Tail water treatment method and system based on industrial aquaculture of South America white shrimps

Abstract

The invention discloses a tail water treatment method and system based on industrial aquaculture of South America white shrimps, and relates to the technical field of aquaculture tail water treatment.The method comprises the steps that a tail water purification integrated device containing a microfilter, a protein separator and other components is obtained, and pretreatment, biological treatment and deep treatment nodes of the tail water treatment process are extracted; corresponding nodes and device components are correlated and mapped to obtain three corresponding processing channels, then tail water component data are collected, control parameters are analyzed based on the three processing channels, and tail water closed-loop processing is carried out. The technical problems that an existing treatment mode for the industrial aquaculture tail water of the South America white shrimps is difficult to achieve standard treatment, high in treatment cost and poor in synergism of all links are solved, and the technical effects that the industrial aquaculture tail water of the South America white shrimps is efficiently treated, the treatment cost is reduced, operation is more convenient and faster, and the synergism of all links is better are achieved.

Description

Technical Field

[0001] This invention relates to the field of aquaculture wastewater treatment technology, and in particular to a wastewater treatment method and system based on the factory farming of white shrimp. Background Technology

[0002] Effective treatment of aquaculture wastewater is crucial for ecological protection and the sustainable development of the aquaculture industry. The wastewater from factory farming of white shrimp, containing uneaten feed, shrimp feces, ammonia nitrogen, and pathogens, presents a particularly urgent need for treatment. Current technologies often employ decentralized filtration and biological purification equipment, relying on experience-based parameter settings. However, these methods have limitations when applied to factory farming wastewater: the treatment process is fragmented, the coordination between stages is poor, and the fixed parameters are difficult to adapt to the dynamic changes in wastewater composition. This results in difficulties in achieving compliant wastewater treatment, excessively high costs, and incomplete treatment data, failing to meet the needs for precise control and efficient purification. Summary of the Invention

[0003] This application provides a method and system for treating wastewater from factory farming of white shrimp, which solves the technical problems of existing treatment methods for wastewater from factory farming of white shrimp, such as difficulty in achieving standard treatment, high treatment costs, and poor coordination among various links.

[0004] The first aspect of this application provides a method for treating wastewater from factory farming of whiteleg shrimp. The method includes: acquiring an integrated wastewater purification device, which comprises a microfilter, a protein separator, an aerator, a biological packing reactor, and an ultraviolet sterilizer; extracting nodes from the wastewater treatment process of factory farming of whiteleg shrimp to obtain pretreatment nodes, biological treatment nodes, and deep treatment nodes; mapping the pretreatment nodes, biological treatment nodes, and deep treatment nodes to the components in the integrated wastewater purification device to obtain a microfilter-pretreatment channel, a protein aerator-biological treatment channel, and an ultraviolet sterilizer-deep treatment channel; collecting wastewater composition data; and performing control parameter analysis and closed-loop wastewater treatment based on the microfilter-pretreatment channel, the protein aerator-biological treatment channel, and the ultraviolet sterilizer-deep treatment channel.

[0005] The second aspect of this application provides a wastewater treatment system based on the industrialized farming of whiteleg shrimp. The system includes: a wastewater purification device construction module for acquiring an integrated wastewater purification device, which comprises a microfilter, a protein separator, an aerator, a biological packing reactor, and an ultraviolet sterilizer; a treatment node acquisition module for extracting nodes from the wastewater treatment process of industrialized whiteleg shrimp farming to obtain pretreatment nodes, biological treatment nodes, and deep treatment nodes; a treatment channel acquisition module for associating the pretreatment nodes, biological treatment nodes, and deep treatment nodes with the components in the integrated wastewater purification device to obtain a microfilter-pretreatment channel, a protein aerator-biological treatment channel, and an ultraviolet sterilizer-deep treatment channel; and a wastewater closed-loop treatment module for collecting wastewater composition data and performing control parameter analysis and closed-loop wastewater treatment based on the microfilter-pretreatment channel, protein aerator-biological treatment channel, and ultraviolet sterilizer-deep treatment channel.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages: This application utilizes an integrated wastewater purification device to extract and map pretreatment, biological treatment, and advanced treatment nodes to construct treatment channels. It collects wastewater composition data and obtains multi-node treatment parameters through channel-based control parameter analysis. Closed-loop adjustments are then made based on the multi-node treatment component data, enabling phased and precise treatment of wastewater from factory farming of white shrimp. This results in more efficient wastewater treatment, more stable compliance with standards, and achieves the technical effects of high-efficiency treatment of wastewater from factory farming of white shrimp, reduced treatment costs, more convenient operation, and better coordination among various stages. Attached Figure Description

[0007] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0008] Figure 1 This is a schematic flowchart of a wastewater treatment method based on the factory farming of white shrimp provided in this application embodiment.

[0009] Figure 2 This is a schematic diagram of the wastewater treatment system based on the factory farming of white shrimp provided in this application embodiment.

[0010] Figure labeling: 1. Wastewater purification device construction module; 2. Processing node acquisition module; 3. Processing channel acquisition module; 4. Wastewater closed-loop treatment module. Detailed Implementation

[0011] This application provides a method and system for treating wastewater from factory farming of white shrimp, which solves the technical problems of existing treatment methods for wastewater from factory farming of white shrimp, such as difficulty in achieving standard treatment, high treatment costs, and poor coordination among various links.

[0012] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0013] It should be noted that the terms "first," "second," etc., in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules not explicitly listed or inherent to such processes, methods, products, or devices.

[0014] Example 1, as Figure 1 As shown, a method for treating wastewater from factory farming of whiteleg shrimp is provided, wherein the method includes: Step A100: Obtain an integrated wastewater purification device, which consists of a microfilter, a protein separator, an aerator, a biological packing reactor, and an ultraviolet sterilizer.

[0015] Specifically, the first step is to acquire an integrated wastewater purification device. This device consists of a microfilter, a protein skimmer, an aerator, a biological packing reactor, and an ultraviolet sterilizer. The overall dimensions are 10m long, 5m wide, and 4m high, suitable for wastewater treatment in factory farming of whiteleg shrimp. The microfilter uses a 200-mesh screen and has a processing capacity of 100m³ / h. It primarily intercepts suspended particulate matter in the wastewater, such as uneaten feed and shrimp feces, providing preliminary purification for subsequent treatment. The protein skimmer also has a processing capacity of 100m³ / h and separates colloidal organic matter and proteinaceous substances from the wastewater, reducing the organic load in the water. The aerator introduces oxygen into the wastewater, providing a suitable living environment for microorganisms in the biological packing reactor. The packing material in the biological packing reactor provides a carrier for microorganisms to attach to, and the microorganisms degrade harmful substances such as ammonia nitrogen and nitrite in the wastewater through metabolism. The ultraviolet sterilizer kills pathogens and harmful microorganisms in the wastewater through ultraviolet irradiation, ensuring the hygiene and safety of the treated wastewater.

[0016] The process begins with pretreatment of the aquaculture wastewater using a microfiltration system. The 200-mesh screen effectively removes most suspended solids. The treated wastewater then enters a protein skimmer, where organic colloids are further separated at a treatment efficiency of 100 m³ / h, reducing the water's pollution load. Subsequently, the wastewater flows into the biological treatment stage, where aerators continuously supply oxygen, working in conjunction with microorganisms on the biological packing material to degrade nitrogen, phosphorus, and other nutrient pollutants. Finally, the biologically treated wastewater undergoes further treatment in an ultraviolet sterilizer for disinfection. Throughout the entire process, all components of the integrated wastewater purification system operate collaboratively, matching the treated water volume to the generated aquaculture wastewater volume, ensuring a continuous and efficient treatment process.

[0017] By systematically combining the various components according to the processes of pretreatment, biological treatment, and advanced treatment, and relying on specific treatment parameters and device dimensions, the efficient purification treatment of wastewater from the factory farming of white shrimp in South America has been achieved.

