurate operation control parameters based on real-time influent water quality monitoring data, ensure that the denitrification device operates in an optimal state, avoid energy waste due to unreasonable parameters, etc., help to reduce the operating cost of sewage treatment, achieve rational utilization of resources and effective cost savings, improve the economy of sewage treatment, and meet the development trend of efficient, stable and intelligent operation of modern sewage treatment plants.

[0057] The embodiment of the present invention is based on the process of training the denitrification device operation control parameter prediction model based on ensemble learning, such as Figure 3 As shown, including:

[0058] S201, obtaining historical operation data of the sewage treatment plant and performing preprocessing, wherein the historical operation data includes: inlet water quality parameters and corresponding operation modes and operation parameters, outlet water quality parameters, and consumables prices.

[0059] Specifically, the influent water quality parameters include: influent COD cr, inlet total nitrogen, inlet flow; outlet water quality parameters, including: outlet COD cr (chemical oxygen demand), effluent total nitrogen, inlet flow; consumables price: including liquid carbon source price, solid carbon source price, sulfur autotrophic composite consumables price. cr , influent total nitrogen, influent flow rate and other key indicators can more comprehensively evaluate the nature of sewage and convert effluent COD cr , effluent total nitrogen and effluent flow rate are included in the historical data, so that the model can establish a complete correlation between the inlet water quality parameters and the final effluent water quality. The model can learn how the denitrification device can operate under different inlet conditions to achieve the expected effluent water quality standards. The price information of various consumables allows the model to consider economic costs during the training process. The model can not only learn how to optimize the water quality treatment effect, but also take into account the impact of different consumable prices on operating costs. For example, when the model predicts operating parameters, it will weigh the cost of using different carbon sources (liquid carbon or solid carbon sources) and the cost corresponding to different amounts of sulfur autotrophic composite fillers, so as to select a more economical operation mode while ensuring that the effluent water quality meets the standards.

[0060] In the embodiment of the present invention, when the total nitrogen in the influent meets the preset standard, the water quality type is up-to-standard water quality, and the operation mode corresponding to the up-to-standard water quality is the bypass override mode. In one embodiment, the structure of the sewage treatment denitrification device is as follows: Figure 4 As shown, the operation process of the bypass override mode includes: the bypass valve 18 is opened, the water outlet 2 of the secondary sedimentation tank enters the packed bed through the bypass water inlet 19, the effluent flows over the overflow weir 36, and is discharged through the outlet 24, the outlet valve 25 is opened, and enters the disinfection tank 35, and then enters the clean water tank 26.

[0061] When the total nitrogen in the inlet water does not meet the preset standards, the water quality carbon-nitrogen ratio and water quality stability are further judged, and the water quality is divided into different sewage types according to the numerical range of the water quality carbon-nitrogen ratio and whether the water quality is stable, and different operating modes and operating parameters are set for different sewage types.

[0062] The embodiment of the present invention first performs preliminary classification based on whether the total nitrogen in the influent meets the preset standard, and adopts the bypass override mode for water quality that meets the standard. This is a way to efficiently utilize resources. When the total nitrogen in the influent does not meet the standard, the sewage type is further subdivided in combination with the carbon-nitrogen ratio, water volume, and water quality stability, and different operating modes and parameters are set. This detailed classification and targeted treatment method can enable the denitrification device to operate in the most appropriate manner when facing sewage of different properties. This refined treatment method helps to control operating costs. For example, for water quality that meets the standard and can adopt the bypass override mode, unnecessary processing processes are avoided, energy and chemical consumption are saved, and operating costs can be reduced while ensuring the sewage treatment effect.

[0063] Furthermore, the embodiment of the present invention divides water quality into different sewage types according to the numerical range of the carbon-nitrogen ratio of water quality, the water volume, and whether the water quality is stable, and sets different operation modes and operation parameters for different sewage types, specifically including:

