Sewage treatment device with functions of carbon source slow release and region-adjustable AOA sludge full-automatic backflow

Abstract

The sewage treatment device comprises a full-automatic control system, a carbon source slow release mechanism, a sludge return system, a sensing detection system, and an adjusting tank, a reaction tank and a sedimentation tank which are connected in sequence, the full-automatic control system comprises a data acquisition module, a soft sensor prediction module and a control module group; the soft sensor prediction module is used for outputting BOD, TN and TP predicted values based on the sewage flow and the sewage parameters acquired by the data acquisition module; according to the region-adjustable accurate switching method, anaerobic / aerobic partitions are dynamically adjusted according to the C / N ratio. According to the invention, COD, TN predicted values, sludge activity and temperature are incorporated into carbon nitrogen ratio calculation, on-demand supply and slow release of a carbon source are realized in combination with a porous ceramic slow release device, the problems of low precision, sudden increase of local concentration and index monitoring lagging of traditional carbon source feeding are solved, and the treatment efficiency, stability and cost effectiveness are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to sewage treatment device technical field, especially to a kind of carbon source slow-release and region adjustable AOA sludge full-automatic reflux sewage treatment device. BACKGROUND

[0002] Anaerobic-Oxic-Anoxic (AOA) process as one of the core technologies of biological nitrogen and phosphorus removal, with its simple process, high nitrogen removal efficiency, relatively low energy consumption advantage, has been widely used in urban domestic sewage, industrial wastewater treatment field.The process releases phosphorus in anaerobic zone, absorbs phosphorus and nitrification in aerobic zone, and denitrification in anoxic zone, to realize the simultaneous removal of nitrogen, phosphorus and organic matter in sewage.

[0003] Although AOA process is widely used, but the existing device and control method still has the following core defects: 1, carbon source adding precision is insufficient, such as patent CN202510732917.9 only relies on COD and TN measured value, without considering sludge activity, temperature and other parameters;2, sludge reflux control is extensive, mostly using fixed reflux ratio, which cannot adapt to sludge activity change (such as SVI abnormality cannot be adjusted in time);3, key water quality monitoring lags behind, BOD, TN and other parameters rely on offline detection, lagging behind in regulation and control;4, sludge activity monitoring relies on manual sampling, which cannot provide real-time feedback;Multi-parameter coordinated control is missing;5, process partition is fixed, which cannot be adjusted dynamically according to water quality.The present application integrates five functions of predicted carbon-nitrogen ratio, carbon source slow-release, sludge activity rapid monitoring, SVI precise reflux and switching zone switching, to realize full-automatic, precision and intelligent sewage treatment. SUMMARY

[0004] The present application aims to overcome the deficiencies of the prior art, and provides a kind of carbon source slow-release and region adjustable AOA sludge full-automatic reflux sewage treatment device.

[0005] The purpose of the present application is achieved by the following technical solutions: The carbon source slow-release and region adjustable AOA sludge full-automatic reflux sewage treatment device comprises a full-automatic control system, a carbon source slow-release mechanism, a sludge reflux system, a sensing and detection system, and a regulating tank, a reaction tank and a sedimentation tank connected in sequence. The reaction tank comprises an anaerobic zone, a switching zone, an aerobic zone and an anoxic zone arranged in sequence, wherein the anaerobic zone and the anoxic zone are provided with stirrers, and the aerobic zone is provided with an aeration disc;The switching zone is isolated from the anaerobic zone and the aerobic zone by a liftable partition. The carbon source slow-release mechanism is used to adjust the carbon source concentration in the anoxic zone. The sludge return system comprises a first sludge return pump connected with the anaerobic zone, a second sludge return pump connected with the anoxic zone, and an auxiliary return pump connected with the switching zone; The sludge return system accurately controls the return amount based on the SVI value, and realizes adaptive regulation of sludge activity. The sensing and detecting system is used for collecting the sewage flow and parameters in the adjusting tank, the reaction tank and the sedimentation tank, and comprises a sludge activity sensor, which quickly monitors SV5 and calculates SVI through laser scanning. The laser scanning quickly monitors the sludge activity (SV n ) of the device and method: the device and method for quickly monitoring the sludge activity (SV n ) of the device and method, the device comprises a graduated transparent sedimentation dish, a laser scanner and an operator, and is further provided with a water inlet and a water inlet valve of the sedimentation dish, an overflow port, a drainage port and a drainage valve; during operation, water is injected through the water inlet and is controlled through the overflow port, the laser scanner emits laser scanning of the water layer (height L1) and the sludge layer (height L2) in the sedimentation dish, and the operator quickly operates through the formula SV n =(l1 / (l1+l2))×100 (usually n=30, and n=5 in the present application) to obtain the sludge activity, realizes efficient and accurate monitoring, and is suitable for the field of sludge property detection.

[0006] The full-automatic control system is a combination of a machine learning processing unit and a PLC controller, and comprises a data acquisition module, a soft sensor prediction module and a control module group. The data acquisition module is used for real-time acquisition of the sewage flow and parameters of the sensing and detecting system. The soft sensor prediction module is used for outputting the predicted values of BOD, TN and TP through the machine learning processing unit based on the sewage flow and parameters acquired by the data module, and calculating the carbon-nitrogen ratio (C / N=BOD / TN) and the biodegradability (B / C=BOD / COD). The control module group is used for controlling the carbon source slow-release mechanism, the sludge return system, the stirrer and the aeration disc based on the predicted values of BOD, TN and TP and the calculated parameters to realize carbon source supply, sludge return and aeration stirring control.

[0007] The switching zone dynamically switches process modes based on the C / N ratio via a liftable baffle, achieving precise zoning adjustment. The dynamic adjustment method for the switching zone (A / O) is as follows: First, monitor and calculate the predicted influent C / N ratio. If C / N ≥ 5 (sufficient carbon source), the A / O switching zone is set to the aerobic zone, aeration is turned on, and stirring is turned off to enhance nitrogen removal. The system flow is anaerobic → aerobic → aerobic → anoxic (AOOA). If C / N < 3.8 (scarce carbon source), the A / O switching zone is set to the anaerobic zone, aeration is turned off, and stirring is turned on to enhance phosphorus removal. The system flow is anaerobic → anaerobic → aerobic → anoxic (AAOA), achieving precise nitrogen and phosphorus removal under carbon source adaptation.

[0008] Furthermore, the carbon source slow-release mechanism includes a carbon source storage tank, a metering pump, and a slow-release device. The carbon source storage tank is connected to the slow-release device via a metering supply pump, and the slow-release device is located at the bottom of the anoxic zone.

[0009] Furthermore, the sustained-release device includes a porous carbon source carrier and a sustained-release pellet disposed within the porous carbon source carrier, the sustained-release pellet being made from a natural carbon source.