[0018] Step A200: Extract nodes from the wastewater treatment process of factory farming of white shrimp to obtain pretreatment nodes, biological treatment nodes, and deep treatment nodes.

[0019] Optionally, the types of pollutants in the effluent and their treatment requirements should be analyzed first. Aquaculture effluent contains suspended particulate matter such as uneaten feed and shrimp feces. These substances need to be removed by physical filtration to reduce the load on subsequent treatment. Based on this, a pretreatment node is extracted. This node mainly corresponds to the microfilter in the integrated effluent purification device. The microfilter uses a 200-mesh screen and has a treatment capacity of 100 m³ / h, which can effectively intercept the above-mentioned suspended particulate matter and complete the initial purification.

[0020] Next, for pollutants such as colloidal organic matter, ammonia nitrogen, and nitrite still present in the pretreated effluent, these substances need to be degraded through biological processes. Therefore, a biological treatment node was identified. This node is connected to a protein separator, an aerator, and a biological packing reactor. The protein separator has a treatment capacity of 100 m³ / h and is responsible for separating colloidal organic matter. The aerator provides sufficient oxygen to create a suitable environment for microorganisms in the biological packing reactor. The biological packing provides a carrier for microbial attachment, and the microorganisms degrade pollutants such as nitrogen and phosphorus through metabolism, jointly completing the biological purification process.

[0021] Finally, pathogens and other harmful microorganisms may remain in the effluent after biological treatment, which needs to be sterilized and disinfected to ensure safe discharge or recycling. This leads to the discovery of a deep treatment node, which corresponds to the ultraviolet sterilizer in the device. The ultraviolet light is used to kill pathogens and achieve deep purification of the effluent.

[0022] By analyzing the characteristics of effluent pollutants and treatment requirements, and combining the functional positioning of each component of the device, pretreatment, biological treatment, and advanced treatment nodes were extracted, providing a clear stage division and functional orientation for the construction of subsequent treatment channels.

[0023] Step A300: Associate and map the pretreatment node, biological treatment node, and deep treatment node with the components in the integrated wastewater purification device to obtain the microfilter-pretreatment channel, the protein aeration packing-biological treatment channel, and the ultraviolet sterilizer-deep treatment channel.

[0024] In one embodiment of this application, the pretreatment node, biological treatment node, deep treatment node and corresponding components of the integrated wastewater purification device are associated and mapped to determine the treatment target of each node. Based on the associated nodes, historical data is mined to obtain the parameter space. Then, the parameter space is analyzed according to the treatment target to construct the microfilter-pretreatment channel, the protein aeration packing-biological treatment channel and the ultraviolet sterilizer-deep treatment channel. The specific steps are described in detail in A310-A340.

[0025] Step A400: Collect aquaculture wastewater composition data, and perform control parameter analysis and closed-loop treatment of the aquaculture wastewater composition data based on the microfilter-pretreatment channel, protein aeration packing-biological treatment channel and ultraviolet sterilizer-deep treatment channel.

[0026] Specifically, firstly, monitoring requirements analysis is conducted on the pretreatment, biological treatment, and deep treatment nodes to obtain the corresponding monitoring requirements parameters. Based on this, the sensor network for each node is configured, and then the composition data of the aquaculture wastewater is obtained based on the sensor network detection. The specific steps are explained in detail in A410-A430.

[0027] Next, a series of microfiltration-pretreatment channels, protein aeration packing-biological treatment channels, and ultraviolet sterilizer-deep treatment channels are connected to build an aquaculture wastewater treatment channel. The aquaculture wastewater treatment channel is used to analyze the composition data of the aquaculture wastewater to determine multi-node control parameters. Based on these parameters, treatment control and node monitoring are carried out to obtain multi-node wastewater treatment composition data and realize closed-loop treatment of wastewater. The specific steps are explained in detail in A440-A460.

[0028] Furthermore, step A300 in the method provided in this application embodiment includes: A310: Associate and map the pretreatment node, biological treatment node, and deep treatment node with the components in the integrated wastewater purification device to obtain the microfilter-pretreatment node, protein separator, aerator, biological packing reactor-biological treatment node, and ultraviolet sterilizer-deep treatment node.

[0029] A320: Based on the pretreatment node, biological treatment node, and advanced treatment node, determine the effluent pretreatment target, effluent biological treatment target, and effluent advanced treatment target.

[0030] A330: Based on the historical data mining of the microfilter-pretreatment node, protein separator, aerator, biological packing reactor-biological treatment node and ultraviolet sterilizer-deep treatment node, respectively, the effluent pretreatment parameter space, effluent biological treatment parameter space and effluent deep treatment parameter space are obtained.

[0031] A340: According to the stated effluent pretreatment target, effluent biological treatment target, and effluent deep treatment target, the control parameters of the effluent pretreatment parameter space, effluent biological treatment parameter space, and effluent deep treatment parameter space are analyzed to construct a microfilter-pretreatment channel, a protein aeration packing-biological treatment channel, and an ultraviolet sterilizer-deep treatment channel.

[0032] Specifically, the pretreatment node, biological treatment node, and advanced treatment node are first mapped to the components of the integrated wastewater purification device. The pretreatment node corresponds to a microfilter, which uses a 200-mesh screen to intercept suspended particles such as uneaten feed and shrimp feces, matching the function of the pretreatment node in removing large particulate impurities, thus forming a microfilter-pretreatment node. The biological treatment node is associated with a protein separator, aerator, and biological packing reactor. The protein separator has a treatment capacity of 100 m³ / h and can separate colloidal organic matter. The aerator provides oxygen, and the biological packing provides a carrier for microorganisms to jointly complete biodegradation, forming a protein separator-aerator-biological treatment node. The advanced treatment node corresponds to an ultraviolet sterilizer, which kills pathogens using ultraviolet light, forming an ultraviolet sterilizer-advanced treatment node.

[0033] Next, the treatment objectives were determined based on the functions of each node. For the microfiltration-pretreatment node, considering its treatment capacity of 100 m³ / h, the effluent pretreatment objective was to reduce the concentration of suspended particulate matter to a range suitable for subsequent treatment. For the biological treatment node, based on the protein separator and the overall unit's treatment capacity of 100 m³ / h, the effluent biological treatment objective was to reduce the concentrations of ammonia nitrogen, nitrite, and organic pollutants. For the advanced treatment node, the effluent advanced treatment objective was to kill pathogens in the effluent to meet discharge or recycling standards.

[0034] For example, the pretreatment target for effluent can be set to reduce the concentration of suspended particulate matter such as uneaten feed and shrimp feces in the effluent to below 10 mg / L, ensuring that suspended particulate matter will not adhere to the surface of the biological packing material and affect the activity of microorganisms in the subsequent biological treatment process; the biological treatment target for effluent can be set to reduce the concentration of ammonia nitrogen in the effluent to below 0.5 mg / L, the concentration of nitrite to below 0.1 mg / L, and the chemical oxygen demand (COD) to below 30 mg / L, where COD reflects the content of organic pollutants and provides a water quality basis for advanced treatment; the advanced treatment target for effluent can be set to reduce the number of pathogens such as Vibrio in the effluent to below 100 CFU / mL, ensuring that the treated effluent meets the health and safety requirements whether it is discharged into natural water bodies or recycled back into the aquaculture pond. The specific pretreatment targets can be adjusted by those skilled in the art according to actual needs.

[0035] Subsequently, historical data mining was performed based on the aforementioned related nodes. For the microfilter-pretreatment node, data such as filtration pressure and backwashing frequency of the 200-mesh microfilter at a treatment capacity of 100 m³ / h were collected under different influent suspended particle concentrations to form a parameter space for effluent pretreatment. For the biological treatment node, historical data such as the separation efficiency of the protein skimmer at a treatment capacity of 100 m³ / h, the oxygen supply intensity of the aerator, and the degradation efficiency of microorganisms on the biological packing were collected to form a parameter space for effluent biological treatment. For the ultraviolet sterilizer-deep treatment node, historical data on different ultraviolet intensities, irradiation times, and pathogen kill rates were collected to form a parameter space for effluent deep treatment.