[0064] 1. When the influent carbon-nitrogen ratio is greater than the first preset threshold value and the influent flow rate and water quality change within the preset unit time are less than the preset change threshold range, the influent carbon-nitrogen ratio is high and the water quality is stable. The corresponding operation mode is heterotrophic denitrification mode, and its control parameters include: central pool level, carbon source dosage in the central pool, and central pool influent flow rate. Among them, the central pool level reflects the content of the liquid in the central pool. When the liquid level is high, the residence time of the wastewater in the packed bed may be extended, which means that the adsorbent has more opportunities to react with the pollutants in the wastewater. By using the central pool level as an input feature, the model can learn the relationship between the liquid level change and the treatment effect. In sewage treatment and other processes, the amount of carbon source dosage in the central pool is a key factor. Carbon sources are mainly used for the growth and metabolic activities of microorganisms. The appropriate amount of carbon addition can promote the reproduction of microorganisms and enhance the ability of microorganisms to decompose and remove pollutants; the hydraulic retention time of sewage in the central pool can be calculated by the influent flow rate and liquid level of the central pool, which directly affects the contact time between sewage and treatment factors such as microorganisms and agents.

[0065] In one embodiment, when the carbon-nitrogen ratio (C / N) of the inlet water is high (for example, above 5) and the water volume and water quality are stable (for example, within 1 hour, the preset lower limit is less than 30%-40%), in order to save sulfur autotrophic filler and improve the economic efficiency of the project, the heterotrophic denitrification mode is adopted. The operation process corresponding to this mode is: the central pool water inlet valve 5 is opened, and the effluent of the secondary sedimentation tank passes through the secondary sedimentation tank outlet 2 and the central pool water inlet 15 at the bottom of the packed bed into the central pool 17. The treated effluent turns over the overflow weir 36 and is discharged through the outlet 24. The outlet valve 25 is opened and enters the disinfection pool 35, and then enters the clean water pool 26. The central pool 17 is air-washed by opening the main air-washing valve 7, the central pool air-washing valve 10 and the air-washing pump 4, and water-washed by opening the central pool water-washing valve 27 and the water-washing pump 14. The air-washing intensity is 45-90m 3 / ㎡·h, water washing intensity is 15m 3 / ㎡·h, the backwash cycle is once every 1 to 2 days. The water source for washing is drawn from the clean water tank 26, the water washing valve 27 of the central tank is opened, and the backwash water is discharged into the secondary sedimentation tank 1 through the water washing sewage outlet 37. The bottom of the central tank 17 is a supporting layer 12, and the filler is quartz sand (with a diameter of 2 to 3 mm), ceramsite (with a diameter of 3 to 5 mm), and a solid carbon source (with a diameter of 3 to 5 mm). Among them, the solid carbon source filling ratio is 15% to 30%, and low molecular weight synthetic carbon sources such as polybutylene succinate PBS and polycaprolactone PCL can be selected. The solid carbon source is not only a carrier of the microbial membrane, but also can slowly release organic matter under the action of microbial enzymes, creating a stable and easy-to-maintain growth environment for microorganisms, and further improving the denitrification efficiency. The solid carbon source has the characteristics of slow release, and the replenishment cycle is 6-12 months. The replenishment process is: enter the consumable replenishment pipe 21 through the filter material replenishment port 20, and finally enter the central tank 17 through the heterotrophic consumable replenishment pipe 23.

[0066] In another embodiment, if the carbon-nitrogen ratio (C / N) of the inlet water is high (for example, above 10) and the water quantity and quality are stable (for example, within 1 hour, it exceeds the preset lower limit value by less than 30%-40%), the solid carbon source in the previous embodiment is changed to a liquid carbon source, and the dosage of the liquid carbon source is set according to the carbon-nitrogen ratio.

[0067] It should be noted that, except for the carbon source dosage of the central pool, the liquid level of the central pool, and the water inlet flow rate of the central pool, which are control parameter variables that need to be trained, other parameters such as air washing intensity, water washing intensity, and backwashing cycle are all pre-set fixed values.

[0068] 2. When the carbon-nitrogen ratio of the influent is greater than the second preset threshold and less than the first preset threshold, and the influent flow rate and water quality change within the preset unit time are less than the preset change threshold range, the influent carbon-nitrogen ratio is low and the water quality is stable. The corresponding operation mode is the autotrophic denitrification mode, and its control parameters include: the liquid level of the peripheral pool, the filling ratio of the peripheral pool filler, and the influent flow of the peripheral pool. Among them, the liquid level height of the peripheral pool is directly related to the residence time of the reaction material in the peripheral pool. The appropriate liquid level can ensure that the reactants have enough time to fully contact with the catalyst or adsorbent and other media in the peripheral pool, thereby improving the reaction efficiency or adsorption effect; the peripheral pool filler filling ratio determines the amount of effective medium in the peripheral pool that can be used to treat the fluid. In the control strategy, it is necessary to determine the appropriate filler filling ratio based on the specific treatment objectives and economic costs; the hydraulic retention time of sewage in the peripheral pool can be calculated through the inlet flow and liquid level of the peripheral pool. The hydraulic retention time of the peripheral pool is one of the key factors affecting the treatment effect. Sufficient hydraulic retention time can ensure that the components in the fluid have enough opportunities to fully react with the filler and the treatment medium in the pool. The model can formulate the optimal operation strategy by analyzing the impact of hydraulic retention time on treatment effect and operating cost.

