Intelligently adjustable micro-ecological base reaction bed

Abstract

The invention discloses an intelligently adjustable micro-ecological base reaction bed, which belongs to the technical field of water treatment, and comprises an ecological base reaction bed, a real-time detection module, a control module, an intelligent aeration module and a fungicide release module, the ecological base reaction bed comprises a reaction bed frame, an MOBR film is attached to the surface of the reaction bed frame, and the reaction bed frame is used for synchronously realizing half-process denitrification and anaerobic ammonia oxidation reaction; the real-time detection module is fixed on the reaction bed frame and is used for monitoring data in the reaction bed and feeding back the data to the control module; the control module is used for realizing remote control, automatic adjustment, fault early warning and dynamic optimization based on real-time detection data; the intelligent aeration module is a nano aeration device and is used for accurately adjusting the aeration rate according to an instruction of the control module, so that the concentration of dissolved oxygen is maintained at 0.2-0.5 mg / L; and the microbial agent release module comprises a plurality of microbial agent slow-release adding devices, is mounted near the intelligent aeration module on the lower layer of the reaction bed frame, and is used for dynamically adding nitrifying bacteria or denitrifying microbial agents according to instructions of the control module.

Description

Technical Field

[0001] The present invention belongs to the technical field of aquaculture tail water purification, and in particular relates to an intelligently adjustable microecological reaction bed. Background Art

[0002] The micro-aerobic eco-based reactor bed is a reaction device used for aquaculture tailwater purification. It mainly uses microorganisms attached to the eco-based bed to decompose and transform pollutants in wastewater in a micro-aerobic environment.

[0003] With the demand for green development in the aquaculture industry, aquaculture tailwater, as a type of slightly polluted water, can have an adverse impact on the surrounding water environment during concentrated discharge periods. Traditional activated sludge methods have many limitations in treating slightly polluted water, such as unsatisfactory treatment results, lack of targeted treatment, and high treatment costs. The micro-aerobic ecological-based reactor bed, a device suitable for aquaculture tailwater treatment, uses a multi-layer sandwich design of biological carriers to specifically remove total nitrogen and organic matter from the tailwater, achieving unidirectional transfer in the upper layer and heterogeneous transfer in the lower layer, thereby enhancing the removal of total nitrogen in the tailwater.

[0004] However, the existing microecological reaction beds have the following problems:

[0005] 1. No bacterial release occurs

[0006] Poor microbial community stability: In an eco-based reactor bed, the microbial community is crucial for tailwater purification. Without the release of microbial agents, the microbial population relies primarily on natural attachment and slow reproduction. This makes it difficult to effectively control the species and abundance of the microbial community. With seasonal changes or changes in water quality, the microorganisms within the reactor bed are unable to adapt promptly to new pollution loads. During the tailwater discharge period, the microbial community, lacking sufficient bacterial species, is unable to quickly adjust its structure, resulting in reduced purification efficiency.

[0007] Long startup time: When a new reactor bed is put into use or the microbial community in the reactor bed is severely damaged (such as due to sudden changes in water quality or disinfection), without the release of bacterial agents, it is necessary to wait for the microorganisms to naturally attach and reproduce on the ecological matrix to re-establish an effective microbial community. This process can last for several months, which is very disadvantageous in actual aquaculture tailwater purification applications, because aquaculture tailwater must be treated in a timely manner to avoid contamination of the surrounding water environment.

[0008] Insufficient functional microorganisms: For example, nitrifying bacteria and denitrifying bacteria. The microbial communities established by natural reproduction are relatively numerous, but lack dominant species. Nitrifying bacteria and denitrifying bacteria are difficult to meet the needs of water denitrification.

[0009] 2. No intelligent aeration exists

[0010] High energy consumption and low efficiency: Aeration cannot be precisely tailored to the actual tailwater quality (such as dissolved oxygen content and pollutant concentration) and the microbial oxygen demand. To maintain a certain purification effect, excessive aeration is often used, which not only wastes energy but can also negatively impact microorganisms. For example, excessive aeration inhibits denitrification, resulting in poor denitrification performance.

[0011] Unable to adapt to dynamically changing water quality conditions: The water quality of aquaculture tailwater fluctuates dynamically, affected by factors such as the culture stage, feeding schedule, and weather. Without intelligent aeration, it is impossible to adapt to these changes in water quality. For example, if the organic matter concentration or ammonia nitrogen concentration in the tailwater suddenly increases, the aeration rate needs to be increased to consume the excess organic matter and ammonia nitrogen. When the total nitrate nitrogen concentration in the tailwater increases, the aeration rate needs to be reduced to create conditions for denitrification. Summary of the Invention

[0012] The technical problem to be solved by the present invention is to overcome the shortcomings of the above-mentioned prior art and provide a microecological reaction bed that can be intelligently adjusted.

