Marine ranch resource regulation and control method and system

Abstract

The invention discloses a marine ranch resource regulation and control method and system, and the method comprises the steps: carrying out the unified scoring of multi-dimensional time series data, such as historical fishing harvesting, species habits and environment information, and mapping the data to a three-dimensional geometric structure, effectively eliminating the subjective deviation of traditional experience judgment, and achieving the quantification and transparency of a decision; a sphere screening and ecological topological graph is constructed by taking a standard score as an anchor point, so that the connectivity of the habitat of the marine ranch and the temporal-spatial resolution of a key ecological patch are improved, and accurate targeting is provided for directional enhancement and release and habitat restoration; furthermore, nonlinear interaction between environmental factors and biocenosis is captured through a graph convolutional network, and real-time dynamic adjustment of release density and environmental regulation measures is realized, so that the response speed to sudden environmental change is increased, and the risk of group death is reduced; therefore, resource utilization optimization, ecological toughness enhancement and sustainable fish catch increase are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of resource regulation, and particularly to a method and system for regulating resources in a marine ranch. Background Art

[0002] Population growth and consumption upgrading have driven a surge in the demand for aquatic products. As a "blue granary", marine ranches have become an important means to solve food security and ecological restoration. Ecological integrity is the core pillar of the sustainable development of marine ranches, and its influence runs through the entire chain of resource production, environmental quality, and economic benefits. Reasonable resource regulation can accelerate the construction of marine ranches and improve the economic construction of marine ranches.

[0003] Most of the existing resource regulation methods for ranches regulate marine ranches through resource assessment, ignoring the ecological integrity of marine ranches. The lack of ecological integrity may lead to "short-sighted economy" and insufficient connectivity of the marine ranch ecosystem. For example, although high-density cage farming can increase production in the short term, it will cause water quality deterioration in the long term, increase the risk of diseases and treatment costs, and have poor disaster resistance. Summary of the Invention The present invention provides a method and system for regulating resources in a marine ranch to reasonably regulate the resources of the marine ranch.

[0004] To solve the above technical problems, the present invention provides a method for regulating resources in a marine ranch, including: Obtaining multi-dimensional time-series data of the marine ranch, performing score evaluation on the multi-dimensional time-series data, and obtaining the fishing score, habit score, and environment score at each moment; constructing an endpoint set based on the fishing score, habit score, and environment score at each moment; Taking a preset standard score as a base point, connecting the base point and all endpoints in the endpoint set to construct a three-dimensional geometric structure; and taking the base point as the center of the sphere, constructing a sphere structure based on a preset radius, and screening the endpoints within the sphere structure as a candidate set; Connecting any two endpoints in the candidate set to construct an initial ecological topology graph; and correcting the initial ecological topology graph based on a preset distance threshold to obtain a target ecological topology graph; Constructing an ecological evaluation model based on a graph convolution model, and evaluating the ecological integrity of the marine ranch based on the target ecological topology graph and the ecological evaluation model; Generating resource regulation parameters based on the fishing score, habit score, environment score, and ecological integrity, and performing resource regulation based on the resource regulation parameters.

[0005] The present invention effectively eliminates the subjective deviation of traditional empirical judgment by uniformly scoring multi-dimensional time-series data such as historical catch, species habits, and environmental information and mapping it to a three-dimensional geometric structure, realizing the quantification and transparency of decision-making. On this basis, a sphere screening and ecological topology map is constructed with the standard score as the anchor point, improving the spatio-temporal resolution of the habitat connectivity and key ecological patches in the marine ranch, and providing a precise target for directional stock enhancement and habitat restoration. Further, the non-linear interaction between environmental factors and biological communities is captured through a graph convolutional network, making the prediction performance of ecological integrity assessment significantly better than that of traditional linear models. The resource regulation parameters generated based on this assessment score can be closed-loop issued through the Internet of Things monitoring platform to realize the real-time dynamic adjustment of the release density and environmental regulation measures, thereby improving the response speed to sudden environmental changes and reducing the risk of mass mortality, so as to achieve the optimization of resource utilization, the enhancement of ecological resilience, and the improvement of sustainable catch.