[0069] In a specific embodiment, when the carbon-nitrogen ratio (C / N) of the influent is low (for example, 3-5) and stable (within a unit time such as 1 hour, the lower limit value is less than 30%-40%), in order to improve the denitrification efficiency, the autotrophic denitrification mode is adopted. The operation process is as follows: the peripheral pool water inlet valve 6 is opened, and the effluent from the secondary sedimentation tank 1 enters the peripheral pool 16 through the peripheral pool water inlet 8 at the bottom of the packed bed. The treated effluent flows over the overflow weir 36 and is discharged through the outlet 24. The outlet valve 25 is opened and enters the disinfection pool 35, and then enters the clear water pool 26. The peripheral pool 16 is provided with air washing and water washing. Air washing is performed by opening the main air washing valve 7, the first peripheral pool air washing valve 9, the second peripheral pool air washing valve 11 and the air washing pump 4. Water washing is performed by opening the peripheral pool water washing valve 13 and the water washing pump 14. The air washing intensity is 45m 3 / ㎡·h, water washing intensity is 15m 3 / ㎡·h, the backwash cycle is once every 4 to 7 days. The water source for washing is drawn from the clean water tank 26, the water washing valve 13 of the peripheral pool is opened, and the backwash water is discharged into the secondary sedimentation tank 1 through the water washing sewage outlet 37. The bottom of the peripheral pool 16 is a supporting layer 12, and the filler is a sulfur autotrophic composite filter material with a diameter of 2-6mm, a packing density of 1.05g / cm3+0.20g / cm3, a smooth surface with holes, and a filling ratio of 70% to 85%.

[0070] Sulfur autotrophic denitrification uses sulfur autotrophic denitrifying bacteria to produce CO under anoxic or anaerobic conditions. 3 2- , HCO 3 - As the carbon source, S 0 , S 2 O 3 2- , S 2- , HS - , FeS 2 , FeS, etc. as electron donors to convert NO in sewage 3 - -N or NO 2 - -N (electron acceptor) is converted to N 2 Sulfur autotrophic filler is easily consumed with water flow, and the replenishment cycle is 3-6 months. The replenishment process is: enter the consumable replenishment pipe 21 through the filter material replenishment port 20, and finally enter the peripheral pool 16 through the autotrophic consumable replenishment pipe 22.

[0071] It should be noted that except for the filling ratio of the peripheral pool filler, the liquid level of the peripheral pool, and the water inlet flow rate of the peripheral pool, which are control parameter variables that need to be trained, other parameters such as air washing intensity, water washing intensity, and backwashing cycle are all pre-set fixed values.

[0072] 3. When the carbon-nitrogen ratio of the influent is less than the second preset threshold value and the influent flow rate and water quality changes within the preset unit time are greater than the preset change threshold range, the carbon-nitrogen ratio of the influent is extremely low and the water quality is unstable. The corresponding operation mode is the autotrophic-heterotrophic synergistic denitrification mode, and its control parameters include: the liquid level of the central pool, the liquid level of the peripheral pool, and the influent ratio of the central pool and the peripheral pool. By controlling the liquid level of the central pool and the peripheral pool, a stable reaction environment can be provided for autotrophic and heterotrophic microorganisms. The influent ratio of the central pool and the peripheral pool will directly affect the nutrient supply of the microorganisms; in the autotrophic-heterotrophic synergistic denitrification mode, different influent ratios will affect the relative number and activity of autotrophic and heterotrophic microorganisms, and then affect the synergistic denitrification effect between them. As a control parameter, the influent ratio can also help the system better adapt to changes in the influent water quality.