[0010] Furthermore, the sensing and detection system includes pH sensors are installed in the anaerobic, aerobic, and anoxic zones to monitor the acidity and alkalinity of the reaction tank. ORP sensors are installed in anaerobic and anoxic zones to monitor oxidation-reduction potentials. DO sensors are installed in anaerobic and anoxic zones to monitor dissolved oxygen concentration. Ammonia nitrogen sensors are installed at the outlet of the aerobic zone and the outlet of the sedimentation tank to monitor ammonia nitrogen concentration. COD sensors are installed at the outlets of the equalization tank and the sedimentation tank to monitor the concentration of chemical oxygen demand. MLSS sensors are installed in the anaerobic, aerobic, and anoxic zones to monitor sludge concentration. A sludge activity sensor, installed in the aerobic zone, is used to monitor microbial activity; A temperature sensor is installed in the aerobic zone to monitor the reaction temperature in the aerobic zone; A flow meter is installed at the outlet and return pipe of the regulating tank to regulate the flow rate at the outlet and return pipe of the regulating tank.

[0011] Furthermore, it also includes an effluent recirculation system, which includes an effluent detection tank and a third recirculation pump. The effluent detection tank is connected to the outlet of the sedimentation tank, and the third recirculation pump is connected to the effluent detection tank and the regulating tank.

[0012] Furthermore, it also includes an automatic control cabinet, which is equipped with a PLC controller, a touch screen and a data transmission module. The PLC controller is connected to the sensing and detection system, agitator and aeration disc via cables.

[0013] Furthermore, the soft sensor prediction module uses the (BP+RFR)-biLSTM neural network algorithm to construct prediction models for BOD, TN, and TP. The input parameters are pH, ORP, DO, ammonia nitrogen, COD, MLSS, temperature, sludge activity, EC, and wastewater flow rate. The output is the predicted values ​​of BOD, TN, and TP, calculated in the cloud.

[0014] Furthermore, the control module group includes a carbon source supply control module, a sludge return control module, an aeration and stirring control module, and an effluent return control module; The carbon source supply control module is used to calculate the carbon source supply amount based on the carbon-nitrogen ratio and control the working time of the metering pump according to the carbon source supply amount. The sludge return control module is used to adjust the return flow of the first return pump and the second return pump based on MLSS and ammonia nitrogen concentration. The aeration and stirring control module is used to coordinate the aeration frequency of the aeration disc and the stirring speed of the stirrer based on DO and ORP. The water return control module is used to control the opening or closing of the third return pump.

[0015] Furthermore, it also includes an intelligent decision-making and digital twin module, which is used to predict the changing trends of BOD, TN, and TP in the next 15-30 minutes.

[0016] The beneficial effects of this invention are: 1) Incorporate predicted COD, TN, sludge activity, and temperature into the carbon-to-nitrogen ratio calculation, and combine with a porous ceramic slow-release device to achieve "on-demand replenishment and slow release" of carbon sources, solving the problems of low precision in traditional carbon source addition and sudden local concentration increases.

[0017] 2) Based on the MLSS regulation of the anaerobic zone, the flow rate of the first reflux pump is controlled, and based on the ammonia nitrogen regulation of the aerobic zone, the flow rate of the second reflux pump is controlled, so as to achieve dynamic coordination of the dual reflux ratio and overcome the limitation that the fixed reflux ratio cannot adapt to water quality fluctuations.

[0018] 3) Using 10 easily measurable parameters (pH, ORP, DO, etc.) as inputs, soft sensors are constructed using BP neural networks and random forests (RFR) respectively. Based on this, a biLSTM model is established to realize real-time prediction of core water quality indicators, replacing expensive online analyzers and reducing equipment costs.

[0019] 4) Establish a linkage model of "water quality parameters-actuators" to achieve coordinated optimization of aeration, stirring, reflux and carbon source addition, and solve the problems of independent parameter control and high energy consumption of existing devices.

[0020] 5) By adjusting the functional zones through adjustable partitions and combining the three-path recirculation of R1 / R2 / R3, dynamic process adaptation can be achieved, solving the problem of poor impact resistance of fixed zones.

[0021] 6) By integrating BP feature extraction and LSTM time series prediction, a digital twin platform is constructed to achieve predictive optimization control of the wastewater treatment process, breaking through the limitations of traditional control lag.

[0022] 7) Using sludge pyrolysis carbon as a carrier, combined with 3D printing and microbial immobilization technology, the synergistic effect of carbon source step release and efficient microbial retention is achieved.

[0023] 8) Integrate multi-dimensional parameters such as process switching, reflux control, carbon source release, and digital simulation to form a collaborative optimization system of "structure-algorithm-material" to improve the overall efficiency of the system.

[0024] 9) Link the effluent detection with the recirculation system to achieve a fully automatic closed loop of "exceeding standards - recirculation - retreatment", avoiding the risk of discharge caused by delayed manual intervention.

[0025] 10) By adopting 'BP+RFR feature fusion (converting flow-concentration into load features) + bisLSTM time series prediction (16 hidden layer neurons)', the prediction error was reduced from 8% to 3% compared with the traditional bisLSTM model.

[0026] 11) Precisely control the sludge return flow rate through the SVI value to avoid sludge bulking or insufficient activity and improve system stability; 12) A sludge activity monitoring method based on laser scanning has a response time of less than 3 minutes, replacing traditional time-consuming detection methods; 13) The switching zone is dynamically switched in conjunction with the carbon-nitrogen ratio to enhance the system's resistance to shock loads and adapt to water quality fluctuations; 14) The five major functions work together to form a closed loop of "monitoring-prediction-control-feedback" to achieve full-process intelligence. Attached Figure Description

[0027] Figure 1 This is a schematic diagram of the overall structure of the wastewater treatment device with carbon source slow release and regionally adjustable AOA sludge fully automatic return in an embodiment of the present invention. Figure 2 Diagram of (BP+RFR)-biLSTM network model; Figure 3 Flowchart of dynamic adjustment of the switching area (A / O); Figure 4Methods and principles of rapid monitoring of sludge activity (SVI) using laser scanning; In the diagram, 1. Carbon source slow-release mechanism; 2. Equalization tank; 3. Anaerobic zone; 4. Switching zone; 5. Aerobic zone; 6. Anoxic zone; 7. Agitator; 8. Aeration disc; 9. First sludge return pump; 10. Second sludge return pump; 11. Auxiliary return pump; 12. Carbon source storage tank; 13. Metering pump; 14. Slow-release device; 15. Third return pump; 16. Automatic control cabinet; 17. Sedimentation tank; 18. Transparent sedimentation dish with graduations; 19. Laser scanner; 20. Calculator; 21. Inlet valve; 22. Overflow port; 23. Drain valve; L1. Water level in sedimentation dish; L2. Sludge level in sedimentation dish. Detailed Implementation

[0028] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0029] See Figure 1 and Figure 2 The present invention provides a technical solution: Example 1: like Figure 1 and Figure 2 As shown, a wastewater treatment device with carbon source slow release and regional adjustable AOA sludge fully automatic return includes a fully automatic control system, a carbon source slow release mechanism 1, a sludge return system, a sensing and detection system, and an equalization tank 2, a reaction tank, and a sedimentation tank 17 connected in sequence. Equalization tank 2: Used for homogenization and equalization, with the inlet connected to the sewage pipe network and the outlet connected to the anaerobic zone 3 of the reaction tank via a booster pump.