[0013] The technical solution adopted to solve the above technical problems is: an intelligently adjustable microecological reaction bed, including an ecological reaction bed, a real-time detection module, a control module, an intelligent aeration module and a bacterial agent release module;

[0014] The ecological-based reactor bed adopts a three-dimensional multi-layer sandwich structure, including a reactor bed frame with a MOBR membrane attached to the surface, which is used to simultaneously achieve half-way denitrification and anaerobic ammonium oxidation reactions;

[0015] The real-time detection module is fixed on the reaction bed frame and includes a dissolved oxygen sensor, a flow sensor, a temperature sensor, a liquid level sensor, a NO3 - -N detector and NH4 + -N detector, used to monitor dissolved oxygen, flow, temperature, liquid level and nitrogen pollutant concentration data in the reaction bed and feed back to the control module;

[0016] The control module includes a programmable logic controller, a data acquisition card, an industrial control computer, an MPC model prediction algorithm module and a BP neural network model, which is used to achieve remote control, automatic adjustment, fault warning and dynamic optimization based on real-time detection data;

[0017] The intelligent aeration module is a nano-aeration device fixed on the reaction bed frame, including nano-aeration equipment and aeration control system, which is used to accurately adjust the aeration volume according to the instructions of the control module to maintain the dissolved oxygen concentration at 0.2-0.5 mg / L;

[0018] The bacterial agent release module includes several bacterial agent slow-release dosing devices, which are installed near the intelligent aeration module in the lower layer of the reaction bed frame and are used to dynamically add nitrifying bacteria or denitrifying bacteria agents according to the instructions of the control module.

[0019] Preferably, the dissolved oxygen sensor comprises at least a first dissolved oxygen detector arranged at the bottom of the reaction bed and a second dissolved oxygen detector arranged in the middle, for detecting the dissolved oxygen content at different points in real time.

[0020] Preferably, the control module calculates the oxygen transfer rate and the oxygen supply of the aeration equipment based on the following formula:

[0021] Oxygen transfer rate:

[0022] dC / dt=K L a(C s -C),

[0023] Where dC / dt is the oxygen transfer rate (unit: mg / (L·h)), which represents the change in dissolved oxygen concentration in water per unit time; K L a is the total mass transfer coefficient (unit: h -1 ), which comprehensively considers the influence of liquid film mass transfer coefficient and gas-liquid contact area on oxygen transfer, and its size is related to the type of aeration equipment and the nature of sewage; C s It is the saturated dissolved oxygen concentration in the water at the current temperature and pressure (unit: mg / L). This value will change with the water temperature and air pressure environmental factors; C is the actual dissolved oxygen concentration in the water (unit: mg / L)

[0024] Aeration equipment oxygen supply:

[0025] Oxygen supply of aeration equipment G s (Unit: kgO2 / h) The calculation formula is:

[0026]

[0027] Where T is the sewage temperature (unit: °C). Because temperature affects the solubility of oxygen in water and the metabolic activity of microorganisms, a temperature correction factor of 1.024 needs to be considered. T-20 , E A It is the oxygen utilization rate of the aeration equipment. Different types of aeration equipment have different oxygen utilization rates. V is the effective volume of the reaction bed (unit: m 3 );

[0028] The specific value of oxygen supply can be calculated through sewage temperature, oxygen utilization rate of aeration equipment, oxygen transfer rate and effective volume of aeration tank, making the control more accurate.

[0029] More preferably, the MPC model prediction algorithm of the control module is constructed based on the following state space model:

[0030] The state variables x(k) include dissolved oxygen concentration and microbial activity;

[0031] The controlled variables u(k) include the aeration flow rate and aeration time of the aeration equipment;

[0032] The output variable y(k) is the measured dissolved oxygen concentration;

[0033] Equation of state:

[0034] x(k+1)=Ax(k)+Bu(k)+w(k),

[0035] Where Ax(k) is the state transition matrix, which describes the natural evolution of the system state when there is no control input; Bu(k) is the control input matrix, which shows the impact of the control input on the state variables; w(k) is the process noise, which is used to account for the unmodeled dynamics and disturbance factors present in the actual system.

[0036] Output equation:

[0037] y(k)=Cx(k)+Du(k)+v(k),

[0038] Where Cx(k) is the output matrix, which maps state variables to output variables; Du(k) is the feedforward matrix; v(k) is the measurement noise, which takes into account the error in the measurement process;

[0039] The objective function of the MPC model is:

[0040]

[0041] where y ref (k+i) is the reference output at time k+i (the set dissolved oxygen concentration target value), and λ is the control weight, which is used to balance the control performance and control energy consumption;

[0042] Constraints:

[0043] The aeration flow rate u(k) has upper and lower limits, namely u min ≤u(k)≤u max ;

[0044] Dissolved oxygen concentration 0.2≤y(k)≤0.5.