[0073] In the process of heterotrophic denitrification, insufficient carbon source will lead to accumulation of nitrate or nitrite, and excessive addition of carbon source will lead to effluent COD cr Excessive secondary pollution. Sulfur autotrophic denitrification produces sulfate as a byproduct, and a large amount of sulfate in the effluent will cause serious secondary pollution; and this process is an acid-producing process, which leads to a decrease in the pH of the system, affecting the activity of microorganisms and the effluent quality.

[0074] In a specific embodiment, sewage with extremely low carbon-nitrogen ratio C / N (below 3) and unstable influent water quality (such as 30%-40% exceeding the preset lower limit value within 1 hour) is operated in an autotrophic-heterotrophic synergistic denitrification mode. Autotrophic denitrification and heterotrophic denitrification work synergistically in the same system, and organic matter and sulfur substrates are used as electron donors to promote the denitrification effect of the system and jointly complete the complete denitrification process. The operation process includes: the central pool inlet valve 5 and the peripheral pool inlet valve 6 are opened at the same time, and the effluent of the secondary sedimentation tank 1 enters the central pool 17 and the peripheral pool 16 of the packed bed respectively. The treated effluent is turned over the overflow weir 36 and discharged through the outlet 24. The outlet valve 25 is opened and enters the disinfection pool 35, and then enters the clear water pool 26. The inlet flow ratio of the central pool 17 and the peripheral pool 16 is adaptively adjusted with the change of C / N. For example, when the C / N ratio is about 3, the water flow ratio of the central pool 17 to the surrounding pool 16 is 1:1. As the C / N ratio decreases, when it is 2.5, the water flow ratio of the central pool 17 to the surrounding pool 16 is 1:2. The air washing and water washing of this process are the same as the operation process of the autotrophic denitrification mode and the heterotrophic denitrification mode, and the two areas are backwashed simultaneously.

[0075] It should be noted that, except for the central pool liquid level, peripheral pool liquid level, and central pool and peripheral pool water inlet ratio, which are control parameter variables that need to be trained, other parameters such as air washing intensity, water washing intensity, and backwashing cycle are all pre-set fixed values.

[0076] S202, taking the historical operation data as sample data, and dividing the sample data into a training set and a test set;

[0077] The embodiment of the present invention ensures that the model has enough data to learn the inherent correlation and law between the inlet water quality parameters, operation mode, operation parameters and outlet water quality parameters during the operation of the denitrification device through the training set. The model can mine the optimal operation control parameters corresponding to different inlet total nitrogen content, carbon-nitrogen ratio and the like based on a large number of samples in the training set, thereby continuously optimizing its own prediction ability. For example, by analyzing the corresponding relationship between different inlet flow rates and corresponding hydraulic retention times in the training set, the model can summarize general laws for subsequent parameter predictions. The test set is equivalent to a part reserved from historical data, which is specifically used to test the final performance of the model, so that while using historical data to train the model, there is also a part of data to verify whether the model has truly learned useful knowledge, ensuring that data resources can be used for both model training and accurate model evaluation, so that the value of the data can be fully utilized. In practical applications, the sample data can be divided into a training set and a test set at a ratio of 7:3 as an example only, and is not limited to this.

[0078] S203, dividing the test set into multiple test data subsets and inputting them into multiple learners respectively, training them with the constraints of meeting the effluent water quality parameters and minimizing the cost of consumables, evaluating the trained model based on preset evaluation indicators, and determining the final learner or learner combination as the final denitrification device operation control parameter prediction model according to the evaluation value.

[0079] The embodiment of the present invention uses the effluent water quality parameters meeting the standards while minimizing the price cost of consumables as the constraint training model, which can not only learn how to optimize the water quality treatment effect, but also take into account the impact of different consumable prices on operating costs. By using preset evaluation indicators to evaluate multiple trained models, the performance of the models can be fully understood from multiple dimensions. For the goals of meeting the effluent water quality parameters and minimizing the price cost of consumables, by comparing the model evaluation results of different learners after training on different data subsets, it can be clearly seen which models perform well in predicting specific water quality parameters and which models have more advantages in cost control. This comprehensive evaluation helps to provide a more sufficient basis for actual water quality treatment decisions, such as selecting appropriate filler types and treatment processes.