[0030] The reaction tank includes an anaerobic zone 3, a switching zone 4, an aerobic zone 5, and an anoxic zone 6 arranged sequentially. Each zone is separated by a partition wall, and the bottom of the partition wall is provided with water passage holes to ensure continuous water flow. The anaerobic zone 3 and the anoxic zone 6 are equipped with a stirrer 7, and the aerobic zone 5 is equipped with an aeration disc 8. The switching zone 4 is isolated from the anaerobic zone 3 and the aerobic zone 5 by a liftable partition. The height of the liftable partition is controlled by a hydraulic device to adjust the volume distribution with the anaerobic zone 3 and the aerobic zone 5. The bottom of the switching zone 4 is equipped with both a submersible stirrer 7 and an aeration disc 8. The aeration disc 8 is connected to a Roots blower through a branch pipe, and an electric regulating valve is installed on the branch pipe.

[0031] Dynamic switching logic for zones: When the sensor system detects that the influent C / N ratio is less than 3, the PLC controls the adjustable baffle to descend, reducing the volume of aerobic zone 5 and expanding the volume of anaerobic zone 3 (the volume of anaerobic zone 3 can be increased by up to 50%). At the same time, the aeration disc 8 in switchable zone 4 is turned off and the stirrer 7 is turned on to enhance the slow release of carbon source and phosphorus release reaction. When the influent TN is greater than 45 mg / L, the baffle rises to expand the volume of anoxic zone 6, the stirrer 7 in switchable zone 4 is turned on, aeration is turned off, and the denitrification space is increased.

[0032] The switchable zone can be dynamically switched between the anaerobic and aerobic zones according to the changes in the carbon-nitrogen ratio (C / N). The logic is as follows: the switching zone switching method is based on the dynamic adjustment of the liftable baffle according to the influent C / N ratio. When C / N < 3.8, the anaerobic zone is expanded to enhance the slow release of carbon source and phosphorus release reaction. When C / N > 4.2, the aerobic zone is expanded to increase the space for nitrification reaction.

[0033] Sedimentation tank 17: adopts a radial flow structure. The inlet is connected to the outlet of the anoxic zone 6 of the AOA reaction tank. A sludge hopper is installed at the bottom. The sludge hopper is connected to the first sludge return pump, the second sludge return pump and the excess sludge pump through pipes.

[0034] It includes a first sludge return pump connected to the anaerobic zone, a second sludge return pump connected to the anoxic zone, and an auxiliary return pump connected to the switching zone. The first sludge return pump is started or stopped according to the calculated and predicted total phosphorus value (TP) and the sludge concentration (MLSS) in the anaerobic zone to enhance the slow release of carbon source and phosphorus release reaction. The second return pump is started or stopped according to the calculated and predicted carbon-nitrogen ratio (C / N) and the MLSS in the anoxic zone to achieve good denitrification.

[0035] The carbon source slow-release mechanism 1 is used to adjust the carbon source concentration in the anoxic zone 6. The carbon source slow-release mechanism 1 includes a carbon source storage tank 12, a metering pump 13, and a slow-release device 14. The carbon source storage tank 12 is connected to the slow-release device 14 through a metering supply pump. The slow-release device 14 is located at the bottom of the anoxic zone 6.

[0036] The slow-release device 14 includes a porous carbon source carrier and slow-release pellets disposed within the porous carbon source carrier. The slow-release pellets are made from a natural carbon source. The natural carbon source consists of wheat straw enzymatic hydrolysate (40%) + municipal sludge pyrolysis carbon powder (20%). The artificial controlled-release framework includes PHA (polyhydroxyalkanoate, 30%), with the remainder being PVA-PAA (10%). Fused deposition modeling (FDM) technology is used, with a nozzle diameter of 0.4 mm and a printing temperature of 160~180℃ to avoid thermal decomposition of the PVA-PAA material. Slow-release pellets with a diameter of 5-8 mm and internal honeycomb pores (pore size 0.1-0.3 mm) are prepared to achieve a stepped release of the carbon source (15%-20% release in the initial 12 hours, 60%-70% release in the middle 12-48 hours, and 10%-15% release in the later 48-72 hours).

[0037] (1) The carbon source slow-release mechanism 1 is an integrated structure of "carrier-carbon source-microorganism".

[0038] 1) Porous synthetic carbon source carrier: New municipal sludge pyrolysis carbon carrier: ① Preparation process: Municipal sludge is anaerobically pyrolyzed at 500℃ for 2 hours, pulverized to a particle size of 2-5mm, and then mixed with ceramic powder at a mass ratio of 3:7. The composite carrier is prepared by pore-forming agent-assisted sintering method, with a porosity ≥80% and a specific surface area >180m² / g. Functional enhancement: Chitosan (5% by mass) is loaded onto the carrier surface, and highly efficient denitrifying bacteria (such as Pseudomonas) are immobilized through amino groups.

[0039] 2) Compound slow-release pellets: upgraded to a PHA-agricultural waste composite system: ① Component optimization: The main natural carbon source was changed from "corn cob / kitchen waste leachate" to "wheat straw enzymatic hydrolysate (40%) + municipal sludge pyrolysis charcoal powder (20%)", and the artificial controlled-release framework was added with PHA (polyhydroxy fatty acid ester, 30%), and the rest was PVA-PAA (10%). ② 3D printing molding: Fused deposition modeling (FDM) technology is used with a nozzle diameter of 0.4 mm and a printing temperature of 160~180℃ to avoid thermal decomposition of PVA-PAA material. Slow-release pellets with a diameter of 5-8 mm and containing honeycomb pores (pore size 0.1-0.3 mm) are prepared to achieve a stepwise release of carbon source (15%-20% release in the first 12 hours, 60%-70% release in the middle 12-48 hours, and 10%-15% release in the later 48-72 hours).

[0040] (2) Working principle A new "microbial immobilization synergistic mechanism" has been added: Chitosan on the surface of the carrier immobilizes denitrifying bacteria through electrostatic adsorption, and the carbon source released by the slow-release pellet directly provides nutrition for the attached microorganisms, forming a "carbon source release-microbial metabolism" microenvironment, which increases the activity of microorganisms by 30%-40% compared with the free state; at the same time, the porous structure of the carrier retains microorganisms and prevents them from being lost, which increases the concentration of functional bacteria in the reaction tank by 20%-25%.