[0045] More preferably, the MPC model solves the optimization problem, and the specific steps include:

[0046] At each sampling moment, a finite-time domain optimization problem is formed based on the prediction model, objective function, and constraints. This optimization problem is solved by the optimization algorithm to obtain the optimal control sequence:

[0047] [u(k|k),u(k+1|k),…,u(k+N c -1|k)]

[0048] The first control variable u(k|k) of the optimized control sequence is applied to the actual system. Then, at the next sampling time k+1, the system state is remeasured, the prediction model is updated, and the optimization problem is solved again. This process is repeated to achieve dynamic control of the intelligent aeration system.

[0049] The specific formula of the optimization algorithm is:

[0050] set up:

[0051] x=[u(k|k),u(k+1|k),…,u(k+N c -1|k)] T

[0052] Rewrite the constraint as g1(x)=u min -u(k|k)≤0,g2(x)=u(k|k)-u max ≤0;

[0053] Introducing the logarithmic barrier function Where μ>0 is the barrier parameter. When x approaches the inequality constraint boundary (i.e. g i (x)→0), ln(-g i (x))→-∞, so that the optimization process tends to proceed within the feasible region and avoids crossing the boundary;

[0054] Step 1: Solve the obstacle problem

[0055] For a given μ, solve the unconstrained optimization problem minB(x,μ), which usually requires calculating the gradient and the Hessian matrix

[0056] The gradient calculation formula is:

[0057]

[0058] The Hessian matrix calculation formula is:

[0059]

[0060] Then use Newton's method to solve this unconstrained optimization problem. The iterative formula of Newton's method is:

[0061]

[0062] where x k is the point of the kth iteration;

[0063] Step 2: Update barrier parameters

[0064] As the iteration proceeds, μ gradually decreases, making the optimization process gradually approach the solution of the original constrained optimization problem. Common update strategies are

[0065] μ k+1 =θμ k , where θ∈(0,1);

[0066] Step 3: Convergence judgment

[0067] Check whether the convergence criterion is met. When , the algorithm is considered to have converged, and x is regarded as an approximate optimal solution.

[0068] Further preferably, the input layer of the BP neural network model is the dissolved oxygen, NO3 - -N、NH4 + -N data, the output layer is the aeration volume of the intelligent aeration module and the dosage of the bacterial agent release module, which are used to coordinately adjust the aeration intensity and bacterial agent release dosage.

[0069] More preferably, the microbial agent slow-release dosing device comprises a driving motor, an inner tube and an outer tube;

[0070] The inner tube is filled with slow-release bacteria and has a plurality of holes on one side; the outer tube is provided with an opening on one side and is sleeved on the outer side of the inner tube;

[0071] The drive motor is connected to the inner tube and controls the rotation angle of the inner tube (0-180°) to adjust the exposure area of ​​the inner tube hole and the outer tube opening, thereby realizing the start and stop of bacterial release and dosage control.

[0072] The drive motor is connected to the control module signal, according to NO3 - -N and NH4 + -N detection concentration triggers the dosing instruction.

[0073] Still further preferably, the bacterial agent slow-release dosing device comprises an independent nitrifying bacteria slow-release dosing device and a denitrifying bacteria slow-release dosing device, both of which have the same structure;

[0074] The rotation angle of the nitrifying bacteria slow-release dosing device is + The corresponding relationship of -N concentration is:

[0075] NH4 + -N concentration < 1mg / L, the adjustment angle is 0° (closed),

[0076] NH4 + -N concentration is 1-2mg / L, the adjustment angle is 45°,

[0077] NH4 + -N concentration is 2-3mg / L, the adjustment angle is 90°,

[0078] NH4 + -N concentration>3mg / L, the adjustment angle is 180° (maximum release);

[0079] According to NO3 - -N detector detected NO3 - -N concentration regulates the opening and closing angle of the denitrifying bacteria device as follows:

[0080] NO3 - -N concentration < 1mg / L, the adjustment angle is 0° (closed),

[0081] NO3 - -N concentration is 1-2mg / L, the adjustment angle is 45°,

[0082] NO3 - -N concentration is 2-3mg / L, the adjustment angle is 90°,

[0083] NO3 - -N concentration>3mg / L, the adjustment angle is 180° (maximum release).