[0006] Further, the multi-dimensional time-series data includes historical catch data, species habit data, and environmental information data; obtaining the multi-dimensional time-series data of the marine ranch, scoring and evaluating the multi-dimensional time-series data, and obtaining the catch score, habit score, and environmental score at each moment; constructing an endpoint set based on the catch score, habit score, and environmental score at each moment, including: Obtaining the historical catch data, species habit data, and environmental information data of the marine ranch; resampling the historical catch data, species habit data, and environmental information data based on a time-series model and a preset standard scale to obtain standard catch data, standard habit data, and standard environmental data; Scoring and evaluating the standard catch data, standard habit data, and standard environmental data based on the fuzzy comprehensive evaluation method to obtain the catch score, habit score, and environmental score at each moment; Constructing a score vector corresponding to each moment based on the catch score, habit score, and environment at each moment, and constructing an endpoint set based on the score vector.

[0007] The present invention realizes the standardization and comprehensive quantification of time-series data by resampling and fuzzy comprehensive evaluation scoring of the historical catch data, species habit data, and environmental information data of the marine ranch. On the one hand, the resampling process ensures the consistency of various heterogeneous data on the time scale, effectively eliminating the information loss and deviation caused by different sampling frequencies; on the other hand, the fuzzy comprehensive evaluation method can take into account the mutual influence between multiple indicators and uniformly measure the catch score, habit score, and environmental score, so as to obtain a score vector that can truly reflect the ecological status of the ranch at different moments. By constructing an endpoint set, this method lays a solid data foundation for the subsequent construction and dynamic evaluation of the ecological topology network, significantly improving the accuracy and operability of resource regulation in the marine ranch.

[0008] Further, connecting any two endpoints in the candidate set to construct an initial ecological topology graph; and correcting the initial ecological topology graph based on a preset distance threshold to obtain a target ecological topology graph, including: Connecting any two endpoints in the candidate set to construct an initial ecological topology graph; Calculating the endpoint distance between any two endpoints in the initial ecological topology graph; Removing the connection between two endpoints in the initial ecological topology graph whose endpoint distance is greater than the preset distance threshold to obtain a target ecological topology graph.

[0009] The present invention adopts a graph topology correction mechanism based on a preset distance threshold. This mechanism effectively filters out the connections of nodes with weak geographical or ecological relevance, avoiding the problem of over - connection leading to a bloated network structure. At the same time, through the adjustable nature of the distance threshold, it can flexibly adapt to different sea area scales and species distribution characteristics, ensuring that the target ecological topology graph not only retains key ecological connections but also reduces noise interference. This method not only improves the accuracy of the spatio - temporal expression of the ecological network but also provides a clearer structural input for the subsequent feature learning of the graph convolutional model on complex networks, enhancing the generalization performance and computational efficiency of the model.

[0010] Further, constructing an ecological evaluation model based on the graph convolutional model, and evaluating the ecological integrity of the marine ranch based on the target ecological topology graph and the ecological evaluation model, including: Constructing an ecological evaluation model based on the graph convolutional model, where the ecological evaluation model includes a first graph convolutional module, a second graph convolutional module, and a spatial attention module; Performing node attribute encoding and edge attribute encoding based on the target ecological topology graph to obtain a node feature matrix, an adjacency matrix, and an edge tensor; Performing feature extraction on the node feature matrix and the adjacency matrix based on the first graph convolutional module to obtain local ecological features; Aggregating the local ecological features and the adjacency matrix based on the second graph convolutional module to obtain regional ecological features; Fusing the regional ecological features based on the spatial attention module and a preset key node to obtain enhanced features; Generating a heterogeneous graph feature map based on the enhanced features and the edge tensor, and performing node prediction on the heterogeneous graph feature map to obtain an ecological integrity score.