[0041] (3) Preparation method 1) Preparation of municipal sludge pyrolysis carbon composite carrier: ① Sludge pretreatment: Municipal sludge is filtered by plate and frame filter press (moisture content reduced to 60%) and crushed to a particle size of <10mm; ② Anaerobic pyrolysis: Under nitrogen protection, pyrolyze in a muffle furnace at 500℃ for 2 hours, then cool and pulverize through an 80-mesh sieve; ③ Composite molding: Pyrolytic carbon powder, ceramic powder, and pore-forming agent (starch) are mixed in a mass ratio of 3:5:2, molded (pressure 12MPa), sintered at 1300℃ for 3h, cooled and then immersed in 2% chitosan solution for 2h, and vacuum dried for later use.

[0042] 2) Preparation of 3D-printed composite sustained-release pellets: ① Material preparation: Mix wheat straw enzymatic hydrolysate, sludge pyrolysis charcoal powder, PHA, and PVA-PAA in a certain proportion, and add deionized water to prepare a molten state (solid content 60%). ② 3D printing: Using an FDM printer, with a layer thickness of 0.2 mm and a printing speed of 50 mm / s, honeycomb structure pellets were prepared; ③ Curing treatment: After printing, place in a vacuum drying oven at 60℃ for 8 hours, and then irradiate with ultraviolet light (wavelength 254nm) for 2 hours to strengthen the crosslinking of PVA.

[0043] 3) Preparation of compound sustained-release pills ① “Pretreated corn cob + PVA-PAA” compound pellets: Step 1: Corn cob pretreatment. Crush the corn cobs to 20-40 mesh, acid-hydrolyze them with 2% dilute sulfuric acid at 80℃ for 2 hours (to destroy the cellulose crystal structure), then add cellulase (enzyme activity 100,000 U / g, added at 5% of the corn cob mass) and enzymatically hydrolyze at 50℃ for 4 hours to obtain a hydrolysate rich in small molecule sugars; spray-dry the hydrolysate (inlet air temperature 180℃, outlet air temperature 80℃) to obtain the main powder of natural carbon source.

[0044] Step 2: Slow-release pellet forming. Mix natural carbon source powder: PVA: PAA = 6:3:1 by mass, add deionized water to make a paste (solid content 40%-50%), and form pellets with a diameter of 5-8 mm using an extrusion spheronizer (speed 300-500 r / min). Dry in a vacuum drying oven at 60℃ for 12 h to allow PVA to form a film and PAA to crosslink to form a controlled-release framework.

[0045] ② “Freeze-dried powder of kitchen waste leachate + modified polyurethane” composite pellets: Step 1: Anaerobic fermentation of kitchen waste. Kitchen waste (moisture content 80%–85%) is put into an anaerobic fermentation tank and fermented at 35℃ for 20 days to produce leachate rich in volatile fatty acids. The leachate is concentrated using a nanofiltration membrane (molecular weight cutoff 200–300 Da) (concentration factor 5–10 times), and then freeze-dried (temperature -50℃, vacuum degree 10 Pa) to obtain a freeze-dried powder with natural carbon source as the main component.

[0046] Step 2: Slow-release pellet molding. The freeze-dried powder and modified polyurethane prepolymer are mixed at a mass ratio of 8:2. A catalyst (such as dibutyltin dilaurate, added at 1% of the mass of the prepolymer) is added and stirred evenly. The mixture is then poured into a spherical mold (6-10 mm in diameter) and cured at 60°C for 6 hours to form a slow-release pellet in which "freeze-dried powder is embedded in a polyurethane skeleton".

[0047] (4) Application effect ① Carbon source release and microbial activity The slow-release carbon source system allows carbon sources to be released into the anoxic zone at a stable rate (0.5–2 g / (L·h)). This avoids the sudden increase in local C / N ratio caused by the instantaneous addition of traditional liquid artificial carbon sources (such as sodium acetate solution) (which can easily exceed 10 in a short time, causing microbial metabolic disorders), and solves the problems of uneven dissolution and carrier clogging caused by the direct addition of natural carbon sources (such as straw fragments). Long-term operation monitoring shows that microbial activity (characterized by ATP content) is 15%–20% higher than that of "pure artificial carbon source addition" (the complex organic matter of natural carbon sources maintains the diversity of the microbial community); the denitrification rate is 20%–30% higher than that of "pure natural carbon source addition" (the artificially synthesized framework ensures the controllability of release).

[0048] ② Wastewater treatment efficiency In a municipal wastewater treatment scenario with a treatment capacity of 1000 m³ / d and an influent C / N ratio of 3: 1) When filling “pretreated corn cob + PVA-PAA” composite pellets (200-300 pellets, loaded on a porous ceramic carrier), the total nitrogen (TN) removal rate is stable at over 85%, and the effluent TN < 10 mg / L; the chemical oxygen demand (COD) removal rate reaches over 90%, and the effluent COD < 30 mg / L; the ammonia nitrogen (NH3-N) removal rate is > 95%, and the effluent NH3-N < 3 mg / L.

[0049] 2) When using “freeze-dried powder of kitchen waste leachate + modified polyurethane” composite pellets (180-250 pellets loaded on graphene modified polyurethane carrier), the TN removal rate can reach 88%-92% because the carbon source has stronger biodegradability, and the carbon source addition cost is reduced by 30%-40% compared with pure artificial carbon source (such as sodium acetate).

[0050] ③ System shock resistance By linking the water flow scouring rate with the stirring speed, the release rate of the slow-release carbon source can be adjusted in real time according to the fluctuation of the influent carbon load: under the condition of influent flow rate change of ±20% and COD fluctuation of ±30%, the effluent TN fluctuation amplitude is <5%, and the system's resistance to shock load is significantly better than the traditional fixed dosing method (the effluent TN fluctuation amplitude of the traditional method can reach 15% to 25%).

[0051] The sludge return system includes a first sludge return pump 9 (R1) connected to the anaerobic zone 3, a second sludge return pump 10 (R2) connected to the anoxic zone 6, and an auxiliary return pump 11 (R3) and a return ratio regulating valve connected to the switching zone 4. The return ratio regulating valve adjusts the return flow rate in real time.

[0052] The auxiliary return pump 11 connects the sludge hopper of the sedimentation tank 17 and the switching zone 4 to form a three-path return system of "R1 (anaerobic zone 3) + R2 (anoxic zone 6) + R3 (switching zone 4)". The return ratio is synergistically controlled: the return flow is dynamically allocated according to the functional positioning of MLSS in anaerobic zone 3, ammonia nitrogen in aerobic zone 5, and switching zone 4. For example, when switching zone 4 is used as an anoxic supplement zone, the return ratio of R3 is set to 30%-50%, which works in synergy with R2 to improve denitrification efficiency.