[0084] Furthermore, the collaborative control logic of the intelligent aeration module and the bacterial agent release module includes:

[0085] When NO3 - When -N>4mg / L and DO<0.5mg / L, the aeration rate was reduced by 30% through the MPC model, and the denitrifying bacteria slow-release dosing device was adjusted to 180°;

[0086] When the dissolved oxygen (DO9) at the bottom is less than 2 mg / L, nano aeration is started;

[0087] When the dissolved oxygen at the bottom (DO9) is greater than 4 mg / L or the dissolved oxygen in the middle (DO8) is greater than 0.5 mg / L, aeration is terminated.

[0088] The beneficial effects of the present invention are as follows:

[0089] 1. This invention dynamically adds nitrifying and denitrifying agents through a microbial agent release module, resolving the problem of traditional reactor beds relying on naturally attached microorganisms. When water quality changes or the microbial community is impacted, functional bacterial species can be promptly replenished, allowing the microbial community structure to quickly adapt to changes in pollution loads and ensuring stable purification efficiency. This invention also uses a microbial agent release module to actively add dominant bacterial species, accelerating the establishment of an ecologically based microbial community and meeting the needs of timely aquaculture tailwater treatment.

[0090] 2. The present invention sets up independent slow-release dosing devices for nitrifying bacteria and denitrifying bacteria, which can be used according to the NH4 + -N and NO3 - -N concentration accurately adjusts the release dose. When NH4 + - When the N concentration is greater than 3 mg / L, the nitrifying bacteria dosing device is adjusted to 180° for maximum release to increase the nitrification reaction rate; NO3 - When the -N concentration is greater than 3 mg / L, the denitrifying bacteria agent is released in maximum amount, which strengthens the denitrification process and solves the problem of insufficient functional microorganisms in traditional reaction beds.

[0091] 3. The intelligent aeration module of this invention utilizes an MPC model prediction algorithm in conjunction with a BP neural network to precisely adjust aeration volume based on real-time data such as dissolved oxygen and nitrogen pollutant concentrations. Maintaining dissolved oxygen concentrations at a micro-aerobic environment of 0.2-0.5 mg / L reduces energy consumption by 30%-50% compared to traditional excessive aeration. This also prevents excessive aeration from inhibiting denitrification, thereby improving denitrification efficiency.

[0092] 4. The control module of this invention combines an MPC model with a BP neural network to respond in real time to dynamic changes in aquaculture tailwater. When organic matter concentration increases, the aeration rate and microbial dosage are automatically adjusted. When nitrogen pollutant concentration fluctuates, microbial release and aeration are coordinated to ensure stable treatment results, addressing the poor adaptability of traditional reactor beds.

[0093] 5. The three-dimensional multi-layer sandwich structure of the eco-based reactor bed, combined with the MOBR membrane, simultaneously achieves mid-stage denitrification and anaerobic ammonium oxidation reactions, while also improving total nitrogen and organic matter removal rates. Dissolved oxygen sensors positioned at the bottom and center of the reactor bed regulate the dissolved oxygen gradient within the reactor bed, optimizing the microbial growth environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0094] Figure 1 It is a system flow chart of the present invention;

[0095] Figure 2 This is the overall structural diagram of the eco-based reactor bed of the present invention;

[0096] Figure 3 It is a structural diagram of the bacteria release device of the present invention.

[0097] Reference numerals: 1. nano-aeration device; 2. slow-release dosing device for nitrifying bacteria; 3. slow-release dosing device for denitrifying bacteria; 4. reaction bed frame; 5. MOBR membrane; 6. NO3 - -N detector; 7, NH4 + -N detector; 8. First dissolved oxygen detector; 9. Second dissolved oxygen detector; 10. Drive motor; 11. Inner tube; 12. Outer tube; 13. Slow-release bacteria. DETAILED DESCRIPTION

[0098] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0099] like Figures 1 to 3 As shown, an intelligently adjustable microecological-based reaction bed of this embodiment includes an ecological-based reaction bed, a real-time detection module, a control module, an intelligent aeration module and a bacterial agent release module.

[0100] The eco-based reactor bed adopts a three-dimensional multi-layer sandwich structure. The reactor bed frame 4 is made of stainless steel and has a rectangular frame structure with dimensions of 5m long, 3m wide, and 4m high. The MOBR membrane 5 is evenly attached to the frame surface. The membrane material is polyvinylidene fluoride (PVDF) with a pore size of 0.1-0.2μm. It is used to provide a carrier for microbial attachment and achieve gas-liquid separation, and simultaneously carry out half-way denitrification and anaerobic ammonium oxidation reactions. The interior of the reactor bed is divided into three layers along the height direction: the upper layer, the middle layer, and the lower layer. Each layer is separated by a partition with a diversion hole (not shown in the figure) on the partition to ensure uniform water distribution.