[0080] In a specific embodiment, four data subsets are randomly extracted from the training set using a self-service sampling method and are respectively input into four learners, namely, support vector machine, decision tree, neural network, and Adaboost, for integration.

[0081] in:

[0082] 1. Decision tree is a model for making decisions based on tree structure. It divides the sample space by making a series of judgments on features, and finally forms leaf nodes to predict the target value. In the prediction of denitrification operation control parameters in sewage treatment, decision tree can process numerical and categorical data, and can adapt well to various types of water quality monitoring data in sewage treatment plants.

[0083] 2. Support vector machine is a classification and regression model based on statistical learning theory. Its basic idea is to find an optimal hyperplane to divide samples of different categories or predicted target values. When dealing with nonlinear problems, support vector machine can map the original data to high-dimensional space through kernel function, so as to find a linearly separable hyperplane in high-dimensional space. For the prediction of denitrification operation control parameters of sewage treatment, such as predicting the influent flow rate, the influent ratio of the central pool and the peripheral pool, the kernel function can be used to deal with the complex nonlinear relationship between the influent water quality parameters, which can have good performance even in the case of small samples and is more effective for high-dimensional data.

[0084] 3. A neural network is a computational model that mimics the structure and function of a biological neural network. It consists of multiple neurons and learns the relationship between input and output by adjusting the connection weights between neurons. In the prediction of denitrification operation control parameters in sewage treatment, such as predicting the amount of carbon source added during denitrification, neural networks can use their powerful nonlinear fitting capabilities to handle complex water quality changes and reaction processes. Deep neural networks (such as multi-layer perceptrons) can automatically extract features from data and can process a large amount of influent water quality monitoring data and other related operation parameter data.

[0085] 4. Adaboost is an iterative algorithm that combines multiple weak learners into a strong learner by changing the weight distribution of training data. In each iteration, more attention will be paid to samples that were misclassified by the previous weak learners or had large prediction errors. For the prediction of denitrification operation control parameters in sewage treatment, such as predicting the appropriate hydraulic retention time in the denitrification tank, all samples have the same weight at the beginning. After the first weak learner (such as a simple linear regression model) predicts the sample, the weights of those samples with large prediction errors will increase. Then the second weak learner (such as a decision tree model) will pay more attention to these high-weighted samples for learning, and so on. Finally, these weak learners are combined according to certain weights to obtain the final prediction model.

[0086] The above four classifiers are used as examples in the embodiment of the present invention, but the present invention is not limited thereto. Other number of other classifiers may also be used for ensemble learning.

[0087] After completing the training of the above four classifiers, the embodiment of the present invention selects R2 The prediction effect is evaluated by at least one of the three error indicators: root mean square error (RMSE) and mean relative error (MAPE). 2 The closer it is to 1, the closer RMSE is to 0, and the closer MAPE is to 0, the better the prediction effect.

[0088] When a single evaluation index is used, the evaluation values ​​are sorted in order of size, and the final coagulant prediction model is determined according to the sorting result. For example, if the mean absolute error (MAE) is selected as the evaluation index, the MAE value of each model is calculated respectively, and then the learners are sorted from small to large according to the MAE value, and the first learner after sorting is selected as the final prediction model. The sorting of the embodiment of the present invention based on a single evaluation index can clearly show the advantages and disadvantages of each learner in specific performance. For example, when predicting the carbon source dosage of the denitrification device, the mean absolute error MAE is used as the evaluation index. According to the sorting of the MAE value, it can be directly judged which learner is better in the accuracy of predicting the carbon source dosage, which is convenient for researchers and operators to quickly screen out the model that best meets the requirements. For operators of sewage treatment plants, they may pay more attention to the performance of a certain aspect of the model in certain specific scenarios. For example, in a period when cost control is more critical, they may pay more attention to evaluation indicators related to cost. Using a single evaluation index allows operators to focus directly on this key performance factor and select the learner that performs best on this factor as the final model, so as to better achieve operational goals.

[0089] When multiple evaluation indicators are used, the average of the prediction results corresponding to the learners with evaluation values ​​greater than the preset threshold is taken as the final prediction result. 2 Greater than 0.8, RMSE less than 5, and MAPE less than 10 are used as screening conditions to determine the learners with good prediction effects, and the average of the prediction results of these learners is taken as the final prediction result.