[0053] The sensing and detection system is used to collect wastewater flow rate and wastewater parameters in the equalization tank 2, reaction tank, and sedimentation tank 17; the sensing and detection system includes 12 types of sensors, as follows: pH sensors are installed in anaerobic zone 3, aerobic zone 5 and anoxic zone 6 to monitor the acidity and alkalinity of the reaction tank. ORP sensors are installed in anaerobic zone 3 and anoxic zone 6 to monitor oxidation-reduction potential; DO sensors are installed in anaerobic zone 3 and anoxic zone 6 to monitor dissolved oxygen concentration; Ammonia nitrogen sensors are installed at the outlet of aerobic zone 5 and the outlet of sedimentation tank 17 to monitor ammonia nitrogen concentration. COD sensors are installed at the outlet of equalization tank 2 and the outlet of sedimentation tank 17 to monitor the concentration of chemical oxygen demand. MLSS sensors are installed in anaerobic zone 3, aerobic zone 5 and anoxic zone 6 to monitor sludge concentration; A sludge activity sensor is installed in aerobic zone 5 to monitor microbial activity; A temperature sensor is installed in aerobic zone 5 to monitor the reaction temperature in aerobic zone 5; The flow meter is installed at the outlet of equalization tank 2 and the return pipe to regulate the flow rate at the outlet of equalization tank 2 and the return pipe. It also includes TN soft sensors and TP soft sensors installed at the outlets of anaerobic zone 3, aerobic zone 5, anoxic zone 6, and sedimentation tank 17, and BOD soft sensors installed in anaerobic zone 3 and aerobic zone 5.

[0054] Among them, the sludge activity monitoring method includes rapid determination of SV5 by laser scanning, and based on the formula SV 30 =SV5×m and SVI=(SV 30 The sludge volume index is calculated as (×10) / MLSS0, where m is an empirical value, taken as (1.2~1.5), and SV 30MLSS0 represents the sludge settling volume over 30 minutes (mL / L), and MLSS0 represents the initial mixed liquor suspended solids concentration (g / L).

[0055] Laser scanning rapid monitoring of sludge activity (SV) n The apparatus and method: The apparatus includes a graduated transparent settling dish 18, a laser scanner 19, and a calculator 20. It also includes a water inlet and inlet valve 21, an overflow outlet 22, and a drain outlet and drain valve 23. During operation, water is injected through the inlet and controlled by the overflow outlet. The laser scanner emits a laser to scan the water layer (height L1) and sludge layer (height L2) inside the settling dish. The calculator then uses formula SV... n = (l1 / (l1+l2))×100 (usually n=30, in this invention n=5) This rapid calculation yields the sludge activity, enabling efficient and accurate monitoring. It is applicable to the field of sludge characteristic detection, such as... Figure 4 .

[0056] According to calculations and predictions, when TP is greater than 0.5 mg / L or MLSS in the anaerobic zone is less than 4500 mg / L, the first sludge return pump will be started to enhance the slow release of carbon source and phosphorus release reaction; otherwise, the pump will stop working. The second sludge return pump will be started according to the calculated and predicted C / N < 4.2 or MLSS in the anoxic zone < 4200 mg / L to carry out good denitrification; otherwise, it will stop working.

[0057] It also includes an effluent return system, which includes an effluent detection tank and a third return pump 15. The effluent detection tank is connected to the outlet of the sedimentation tank 17, and the third return pump 15 is connected to the effluent detection tank and the regulating tank 2.

[0058] It also includes an automatic control cabinet 16, which is equipped with a PLC controller, a touch screen and a data transmission module. The PLC controller is connected to the sensing and detection system, the agitator 7, the aeration disc 8 and the metering pump 13 and the return pump via cables.

[0059] The fully automatic control system is centered on a PLC controller and operates through a closed-loop logic of "data acquisition - model prediction - decision control - execution feedback". Specifically, it includes the following modules: a data acquisition module, a soft sensor prediction module, a control module group, and an intelligent decision and digital twin module. The data acquisition module collects 12 types of parameters from the sensing and detection system in real time, with a sampling frequency of once per minute. The data is preprocessed (denoising and normalization) and then stored in the PLC memory.

[0060] The soft sensor prediction module is used to output predicted values ​​of BOD, TN, and TP based on the sewage flow rate and sewage parameters collected by the data module, and to calculate the carbon-nitrogen ratio C / N=BOD / TN and the biodegradability B / C=BOD / COD. The (BP+RFR)-biLSTM neural network algorithm (e.g.) is adopted. Figure 2 As shown, a prediction model for BOD, TN, and TP is constructed. (BP+RFR is used for parallel computation, simultaneously predicting BOD, TN, and TP. These predicted values ​​are then used as input to a biLSTM, which further predicts BOD, TN, and TP.) Input parameters include pH, ORP, DO, ammonia nitrogen, COD, MLSS, temperature, and sludge activity. Outputs are the predicted values ​​for BOD, TN, and TP. The model training process is as follows: 1) Collect 1000 sets of historical data (including input parameters and offline detection values ​​of BOD, TN, and TP) from a wastewater treatment plant in a certain town. 2) Data preprocessing: Outliers were removed (3σ criterion), and Min-Max normalization was used to compress the data to the [0,1] interval; 3) Construct a 3-layer BP+biLSTM neural network: 10 neurons in the input layer (ORP, pH, EC, COD, NH3-N, DO, MLSS, wastewater flow rate, temperature, sludge activity), 12 neurons in the hidden layer (activation function is Sigmoid), and 3 neurons in the output layer (BOD, TN, TP). 4) Gradient descent is used to optimize the weights, and the training error is controlled within 5%. 5) Online update: The model parameters are corrected every 24 hours using 10 newly collected sets of offline calibration data to ensure prediction accuracy.

[0061] The control module group is used to control the carbon source slow release mechanism 1, sludge return system, agitator 7 and aeration disc 8 based on the predicted values ​​of BOD, TN and TP to realize carbon source replenishment, sludge return and aeration mixing control.

[0062] Furthermore, the control module group includes a carbon source supply control module, a sludge return control module, an aeration and stirring control module, and an effluent return control module; The carbon source supply control module is used to calculate the carbon source supply amount based on the carbon-nitrogen ratio, and to control the working time of the metering pump 13 according to the carbon source supply amount, specifically: The carbon source supply is calculated based on the carbon-to-nitrogen ratio (C / N) using the following formula: 1) Calculation of carbon-nitrogen ratio: C / N = [(BOD predicted value) / (TN predicted value) + (m*COD measured value) / (n*NH3-N predicted value)] / 2, where m is the biodegradability coefficient, which is between 0.2 and 1.0 and can be calculated by a model; n is the ratio of total nitrogen to ammonium nitrogen, usually n is between 1.2 and 1.5, and the specific value is determined based on the analysis of metallic nitrogen speciation. 2) Target C / N setting: dynamically adjusted according to sludge activity and temperature. When sludge activity > 200 μg / L and temperature 20~25℃, C / N = 4~5; when sludge activity < 100 μg / L or temperature < 15℃, C / N = 5~6. 3) Calculation of carbon source replenishment: Q = (Target C / N × (Predicted TN value + Measured ammonia nitrogen value) - Measured COD value) / Carbon source concentration; 4) Working time of the replenishment pump: t = Q / flow rate of the metering replenishment pump.