[0011] Through the constructed ecological evaluation model, the present invention integrates a double-layer graph convolution module and a spatial attention module to form a multi-scale and multi-perspective ecological feature extraction and fusion architecture. The first graph convolution module focuses on the local node neighborhood information and meticulously depicts the microscopic ecological interactions; the second graph convolution module conducts cross-regional feature aggregation to capture the macroscopic ecological distribution patterns; the spatial attention module then weights and fuses the regional ecological features based on the preset key nodes, strengthening the expression of important ecological information. This composite structure can generate highly expressive feature maps on heterogeneous graphs and obtain accurate ecological integrity scores through node prediction, effectively improving the accuracy and reliability of the ecological health assessment of marine ranches.

[0012] Further, generating a heterogeneous graph feature map based on the enhanced features and performing node prediction on the heterogeneous graph feature map to obtain an ecological integrity score includes: Generating a heterogeneous graph feature map based on the enhanced features, performing node prediction on each node in the heterogeneous graph feature map, and obtaining the integrity scores of each node; Performing weighted summation on the integrity scores of each node to obtain an ecological integrity score.

[0013] The present invention predicts the integrity score of each node independently to finely reflect the contributions and potential risks of different nodes in the ecosystem; then, through preset or dynamically adjusted weights, the scores of each node are weighted and summed to achieve the effective aggregation of the global ecological integrity score. Taking into account the differences of local nodes and the globality of the overall network, it can not only provide detailed node-level health diagnoses but also output clear global integrity indicators, providing a scientific basis for accurate resource regulation and management decisions.

[0014] Further, generating resource regulation parameters based on the fishing score, habit score, environment score, and ecological integrity and performing resource regulation based on the resource regulation parameters includes: Determining the regional regulation objectives and regulation directions based on the fishing score, habit score, environment score, and ecological integrity; Matching resource regulation parameters in a preset parameter rule base based on the regulation objectives and regulation directions and performing resource regulation based on the resource regulation parameters.

[0015] The present invention determines the regulation objectives and regulation directions of each region according to the integrity score and matches the optimal regulation parameters in a preset parameter rule base. Transforming the complex ecological evaluation results into implementable regulation schemes not only improves the scientificity and accuracy of regulation decisions but also enhances the automation and intelligence of resource regulation, contributing to the sustainable utilization of marine ranch resources and the maintenance of ecological balance.

[0016] Second aspect, the present invention provides a marine ranch resource regulation system, including: a data acquisition module, a screening module, a topology generation module, an evaluation module, and a resource regulation module; The data acquisition module is used to acquire multi-dimensional time-series data of the marine ranch, evaluate the scores of the multi-dimensional time-series data, and obtain the fishing score, habit score, and environment score at each moment; construct an endpoint set based on the fishing score, habit score, and environment score at each moment; The screening module is used to connect all endpoints in the endpoint set with a preset standard score as the base point to construct a three-dimensional geometric structure; and construct a spherical structure with the base point as the center of the sphere based on a preset radius, and screen the endpoints within the spherical structure as the candidate set; The topology generation module is used to connect any two endpoints in the candidate set to construct an initial ecological topology map; and correct the initial ecological topology map based on a preset distance threshold to obtain a target ecological topology map; The evaluation module is used to construct an ecological evaluation model based on a graph convolutional model, and evaluate the ecological integrity of the marine ranch based on the target ecological topology map and the ecological evaluation model; The resource regulation module is used to generate resource regulation parameters based on the fishing score, habit score, environment score, and ecological integrity, and perform resource regulation based on the resource regulation parameters.