[0090] By integrating multiple evaluation indicators, the final prediction results of the embodiments of the present invention can meet certain standards in multiple important performance dimensions. For example, a good denitrification device operation control parameter prediction model must not only be able to accurately predict the operation parameters to ensure that the effluent water quality meets the standards, but also be able to reasonably consider cost factors such as the price of consumables, and maintain stable prediction performance when facing different influent water quality conditions. This comprehensive consideration approach can improve the reliability and practicality of the model in actual applications, making it more in line with the complex and changeable actual operational needs of sewage treatment plants.

[0091] When training the model, the embodiment of the present invention obtains comprehensive historical operating data of the sewage treatment plant, including information such as inlet water quality parameters, operating mode, operating parameters, effluent water quality parameters and consumables prices, so that the integrated learning model can learn the operating rules of the denitrification device under different working conditions. At the same time, the training is performed with the effluent water quality parameters meeting the standards and the consumables price cost being minimized as constraints, so that the model closely revolves around the actual operating goals of the sewage treatment plant. This training method does not simply pursue prediction accuracy, but comprehensively considers the two key factors of economic cost and water quality compliance. In the actual sewage treatment process, substandard effluent water quality will lead to environmental problems and regulatory risks, while excessively high consumables prices will increase the operational burden, which can provide sewage treatment plants with more economical and environmentally beneficial operation control parameter predictions.

[0092] In this embodiment, an operation control system of a sewage treatment denitrification device is also provided, which is used to implement the above-mentioned embodiments and preferred embodiments, and the descriptions that have been made will not be repeated. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the system described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0093] This embodiment provides an operation control system for a sewage treatment denitrification device, such as Figure 5 As shown, including:

[0094] A real-time water quality parameter monitoring module 501 is used to obtain the influent water quality monitoring data of the sewage treatment plant in real time;

[0095] The operation mode prediction module 502 is used to pre-process the influent water quality monitoring data and input it into the denitrification device operation control parameter prediction model trained based on ensemble learning to obtain the predicted operation control parameters;

[0096] The operation control module 503 is used to control the operation of the sewage treatment denitrification device based on the predicted operation control parameters.

[0097] In some optional embodiments, the process of training a denitrification device operation control parameter prediction model based on ensemble learning includes:

[0098] Obtaining historical operation data of the sewage treatment plant and preprocessing the data, the historical operation data including: inlet water quality parameters and corresponding operation modes and operation parameters, outlet water quality parameters, and consumables prices;

[0099] Use historical running data as sample data, and divide the sample data into a training set and a test set;

[0100] The test set is divided into multiple test data subsets and input into multiple learners respectively. The training is carried out with the constraints of meeting the effluent water quality parameters and minimizing the cost of consumables. The trained model is evaluated based on preset evaluation indicators, and the final learner or learner combination is determined according to the evaluation value as the final denitrification device operation control parameter prediction model.

[0101] In some optional embodiments, the influent water quality parameters include: influent COD cr , influent total nitrogen, influent flow rate; and according to influent COD cr and total nitrogen in the influent to determine the carbon-nitrogen ratio of the water quality, and determine the stability of the water quality based on the change in the influent flow rate within a preset unit time;

[0102] When the total nitrogen in the inlet water meets the preset standard, the water quality type is up-to-standard water quality, and the operation mode corresponding to the up-to-standard water quality is the bypass override mode;

[0103] When the total nitrogen in the inlet water does not meet the preset standards, the water quality carbon-nitrogen ratio, water volume and water quality stability are further judged, and the water quality is divided into different sewage types according to the numerical range of the water quality carbon-nitrogen ratio and whether the water quality is stable, and different operating modes and operating parameters are set for different sewage types.

[0104] In some optional implementations, the water quality is divided into different sewage types according to the numerical range of the water quality carbon-nitrogen ratio and whether the water quality is stable, and different operation modes and operation parameters are set for different sewage types, including:

[0105] When the total nitrogen in the influent does not meet the preset standard, the carbon-nitrogen ratio of the water quality, water volume, and water quality stability are further determined;

[0106] When the influent carbon-nitrogen ratio is greater than the first preset threshold value and the influent flow rate and water quality change within the preset unit time are less than the preset change threshold range, the influent carbon-nitrogen ratio is high and the water quality is stable, and the corresponding operation mode is the heterotrophic denitrification mode, and its control parameters include: central pool liquid level, central pool carbon source dosage, and central pool influent flow rate;