[0101] The sensors of the real-time detection module of this embodiment are fixed on the reaction bed frame 4, wherein:

[0102] Dissolved oxygen sensor: including a first dissolved oxygen detector 8 arranged at the bottom of the reaction bed and a second dissolved oxygen detector 9 in the middle, model is Mettler-Toledo InPro6050, with a detection accuracy of ±0.01 mg / L, used for real-time detection of dissolved oxygen content at different points, providing data support for intelligent aeration.

[0103] Flow sensor: installed on the water inlet pipe of the reaction bed, model is E+H Promag W 400, used to monitor the water inlet flow, with a range of 0-100m 3 / h, accuracy ±0.5%.

[0104] Temperature sensor: It is located in the middle of the reaction bed, model is PT100, used to detect sewage temperature, with an accuracy of ±0.5℃, providing a basis for temperature correction in oxygen supply calculation.

[0105] Liquid level sensor: installed on the side wall of the reaction bed, the model is a immersion static pressure level gauge with a range of 0-4m, used to monitor the liquid level height in the reaction bed to prevent overflow.

[0106] NO3 --N detector 6: model is Hach Lange LCK 350, with a detection range of 0-100 mg / L, used for real-time monitoring of nitrate nitrogen concentration in the reaction bed.

[0107] NH4 + -N detector 7: Model is Hach Lange LCK 303, detection range is 0-50mg / L, used for real-time monitoring of ammonia nitrogen concentration.

[0108] Each detection sensor is connected to the data acquisition card of the control module through a shielded cable. The sampling frequency is 1 time / 10s, and the detection data is transmitted to the control module in real time.

[0109] The control module of this embodiment is integrated into a control cabinet (not shown in the figure), including a programmable logic controller (PLC, model Siemens S7-1200), a data acquisition card (ADAM-4017), an industrial control computer, an MPC model prediction algorithm module and a BP neural network model. The specific control calculation is as follows:

[0110] 1. Calculation of oxygen transfer rate and oxygen supply

[0111] The control module calculates the oxygen transfer rate and the oxygen supply of the aeration equipment based on the following formula:

[0112] Oxygen transfer rate: dC / dt = K L a(C s -C)

[0113] Where, dC / dt is the oxygen transfer rate (unit: mg / (L·h)), K L a is the total mass transfer coefficient (unit: h -1 ), in this embodiment, the value is 0.3h -1 (determined by the type of nano aeration equipment and the nature of the sewage), C s It is the saturated dissolved oxygen concentration in water (unit: mg / L) at the current temperature and pressure, which is obtained by looking up a table or calculating the formula. C is the actual dissolved oxygen concentration in water (unit: mg / L), which is detected in real time by the dissolved oxygen sensor.

[0114] Aeration equipment oxygen supply: (Unit: kgO2 / h),

[0115] Where, T is the sewage temperature (unit: °C), which is detected by the temperature sensor; E A is the oxygen utilization rate of the aeration equipment. In this embodiment, the oxygen utilization rate of the nano-aeration equipment E A Take 25%; V is the effective volume of the reaction bed.

[0116] 2.MPC model prediction algorithm

[0117] The MPC model is built on the following state-space model:

[0118] State variables x(k): include dissolved oxygen concentration and microbial activity (indirectly represented by the removal rates of ammonia nitrogen and nitrate nitrogen).

[0119] Control variables u(k): include aeration flow and aeration time of aeration equipment.

[0120] Output variable y(k): is the measured dissolved oxygen concentration.

[0121] State equation: x(k+1)=Ax(k)+Bu(k)+w(k)

[0122] Among them, Ax(k) is the state transfer matrix, which is determined by system identification; Bu(k) is the control input matrix; w(k) is the process noise, which is Gaussian white noise with mean 0 and variance 0.01.

[0123] Output equation: y(k) = Cx(k) + Du(k) + v(k)

[0124] Where Cx(k) is the output matrix; Du(k) is the feedforward matrix; v(k) is the measurement noise, which is Gaussian white noise with mean 0 and variance 0.005.

[0125] Objective function:

[0126]

[0127] Among them, y ref (k+i) is the set dissolved oxygen concentration target value (taken as 0.3 mg / L in this embodiment), Np is the prediction time domain (taken as 10), N e is the control time domain (taken as 5), and λ is the control weight (taken as 0.1), which is used to balance the control performance and energy consumption.

[0128] Constraints:

[0129] Aeration flow u min ≤u(k)≤u max In this embodiment, umin=0m 3 / h,umax=50m 3 / h.

[0130] Dissolved oxygen concentration y(k): 0.2≤y(k)≤0.5mg / L.

[0131] MPC model solution steps:

[0132] At each sampling moment, a finite-time domain optimization problem is formed based on the prediction model, objective function and constraints.

