A rice-snail-fish symbiotic system constructed using tilapia aquaculture wastewater and a breeding method based on the symbiotic system.

Abstract

This invention discloses a rice-snail-fish symbiotic system constructed using tilapia aquaculture wastewater and a breeding method based on the symbiotic system. It belongs to the field of wastewater symbiotic aquaculture technology. The rice-snail-fish symbiotic system includes experimental terraces; a rice planting area; a snail aquaculture area; a fish aquaculture area; a wastewater treatment unit; and a recirculating aquaculture system. The breeding method includes constructing a "rice-snail-fish" symbiotic ecosystem; ex-situ treatment of tilapia aquaculture wastewater; and using deep learning to construct a digital model to accurately screen genes beneficial to wastewater treatment, thereby achieving molecular-level regulation of wastewater treatment. This invention uses aquaculture wastewater as a living environment for rice and symbiotic organisms, reducing the loss of freshwater resources and other raw materials, improving the efficiency of material reuse, and breeding "rice-snail-fish" biological species to obtain species more suitable for this ecosystem.

Description

Technical Field

[0001] This invention relates to the field of tailwater symbiotic aquaculture technology, specifically to a rice-snail-fish symbiotic system constructed using tilapia aquaculture tailwater and a breeding method based on the symbiotic system. Background Technology

[0002] This study aimed to predict key water quality parameters for red tilapia farming in an outdoor recirculating aquaculture system, including dissolved oxygen, pH, total ammonia nitrogen, nitrite nitrogen, and alkalinity. First, a random forest model was used to identify the key features for each parameter. Then, convolutional neural networks (CNN), long short-term memory networks (LSTM), and CNN-LSTM models were used for prediction. The results showed that the 5000-episode CNN-LSTM model performed well in predicting dissolved oxygen, total ammonia nitrogen, nitrite nitrogen, and alkalinity, but its pH prediction efficiency was low.

[0003] The aforementioned algorithms have poor adaptability to wastewater treatment scenarios. Based on aquaculture farm environments, they do not demonstrate specificity for the "rice-snail-fish" ecosystem. Furthermore, the materials and methods applied under different treatment conditions can significantly impact the final results. Existing algorithms for water quality prediction or regulation are mostly designed for general aquaculture water bodies, failing to consider the unique characteristics of the "planting field-wastewater treatment" linkage scenario. For example, the algorithms do not consider the dynamic impacts of nitrogen and phosphorus absorption by rice roots, snail filter feeding on suspended matter, and fish activity on dissolved oxygen in the planting field. They only model based on the temporal changes or spatial distribution of water quality parameters, leading to a disconnect between regulatory commands and the needs of the symbiotic system, failing to achieve synergistic optimization of wastewater purification and ecological field productivity. Existing algorithms do not incorporate data on abiotic disturbances (such as the impact of light intensity and rainfall on rice transpiration) and biotic disturbances (such as algal competition for nitrogen and phosphorus and the impact of microbial community changes on ammonia nitrogen degradation) in the planting field. For example, when predicting the total phosphorus concentration in the wastewater, the feedback effect of phosphorus loss from the soil of the planting fields caused by rainwater erosion on the wastewater retreatment was not considered, which led to the idealization of the algorithm control results. In actual applications, the wastewater purification effect fluctuated by more than 20%, making it difficult to stably meet the water quality requirements of the "rice-snail-fish" symbiotic system. Therefore, it is necessary to design a rice-snail-fish symbiotic system constructed using tilapia farming wastewater and a breeding method based on the symbiotic system. Summary of the Invention

[0004] The purpose of this invention is to provide a rice-snail-fish symbiotic system constructed using tilapia aquaculture wastewater and a breeding method based on the symbiotic system, so as to solve the problems mentioned in the background art.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a rice-snail-fish symbiotic system constructed using tilapia farming wastewater, comprising:

[0006] Experimental terraced fields;

[0007] A rice-growing area, which is set up in the experimental terraces for growing rice;

[0008] A snail farming area, which is set up in the experimental terraced fields for the purpose of raising snails;

[0009] A fish farming area, which is set up in the experimental terraces for raising fish;

[0010] The wastewater treatment unit is used to perform graded sedimentation and filtration treatment on the wastewater from tilapia farming.