[0017] Furthermore, the multi-dimensional time-series data includes historical fishing data, species habit data, and environmental information data; the data acquisition module is used to acquire the multi-dimensional time-series data of the marine ranch, evaluate the scores of the multi-dimensional time-series data, and obtain the fishing score, habit score, and environment score at each moment; constructing an endpoint set based on the fishing score, habit score, and environment score at each moment includes: Obtain the historical fishing data, species habit data, and environmental information data of the marine ranch; resample the historical fishing data, species habit data, and environmental information data based on a time-series model and a preset standard scale to obtain standard fishing data, standard habit data, and standard environmental data; Evaluate the scores of the standard fishing data, standard habit data, and standard environmental data based on the fuzzy comprehensive evaluation method to obtain the fishing score, habit score, and environment score at each moment; Construct a score vector corresponding to each moment based on the fishing score, habit score, and environment at each moment, and construct an endpoint set based on the score vector.

[0018] Further, the screening module is used to connect any two endpoints in the candidate set to construct an initial ecological topology map; and correct the initial ecological topology map based on a preset distance threshold to obtain a target ecological topology map, including: Connect any two endpoints in the candidate set to construct an initial ecological topology map; Calculate the endpoint distance between any two endpoints in the initial ecological topology map; Remove the connection between the two endpoints in the initial ecological topology map whose endpoint distance is greater than the preset distance threshold to obtain the target ecological topology map.

[0019] Further, the evaluation module is used to construct an ecological evaluation model based on a graph convolutional model, and evaluate the ecological integrity of the marine ranch based on the target ecological topology map and the ecological evaluation model, including: Construct an ecological evaluation model based on a graph convolutional model, where the ecological evaluation model includes a first graph convolutional module, a second graph convolutional module, and a spatial attention module; Based on the target ecological topology map, obtain a node feature matrix and an adjacency matrix; Extract features from the node feature matrix and the adjacency matrix based on the first graph convolutional module to obtain local ecological features; Aggregate the local ecological features and the adjacency matrix based on the second graph convolutional module to obtain regional ecological features; Fuse the regional ecological features based on the spatial attention module and preset key nodes to obtain enhanced features; Generate a heterogeneous graph feature map based on the enhanced features, and perform node prediction on the heterogeneous graph feature map to obtain an ecological integrity score. Description of the Drawings

[0020] Figure 1 It is a schematic flowchart of a method for regulating marine ranch resources provided by an embodiment of the present invention. Detailed Embodiments

[0021] The following combines the drawings and embodiments to further describe in detail the specific embodiments of the present invention. The following embodiments are used to illustrate the present invention, but are not used to limit the scope of the present invention.

[0022] In the description, claims, and drawings of this application, terms such as "first" and "second" are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally further include steps or units not listed, or may optionally further include other steps or units inherent to these processes, methods, products, or devices.

[0023] Reference to "embodiment" herein means that a specific feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of this application. The phrase appearing in various positions in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art will explicitly and implicitly understand that the embodiments described herein can be combined with other embodiments.

[0024] Embodiment 1 See Figure 1 , Figure 1 which is a schematic flowchart of a method for regulating and controlling marine ranch resources provided by an embodiment of the present invention. An embodiment of the present invention provides a method for regulating and controlling marine ranch resources, including steps 101 to 105, specifically as follows: Step 101: Obtain multi-dimensional time-series data of the marine ranch, perform score evaluation on the multi-dimensional time-series data, and obtain the fishing score, habit score, and environment score at each moment; construct an endpoint set based on the fishing score, habit score, and environment score at each moment; In this embodiment, the multi-dimensional time-series data includes historical fishing data, species habit data, and environmental information data; the obtaining of the multi-dimensional time-series data of the marine ranch, performing score evaluation on the multi-dimensional time-series data, obtaining the fishing score, habit score, and environment score at each moment; and constructing an endpoint set based on the fishing score, habit score, and environment score at each moment includes: Obtain the historical fishing data, species habit data, and environmental information data of the marine ranch; resample the historical fishing data, species habit data, and environmental information data based on a time-series model and a preset standard scale to obtain standard fishing data, standard habit data, and standard environmental data; Perform score evaluation on the standard fishing data, standard habit data, and standard environmental data based on the fuzzy comprehensive evaluation method to obtain the fishing score, habit score, and environment score at each moment; Construct a score vector corresponding to each moment based on the fishing score, habit score, and environment at each moment, and construct an endpoint set based on the score vector.