[0036] Finally, the corresponding effect index sets are obtained by sequentially extracting indicators for the targets of effluent pretreatment, biological treatment, and advanced treatment. The effect sets are then evaluated on the corresponding parameter spaces according to the effect index sets to obtain effect sets. The parameter spaces are then screened using the effect sets to obtain usable parameter sets. Finally, the usable parameter sets are controlled, trained, and optimized to obtain the microfilter-pretreatment channel, the protein aeration packing-biological treatment channel, and the ultraviolet sterilizer-advanced treatment channel. The specific steps are explained in detail in A341-A344.

[0037] By linking and mapping nodes and components, identifying targets, mining data, and analyzing parameters in a coherent manner, three targeted processing channels were constructed, providing a path for the phased and precise treatment of tailwater.

[0038] Furthermore, step A340 in the method provided in this application embodiment includes: A341: Extract indicators sequentially from the wastewater pretreatment target, wastewater biological treatment target, and wastewater advanced treatment target to obtain a set of indicators for wastewater pretreatment effect, a set of indicators for wastewater biological treatment effect, and a set of indicators for wastewater advanced treatment effect.

[0039] A342: Evaluate the effectiveness of the effluent pretreatment parameter space, effluent biological treatment parameter space, and effluent deep treatment parameter space according to the effluent pretreatment effect index set, effluent biological treatment effect index set, and effluent deep treatment effect index set, to obtain the effluent pretreatment effect set, effluent biological treatment effect set, and effluent deep treatment effect set.

[0040] A343: The tailwater pretreatment effect set, tailwater biological treatment effect set, and tailwater deep treatment effect set are used to screen the control parameters of the tailwater pretreatment parameter space, tailwater biological treatment parameter space, and tailwater deep treatment parameter space, respectively, to obtain the usable tailwater pretreatment parameter set, usable tailwater biological treatment parameter set, and usable tailwater deep treatment parameter set.

[0041] A344: Control training and optimization are performed on the available effluent pretreatment parameter set, available effluent biological treatment parameter set, and available effluent deep treatment parameter set respectively to obtain the microfilter-pretreatment channel, protein aeration packing-biological treatment channel, and ultraviolet sterilizer-deep treatment channel.

[0042] Optionally, firstly, indicators are extracted for the wastewater pretreatment targets, wastewater biological treatment targets, and wastewater advanced treatment targets in sequence. For the effluent pretreatment target, the extracted performance indicator set includes suspended particulate matter removal rate and post-treatment suspended particulate matter concentration, where the removal rate must be no less than 90% and the post-treatment concentration must be controlled below 10 mg / L. For the effluent biological treatment target, the performance indicator set covers ammonia nitrogen removal rate, nitrite removal rate, chemical oxygen demand (COD) removal rate, and the corresponding concentrations after treatment. The requirements are: ammonia nitrogen removal rate no less than 90% and post-treatment concentration ≤ 0.5 mg / L; nitrite removal rate no less than 90% and post-treatment concentration ≤ 0.1 mg / L; COD removal rate no less than 40% and post-treatment concentration ≤ 30 mg / L. For the effluent advanced treatment target, the performance indicator set includes pathogen kill rate and post-treatment pathogen quantity, where the kill rate must be no less than 90% and the post-treatment quantity ≤ 100 CFU / mL. Thus, the effluent pretreatment performance indicator set, effluent biological treatment performance indicator set, and effluent advanced treatment performance indicator set are obtained.

[0043] Next, the effectiveness of the parameters in the effluent pretreatment parameter space, effluent biological treatment parameter space, and effluent advanced treatment parameter space was evaluated according to the three sets of effectiveness indicators mentioned above. For the effluent pretreatment parameter space, this space includes parameters such as the filtration time of the microfilter and the backwashing frequency. During the evaluation, the suspended particulate matter removal rate and the post-treatment concentration under different parameter combinations need to be tested. For example, when the filtration time is 30 minutes and the backwashing frequency is 2 hours / time, if the suspended particulate matter removal rate reaches 92% and the post-treatment concentration is 8 mg / L, then it meets the pretreatment effectiveness indicators.

[0044] The parameter space for biological treatment of effluent includes parameters such as the operating pressure of the protein separator, the oxygen supply intensity of the aerator, and the filling rate of the biological packing material. The removal efficiency of ammonia nitrogen, nitrite, and chemical oxygen demand under different parameter combinations is evaluated. For example, when the protein separator pressure is 0.2 MPa, the aeration intensity is 2 m³ / h, and the packing material filling rate is 70%, if the ammonia nitrogen removal rate is 93%, the nitrite removal rate is 91%, and the chemical oxygen demand removal rate is 45%, and the concentrations after treatment all meet the standards, then the biological treatment efficiency index is met.

[0045] The parameter space for advanced effluent treatment includes parameters such as the irradiation time and UV intensity of the UV sterilizer. It evaluates the pathogen killing effect under different parameter combinations. For example, if the irradiation time is 30 seconds and the intensity is 30 mW / cm², and the pathogen killing rate is 92% and the number of bacteria after treatment is 80 CFU / mL, then it meets the advanced treatment effect index, and thus the effluent pretreatment effect set, effluent biological treatment effect set, and effluent advanced treatment effect set are obtained.

[0046] Subsequently, the aforementioned effect sets were used to screen control parameters for the corresponding parameter spaces. From the effluent pretreatment effect set, all parameter combinations that met the requirements of suspended particulate matter removal rate ≥90% and post-treatment concentration ≤10mg / L were selected, such as filtration time 25-40 minutes and backwashing frequency 1.5-3 hours / time, forming a usable effluent pretreatment parameter set. From the effluent biological treatment effect set, parameter combinations that met the removal rate and concentration requirements of ammonia nitrogen, nitrite, and chemical oxygen demand were selected, such as protein separator pressure 0.15-0.25MPa, aeration intensity 1.5-2.5m³ / h, and packing filler rate 60%-75%, forming a usable effluent biological treatment parameter set. From the effluent advanced treatment effect set, parameter combinations that met the requirements of pathogen kill rate ≥90% and post-treatment quantity ≤100CFU / mL were selected, such as irradiation time 20-40 seconds and ultraviolet intensity 20-40mW / cm², forming a usable effluent advanced treatment parameter set.

[0047] Finally, the available parameters for wastewater pretreatment, biological treatment, and advanced treatment are integrated according to time series to obtain three corresponding sequence parameter sets. These are then controlled and labeled to obtain three corresponding sequence sample sets. A recurrent neural network is then used to train and optimize each sequence sample set to obtain the microfilter-pretreatment channel, the protein aeration packing-biological treatment channel, and the ultraviolet sterilizer-advanced treatment channel. The specific steps are explained in detail in A344-1-A344-3.

[0048] By establishing clear evaluation criteria through indicator extraction, verifying the effectiveness of parameters through effect evaluation, and obtaining applicable parameters through parameter screening, a precise and feasible parameter foundation is provided for the construction of subsequent processing channels.

[0049] Furthermore, step A344 in the method provided in this application embodiment includes: A344-1: Integrate the available wastewater pretreatment parameter set, available wastewater biological treatment parameter set, and available wastewater advanced treatment parameter set according to the time series to obtain the wastewater pretreatment sequence parameter set, wastewater biological treatment sequence parameter set, and wastewater advanced treatment sequence parameter set.

[0050] A344-2: Control and identify the tailwater pretreatment sequence parameter set, tailwater biological treatment sequence parameter set, and tailwater advanced treatment sequence parameter set respectively to obtain tailwater pretreatment sequence sample set, tailwater biological treatment sequence sample set, and tailwater advanced treatment sequence sample set.

[0051] A344-3: A recurrent neural network is used to control, train, and optimize the tailwater pretreatment sequence sample set, tailwater biological treatment sequence sample set, and tailwater deep treatment sequence sample set, respectively, to obtain the microfilter-pretreatment channel, protein aeration packing-biological treatment channel, and ultraviolet sterilizer-deep treatment channel.