[0133] The logarithmic barrier function is introduced to transform the constrained optimization problem into an unconstrained optimization problem: Where μ is the obstacle parameter, the initial value is 100, and gi(x) is the constraint condition.

[0134] Compute the gradient and Hessian matrix:

[0135] gradient:

[0136] Hessian matrix:

[0137] Iteratively solve an unconstrained optimization problem using Newton's method:

[0138]

[0139] Update barrier parameter: μ k+1 =θμ k

[0140] when When , the algorithm converges, the optimal control sequence is obtained, and the first control variable is applied to the system.

[0141] 3. BP neural network model

[0142] The input layer of the BP neural network model is the dissolved oxygen and NO3 - -N、NH4 + -N data, with an input layer of 3 nodes; two hidden layers, each with 6 nodes; and an output layer containing the aeration rate of the intelligent aeration module and the dosage of the microbial agent release module, with 2 nodes. The activation function uses the ReLU function, and the training algorithm is the Levenberg-Marquardt algorithm, which is used to coordinately adjust the aeration intensity and microbial agent release dosage to improve system response speed and control accuracy.

[0143] (4) Intelligent aeration module

[0144] The intelligent aeration module is a nano aeration device 1 (such as Figure 2 、 Figure 3 The aeration system (shown in Figure 4) is fixed to the bottom of the reactor bed frame 4 and includes a nano-aerator (model JS-NM-10, aeration pore size 50-100 nm) and an aeration control system. The aeration control system receives instructions from the control module and adjusts the aeration device's air flow rate and aeration time to precisely control the aeration volume, maintaining a dissolved oxygen concentration of 0.2-0.5 mg / L.

[0145] When the dissolved oxygen (DO9) at the bottom is less than 2mg / L, nano-aeration is activated; when the dissolved oxygen (DO9) at the bottom is greater than 4mg / L or the dissolved oxygen (DO8) in the middle is greater than 0.5mg / L, aeration is terminated. The air supply of the aeration equipment is adjusted according to the Gs value calculated by the control module to ensure that the dissolved oxygen concentration in the reactor bed is stable within the target range.

[0146] (5) Bacterial agent release module

[0147] The bacterial agent release module includes several bacterial agent slow-release dosing devices, which are installed near the intelligent aeration module on the lower layer of the reaction bed frame 4, including an independent nitrifying bacteria slow-release dosing device 2 and a denitrifying bacteria slow-release dosing device 3, both of which have the same structure: Drive motor 10: Model is stepper motor 42BYGH40-1704A, which is connected to the control module signal to receive dosing instructions. Inner tube 11: Made of PVC, with a diameter of 100mm and a length of 300mm. There are 10 holes with a diameter of 5mm on one side, which are filled with slow-release bacteria 13 (nitrifying bacteria or denitrifying bacteria). Outer tube 12: Made of PVC, with a diameter of 110mm and a length of 300mm. An opening is set on one side and is sleeved on the outside of the inner tube 11. The drive motor 10 adjusts the exposed area of ​​the inner tube hole and the outer tube opening by controlling the rotation angle of the inner tube 11 (0-180°), thereby realizing the start and stop of bacterial release and dosage control:

[0148] Nitrifying bacteria slow-release dosing device 2: According to NH4 + -N concentration adjusts the rotation angle:

[0149] NH4 + -N concentration < 1 mg / L, the adjustment angle is 0° (closed);

[0150] NH4 + -N concentration is 1-2 mg / L, and the adjustment angle is 45°;

[0151] NH4 + -N concentration is 2-3 mg / L, and the adjustment angle is 90°;

[0152] NH4 + -N concentration>3mg / L, the adjustment angle is 180° (maximum release).

[0153] Denitrifying bacteria slow-release dosing device 3: According to NO3 - -N concentration adjusts the rotation angle:

[0154] NO3 - -N concentration < 1 mg / L, the adjustment angle is 0° (closed);

[0155] NO3 - -N concentration is 1-2 mg / L, and the adjustment angle is 45°;

[0156] NO3 - -N concentration is 2-3 mg / L, and the adjustment angle is 90°;

[0157] NO3 - -N concentration>3mg / L, the adjustment angle is 180° (maximum release).

[0158] (6) Collaborative control logic

[0159] The collaborative control logic of the intelligent aeration module and the bacterial agent release module is as follows:

[0160] 1. When NO3 - When -N>4mg / L and DO<0.5mg / L, the control module reduces the aeration volume by 30% through the MPC model and adjusts the denitrifying bacteria slow-release dosing device 3 to 180° to create denitrification conditions and accelerate the removal of nitrate nitrogen.