[0063] The sludge return control module is used to adjust the return flow rate of the first and second return pumps based on the SVI value, MLSS, and ammonia nitrogen concentration. Specifically: The control logic for adjusting the reflux flow rates of R1 and R2 based on SVI value, MLSS, and ammonia nitrogen concentration is as follows: 1) Adjust the total sludge return flow rate based on the SVI value: when SVI < 80 mL / g, increase the sludge return flow rate; when SVI > 150 mL / g, stop sludge return. 2) R1 (anaerobic zone 3 reflux) is controlled based on MLSS: When MLSS in anaerobic zone 3 is <2.5 g / L, increase the reflux flow rate (reflux ratio 100%-120%); when MLSS is >4.0 g / L, decrease the reflux flow rate (reflux ratio 50%-80%). 3) R2 (6-fold recirculation in the anoxic zone) is adjusted based on the C / N ratio and ammonia nitrogen value: when the C / N ratio in the aerobic zone is <4.2 or the ammonia nitrogen is >5 mg / L, increase the R2 recirculation flow rate (recirculation ratio 150%-180%) to improve denitrification efficiency; when the ammonia nitrogen is <2 mg / L, decrease the R2 recirculation flow rate (recirculation ratio 100%-120%). 4) Reflux pump operating time: t1 = (target MLSS - measured MLSS) × reaction tank volume / R1 flow rate; t2 = (target ammonia nitrogen - measured ammonia nitrogen) × reaction tank volume / R2 flow rate.

[0064] The aeration and stirring control module is used to coordinate the aeration frequency of the aeration disc 8 and the stirring rate of the stirrer 7 based on DO and ORP. Specifically: 1) DO control in the aerobic zone: When DO < 2 mg / L, increase the aeration frequency to the upper limit (50 Hz); when DO > 3 mg / L, decrease the aeration frequency to the lower limit (20 Hz). 2) ORP control in the anaerobic zone: When ORP > -100 mV, increase the stirring rate to 150 r / min; when ORP < -200 mV, decrease the stirring rate to 80 r / min. 3) ORP control in the anoxic zone: When ORP > -50 mV, increase the stirring rate to 80 r / min; when ORP < -110 mV, decrease the stirring rate to 40 r / min. The effluent recirculation control module is used to control the opening or closing of the third recirculation pump 15, specifically: When the COD in the effluent testing tank is ≥30mg / L, NH3-N is ≥3mg / L, or TN is ≥10mg / L, the PLC automatically starts the third reflux pump 15 and the electric valve on the effluent reflux system to return the unqualified water to the equalization tank 2; when all indicators meet the standards, the reflux system is closed and the effluent valve is opened.

[0065] The intelligent decision-making and digital twin module runs on an industrial-grade server (configuration: Intel Xeon Silver4210 CPU, 64 GB RAM, NVIDIA T4 GPU), and communicates with the PLC controller in real time via the OPC UA protocol, ensuring a data synchronization frequency of ≤5 seconds. This module is used to predict the BOD, TN, and TP change trends and the digital mirror of the component device over the next 15-30 minutes, synchronizing the operating parameters of the physical device in real time. Specifically: This includes BP+biLSTM neural network models and digital twin systems.

[0066] 1) BP+biLSTM neural network model: Input parameters: In addition to the original 10 parameters, NO3⁻-N (a new nitrate sensor is added for monitoring, installed at the outlet of the anoxic zone, with an accuracy of ±0.05mg / L), and switchable zone functional states (0-1 variables, 0 for anaerobic / anoxic state, 1 for aerobic state). ① Model structure: The model adopts a two-layer architecture of "simultaneous feature extraction by BP and RFR + biLSTM time series prediction". BP and RFR perform feature fusion on multi-source heterogeneous data at the same time (such as converting parameters such as flow and concentration into load features). The biLSTM layer (with the hidden layer increased to 16 neurons) predicts the trend of BOD, TN and TP changes in the next 15-30 minutes. ② Training optimization: Adaptive learning rate gradient descent is used to control the training error within 3%, and the model parameters are corrected online with new data every 12 hours.

[0067] 2) Digital Twin System: Based on the Unity3D platform and integrating fluid mechanics and bioreaction kinetics models, a 1:1 digital mirror of the device is constructed. Operating parameters (such as flow rate, concentration, valve opening, pump frequency, etc.) are read in real-time from the PLC database via the OPC UA protocol, driving the digital model to run synchronously. Simultaneously, simulation-optimized control parameters (such as suggested aeration frequency and reflux ratio) are written into the PLC database for the control module group to call and execute. ① Virtual simulation and optimization: Simulate parameter adjustment schemes under different water inflow conditions (such as sudden drop in C / N ratio and sudden increase in flow rate) in a digital twin environment, output the optimal control strategy and send it to the PLC for execution; ② Cost accounting function: Real-time calculation of carbon source consumption and energy consumption costs, and optimization of control parameters to achieve the optimal dual objectives of "treatment effect - operating cost".

[0068] This invention achieves: 1. Precise carbon source addition reduces operating costs.

[0069] The dynamic carbon-nitrogen ratio is calculated by predicting TN and real-time COD using soft sensors. Combined with sludge activity and temperature, the target C / N ratio is corrected, and the carbon source addition error is controlled within ±5%. Compared with traditional manual addition, carbon source consumption is reduced by 20%-30% (based on a 1000m³ / d wastewater treatment plant, the annual carbon source cost is reduced by about 50,000-80,000 yuan). The porous synthetic carbon source carrier of the slow release device enables continuous carbon source release, avoids local sudden increases in C / N, and improves microbial activity by 15%-20%.

[0070] 2. Adaptive sludge return enhances nitrogen and phosphorus removal efficiency. The fully automatic control of dual sludge return stabilizes the MLSS in the anaerobic zone at 2.5-4.0 g / L and the ammonia nitrogen in the anoxic zone at 2-5 mg / L. Compared with the fixed return ratio mode, the TN removal rate increases from 65%-75% to 85%-90%, and the TP removal rate increases from 70%-75% to 80%-85%.

[0071] 3. Real-time monitoring of key indicators supports dynamic adjustment. The soft sensor model has a prediction response time of less than 1 minute and a prediction error of less than 8% for BOD, TN, and TP. It replaces traditional offline detection (cycle 4-24 hours), solves the problem of "regulation lag", and transforms parameter adjustment from "post-event remediation" to "pre-event prevention".