[0052] Specifically, when constructing the microfiltration-pretreatment channel, the available effluent pretreatment parameter set is first integrated according to a time series. This set includes screened microfiltration operating parameters such as filtration time, backwashing frequency, and inlet pressure. These parameters all meet the requirements of suspended particulate matter removal rate ≥90% and post-treatment concentration ≤10 mg / L. During time series integration, parameter combinations are recorded at hourly intervals to form a sequence arranged chronologically. For example, hour 1: filtration time 30 minutes, backwashing frequency 2 hours / time, inlet pressure 0.12 MPa; hour 2: filtration time 28 minutes, backwashing frequency 1.8 hours / time, inlet pressure 0.13 MPa; hour 3: filtration time 35 minutes, backwashing frequency 2.5 hours / time, inlet pressure 0.11 MPa, etc. This yields the effluent pretreatment sequence parameter set, which reflects the dynamic changes in microfiltration operating parameters at different time points. Similarly, the effluent biological treatment sequence parameter set and the effluent advanced treatment sequence parameter set are obtained.

[0053] Next, the parameter set for the effluent pretreatment sequence was controlled and labeled. For each time point, the corresponding actual treatment effect, such as the removal rate of suspended particulate matter and the concentration after treatment, was labeled. The parameter sequence was associated with the effect label to form a sample set of effluent pretreatment sequences. Each sample contains both the operating parameters arranged in chronological order and the corresponding treatment effect, providing input and reference standards for subsequent model training. Similarly, sample sets of effluent biological treatment sequences and effluent advanced treatment sequences were obtained.

[0054] Subsequently, a recurrent neural network (RNN) was used to control and train the effluent pretreatment sequence sample set for optimization. The RNN captures the dependencies between time-series data through memory units, taking the parameter sequence from the effluent pretreatment sequence sample set as input and the corresponding treatment effect as the output target. During training, the RNN continuously learns the influence of parameter changes over time on the treatment effect, such as identifying how extending filtration time and reducing backwashing frequency synergistically improve the removal rate. By iteratively adjusting the network weights and biases, the error between the predicted treatment effect and the actual labeled effect is gradually reduced. When the error stabilizes within a preset range, such as ≤5%, training is complete, ultimately resulting in a microfilter-pretreatment channel that can dynamically output the pretreatment effect based on real-time operating parameters.

[0055] Furthermore, the construction process of the protein aeration packing-biological treatment channel and the ultraviolet sterilizer-deep treatment channel is similar to that of the microfilter-pretreatment channel, both obtained through time-series parameter integration, control labeling samples, and recurrent neural network training. The inputs and outputs of the three channels are as follows: the input of the microfilter-pretreatment channel is the real-time collected concentration of suspended particulate matter in the influent of the pretreatment node, and the filtration time and backwashing frequency in the microfilter operating parameters; the output is the concentration of suspended particulate matter and removal rate after pretreatment. The input of the protein aeration packing-biological treatment channel is the concentration of ammonia nitrogen and nitrite in the influent of the biological treatment node, and the operating parameters of the protein separator and aerator; the output is the concentration of ammonia nitrogen and nitrite after biological treatment and the corresponding removal rate. The input of the ultraviolet sterilizer-deep treatment channel is the number of pathogens in the influent of the deep treatment node, and the irradiation time and intensity in the ultraviolet sterilizer operating parameters; the output is the number of pathogens and the kill rate after treatment.

[0056] By integrating parameters through time series analysis, identifying samples, and training recurrent neural networks, three channels that can dynamically respond to parameter changes were constructed. The input and output of the three channels were also defined, providing model support for the precise control of the treatment effect at each stage of the wastewater treatment.

[0057] Furthermore, step A400 in the method provided in this application embodiment includes: A410: Perform monitoring requirement analysis on the pretreatment node, biological treatment node and deep treatment node respectively to obtain the monitoring requirement parameters for pretreatment, biological treatment and deep treatment.

[0058] A420: Configure the pretreatment sensor network, biological treatment sensor network, and deep treatment sensor network according to the pretreatment monitoring requirement parameters, biological treatment monitoring requirement parameters, and deep treatment monitoring requirement parameters.

[0059] A430: Based on the pretreatment sensor network, biological treatment sensor network and deep treatment sensor network, the composition of the wastewater is detected to obtain the composition data of the aquaculture wastewater.

[0060] Specifically, when collecting data on the composition of aquaculture wastewater, the monitoring requirements for the pretreatment, biological treatment, and advanced treatment nodes are first analyzed. The pretreatment node removes suspended particulate matter using a microfiltration machine. Its core requirement is to know the concentration of suspended particulate matter before and after treatment. Therefore, the monitoring parameters for pretreatment include the concentration of suspended particulate matter in the influent and the concentration of suspended particulate matter in the effluent. The influent concentration needs to cover the common range of 50-100 mg / L, and the effluent concentration needs to be able to detect a compliant value below 10 mg / L.

[0061] The biological treatment node treats colloidal organic matter, ammonia nitrogen, and nitrite using protein separators, aerators, and biological packing reactors. Monitoring parameters for biological treatment include influent ammonia nitrogen concentration, effluent ammonia nitrogen concentration, influent nitrite concentration, effluent nitrite concentration, and chemical oxygen demand (COD). These parameters must cover the common ranges for ammonia nitrogen (2-5 mg / L and below 0.5 mg / L), nitrite (0.5-2 mg / L and below 0.1 mg / L), and COD (50-100 mg / L and below 30 mg / L), respectively. The advanced treatment node uses ultraviolet (UV) sterilizers to kill pathogens. Monitoring parameters for advanced treatment include influent pathogen count and effluent pathogen count, covering ranges of 1000-5000 CFU / mL and below 100 CFU / mL.

[0062] Next, based on the aforementioned monitoring requirements, configure the corresponding sensor network. The pretreatment sensor network requires suspended particulate matter sensors to be installed at the inlet and outlet of the microfilter, with a sensor accuracy of 1 mg / L to ensure accurate detection of concentration changes. The biological treatment sensor network requires ammonia nitrogen and nitrite sensors to be installed at the inlet of the protein separator and the outlet of the biological packing reactor, respectively. Chemical oxygen demand (COD) sensors should be installed at the inlet and outlet of the biological treatment unit, with ammonia nitrogen sensor accuracy of 0.01 mg / L, nitrite sensor accuracy of 0.001 mg / L, and COD sensor accuracy of 1 mg / L. The advanced treatment sensor network requires pathogen sensors to be installed at the inlet and outlet of the ultraviolet sterilizer, with an accuracy of 1 CFU / mL to accurately detect changes in pathogen quantity.

[0063] Finally, the composition of the wastewater is detected based on the configured sensor network. Each sensor network collects real-time data at a set frequency: the pretreatment sensor network records the concentration of suspended particulate matter in the influent and effluent every 5 minutes; the biological treatment sensor network records ammonia nitrogen, nitrite, and chemical oxygen demand data every 10 minutes; and the deep treatment sensor network records the number of pathogens every 30 minutes. These real-time collected data are summarized and integrated to form aquaculture wastewater composition data that includes key component indicators before and after each treatment stage, fully reflecting the changes in the aquaculture wastewater during the treatment process.

[0064] By analyzing the monitoring needs of each node, configuring a matching sensor network, and conducting real-time monitoring, comprehensive and accurate collection of aquaculture wastewater components was achieved, providing a reliable data foundation for subsequent treatment parameter analysis and closed-loop control.

[0065] Furthermore, step A400 in the method provided in this application embodiment includes: A440: The microfilter-pretreatment channel, protein aeration packing-biological treatment channel, and ultraviolet sterilizer-deep treatment channel were analyzed and merged in series to build an aquaculture wastewater treatment channel.

[0066] A450: The aquaculture wastewater treatment channel is used to analyze the control parameters of the aquaculture wastewater composition data to determine the multi-node wastewater treatment control parameters.