[0011] A recirculating aquaculture system is used to irrigate the experimental terraced fields.

[0012] In a further embodiment, the wastewater treatment unit includes a sedimentation tank and a filtration device. The sedimentation tank is used to perform graded sedimentation on the tilapia farming wastewater, and the filtration device is used to perform fine filtration on the graded sedimented tilapia farming wastewater.

[0013] In a further embodiment, a flow velocity sensor is provided at the inlet and outlet of the experimental terrace.

[0014] A breeding method for selecting and breeding tilapia symbiotic systems constructed using tilapia aquaculture wastewater includes the following steps:

[0015] Step 1: Select a suitable area to build experimental terraced fields, and perform graded sedimentation treatment on the aquaculture wastewater. In the spring season, select indica rice varieties to plant in the rice planting area, establish a recirculating aquaculture system, use normal water sources for irrigation, and add snails such as field snails to the snail breeding area of ​​the experimental terraced fields. When the rice grows to the waxy maturity stage, add carp, crucian carp and other fish to the fish breeding area of ​​the experimental terraced fields. Set up flow velocity sensors at the inlet and outlet of the experimental terraced fields to measure indicators such as suspended solids concentration and snail filter feeding amount, thereby building a "rice-snail-fish" symbiotic ecosystem. During the construction process, control the biological population density of the symbiotic ecosystem, set up at least 5 experimental groups and perform 3 replicates.

[0016] Step 2: The tilapia farming wastewater is pretreated through a wastewater treatment unit. After grading and sedimentation, the wastewater is treated into a suitable water source for the symbiotic ecosystem through fine filtration, oxygenation and oxidation.

[0017] When the "rice-snail-fish" symbiotic ecosystem is stable, the water source of the experimental group is replaced with tilapia farming wastewater for ex-situ treatment. A unique blank group is set up, and corresponding indicators of rice and fish growth, snail filter feeding amount and ammonia nitrogen degradation rate, as well as chemical indicators such as biological oxygen demand, chemical oxygen demand, total nitrogen, total phosphorus, suspended solids and heavy metals are measured. Bioinformatics technology is used to analyze and identify relevant genes that are beneficial to the treatment of aquaculture wastewater.

[0018] Step 3: Preprocess the collected dataset by fusing three types of data: physical parameters of the wastewater treatment device, ecological parameters of the experimental terraces, and environmental disturbance parameters. Use a three-step logical process of spatiotemporal alignment, outlier removal, and feature mapping. Calculate the importance of each feature to the wastewater purification efficiency using a random forest model and assign fusion weights.

[0019] Step 4: Abstract the "rice-snail-fish" symbiotic ecosystem into a graph structure, then construct a graph neural network, and use the graph attention mechanism to analyze key substances and their pathways. Combining bioinformatics and transcriptomics methods, and using visualized data such as differentially expressed genes, accurately screen genes that are beneficial to wastewater treatment, and achieve regulation of wastewater treatment at the molecular level.

[0020] In a further embodiment, in step one, to ensure the quality of the rice-snail-fish, during the construction of the "rice-snail-fish" symbiotic ecosystem, the fish were allowed to feed naturally during the experimental period, that is, to consume rice and aquatic microorganisms, etc., and were raised for 3 months. After the experiment, the fish were anesthetized and their weight was measured.

[0021] In a further embodiment, in step one, during the construction of the "rice-snail-fish" symbiotic ecosystem, the number of fish in the group should not exceed the upper limit of the symbiotic ecosystem, with each group consisting of 100-150 fish, the rice planting density being appropriate, and the snails and various plankton varying.

[0022] In a further embodiment, in step three, regarding spatiotemporal alignment, the data from the wastewater treatment device is downsampled based on the data acquisition time of the planting field sensors, and the environmental data is interpolated to ensure uniformity in the time dimension; spatially, the inlet and outlet data of the wastewater treatment device are associated with the corresponding experimental terrace inlet data to establish a spatial mapping relationship between the treatment unit and the planting area; regarding outlier removal, the 3σ rule based on standard deviation is adopted to remove data that exceeds the range of "mean ± 3σ", and the data is supplemented by linear interpolation of adjacent time data.