[0107] When the inlet carbon-nitrogen ratio is greater than the second preset threshold and less than the first preset threshold, and the inlet flow rate and water quality change within the preset unit time are less than the preset change threshold range, the inlet carbon-nitrogen ratio is low and the water quality is stable, and the corresponding operation mode is the autotrophic denitrification mode, and its control parameters include: peripheral pool liquid level, peripheral pool filler filling ratio, and peripheral pool inlet flow rate;

[0108] When the inlet carbon-nitrogen ratio is less than the second preset threshold value and the inlet flow rate and water quality changes within the preset unit time are greater than the preset change threshold range, the inlet carbon-nitrogen ratio is extremely low and the water quality is unstable. The corresponding operation mode is the autotrophic-heterotrophic coordinated denitrification mode, and its control parameters include: central tank liquid level, peripheral tank liquid level, central tank 17 and peripheral tank inlet ratio.

[0109] In some optional embodiments, the effluent water quality parameters include: effluent COD cr , total nitrogen in effluent, effluent flow rate; the prices of consumables: including liquid carbon source price, solid carbon source price, and sulfur autotrophic composite consumables price.

[0110] In some optional embodiments, determining the final learner or learner combination as the final denitrification device operation control parameter prediction model according to the evaluation value includes:

[0111] When a single evaluation index is used, the evaluation values ​​are sorted in order of size, and the final coagulant prediction model is determined based on the sorting results;

[0112] When multiple evaluation indicators are used, the average of the prediction results corresponding to the learners with evaluation values ​​greater than the preset threshold is taken as the final prediction result.

[0113] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.

[0114] The operation control system of the sewage treatment denitrification device in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0115] The embodiment of the present invention also provides a computer device having the above Figure 5 The operation control system of the sewage treatment denitrification device shown.

[0116] See also Figure 6 , Figure 6 is a schematic diagram of the structure of a computer device provided by an optional embodiment of the present invention, such as Figure 6As shown, the computer device includes: one or more processors, memory, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components utilize different buses to communicate with each other, and can be installed on a common mainboard or installed in other ways as required. The processor can process the instructions executed in the computer device, including instructions stored in the memory or on the memory to display the graphical information of the GUI on an external input / output device (such as, a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides part of the necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 6 A processor is taken as an example.

[0117] The processor may be a central processing unit, a network processor or a combination thereof. The processor may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0118] The memory stores instructions executable by at least one processor, so that the at least one processor executes the method shown in the above embodiment.

[0119] The memory may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory may optionally include a memory remotely arranged relative to the processor, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0120] The memory may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, hard disk or solid state drive; the memory may also include a combination of the above types of memory.

[0121] The computer device also includes a communication interface, which is used for the computer device to communicate with other devices or a communication network.

[0122] The embodiment of the present invention also provides a computer-readable storage medium. The method according to the embodiment of the present invention can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium through a network download, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state hard disk, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor central control system or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor or hardware, the method shown in the above embodiment is implemented.

[0123] A part of the present invention may be applied as a computer program product, such as a computer program instruction, which, when executed by a computer, can call or provide the method and / or technical solution according to the present invention through the operation of the computer. Those skilled in the art should understand that the existence of the computer program instruction in a computer-readable medium includes, but is not limited to, a source file, an executable file, an installation package file, etc., and accordingly, the way in which the computer program instruction is executed by the computer includes, but is not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer-readable medium may be any available computer-readable storage medium or communication medium accessible to the computer.

[0124] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A method for controlling the operation of a sewage treatment denitrification device, characterized in that: include: Obtain real-time monitoring data on the influent water quality of sewage treatment plants; After preprocessing the influent water quality monitoring data, the data is input into the denitrification device operation control parameter prediction model trained based on ensemble learning to obtain the predicted operation control parameters; The operation of the sewage treatment denitrification device is controlled based on the predicted operation control parameters.

2. The method according to claim 1, characterized in that The process of training the denitrification device operation control parameter prediction model based on ensemble learning includes: Obtaining historical operation data of the sewage treatment plant and preprocessing the data, the historical operation data including: inlet water quality parameters and corresponding operation modes and operation parameters, outlet water quality parameters, and consumables prices; Use historical running data as sample data, and divide the sample data into a training set and a test set; The test set is divided into multiple test data subsets and input into multiple learners respectively. The training is carried out with the constraints of meeting the effluent water quality parameters and minimizing the cost of consumables. The trained model is evaluated based on preset evaluation indicators, and the final learner or learner combination is determined according to the evaluation value as the final denitrification device operation control parameter prediction model.