[0067] A460: Based on the multi-node effluent treatment control parameters, perform effluent treatment control and node component monitoring to obtain multi-node effluent treatment component data, and perform closed-loop effluent treatment using the multi-node effluent treatment component data.

[0068] Specifically, firstly, the microfilter-pretreatment channel, the protein aeration packing-biological treatment channel, and the ultraviolet sterilizer-deep treatment channel are used as sub-model layers. An attention mechanism is introduced to dynamically allocate the sub-model weight factors and construct a parameter optimization layer. Then, the data input layer, sub-model layer, parameter optimization layer, and result output layer are connected in series to build an aquaculture wastewater treatment channel. The specific steps are explained in detail in A441-A443.

[0069] Next, the aquaculture wastewater treatment channel is used to analyze the control parameters of the aquaculture wastewater composition data. The aquaculture wastewater composition data collected in step A430 is input into the data input layer of the aquaculture wastewater treatment channel. This data covers key indicators such as the suspended particulate matter concentration at the pretreatment node, the ammonia nitrogen and nitrite concentration at the biological treatment node, and the number of pathogens at the deep treatment node, fully reflecting the initial state of the wastewater at each treatment stage.

[0070] After the aquaculture wastewater composition data enters the sub-model layer, the microfiltration machine-pretreatment channel, the protein aeration packing-biological treatment channel, and the ultraviolet sterilizer-deep treatment channel will analyze the corresponding input composition data, and each channel will output the pollutant concentration and removal rate after treatment.

[0071] Subsequently, these preliminary parameters enter the parameter optimization layer. The optimization layer first uses an attention mechanism to dynamically adjust the sub-model weights based on the current state of the component data at each node, weighting and fusing the preliminary parameters to form a comprehensive parameter set that takes into account the needs of each stage. Then, through NSGA-II multi-objective optimization, using the overall wastewater treatment objective as a benchmark, the comprehensive parameter set is globally optimized to balance treatment effectiveness, efficiency, and cost requirements, selecting a better parameter combination.

[0072] Subsequently, the specific operating parameters of each node were analyzed from the optimized parameter combination, which are the control parameters for multi-node effluent treatment. These parameters clarify the operating parameters of the microfilter in the pretreatment stage, the aeration and packing reaction parameters in the biological treatment stage, and the ultraviolet sterilization parameters in the advanced treatment stage, ensuring the coordinated operation of each node.

[0073] Finally, the state prediction and evaluation of the multi-node effluent treatment component data are used to obtain the predicted component parameters of the aquaculture effluent. The PID controller is used to adjust and analyze the predicted parameters based on the multi-node effluent treatment control parameters to determine the control parameter correction amount, and then the effluent is treated in a closed loop. The specific steps are explained in detail in A461-A462.

[0074] By inputting the wastewater composition data into the processing channel, and after preliminary analysis by the sub-model and global optimization by the optimization layer, the control parameters for multi-node wastewater treatment were determined, providing an operational basis for the precise treatment of wastewater.

[0075] Furthermore, step A440 in the method provided in this application embodiment includes: A441: The microfilter-pretreatment channel, protein aeration packing-biological treatment channel, and ultraviolet sterilizer-deep treatment channel are used as sub-model layers.

[0076] A442: Introduce an attention mechanism to dynamically allocate sub-model weight factors, and perform weighted fusion optimization on the sub-model layer according to the sub-model weight factors to construct a parameter optimization layer.

[0077] A443: Connect the data input layer, the sub-model layer, the parameter optimization layer, and the result output layer in series to build the aquaculture wastewater treatment channel.

[0078] In one embodiment, the microfilter-pretreatment channel, the protein aeration packing-biological treatment channel, and the ultraviolet sterilizer-deep treatment channel are first used as sub-model layers. The microfilter-pretreatment channel outputs corresponding treatment parameters based on the influent suspended particulate matter concentration to ensure the treated concentration is reduced to below 10 mg / L. The protein aeration packing-biological treatment channel outputs operating parameters that reduce pollutant concentrations such as ammonia nitrogen and nitrite to below 0.5 mg / L and 0.1 mg / L, respectively, based on influent parameters. The ultraviolet sterilizer-deep treatment channel outputs treatment parameters that reduce pathogen counts to below 100 CFU / mL based on the influent pathogen count. These three sub-model layers correspond to the three key treatment stages of effluent treatment, each undertaking a specific treatment function.

[0079] Subsequently, an attention mechanism is introduced to dynamically allocate the sub-model weight factors. The parameters of the model layer are weighted and fused according to the sub-model weight factors to obtain the parameter fusion functional layer. The output parameters of this layer are optimized by NSGA-II multi-objective optimization according to the overall goal of aquaculture wastewater treatment to obtain the global optimization functional layer. The two are then nested and connected to construct the parameter optimization layer. The specific steps are explained in detail in A442-1-A442-3.

[0080] Finally, the data input layer, sub-model layer, parameter optimization layer, and result output layer are connected in series. The data input layer receives real-time composition data of aquaculture wastewater collected hourly from the aforementioned sensor network, including indicators such as suspended particulate matter, ammonia nitrogen, and pathogens. The sub-model layer outputs the processing parameters for each stage based on the input data. The parameter optimization layer performs weighted fusion and optimization of these parameters to ensure that the overall treatment effect meets the target. The result output layer outputs the various indicators of the final treated wastewater, such as suspended particulate matter ≤10mg / L, ammonia nitrogen ≤0.5mg / L, and pathogens ≤100CFU / mL, forming a complete aquaculture wastewater treatment channel.

[0081] By treating each stage of the treatment channel as a sub-model, dynamically allocating weights to optimize parameters, and merging each functional layer in series, a treatment channel for aquaculture wastewater that can adapt to the dynamic changes in wastewater composition was built, achieving coordinated and efficient operation of each treatment stage.

[0082] Furthermore, step A442 in the method provided in this application embodiment includes: A442-1: Perform parameter weighting and fusion on the sub-model layer according to the sub-model weight factor to obtain the parameter fusion functional layer.

[0083] A442-2: Based on the overall goal of aquaculture wastewater treatment, the output parameters of the parameter fusion functional layer are optimized using NSGA-II multi-objective optimization to obtain a global optimization functional layer.

[0084] A442-3: The parameter fusion function layer and the global optimization function layer are nested and connected to construct the parameter optimization layer.

[0085] Optionally, when constructing the parameter optimization layer, the parameters of the sub-model layer are first weighted and fused according to the sub-model weight factors to obtain the parameter fusion functional layer. The sub-model layer consists of a microfilter-pretreatment channel, a protein aeration packing-biological treatment channel, and an ultraviolet sterilizer-deep treatment channel. Its output parameters include the pollutant concentration and removal rate after treatment in each channel. For example, the microfilter-pretreatment channel outputs a suspended particulate matter concentration of 8 mg / L and a removal rate of 92%, the protein aeration packing-biological treatment channel outputs an ammonia nitrogen concentration of 0.4 mg / L and a removal rate of 92%, and the ultraviolet sterilizer-deep treatment channel outputs a pathogen count of 90 CFU / mL and a kill rate of 91%.

[0086] The sub-model weight factors are dynamically allocated by the attention mechanism. Assuming the current weight factors are 0.3, 0.5, and 0.2, during weighted fusion, the output parameters of each channel are calculated into a comprehensive value according to their weights. For example, the comprehensive influence value of suspended particulate matter is 8 × 0.3 = 2.4, the comprehensive influence value of ammonia nitrogen is 0.4 × 0.5 = 0.2, and the comprehensive influence value of pathogens is 90 × 0.2 = 18. These comprehensive values ​​are integrated to form the output parameter set of the parameter fusion functional layer. This output parameter set not only retains the processing characteristics of each channel but also reflects the relative importance of different stages.