[0023] In a further embodiment, the calculation formula for the random forest model in step three is as follows:

[0024] Among them, X fusion These are the fused feature values. i Let X be the weight of the i-th feature. i The standardized value of the i-th feature is given, and the weights are obtained through training with 1000 sets of experimental data. In a further embodiment, the graph structure in step four includes nodes representing tailwater treatment units, experimental terraces, functional areas, and environmental factors. Edges represent the material or energy flow relationships between nodes, and edge weights are calculated based on the material and energy transfer efficiency between nodes.

[0025] In a further embodiment, the graph neural network construction formula in step four is as follows:

[0026] ;

[0027] in, Let be the attention coefficient of node i to node j. Let W be the attention vector, W be the weight matrix, and h be the weight vector. i h j These are the feature vectors of nodes i and j, respectively, and N is the sum of the feature vectors of nodes i and j. i Let i be the set of neighboring nodes of node i. Using this formula, the filter device → rice paddy area is identified as the key material flow path, and the parameters of this path are adjusted preferentially.

[0028] Compared with the prior art, the beneficial effects achieved by the present invention are:

[0029] 1. Construct a "rice-snail-fish" multitrophic ecosystem, integrate tilapia farming wastewater, and use the farming wastewater as a living environment for rice and symbiotic organisms. This reduces the loss of raw materials such as freshwater resources, improves the efficiency of material reuse, and allows for a more comprehensive analysis of the interactions between various biological members and between biotic and abiotic factors by using the ecosystem to study the wastewater.

[0030] 2. A dynamic and visualized mathematical model is established using deep learning algorithms. By monitoring key parameters such as nitrogen and phosphorus in the water quality, the model predicts data. Through big data prediction and dynamic visualization of the model, changes occurring in the ecosystem are tracked, key genes in the tailwater are screened and treated, and the entire related gene family is further analyzed. The biological species of "rice-snail-fish" are selected and bred to obtain species more suitable for this ecosystem. Attached Figure Description

[0032] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0034] In the following description, numerous specific details are set forth in order to provide a more thorough understanding of the invention. However, it will be apparent to those skilled in the art that the invention can be practiced without one or more of these details. In other instances, certain technical features well-known in the art have not been described in order to avoid obscuring the invention.

[0035] Please see Figure 1 This invention provides a technical solution: a rice-snail-fish symbiotic system constructed using tilapia aquaculture wastewater, comprising:

[0036] Experimental terraced fields;

[0037] A rice-growing area, which is set up in the experimental terraces for growing rice;

[0038] A snail farming area, which is set up in the experimental terraced fields for the purpose of raising snails;

[0039] A fish farming area, which is set up in the experimental terraces for raising fish;

[0040] The wastewater treatment unit is used to perform graded sedimentation and filtration treatment on the wastewater from tilapia farming.

[0041] A recirculating aquaculture system is used to irrigate the experimental terraced fields.

[0042] In a further embodiment, the wastewater treatment unit includes a sedimentation tank and a filtration device. The sedimentation tank is used to perform graded sedimentation on the tilapia farming wastewater, and the filtration device is used to perform fine filtration on the graded sedimented tilapia farming wastewater.

[0043] In a further embodiment, a flow velocity sensor is provided at the inlet and outlet of the experimental terrace.

[0044] A breeding method for selecting and breeding tilapia symbiotic systems constructed using tilapia aquaculture wastewater includes the following steps:

[0045] Establishing a "rice-snail-fish" symbiotic ecosystem: Select a suitable region, such as a rice-growing area in southern China, and construct experimental terraced fields. Treat the aquaculture wastewater through graded sedimentation. In spring, plant indica rice varieties in the rice-growing area and establish a recirculating aquaculture system. Irrigate using normal water sources, including rainwater, well water, or tap water. Introduce snails such as field snails to the snail farming area of ​​the experimental terraces. Once the rice reaches the waxy ripening stage, introduce carp, crucian carp, and other fish to the fish farming area of ​​the experimental terraces. Install flow velocity sensors at the inlet and outlet of the experimental terraces to measure the suspended solids concentration—specifically, the snail-specific solids concentration. The "rice-snail-fish" symbiotic ecosystem was constructed based on indicators such as filter feed volume. To ensure the quality of the rice-snail-fish, the fish were allowed to feed naturally during the experimental period, consuming rice and aquatic microorganisms. After 3 months of rearing, the fish were anesthetized and weighed at the end of the experiment. During this process, the population density of the symbiotic ecosystem was controlled, with at least 5 experimental groups and 3 replicates. The number of fish in the group should not exceed the upper limit of the symbiotic ecosystem, with each group consisting of 100-150 fish. The rice planting density was appropriate, and the density of snails and various plankton varied.

[0046] Ex-situ treatment of tilapia farming wastewater: Tilapia farming wastewater is pretreated through a wastewater treatment unit. After graded sedimentation, the wastewater is treated into a suitable water source for a symbiotic ecosystem through fine filtration, oxygenation and oxidation.

[0047] When the "rice-snail-fish" symbiotic ecosystem is stable, the water source of the experimental group is replaced with tilapia farming wastewater for ex-situ treatment. A unique blank group is set up, and corresponding indicators of rice and fish growth, snail filter feeding amount and ammonia nitrogen degradation rate, as well as chemical indicators such as biological oxygen demand, chemical oxygen demand, total nitrogen, total phosphorus, suspended solids and heavy metals are measured. Bioinformatics technology is used to analyze and identify relevant genes that are beneficial to the treatment of aquaculture wastewater.

[0048] Deep learning is used to construct a digital model: The collected dataset is preprocessed, fusing three types of data: physical parameters of the wastewater treatment device (sedimentation tank flow rate, filtration device pressure, aeration power), ecological parameters of the experimental terraces (rice plant height, snail filter feed volume, fish survival rate), and environmental disturbance parameters (light intensity, rainfall). A three-step logical processing approach is employed: spatiotemporal alignment, outlier removal, and feature mapping. For spatiotemporal alignment, the data acquisition time of the planting field sensors is used as a benchmark (5 min / time), and the wastewater treatment device data is downsampled (1 min / time), while environmental data is interpolated (15 min / time) to ensure temporal consistency. Spatially, the inlet and outlet data of the wastewater treatment device are correlated with the corresponding inlet data of the experimental terraces to establish a spatial mapping relationship between "treatment unit" and "planting area." For outlier removal, the 3σ rule based on standard deviation is used to remove data exceeding the "mean ± 3σ" range, and the outliers are filled in by linear interpolation of adjacent time points. A random forest model is used to calculate the importance of each feature to the wastewater purification efficiency and assign fusion weights, as shown in the following formula:

[0049] ;

[0050] Among them, X fusion These are the fused feature values. i X represents the weight of the i-th feature (e.g., sedimentation tank flow velocity weight 0.25, rice plant height weight 0.18, light intensity weight 0.08). i The standardized value of the i-th feature (standardization range: [0,1]) is used. The weights are obtained through training with 1000 sets of experimental data, and the accuracy of feature importance ranking is >90%.

[0051] A graph neural network model is then established, abstracting the "rice-snail-fish" symbiotic ecosystem into a graph structure G=VE. Nodes V include: wastewater treatment units (sedimentation tanks, filtration devices), experimental terraced field functional areas (rice planting area, snail farming area, fish farming area), and environmental factors (sunlight, rainfall). Edges E represent the material or energy flow relationships between nodes (e.g., wastewater flow from sedimentation tank to rice paddy, energy flow from sunlight to rice paddy). Edge weights are calculated based on the material / energy transfer efficiency between nodes (e.g., wastewater transfer efficiency from sedimentation tank to rice paddy = inflow rate to rice paddy / outflow rate from sedimentation tank). Subsequently, a graph neural network is constructed using a graph attention mechanism (GAT), as shown in the following formula:

[0052] ;

[0053] in, Let be the attention coefficient of node i to node j. The attention vector (dimension: 1×2F, where F is the node feature dimension), W is the weight matrix (dimension: F×F), and hi h j These are the feature vectors of nodes i and j, respectively, and N is the sum of the feature vectors of nodes i and j. i Let i be the set of neighboring nodes of node i. Using this formula, the algorithm can identify "filter device → rice paddy" as the key material flow path, with an attention coefficient accounting for 35%. The parameters of this path are preferentially controlled, and the key substances and their pathways are analyzed. Combining bioinformatics and transcriptomics methods, and using visual data such as differentially expressed genes, genes beneficial to wastewater treatment are accurately screened, thereby achieving regulation of wastewater treatment at the molecular level.