[0072] 4. Multi-parameter coordination reduces energy consumption. The coordinated regulation of DO-aeration-stirring keeps the DO in the aerobic zone stable at 2-3 mg / L, reducing the energy consumption of the Roots blower by 15%-20%; the dynamic adjustment of the stirring rate reduces the energy consumption in the anaerobic / anoxic zone by 10%-15%, and the overall energy consumption is reduced by 12%-18% compared with the existing equipment.

[0073] 5. Ensure effluent meets and stabilizes standards, guaranteeing safe discharge. The automatic detection-recirculation closed-loop system increases the compliance rate of effluent COD, NH3-N, and TN from 85%-90% to over 99%, eliminating the risk of exceeding emission standards due to untimely manual intervention, and meeting the Class A standard of the "Discharge Standard of Pollutants for Municipal Wastewater Treatment Plants" (GB18918-2002).

[0074] 6. Improved dynamic process adaptability The AO switchable zone works in synergy with the R3 return system to achieve a TN removal rate fluctuation of less than 3% under conditions of influent C / N ratio of 2-6 and flow fluctuation of ±30%, which is 80% higher than the traditional fixed zoning process (fluctuation of 15%-25%).

[0075] 7. Improved precision of intelligent control The (BP+RFR)-biLSTM model has a TN prediction error of <3% and a response time of <30 seconds, enabling early intervention through "prediction-control" and reducing carbon source waste by 10%-15% compared to traditional feedback control.

[0076] 8. Synergistic effect of carbon source and microorganism The integrated carbon source slow-release mechanism improves carbon source utilization efficiency by 40%-50% and denitrification rate to 3.5-4.0 g / (L・d), which is 30%-35% higher than the original system.

[0077] 9. Digital twins reduce costs and increase efficiency Through virtual simulation optimization, the experimental debugging cycle is shortened by 50%, saving 30,000 to 50,000 yuan in debugging costs annually.

[0078] Example 2: This embodiment is a specific application of the wastewater treatment device with carbon source slow release and regionally adjustable AOA sludge fully automatic return in Example 1: A municipal wastewater treatment plant has a treatment capacity of 1000 m³ / d. The influent consists of a mixture of domestic sewage and a small amount of industrial wastewater. The influent water quality fluctuates greatly: COD 150-300 mg / L, NH3-N 20-35 mg / L, TN 30-45 mg / L, C / N ratio 2-4. The original AOA device uses manual control of carbon source addition and fixed sludge return ratio. The effluent TN often exceeds the standard (12-18 mg / L). The annual cost of carbon source consumption is about 250,000 yuan, and the comprehensive energy consumption is about 150,000 kWh / year.

[0079] (1) Device configuration: 1. Reactor dimensions: Equalization tank (10m×5m×3m); AOA reactor: anaerobic zone (8m×5m×3m), aerobic zone (15m×5m×3m), anoxic zone (10m×5m×3m); Sedimentation tank (12m×5m×3m); 2. Actuator parameters: Booster pump (Q=50m³ / h, H=15m); R1 / R2 return pump (Q=30-80m³ / h, H=10m); Metering supply pump (Q=0.5-2m³ / h, accuracy ±1%); Roots blower (air volume 10m³ / min, air pressure 50kPa). 3. Sensor models: pH / ORP sensor (Hash HQ40d), DO sensor (Hash LDOII), ammonia nitrogen / COD sensor (Shimadzu TOC-V), MLSS sensor (Endress+Hauser Turbimax CUS51D). 4. PLC controller: Siemens S7-1200 series, equipped with 16 analog input / output modules.

[0080] 5. Switchable reaction zone equipment: submersible agitator (5.5kW power), aeration disc (2mm aperture), hydraulic lifting baffle (0-1.5m stroke); its working process diagram is as follows. Figure 3 The dynamic adjustment method for the switching zone (A / O) is as follows: First, monitor and calculate the predicted carbon-nitrogen ratio (C / N) of the influent. If C / N ≥ 5 (sufficient carbon source), then set the switching zone A / O to the aerobic zone, turn on aeration and turn off stirring to enhance nitrogen removal. The system flow is anaerobic → aerobic → aerobic → anoxic (AOOA). If C / N < 3.8 (scarce carbon source), then set the switching zone A / O to the anaerobic zone, turn off aeration and turn on stirring to enhance phosphorus removal. The system flow is anaerobic → anaerobic → aerobic → anoxic (AAOA), thus achieving precise nitrogen and phosphorus removal under carbon source adaptation.

[0081] 6. Intelligent control equipment: Nitrate sensor (Hash NitraVisII, accuracy ±0.05mg / L), digital twin terminal (industrial-grade touch screen, equipped with Unity3D simulation software); 7. 3D printing equipment: FDM 3D printer (nozzle diameter 0.4mm, printing temperature 180-220℃).

[0082] (2) Setting operating parameters 1. Carbon source control: Sodium acetate (concentration 30%) is used as the carbon source, with a target C / N ratio of 4-6, which is adjusted according to sludge activity (4 when ATP > 200 μg / L, and 6 when ATP < 100 μg / L); 2. Sludge return: R1 return ratio 50%-120% (MLSS target 3.0±0.5g / L in anaerobic zone), R2 return ratio 100%-180% (ammonia nitrogen target 3±1mg / L in aerobic zone); 3. Aeration and mixing: DO target in aerobic zone: 2-3 mg / L, aeration frequency: 20-50 Hz; ORP target in anaerobic zone: -150 ± 50 mV, mixing rate: 80-150 r / min; ORP target in anoxic zone: -80 ± 30 mV, mixing rate: 40-80 r / min. 4. Effluent recirculation: Recirculation is started when COD≥30mg / L, NH3-N≥3mg / L or TN≥10mg / L, with a recirculation flow rate of 50m³ / h.

[0083] 5. Switchable zone control: When C / N < 3, the baffle descends by 1m to switch to anaerobic state; when TN > 45mg / L, the baffle rises by 1m to switch to anoxic state. 6. BP+biLSTM model parameters: sampling frequency 1 time / 30 seconds, prediction step size 30 minutes, learning rate 0.001; Sustained-release pellet parameters: 3D printed honeycomb pore diameter 0.2mm, porosity 60%, microbial immobilization capacity 10. 8 -10 9 CFU / g vector.

[0084] (3) Operational effect Table 1 shows a comparison of the treatment effect and operating cost after the device has been running for 6 months: Table 1 Comparison of Effects Before and After Implementation The above description is merely a preferred embodiment of the present invention. It should be understood that the present invention is not limited to the forms disclosed herein and should not be construed as excluding other embodiments. It can be used in various other combinations, modifications, and environments, and can be altered within the scope of the concept described herein through the above teachings or related technologies or knowledge. Modifications and variations made by those skilled in the art that do not depart from the spirit and scope of the present invention should be within the protection scope of the appended claims.