[0087] Next, the NSGA-II multi-objective optimization uses the overall objective of aquaculture wastewater treatment as a benchmark to evaluate the output parameter set from multiple dimensions. The core objectives of the overall aquaculture wastewater treatment objective are: suspended particulate matter ≤10mg / L, ammonia nitrogen ≤0.5mg / L, and pathogens ≤100CFU / mL. Simultaneously, it can consider treatment efficiency ≥100m³ / h and energy consumption ≤80kW·h for operating costs. When the output parameter set shows an ammonia nitrogen concentration of 0.6mg / L that does not meet the target, the algorithm identifies the actual index corresponding to that parameter deviating from the target and then adjusts the parameter combination of the sub-model layer accordingly: for example, increasing the aeration intensity of the biological treatment channel or increasing the biological packing filling rate to reduce the ammonia nitrogen concentration to 0.4mg / L. The corresponding comprehensive value is then updated to 0.4×0.5=0.2, meeting the ammonia nitrogen compliance requirement.

[0088] During the adjustment process, the algorithm simultaneously verifies whether the actual indicators corresponding to other comprehensive values ​​still meet the objectives: if the comprehensive value of suspended particulate matter of 2.4 corresponds to an actual concentration of 8 mg / L ≤ 10 mg / L, and the comprehensive value of pathogens of 18 corresponds to an actual quantity of 90 CFU / mL ≤ 100 CFU / mL, and the treatment efficiency is maintained at 100 m³ / h while energy consumption is reduced to 75 kW·h, then this parameter combination is selected as the global optimization parameter through non-dominated ranking and crowding calculation. Non-dominated ranking means that no other combination is better than the others in all objectives. The final output of the global optimization functional layer is the comprehensive value and corresponding actual parameters selected through the above process, balancing all objectives. Examples include a comprehensive value of suspended particulate matter of 2.4, ammonia nitrogen of 0.2, and a comprehensive value of pathogens of 18.

[0089] Finally, the parameter fusion layer and the global optimization layer are nested and connected. This nesting means that the output of the global optimization layer is based on the basic parameters of the parameter fusion layer. Simultaneously, when the composition of the effluent fluctuates, the parameter fusion layer updates the basic parameters in real time, and the global optimization layer re-optimizes accordingly, forming a dynamic response mechanism. For example, when the concentration of suspended particulate matter after pretreatment increases, the parameter fusion layer updates the overall value, and the global optimization layer immediately recalculates and adjusts the correlation parameters between biological treatment and advanced treatment to ensure that the overall treatment effect still meets the overall target.

[0090] By weighted fusion of sub-model parameters, multi-objective optimization to balance global objectives, and nested connections to achieve dynamic response, a parameter optimization layer is constructed that can take into account the characteristics of each treatment stage and the overall objectives, thereby improving the comprehensive efficiency of wastewater treatment.

[0091] Furthermore, step A400 in the method provided in this application embodiment includes: A410: Perform state prediction and evaluation on the multi-node effluent treatment component data to obtain predicted component parameters of aquaculture effluent.

[0092] A420: A PID controller is used to regulate and analyze the predicted component parameters of the aquaculture wastewater based on the multi-node wastewater treatment control parameters, determine the control parameter correction amount, and perform closed-loop treatment of the wastewater through the control parameter correction amount.

[0093] In one embodiment, the aquaculture wastewater composition data acquired in step A430 at the pretreatment, biological treatment, and deep treatment nodes are first integrated, including the concentration of suspended particulate matter after pretreatment, the concentrations of ammonia nitrogen and nitrite after biological treatment, and the number of pathogens after deep treatment. Next, a time-series prediction model is constructed: first, historical time-series data of the target object is collected, such as data on the changes in wastewater composition at each node over time; the data is preprocessed, including cleaning and normalization; then, a recurrent neural network-based architecture is selected, and the time-series prediction model is trained using the preprocessed historical data; the model is optimized by iteratively adjusting the network parameters so that the time-series prediction model can capture the temporal dependencies of the data; finally, the model's prediction accuracy is evaluated by validating the data, and a usable time-series prediction model is determined.

[0094] Based on the aforementioned real-time aquaculture wastewater composition data, combined with the composition change patterns formed during historical treatment processes, a pre-constructed time series prediction model is used to predict the wastewater composition for the next 1-2 hours, obtaining predicted composition parameters for aquaculture wastewater. For example, if the current ammonia nitrogen concentration at the biological treatment node is 0.4 mg / L, and considering the upward trend of 0.05 mg / L per hour over the past 3 hours, it can be predicted that the ammonia nitrogen concentration will reach 0.45 mg / L after 1 hour. This value is one part of the predicted composition parameters for aquaculture wastewater. Complete predicted composition parameters for aquaculture wastewater also include predicted values ​​for suspended particulate matter, pathogens, etc., during the same period.

[0095] Next, a PID controller was used to analyze and regulate the predicted component parameters of the aquaculture wastewater based on the control parameters of the multi-node wastewater treatment. First, the target parameters for each node were defined: ammonia nitrogen ≤ 0.5 mg / L, suspended particulate matter ≤ 10 mg / L, and pathogens ≤ 100 CFU / mL. The predicted component parameters of the aquaculture wastewater were then compared with the target parameters, and the deviation was calculated. For example, if the predicted ammonia nitrogen concentration after 1 hour is 0.55 mg / L, and the target is 0.5 mg / L, the deviation is 0.05 mg / L; if the predicted suspended particulate matter concentration is 9 mg / L, and the target is 10 mg / L, the deviation is -1 mg / L.

[0096] The PID controller then calculates the correction amount for the control parameters based on the proportional, integral, and derivative terms of these deviations. The proportional, integral, and derivative terms represent the current deviation magnitude, the cumulative deviation over a period of time, and the rate of change of the deviation, respectively. In the example above, for an ammonia nitrogen deviation of 0.05 mg / L, if its rate of change is 0.03 mg / L per hour, the PID controller can output a correction amount for the aeration intensity of the biological treatment node, such as increasing it by 0.2 m³ / h to enhance the degradation effect of microorganisms. For negative deviations in suspended particulate matter, the controller can maintain the backwashing frequency of the pretreatment node unchanged, without requiring additional correction.

[0097] After determining the control parameter correction amounts, they are applied to adjust the operating parameters of each treatment node: for example, adjusting the aerator operating parameters of the biological treatment node based on the correction amount for aeration intensity, and adjusting the microfilter parameters of the pretreatment node based on the possible backwashing frequency correction amount. After adjustment, the treated component data is monitored again through the sensor network of each node to form new multi-node effluent treatment component data, and the system re-enters the state prediction and assessment stage to begin the next round of prediction and control, thereby achieving closed-loop effluent treatment.

[0098] By predicting the changing trends of effluent composition in advance and combining this with the dynamic calculation of parameter corrections by the PID controller and the implementation of closed-loop adjustments, real-time optimization of the effluent treatment process is achieved, ensuring that the effluent composition is stably controlled within the target range.

[0099] In summary, the wastewater treatment method based on the factory farming of whiteleg shrimp provided in this application has the following technical effects: This application collects component data from the pretreatment, biological treatment, and deep treatment nodes of the wastewater from the factory farming of white shrimp. Through index extraction, parameter screening, recurrent neural network training, and NSGA-II multi-objective optimization, it obtains the treatment parameters and effect data for each node. It calculates the control parameters and corrections for multiple nodes and adjusts them in conjunction with the overall wastewater treatment objective and the treatment objectives of each node. This allows for the precise treatment of pollutants such as suspended particulate matter, ammonia nitrogen, and pathogens in the wastewater, ensuring stable, efficient, and reliable wastewater treatment. The application achieves the technical benefits of high-efficiency treatment of white shrimp factory farming wastewater, reduced treatment costs, easier operation, and better synergy among various stages.

[0100] Example 2, as Figure 2 As shown, based on the same inventive concept as in Embodiment 1 above, this application provides a wastewater treatment system for factory farming of whiteleg shrimp, the system comprising: Wastewater purification device construction module 1 is used to obtain an integrated wastewater purification device, which consists of a microfilter, a protein separator, an aerator, a biological packing reactor, and an ultraviolet sterilizer.