[0025] In this embodiment, time-series data of the catch in the marine ranch (including catch volume, operation time, etc.), migration trajectories of species electronic tags, and acoustic monitoring records, as well as environmental information data (sea temperature, salinity, dissolved oxygen, etc.) are obtained.

[0026] In this embodiment, the three types of original time-series data are resampled on a daily or weekly basis: using the time-series resampling method provided by platforms such as Dataiku, linear or Lagrangian interpolation is performed on the missing dates to ensure that the time intervals are uniformly consistent.

[0027] In this embodiment, Z-score normalization is performed on the resampled standard catch data, standard habit data, and standard environmental data respectively to map the values of each dimension to the same magnitude and eliminate the influence of different index dimensions.

[0028] In this embodiment, an evaluation index set is constructed: the standard catch volume, standard habit data, standard environmental data (such as dissolved oxygen, pH, suspended solid concentration), etc. are incorporated into the fuzzy evaluation system. For each evaluation index, a semi-sine type or triangular membership function is selected to map the normalized values to the membership degree interval [0,1] to describe the "poor-medium-excellent" grade division, and the entropy weight method (Entropy Weight Method) or the analytic hierarchy process (AHP) is used to determine the weights of each index and construct the weight matrix W.

[0029] In this embodiment, for the index membership degree vector at each moment and the weight vector W are weighted and synthesized to obtain the catch score 、habit score and environmental score , and the ternary score ( , , ) at each moment is regarded as an endpoint in the three-dimensional space , thus generating an endpoint set; a preset standard score point =( , , ) is used as the base point.

[0030] In this embodiment, by resampling and performing fuzzy comprehensive evaluation scores on the historical catch data, species habit data, and environmental information data of the marine ranch, the standardization and comprehensive quantification of time-series data are achieved. On the one hand, the resampling process ensures the consistency of various heterogeneous data on the time scale, effectively eliminating information loss and bias caused by different sampling frequencies; on the other hand, the fuzzy comprehensive evaluation method can take into account the mutual influence between multiple indicators, uniformly measure the catch score, habit score, and environmental score, so as to obtain a score vector that can truly reflect the ecological status of the ranch at different times. By constructing the endpoint set, this method lays a solid data foundation for the subsequent construction and dynamic evaluation of the ecological topology network, significantly improving the accuracy and operability of marine ranch resource regulation.

[0031] Step 102: Connect the base point and all endpoints in the endpoint set with a preset standard score as the base point to construct a three-dimensional geometric structure; and with the base point as the center of the sphere, construct a sphere structure based on a preset radius, and screen the endpoints within the sphere structure as the candidate set; In this embodiment, a three-dimensional vector is defined =( , , ), which respectively represent the standard values of the catch score, habit score, and environmental score, and serve as the reference points for the ecological ideal state.

[0032] In this embodiment, connect all endpoints with the base point to generate a radial three-dimensional line segment set, which can be regarded as a set of three-dimensional vector clusters with the standard ecological state as the origin, and construct a three-dimensional geometric structure: (1) In this embodiment, with the base point as the center of the sphere, set the radius r, where the preset radius is an adjustable parameter used to remove the outlier set through the preset radius.

[0033] In this embodiment, screen all points that satisfy as the candidate point set .

[0034] In this embodiment, map the multi-dimensional index scores to a three-dimensional space structure, effectively integrating the three types of indicators of catch, habit, and environment, facilitating visualization and subsequent geometric operations; use the spatial spherical region to limit the deviation range to achieve the clustering and extraction of "ecologically similar states".