[0161] 2. The real-time detection module continuously monitors various parameters in the reaction bed, and the control module calculates the optimal aeration volume and bacterial agent dosage based on the MPC model and BP neural network model to achieve dynamic coordinated regulation.

[0162] 3. When water quality parameters change, such as NH4 + -N or NO3 - As the -N concentration increases, the control module automatically increases the dosage of the corresponding bacterial agent and adjusts the aeration volume to ensure that the microbial activity matches the reaction environment.

[0163] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention.

Claims

1. An intelligently adjustable microecological reaction bed, characterized in that: It includes an ecological-based reaction bed, a real-time detection module, a control module, an intelligent aeration module and a bacterial agent release module; The ecological-based reactor bed adopts a three-dimensional multi-layer sandwich structure, including a reactor bed frame with a MOBR membrane attached to the surface, which is used to simultaneously achieve half-way denitrification and anaerobic ammonium oxidation reactions; The real-time detection module is fixed on the reaction bed frame and includes a dissolved oxygen sensor, a flow sensor, a temperature sensor, a liquid level sensor, a NO3 - -N detector and NH4 + -N detector, used to monitor dissolved oxygen, flow, temperature, liquid level and nitrogen pollutant concentration data in the reaction bed and feed back to the control module; The control module includes a programmable logic controller, a data acquisition card, an industrial control computer, an MPC model prediction algorithm module and a BP neural network model, which is used to achieve remote control, automatic adjustment, fault warning and dynamic optimization based on real-time detection data; The intelligent aeration module is a nano-aeration device fixed on the reaction bed frame, including nano-aeration equipment and aeration control system, which is used to accurately adjust the aeration volume according to the instructions of the control module to maintain the dissolved oxygen concentration at 0.2-0.5 mg / L; The bacterial agent release module includes several bacterial agent slow-release dosing devices, which are installed near the intelligent aeration module in the lower layer of the reaction bed frame and are used to dynamically add nitrifying bacteria or denitrifying bacteria agents according to the instructions of the control module.

2. The intelligently adjustable microecological reaction bed according to claim 1, characterized in that: The dissolved oxygen sensor comprises at least a first dissolved oxygen detector arranged at the bottom of the reaction bed and a second dissolved oxygen detector in the middle, and is used for real-time detection of dissolved oxygen content at different points.

3. The intelligently adjustable microecological reaction bed according to claim 1, characterized in that: The control module calculates the oxygen transfer rate and the oxygen supply of the aeration equipment based on the following formula: Oxygen transfer rate: dC / dt=K L a(C s -C), Where dC / dt is the oxygen transfer rate, which represents the change in dissolved oxygen concentration in water per unit time; K L a is the total mass transfer coefficient; C s is the saturated dissolved oxygen concentration in water at the current temperature and pressure; C is the actual dissolved oxygen concentration in water; Aeration equipment oxygen supply: Oxygen supply of aeration equipment G s , the calculation formula is: Where T is the sewage temperature. Since temperature affects the solubility of oxygen in water and the metabolic activity of microorganisms, a temperature correction factor of 1.024 needs to be considered. T-20 , E A is the oxygen utilization rate of the aeration equipment. Different types of aeration equipment have different oxygen utilization rates. V is the effective volume of the reaction bed. The specific value of oxygen supply can be calculated through sewage temperature, oxygen utilization rate of aeration equipment, oxygen transfer rate and effective volume of aeration tank, making the control more accurate.

4. The intelligently adjustable microecological reaction bed according to claim 1, characterized in that: The MPC model prediction algorithm of the control module is built based on the following state space model: The state variables x(k) include dissolved oxygen concentration and microbial activity; The controlled variables u(k) include the aeration flow rate and aeration time of the aeration equipment; The output variable y(k) is the measured dissolved oxygen concentration; Equation of state: x(k+1)=Ax(k)+Bu(k)+w(k), Where Ax(k) is the state transition matrix, which describes the natural evolution of the system state when there is no control input; Bu(k) is the control input matrix, which shows the impact of the control input on the state variables; w(k) is the process noise, which is used to account for the unmodeled dynamics and disturbance factors present in the actual system. Output equation: y(k)=Cx(k)+Du(k)+v(k), Where Cx(k) is the output matrix, which maps state variables to output variables; DU(K) is the feedforward matrix; v(k) is the measurement noise, which takes into account the error in the measurement process; The objective function of the MPC model is: where y ref (k+i) is the reference output at time k+i, N p is the prediction time domain (take 10), N e is the control time domain (take 5), λ is the control weight, which is used to balance the control performance and control energy consumption; Constraints: The aeration flow rate u(k) has upper and lower limits, namely u min ≤u(k)≤u max ; Dissolved oxygen concentration 0.2≤y(k)≤0.

5.