[0101] Processing node acquisition module 2 is used to extract nodes from the wastewater treatment process of factory farming of white shrimp in South America, and obtain pretreatment nodes, biological treatment nodes and deep treatment nodes.

[0102] The processing channel acquisition module 3 is used to associate and map the pretreatment node, biological treatment node and deep treatment node with the components in the integrated wastewater purification device to obtain the microfilter-pretreatment channel, the protein aeration packing-biological treatment channel and the ultraviolet sterilizer-deep treatment channel.

[0103] The wastewater closed-loop treatment module 4 is used to collect aquaculture wastewater composition data and perform control parameter analysis and wastewater closed-loop treatment based on the microfilter-pretreatment channel, protein aeration packing-biological treatment channel and ultraviolet sterilizer-deep treatment channel.

[0104] Furthermore, the processing channel acquisition module 3 is used to perform the following steps: The pretreatment node, biological treatment node, and deep treatment node are mapped to the components in the integrated effluent purification device to obtain the microfilter-pretreatment node, protein separator, aerator, biological packing reactor-biological treatment node, and ultraviolet sterilizer-deep treatment node. Based on these nodes, the effluent pretreatment target, effluent biological treatment target, and effluent deep treatment target are determined. Historical data mining is performed on the microfilter-pretreatment node, protein separator, aerator, biological packing reactor-biological treatment node, and ultraviolet sterilizer-deep treatment node to obtain the effluent pretreatment parameter space, effluent biological treatment parameter space, and effluent deep treatment parameter space. Control parameters are analyzed according to the effluent pretreatment target, effluent biological treatment target, and effluent deep treatment target to construct the microfilter-pretreatment channel, protein aeration packing-biological treatment channel, and ultraviolet sterilizer-deep treatment channel.

[0105] Furthermore, the processing channel acquisition module 3 is used to perform the following steps: The targets for effluent pretreatment, biological treatment, and advanced treatment are sequentially indexed to obtain sets of effluent pretreatment effect indicators, biological treatment effect indicators, and advanced treatment effect indicators. The effluent pretreatment parameter space, biological treatment parameter space, and advanced treatment parameter space are then evaluated according to these sets to obtain sets of effluent pretreatment effect indicators, biological treatment effect indicators, and advanced treatment effect indicators. Finally, the effluent pretreatment effect indicators are applied to these sets. The treatment effect set, effluent biological treatment effect set, and effluent deep treatment effect set are used to screen control parameters in the effluent pretreatment parameter space, effluent biological treatment parameter space, and effluent deep treatment parameter space, respectively, to obtain usable effluent pretreatment parameter sets, usable effluent biological treatment parameter sets, and usable effluent deep treatment parameter sets. The usable effluent pretreatment parameter sets, usable effluent biological treatment parameter sets, and usable effluent deep treatment parameter sets are then subjected to control training and optimization to obtain the microfilter-pretreatment channel, protein aeration packing-biological treatment channel, and ultraviolet sterilizer-deep treatment channel.

[0106] Furthermore, the processing channel acquisition module 3 is used to perform the following steps: The available effluent pretreatment parameter set, available effluent biological treatment parameter set, and available effluent advanced treatment parameter set are integrated according to time series to obtain effluent pretreatment sequence parameter set, effluent biological treatment sequence parameter set, and effluent advanced treatment sequence parameter set. These sets are then controlled and labeled to obtain effluent pretreatment sequence sample set, effluent biological treatment sequence sample set, and effluent advanced treatment sequence sample set. A recurrent neural network is used to control, train, and optimize these sets to obtain the microfilter-pretreatment channel, protein aeration packing-biological treatment channel, and ultraviolet sterilizer-advanced treatment channel.

[0107] Furthermore, the tailwater closed-loop treatment module 4 is used to perform the following steps: Monitoring requirements are analyzed for the pretreatment node, biological treatment node, and deep treatment node respectively to obtain monitoring requirement parameters for pretreatment, biological treatment, and deep treatment. Based on these parameters, a pretreatment sensor network, a biological treatment sensor network, and a deep treatment sensor network are configured. Wastewater composition is detected using these sensor networks to obtain aquaculture wastewater composition data.

[0108] Furthermore, the tailwater closed-loop treatment module 4 is used to perform the following steps: The microfiltration machine-pretreatment channel, protein aeration packing-biological treatment channel, and ultraviolet sterilizer-deep treatment channel were analyzed and merged in series to construct an aquaculture wastewater treatment channel. The aquaculture wastewater composition data were analyzed using this channel to determine multi-node wastewater treatment control parameters. Based on these multi-node control parameters, wastewater treatment control and node component monitoring were performed to obtain multi-node wastewater composition data. Finally, closed-loop wastewater treatment was implemented using this multi-node composition data.

[0109] Furthermore, the tailwater closed-loop treatment module 4 is used to perform the following steps: The microfilter-pretreatment channel, protein aeration packing-biological treatment channel, and ultraviolet sterilizer-deep treatment channel are used as sub-model layers. An attention mechanism is introduced to dynamically allocate sub-model weight factors. The sub-model layers are then weighted and fused according to the sub-model weight factors to construct a parameter optimization layer. The data input layer, the sub-model layers, the parameter optimization layer, and the result output layer are connected in series to build the aquaculture wastewater treatment channel.

[0110] Furthermore, the tailwater closed-loop treatment module 4 is used to perform the following steps: The parameters of the sub-model layer are weighted and fused according to the sub-model weight factors to obtain the parameter fusion functional layer; the output parameters of the parameter fusion functional layer are optimized by NSGA-II multi-objective optimization according to the overall goal of aquaculture wastewater treatment to obtain the global optimization functional layer; the parameter fusion functional layer and the global optimization functional layer are nested and connected to construct the parameter optimization layer.

[0111] Furthermore, the tailwater closed-loop treatment module 4 is used to perform the following steps: The state prediction and evaluation of the multi-node effluent treatment component data are performed to obtain the predicted component parameters of aquaculture effluent; a PID controller is used to regulate and analyze the predicted component parameters of aquaculture effluent based on the control parameters of the multi-node effluent treatment, to determine the control parameter correction amount, and to perform closed-loop treatment of effluent through the control parameter correction amount.

[0112] The wastewater treatment system based on the industrialized farming of white shrimp provided in this embodiment of the invention can execute the wastewater treatment method based on the industrialized farming of white shrimp provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0113] Although this application makes various references to certain modules in the system according to the embodiments of this application, any number of different modules can be used and run on user terminals and / or servers. The various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy distinction between each other and are not used to limit the scope of protection of this invention.

[0114] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application. In some cases, the actions or steps described in this application can be performed in a different order than that shown in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A method for treating wastewater from factory farming of whiteleg shrimp, characterized in that, The method includes: An integrated wastewater purification device is obtained, which consists of a microfilter, a protein separator, an aerator, a bio-packed reactor, and an ultraviolet sterilizer. The wastewater treatment process of factory farming of white shrimp in South America was analyzed to extract nodes, namely pretreatment nodes, biological treatment nodes, and deep treatment nodes. The pretreatment node, biological treatment node and deep treatment node are associated and mapped with the components in the integrated wastewater purification device to obtain the microfilter-pretreatment channel, the protein aeration packing-biological treatment channel and the ultraviolet sterilizer-deep treatment channel. Data on the composition of aquaculture wastewater is collected, and control parameters are analyzed and closed-loop treatment of the wastewater is performed based on the microfilter-pretreatment channel, protein aeration packing-biological treatment channel, and ultraviolet sterilizer-deep treatment channel.