[0035] In this embodiment, the candidate points represent historical state samples that are "close to the standard ecological state", which helps to model a stable subset of the ecological state and enhance the robustness of subsequent graph modeling and graph convolution. At the same time, the preset radius r can be adjusted according to the strategy, corresponding to different tolerance requirements, and supporting the selection of ecological management strategies from loose to strict.

[0036] Step 103: Connect any two endpoints in the candidate set to construct an initial ecological topology graph; and correct the initial ecological topology graph based on a preset distance threshold to obtain a target ecological topology graph. In this embodiment, the connecting any two endpoints in the candidate set to construct an initial ecological topology graph; and correcting the initial ecological topology graph based on a preset distance threshold to obtain a target ecological topology graph includes: Connect any two endpoints in the candidate set to construct an initial ecological topology graph. Calculate the endpoint distance between any two endpoints in the initial ecological topology graph. Delete the connection between the two endpoints in the initial ecological topology graph whose endpoint distance is greater than the preset distance threshold to obtain a target ecological topology graph.

[0037] In this embodiment, the candidate set , where .

[0038] In this embodiment, traverse all unordered pairs (Pi, Pj), where 1 ≤ i < j ≤ N.

[0039] In this embodiment, in the initial graph add an undirected edge for each pair of endpoints, so as to obtain an initial ecological topology graph .

[0040] In this embodiment, for each edge (Pi, Pj) of the initial ecological topology graph, calculate the Euclidean distance between the endpoints: (2) where , , , , , are the projected distances of the endpoints and in the three-dimensional coordinate plane, respectively.

[0041] In this embodiment, through the preset distance , when , then retain the endpoints and Edges between them, otherwise, remove the endpoints and Edges between them, thereby generating the target ecological topology graph .

[0042] In this embodiment, by removing the edges with a distance greater than the preset distance threshold, pairs of points with too large score differences are removed, avoiding misidentifying samples with far - apart ecological states as connected. Removing pairs of points with too large score differences avoids misidentifying samples with far - apart ecological states as connected.

[0043] In this embodiment, the target topology graph depicts the real ecological connection path, providing an accurate network structure basis for resource regulation.

[0044] In this embodiment, through the graph topology correction mechanism based on the preset distance threshold, this mechanism effectively filters out the node connections with weak geographical or ecological relevance, avoiding the problem of bloated network structure caused by over - connection; meanwhile, through the adjustability of the distance threshold, it can flexibly adapt to different sea area scales and species distribution characteristics, ensuring that the target ecological topology graph not only retains the key ecological connections but also reduces noise interference. This method not only improves the accuracy of the spatio - temporal expression of the ecological network but also provides a clearer structural input for the subsequent feature learning of the graph convolutional model on complex networks, enhancing the generalization performance and computational efficiency of the model.

[0045] Step 104: Construct an ecological assessment model based on the graph convolutional model, and evaluate the ecological integrity of the marine ranch based on the target ecological topology graph and the ecological assessment model; In this embodiment, the constructing an ecological assessment model based on the graph convolutional model and evaluating the ecological integrity of the marine ranch based on the target ecological topology graph and the ecological assessment model includes: Construct an ecological assessment model based on the graph convolutional model, where the ecological assessment model includes a first graph convolutional module, a second graph convolutional module, and a spatial attention module; Perform node attribute encoding and edge attribute encoding based on the target ecological topology graph to obtain a node feature matrix, an adjacency matrix, and an edge tensor; Extract features from the node feature matrix and the adjacency matrix based on the first graph convolutional module to obtain local ecological features; Aggregate the local ecological features and the adjacency matrix based on the second graph convolutional module to obtain regional ecological features; Fuse the regional ecological features based on the spatial attention module and the preset key nodes to obtain enhanced features; Generate a heterogeneous graph feature map based on the enhanced features and the edge tensor, and perform node prediction on the heterogeneous graph feature map to obtain an ecological integrity score.