Claims

1. A wastewater treatment device with carbon source slow release and regionally adjustable AOA sludge fully automatic return, characterized in that: It includes a fully automatic control system, a carbon source slow release mechanism, a sludge return system, an effluent return system, a sensor detection system, and an equalization tank, a reaction tank, and a sedimentation tank connected in sequence; The reaction tank includes an anaerobic zone, a switching zone, an aerobic zone and an anoxic zone arranged in sequence. The anaerobic zone and the anoxic zone are equipped with agitators, and the aerobic zone is equipped with an aeration disc. The switching zone is isolated from the anaerobic zone and the aerobic zone by a liftable partition. The carbon source slow-release mechanism is used to adjust the carbon source concentration in the hypoxic zone; The sludge return system includes a first sludge return pump connected to the anaerobic zone, a second sludge return pump connected to the anoxic zone, and an auxiliary return pump connected to the switching zone. The sludge return system controls the start / stop and return flow rate of the return pump based on the sludge volume index (SVI). When SVI < 80 mL / g, the sludge return flow rate is increased, and when SVI > 150 mL / g, the sludge return is stopped. The effluent return system connects the sedimentation tank and the equalization tank; The sensing and detection system is used to collect wastewater flow rate and wastewater parameters in the equalization tank, reaction tank, and sedimentation tank, including sensors for monitoring sludge activity, and to calculate SVI value based on SV5 and MLSS. The fully automatic control system includes a data acquisition module, a soft sensor prediction module, and a control module group; The data acquisition module is used to collect the sewage flow rate and sewage parameters of the sensor detection system in real time. The soft sensor prediction module is used to output predicted values ​​of BOD, TN, and TP based on the sewage flow rate and sewage parameters collected by the data acquisition module, and to calculate the carbon-nitrogen ratio C / N=BOD / TN and the biodegradability B / C=BOD / COD. The control module group is used to control the carbon source slow release mechanism, sludge return system, agitator and aeration disc based on the predicted values ​​of BOD, TN and TP to realize carbon source replenishment, sludge return and aeration mixing control.

2. The wastewater treatment device with carbon source slow release and regionally adjustable AOA sludge fully automatic return as described in claim 1, characterized in that: The carbon source slow-release mechanism includes a carbon source storage tank, a metering pump, and a slow-release device. The carbon source storage tank is connected to the slow-release device via a metering supply pump, and the slow-release device is located at the bottom of the anoxic zone.

3. The wastewater treatment device with carbon source slow release and regionally adjustable AOA sludge fully automatic return as described in claim 2, characterized in that: The sustained-release device includes a porous carbon source carrier and a sustained-release pellet disposed within the porous carbon source carrier. The sustained-release pellet is made from a natural carbon source, and a microbial immobilization carrier is disposed within the sustained-release pellet, with a microbial immobilization amount ≥10. 8 CFU / g.

4. The wastewater treatment device with carbon source slow release and regionally adjustable AOA sludge fully automatic return as described in claim 1, characterized in that: The sensing and detection system includes pH sensors are installed in the anaerobic, aerobic, and anoxic zones to monitor the acidity and alkalinity of the reaction tank. ORP sensors are installed in anaerobic and anoxic zones to monitor oxidation-reduction potentials. DO sensors are installed in anaerobic and anoxic zones to monitor dissolved oxygen concentration. Ammonia nitrogen sensors are installed at the outlet of the aerobic zone and the outlet of the sedimentation tank to monitor ammonia nitrogen concentration. COD sensors are installed at the outlets of the equalization tank and the sedimentation tank to monitor the concentration of chemical oxygen demand. MLSS sensors are installed in the anaerobic, aerobic, and anoxic zones to monitor sludge concentration. A sludge activity sensor, installed in the aerobic zone, is used to monitor microbial activity; A temperature sensor is installed in the aerobic zone to monitor the reaction temperature in the aerobic zone; EC sensor, installed at the outlet of the equalization tank; A flow meter is installed at the outlet and return pipe of the regulating tank to regulate the flow rate at the outlet and return pipe of the regulating tank. It also includes soft sensors predicted by models: TN soft sensor, TP soft sensor, and BOD soft sensor, which are used to calculate the carbon-nitrogen ratio and accurately control the amount of carbon source replenishment.

5. The wastewater treatment device with carbon source slow release and regionally adjustable AOA sludge fully automatic return according to claim 4, characterized in that: It also includes an effluent recirculation system, which includes an effluent detection tank and a third recirculation pump. The effluent detection tank is connected to the outlet of the sedimentation tank, and the third recirculation pump is connected to the effluent detection tank and the equalization tank.

6. The wastewater treatment device with carbon source slow release and regionally adjustable AOA sludge fully automatic return as described in claim 1, characterized in that: It also includes an automatic control cabinet, which is equipped with a PLC controller, a touch screen and a data transmission module. The PLC controller is connected to the sensing and detection system, agitator and aeration disc via cables.

7. The wastewater treatment device with carbon source slow release and regionally adjustable AOA sludge fully automatic return according to claim 1, characterized in that, The sensing and detection system includes a pH sensor, an ORP sensor, a DO sensor, an ammonia nitrogen sensor, a COD sensor, an MLSS sensor, a sludge activity sensor, and a temperature sensor. Sludge activity monitoring utilizes laser scanning to rapidly determine SV5, based on the formula SV... 30 =SV5×m and SVI=(SV 30 The sludge volume index (SV) is calculated as (×10) / MLSS, where m is an empirical value ranging from 1.2 to 1.

5. 30 MLSS0 represents the sludge settling volume over 30 minutes (mL / L), and MLSS0 represents the initial mixed liquor suspended solids concentration (g / L).

8. The wastewater treatment device with carbon source slow release and regionally adjustable AOA sludge fully automatic return according to claim 4, characterized in that: The soft sensor prediction module uses the (BP+RFR)-biLSTM neural network algorithm to construct prediction models for BOD, TN, and TP. The input parameters are pH, ORP, DO, ammonia nitrogen, COD, MLSS, temperature, sludge activity, EC, and wastewater flow rate. The output is the predicted values ​​of BOD, TN, and TP.

9. The wastewater treatment device with carbon source slow release and regionally adjustable AOA sludge fully automatic return according to claim 5, characterized in that: The control module group includes a carbon source supply control module, a sludge return control module, an aeration and stirring control module, and an effluent return control module. The carbon source supply control module is used to calculate the carbon source supply amount based on the carbon-nitrogen ratio and control the working time of the metering pump according to the carbon source supply amount. The sludge return control module is used to adjust the return flow of the first return pump and the second return pump based on MLSS and ammonia nitrogen concentration. The aeration and stirring control module is used to coordinate the aeration frequency of the aeration disc and the stirring speed of the stirrer based on DO and ORP. The water return control module is used to control the opening or closing of the third return pump.

10. The wastewater treatment device with carbon source slow release and regionally adjustable AOA sludge fully automatic return according to claim 1, characterized in that: It also includes an intelligent decision-making and digital twin module, which is used to predict the changing trends of BOD, TN, and TP in the next 15-30 minutes.

Images