2. The wastewater treatment method based on the industrialized farming of whiteleg shrimp as described in claim 1, characterized in that, The obtained microfiltration machine-pretreatment channel, protein aeration packing-biological treatment channel, and ultraviolet sterilizer-deep treatment channel include: The pretreatment node, biological treatment node, and deep treatment node are associated and mapped with the components in the integrated wastewater purification device to obtain the microfilter-pretreatment node, protein separator, aerator, biological packing reactor-biological treatment node, and ultraviolet sterilizer-deep treatment node. Based on the pretreatment nodes, biological treatment nodes, and advanced treatment nodes, determine the effluent pretreatment targets, effluent biological treatment targets, and effluent advanced treatment targets; Historical data mining was performed on the microfilter-pretreatment node, protein separator, aerator, biological packing reactor-biological treatment node, and ultraviolet sterilizer-deep treatment node to obtain the effluent pretreatment parameter space, effluent biological treatment parameter space, and effluent deep treatment parameter space. Based on the stated effluent pretreatment objectives, effluent biological treatment objectives, and effluent advanced treatment objectives, control parameters are analyzed for the effluent pretreatment parameter space, effluent biological treatment parameter space, and effluent advanced treatment parameter space to construct a microfilter-pretreatment channel, a protein aeration packing-biological treatment channel, and an ultraviolet sterilizer-advanced treatment channel.

3. The wastewater treatment method based on the industrialized farming of whiteleg shrimp as described in claim 2, characterized in that, The construction of the microfiltration machine-pretreatment channel, the protein aeration packing-biological treatment channel, and the ultraviolet sterilizer-deep treatment channel includes: The indicators for the wastewater pretreatment target, wastewater biological treatment target and wastewater advanced treatment target are extracted sequentially to obtain the wastewater pretreatment effect indicator set, wastewater biological treatment effect indicator set and wastewater advanced treatment effect indicator set. The effect of the effluent pretreatment effect index set, effluent biological treatment effect index set, and effluent deep treatment effect index set is evaluated on the effluent pretreatment parameter space, effluent biological treatment parameter space, and effluent deep treatment parameter space to obtain the effluent pretreatment effect set, effluent biological treatment effect set, and effluent deep treatment effect set. The tailwater pretreatment effect set, tailwater biological treatment effect set, and tailwater deep treatment effect set are used to screen control parameters for the tailwater pretreatment parameter space, tailwater biological treatment parameter space, and tailwater deep treatment parameter space, respectively, to obtain the usable tailwater pretreatment parameter set, usable tailwater biological treatment parameter set, and usable tailwater deep treatment parameter set. The available effluent pretreatment parameter set, available effluent biological treatment parameter set, and available effluent deep treatment parameter set were controlled, trained, and optimized respectively to obtain the microfilter-pretreatment channel, protein aeration packing-biological treatment channel, and ultraviolet sterilizer-deep treatment channel.

4. The wastewater treatment method based on the industrialized farming of whiteleg shrimp as described in claim 3, characterized in that, The process of obtaining the microfiltration machine-pretreatment channel, the protein aeration packing-biological treatment channel, and the ultraviolet sterilizer-deep treatment channel includes: The available wastewater pretreatment parameter set, available wastewater biological treatment parameter set, and available wastewater advanced treatment parameter set are integrated according to time series to obtain the wastewater pretreatment sequence parameter set, wastewater biological treatment sequence parameter set, and wastewater advanced treatment sequence parameter set; The tailwater pretreatment sequence parameter set, tailwater biological treatment sequence parameter set, and tailwater advanced treatment sequence parameter set are respectively controlled and identified to obtain tailwater pretreatment sequence sample set, tailwater biological treatment sequence sample set, and tailwater advanced treatment sequence sample set; Recurrent neural networks were used to control, train, and optimize the tailwater pretreatment sequence sample set, tailwater biological treatment sequence sample set, and tailwater deep treatment sequence sample set, respectively, to obtain the microfilter-pretreatment channel, protein aeration packing-biological treatment channel, and ultraviolet sterilizer-deep treatment channel.

5. The wastewater treatment method based on the industrialized farming of whiteleg shrimp as described in claim 1, characterized in that, The data collected on the composition of aquaculture wastewater includes: Monitoring requirements were analyzed for the pretreatment node, biological treatment node, and deep treatment node respectively, and monitoring requirements parameters for pretreatment, biological treatment, and deep treatment were obtained. Configure the pretreatment sensor network, biological treatment sensor network, and deep treatment sensor network according to the pretreatment monitoring requirement parameters, biological treatment monitoring requirement parameters, and deep treatment monitoring requirement parameters. The composition of the aquaculture wastewater is detected based on the pretreatment sensor network, biological treatment sensor network, and deep treatment sensor network to obtain the composition data of the aquaculture wastewater.

6. The wastewater treatment method based on the industrialized farming of whiteleg shrimp as described in claim 1, characterized in that, The process of analyzing control parameters and performing closed-loop treatment of the aquaculture wastewater composition data based on the microfiltration-pretreatment channel, protein aeration packing-biological treatment channel, and ultraviolet sterilizer-deep treatment channel includes: The microfilter-pretreatment channel, protein aeration packing-biological treatment channel, and ultraviolet sterilizer-deep treatment channel were analyzed and combined in series to construct an aquaculture wastewater treatment channel; The aquaculture wastewater treatment channel is used to analyze the composition data of the aquaculture wastewater and determine the control parameters for multi-node wastewater treatment. Based on the multi-node effluent treatment control parameters, effluent treatment control and node component monitoring are performed to obtain multi-node effluent treatment component data, and effluent closed-loop treatment is performed using the multi-node effluent treatment component data.

7. The wastewater treatment method based on the industrialized farming of whiteleg shrimp as described in claim 6, characterized in that, The construction of the aquaculture wastewater treatment channel includes: The microfilter-pretreatment channel, the protein aeration packing-biological treatment channel, and the ultraviolet sterilizer-deep treatment channel are used as sub-model layers; An attention mechanism is introduced to dynamically allocate sub-model weight factors, and the sub-model layer is weighted and fused according to the sub-model weight factors to construct a parameter optimization layer; The data input layer, the sub-model layer, the parameter optimization layer, and the result output layer are connected in series to build the aquaculture wastewater treatment channel.

8. The wastewater treatment method based on the industrialized farming of whiteleg shrimp as described in claim 7, characterized in that, The construction parameter optimization layer includes: The sub-model layer is weighted and fused according to the sub-model weight factor to obtain the parameter fusion functional layer; the output parameters of the parameter fusion functional layer are optimized by NSGA-II multi-objective optimization according to the overall goal of aquaculture wastewater treatment to obtain the global optimization functional layer. The parameter fusion function layer and the global optimization function layer are nested and connected to construct the parameter optimization layer.

9. A wastewater treatment method based on factory farming of whiteleg shrimp as described in claim 6, characterized in that, The closed-loop treatment of wastewater using the multi-node wastewater treatment component data includes: The state prediction and evaluation of the multi-node effluent treatment component data are performed to obtain the predicted component parameters of aquaculture effluent. A PID controller is used to regulate and analyze the predicted component parameters of the aquaculture wastewater based on the multi-node wastewater treatment control parameters, determine the control parameter correction amount, and perform closed-loop treatment of the wastewater using the control parameter correction amount.

10. A wastewater treatment system based on the industrialized farming of whiteleg shrimp, characterized in that, A system for implementing the wastewater treatment method based on the industrialized farming of whiteleg shrimp according to any one of claims 1-9, the system comprising: A tailwater purification device construction module is used to obtain an integrated tailwater purification device, which consists of a microfilter, a protein separator, an aerator, a biological packing reactor, and an ultraviolet sterilizer. The processing node acquisition module is used to extract nodes from the wastewater treatment process of factory farming of white shrimp in South America, and obtain pretreatment nodes, biological treatment nodes and deep treatment nodes. The processing channel acquisition module is used to associate and map the pretreatment node, biological treatment node, and deep treatment node with the components in the integrated wastewater purification device to obtain the microfilter-pretreatment channel, the protein aeration packing-biological treatment channel, and the ultraviolet sterilizer-deep treatment channel; the wastewater closed-loop treatment module is used to collect aquaculture wastewater composition data and perform control parameter analysis and wastewater closed-loop treatment on the aquaculture wastewater composition data based on the microfilter-pretreatment channel, the protein aeration packing-biological treatment channel, and the ultraviolet sterilizer-deep treatment channel.

Images