3. The method according to claim 2, characterized in that The influent water quality parameters include: influent COD cr , influent total nitrogen, influent flow rate; and according to influent COD cr Determine the carbon-nitrogen ratio of water quality based on the total nitrogen in the influent, and determine the water volume and water quality stability based on the changes in the influent flow rate and influent water quality within a preset unit time; When the total nitrogen in the inlet water meets the preset standard, the water quality type is up-to-standard water quality, and the operation mode corresponding to the up-to-standard water quality is the bypass override mode; When the total nitrogen in the inlet water does not meet the preset standards, the water quality carbon-nitrogen ratio and water quality stability are further judged, and the water quality is divided into different sewage types according to the numerical range of the water quality carbon-nitrogen ratio, water volume, and whether the water quality is stable, and different operating modes and operating parameters are set for different sewage types.

4. The method according to claim 3, characterized in that The water quality is divided into different sewage types according to the numerical range of the carbon-nitrogen ratio of the water quality, the water volume, and whether the water quality is stable, and different operation modes and operation parameters are set for different sewage types, including: When the total nitrogen in the influent does not meet the preset standard, the carbon-nitrogen ratio and water quality stability of the water quality are further judged; When the influent carbon-nitrogen ratio is greater than the first preset threshold value and the influent flow rate and water quality change within the preset unit time are less than the preset change threshold range, the influent carbon-nitrogen ratio is high and the water quality is stable, and the corresponding operation mode is the heterotrophic denitrification mode, and its control parameters include: central pool liquid level, central pool carbon source dosage, and central pool influent flow rate; When the inlet carbon-nitrogen ratio is greater than the second preset threshold and less than the first preset threshold, and the inlet flow rate and water quality change within the preset unit time are less than the preset change threshold range, the inlet carbon-nitrogen ratio is low and the water quality is stable, and the corresponding operation mode is the autotrophic denitrification mode, and its control parameters include: peripheral pool liquid level, peripheral pool filler filling ratio, and peripheral pool inlet flow rate; When the inlet carbon-nitrogen ratio is less than the second preset threshold value and the inlet flow rate and water quality changes within the preset unit time are greater than the preset change threshold range, the inlet carbon-nitrogen ratio is extremely low and the water quality is unstable. The corresponding operation mode is the autotrophic-heterotrophic coordinated denitrification mode, and its control parameters include: central tank liquid level, peripheral tank liquid level, and central tank and peripheral tank inlet ratio.

5. The method according to claim 2, characterized in that: The effluent water quality parameters include: effluent COD cr , total nitrogen in effluent, effluent flow rate; the prices of consumables: including liquid carbon source price, solid carbon source price, and sulfur autotrophic composite consumables price.

6. The method according to claim 1, characterized in that The method of determining the final learner or learner combination as the final denitrification device operation control parameter prediction model according to the evaluation value includes: When a single evaluation index is used, the evaluation values ​​are sorted in order of size, and the final coagulant prediction model is determined based on the sorting results; When multiple evaluation indicators are used, the average of the prediction results corresponding to the learners with evaluation values ​​greater than the preset threshold is taken as the final prediction result.

7. An operation control system for a sewage treatment denitrification device, characterized in that: include: Real-time water quality parameter monitoring module, used to obtain real-time influent water quality monitoring data of sewage treatment plants; The operation mode prediction module is used to pre-process the influent water quality monitoring data and input it into the denitrification device operation control parameter prediction model trained based on ensemble learning to obtain the predicted operation control parameters; An operation control module is used to control the operation of the sewage treatment denitrification device based on the predicted operation control parameters.

8. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the operation control method of the sewage treatment denitrification device according to any one of claims 1 to 6 by executing the computer instructions.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the operation control method of the sewage treatment denitrification device according to any one of claims 1 to 6.

10. A computer program product, characterized in that It comprises computer instructions, and the computer instructions are used to enable a computer to execute the operation control method of the sewage treatment denitrification device according to any one of claims 1-6.

Images