[0054] This invention addresses the characteristics of tilapia farming wastewater, which is rich in suspended solids, organic matter, and nutrients, as well as the specific water quality requirements (such as transparency, dissolved oxygen, and concentration of specific substances) of the "rice-snail-fish" system. It provides a specially adapted graded sedimentation and filtration solution. This includes designing targeted multi-stage sedimentation structures (such as flow-controlled sedimentation tanks / zones) to efficiently remove suspended solids of different particle sizes, configuring filtration devices to finely remove small particles, and installing sensors and other equipment to monitor different material components in real time, ensuring that the water entering the symbiotic field meets the basic requirements for the symbiotic relationship between rice, snails, and fish. Solving the problem of existing technologies where wastewater treatment facilities are separated from or awkwardly integrated with the planting field, this invention tightly connects core treatment units such as graded sedimentation tanks and filtration devices into the existing structure of the "rice-snail-fish" symbiotic field (such as field ridges, ditches, and inlets / outlets), optimizing the water flow path and reducing the land area required. Through the aforementioned targeted adaptation and structural integration, the organic matter and nutrients (nitrogen, phosphorus, etc.) in tilapia farming wastewater are ultimately recycled within the "rice-snail-fish" system. The treated wastewater serves as irrigation water and provides nutrients for rice growth, while simultaneously offering a suitable living environment for snails and fish. Rice absorbs nutrients to purify the water, while snails and fish consume organic debris and microorganisms, creating a synergistic purification effect. This significantly reduces the input of external fertilizers, lowers pollution levels from aquaculture wastewater discharge, and enhances the ecological and economic benefits of the entire system.

[0055] The preferred embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the specific details in the above embodiments. Within the scope of the technical concept of the present invention, various equivalent transformations can be made to the technical solutions of the present invention, and all such equivalent transformations fall within the protection scope of the present invention.

Claims

1. A rice-snail-fish symbiotic system constructed using tilapia aquaculture wastewater, characterized in that, include: Experimental terraced fields; A rice-growing area, which is set up in the experimental terraces for growing rice; A snail farming area, which is set up in the experimental terraced fields for the purpose of raising snails; A fish farming area, which is set up in the experimental terraces for raising fish; The wastewater treatment unit is used to perform graded sedimentation and filtration treatment on the wastewater from tilapia farming. A recirculating aquaculture system is used to irrigate the experimental terraced fields.

2. The rice-snail-fish symbiotic system constructed using tilapia aquaculture wastewater according to claim 1, characterized in that: The wastewater treatment unit includes a sedimentation tank and a filtration device. The sedimentation tank is used to classify and settle the wastewater from tilapia farming, and the filtration device is used to perform fine filtration treatment on the tilapia farming wastewater after classification and sedimentation.

3. The rice-snail-fish symbiotic system constructed using tilapia aquaculture wastewater according to claim 1, characterized in that: The experimental terraces are equipped with flow velocity sensors at their inlets and outlets.