5. The intelligently adjustable microecological reaction bed according to claim 4, characterized in that: The MPC model solves the optimization problem, and the specific steps include: At each sampling moment, a finite-time domain optimization problem is formed based on the prediction model, objective function, and constraints. This optimization problem is solved by the optimization algorithm to obtain the optimal control sequence: [u(k|k),u(k+1|k),…,u(k+N c (-1|k)] The first control variable u(k|k) of the optimized control sequence is applied to the actual system. Then, at the next sampling time k+1, the system state is remeasured, the prediction model is updated, and the optimization problem is solved again. This process is repeated to achieve dynamic control of the intelligent aeration system. The specific formula of the optimization algorithm is: set up: x=[u(k|k),u(k+1|k),…,u(k+N c -1|k)] T Rewrite the constraint as g1(x)=u min -u(k|k)≤0,g2(x)=u(k|k)-u max ≤0; Introducing the logarithmic barrier function Where μ>0 is the barrier parameter. When x approaches the inequality constraint boundary (i.e. g i (x)→0), ln(-g i (x))→-∞, so that the optimization process tends to proceed within the feasible region and avoids crossing the boundary; Step 1: Solve the obstacle problem For a given μ, solve the unconstrained optimization problem minB(x,μ), which usually requires calculating the gradient and the Hessian matrix The gradient calculation formula is: The Hessian matrix calculation formula is: Then use Newton's method to solve this unconstrained optimization problem. The iterative formula of Newton's method is: where x k is the point of the kth iteration; Step 2: Update barrier parameters As the iteration proceeds, μ gradually decreases, making the optimization process gradually approach the solution of the original constrained optimization problem. Common update strategies are m k+1 =thm k , among themθ∈(0,1); Step 3: Convergence judgment Check whether the convergence criterion is met. When , the algorithm is considered to have converged, and x is regarded as an approximate optimal solution.

6. The intelligently adjustable microecological reaction bed according to claim 1, characterized in that: The input layer of the BP neural network model is the dissolved oxygen, NO3 - -N、NH4 + -N data, the output layer is the aeration volume of the intelligent aeration module and the dosage of the bacterial agent release module, which are used to coordinately adjust the aeration intensity and bacterial agent release dosage.

7. The intelligently adjustable microecological reaction bed according to claim 1, characterized in that: The microbial agent slow-release dosing device comprises a driving motor, an inner tube and an outer tube; The inner tube is filled with slow-release bacteria and has a plurality of holes on one side; the outer tube is provided with an opening on one side and is sleeved on the outer side of the inner tube; The drive motor is connected to the inner tube and the exposed area of ​​the inner tube hole and the outer tube opening is adjusted by controlling the rotation angle of the inner tube to realize the start and stop of bacterial release and dosage control. The rotation angle of the inner tube is 0-180 degrees. The drive motor is connected to the control module signal, according to NO3 - -N and NH4 + -N detection concentration triggers the dosing instruction.

8. The intelligently adjustable microecological reaction bed according to claim 7, characterized in that: The bacterial agent slow-release dosing device includes an independent nitrifying bacteria slow-release dosing device and a denitrifying bacteria slow-release dosing device, both of which have the same structure; The rotation angle of the nitrifying bacteria slow-release dosing device is + The corresponding relationship of -N concentration is: NH4 + -N concentration is less than 1mg / L, the adjustment angle is 0°, which means it is closed. NH4 + -N concentration is 1-2mg / L, the adjustment angle is 45°, NH4 + -N concentration is 2-3mg / L, the adjustment angle is 90°, NH4 + -N concentration>3mg / L, the adjustment angle is 180°, at this time the maximum release amount; According to NO3 - -N detector detected NO3 - -N concentration regulates the opening and closing angle of the denitrifying bacteria device as follows: NO3 - -N concentration is less than 1mg / L, the adjustment angle is 0°, which means it is closed. NO3 - -N concentration is 1-2mg / L, the adjustment angle is 45°, NO3 - -N concentration is 2-3mg / L, the adjustment angle is 90°, NO3 - -N concentration>3mg / L, the adjustment angle is 180°, at which point the release is maximum.

9. The microecological reaction bed according to claim 1, characterized in that: The collaborative control logic of the intelligent aeration module and the bacterial agent release module includes: When NO3 - When -N>4mg / L and DO<0.5mg / L, the aeration rate was reduced by 30% through the MPC model, and the denitrifying bacteria slow-release dosing device was adjusted to 180°; When the dissolved oxygen DO9 at the bottom is less than 2 mg / L, start nano aeration; When the dissolved oxygen DO9 at the bottom is greater than 4 mg / L or the dissolved oxygen DO8 in the middle is greater than 0.5 mg / L, aeration is terminated.

Images