4. A breeding method using a rice-snail-fish symbiotic system constructed using tilapia aquaculture wastewater as described in claims 1-3, characterized in that, Includes the following steps: Step 1: Select a suitable area to build experimental terraced fields, and perform graded sedimentation treatment on the aquaculture wastewater. In the spring season, select indica rice varieties to plant in the rice planting area, establish a recirculating aquaculture system, use normal water sources for irrigation, and add snails such as field snails to the snail breeding area of ​​the experimental terraced fields. When the rice grows to the waxy maturity stage, add carp, crucian carp and other fish to the fish breeding area of ​​the experimental terraced fields. Set up flow velocity sensors at the inlet and outlet of the experimental terraced fields to measure indicators such as suspended solids concentration and snail filter feeding amount, thereby building a "rice-snail-fish" symbiotic ecosystem. During the construction process, control the biological population density of the symbiotic ecosystem, set up at least 5 experimental groups and perform 3 replicates. Step 2: The tilapia farming wastewater is pretreated through a wastewater treatment unit. After grading and sedimentation, the wastewater is treated into a suitable water source for the symbiotic ecosystem through fine filtration, oxygenation and oxidation. When the "rice-snail-fish" symbiotic ecosystem is stable, the water source of the experimental group is replaced with tilapia farming wastewater for ex-situ treatment. A unique blank group is set up, and corresponding indicators of rice and fish growth, snail filter feed volume and ammonia nitrogen degradation rate, as well as chemical indicators such as biological oxygen demand, chemical oxygen demand, total nitrogen, total phosphorus, suspended solids and heavy metals are measured. Bioinformatics technology is used to analyze and identify relevant genes that are beneficial to the treatment of aquaculture wastewater. Step 3: Preprocess the collected dataset by fusing three types of data: physical parameters of the wastewater treatment device, ecological parameters of the experimental terraces, and environmental disturbance parameters. Use a three-step logical process of spatiotemporal alignment, outlier removal, and feature mapping. Calculate the importance of each feature to the wastewater purification efficiency using a random forest model and assign fusion weights. Step 4: Abstract the "rice-snail-fish" symbiotic ecosystem into a graph structure, then construct a graph neural network, and use the graph attention mechanism to analyze key substances and their pathways. Combining bioinformatics and transcriptomics methods, and using visualized data such as differentially expressed genes, accurately screen genes that are beneficial to wastewater treatment, and achieve regulation of wastewater treatment at the molecular level.

5. The breeding method according to claim 4, characterized in that: In step one, to ensure the quality of the rice-snail-fish, during the construction of the "rice-snail-fish" symbiotic ecosystem, the fish were allowed to feed naturally during the experimental period, that is, to consume rice and aquatic microorganisms, etc., and were raised for 3 months. After the experiment, the fish were anesthetized and their weight was measured.

6. The breeding method according to claim 4, characterized in that: In step one, during the construction of the "rice-snail-fish" symbiotic ecosystem, the number of fish should not exceed the upper limit of the symbiotic ecosystem, with each group consisting of 100-150 fish. The rice planting density should be appropriate, and the number of snails and other plankton should vary.

7. The breeding method according to claim 4, characterized in that: In step three, regarding spatiotemporal alignment, the data from the wastewater treatment device is downsampled based on the data acquisition time of the planting field sensors, and the environmental data is interpolated to ensure consistency in the time dimension. Spatially, the inlet and outlet data of the wastewater treatment device are associated with the corresponding data from the inlet of the experimental terrace to establish a spatial mapping relationship between the treatment unit and the planting area. Regarding outlier removal, the 3σ rule based on standard deviation is adopted to remove data that exceed the range of "mean ± 3σ", and the data is supplemented by linear interpolation of adjacent time data.

8. The breeding method according to claim 4, characterized in that: In step three, the calculation formula for the random forest model is as follows: ； Among them, X fusion These are the fused feature values. i Let X be the weight of the i-th feature. i The standardized value of the i-th feature is given by weights obtained through training with 1000 sets of experimental data.

9. The breeding method according to claim 4, characterized in that: In step four, the graph structure G=VE, where nodes V include: tailwater treatment unit, experimental terrace functional area, and environmental factors; edges E represent the material or energy flow relationship between nodes, and edge weights are calculated through the material / energy transfer efficiency between nodes.

10. The breeding method according to claim 9, characterized in that: In step four, the formula for constructing the graph neural network is as follows: ； in, Let be the attention coefficient between node i and node j. Let W be the attention vector, W be the weight matrix, and h be the weight vector. i h j These are the feature vectors of nodes i and j, respectively, and N is the sum of the feature vectors of nodes i and j. i Let i be the set of neighboring nodes of node i. Using this formula, the filter device → rice paddy area is identified as the key material flow path, and the parameters of this path are adjusted preferentially.

Images