[0046] In this embodiment, the target ecological topology graph , where is the node list, is the edge list.

[0047] In this embodiment, all node numbers and corresponding spatial or temporal identifiers are directly extracted from the original node set to construct an adjacency matrix , if .

[0048] In this embodiment, after directly normalizing (such as Z-score) the continuous three-dimensional scores in the node list, they are mapped to a 64-dimensional space using a linear layer to obtain continuous features.

[0049] In this embodiment, the species categories of the marine ranch can be mapped to 64-dimensional vectors through a learnable Embedding layer to obtain classification features, and the water quality level can be mapped to 64-dimensional vectors through piecewise quantization and then also through Embedding or positional encoding to obtain ordinal features; finally, the continuous features, classification features, and ordinal features are concatenated or added dimension by dimension to obtain the final 64-dimensional features per node, thereby generating a node feature matrix .

[0050] In this embodiment, generating an heterogeneous graph feature map based on the enhanced features and performing node prediction on the heterogeneous graph feature map to obtain an ecological integrity score includes: Generating an heterogeneous graph feature map based on the enhanced features, performing node prediction on each node in the heterogeneous graph feature map to obtain the integrity scores of each node; Performing weighted summation on the integrity scores of each node to obtain an ecological integrity score.

[0051] In this embodiment, the ecological assessment model consists of three major modules: The first graph convolution module: aggregating the features of the first-order neighbors of each node (moment) to extract the local ecological situation; the second graph convolution module: performing second-order neighbor aggregation on the basis of the first-layer output to capture the ecological associations in a larger range; the spatial attention module: dynamically adjusting the neighbor weights for preset key nodes (such as the spawning peak moment or the artificial fish reef deployment moment) to strengthen their influence in the assessment.

[0052] In this embodiment, the first graph convolution module constructs an adjacency matrix with self-connections , calculates the degree matrix , and thus obtains the local ecological features according to the degree matrix and the adjacency matrix: (3) Where are trainable parameters.

[0053] In this embodiment, the second graph convolution module aggregates the local ecological features and the adjacency matrix to obtain regional ecological features: (4) where are trainable parameters.

[0054] In this embodiment, based on the spatial attention module, the attention energy is calculated for each pair of neighbors (i, j): (5) where represents the attention energy of node i to node j, which is used to measure the initial weight when aggregating information from node i to node j. LeakyReLU(⋅) represents the leaky rectified linear unit activation function, is the leakage coefficient, represents the feature of node i output by the second graph convolution module, the feature of node j output by the second graph convolution module.

[0055] In this embodiment, the key nodes include the high-value spawning moment or the artificial reef deployment moment. When the nodes in the target ecological topology map are at the key nodes, an amplification coefficient is applied to them, so as to amplify their fusion weights and obtain enhanced features: (6) In this embodiment, where is the edge weight between node i and node j.

[0056] In this embodiment, is input into a fully connected layer or a small MLP to predict the integrity score node by node: (7) where is the bias term, which is used to adjust the overall level of the prediction, is the weight vector, and its dimension is the same as the input feature dimension (i.e., the same as the dimension of ).

[0057] In this embodiment, the constructed ecological evaluation model integrates a double-layer graph convolutional module and a spatial attention module to form a multi-scale and multi-perspective ecological feature extraction and fusion architecture. The first graph convolutional module focuses on the local node neighborhood information to meticulously depict the microscopic ecological interactions; the second graph convolutional module conducts cross-regional feature aggregation to capture the macroscopic ecological distribution patterns; the spatial attention module then weights and fuses the regional ecological features based on the preset key nodes, enhancing the expression of important ecological information. This composite structure can generate highly expressive feature maps on the heterogeneous graph and obtain accurate ecological integrity scores through node prediction, effectively improving the accuracy and reliability of the ecological health assessment of the marine ranch.

[0058] Step 105: Generate resource regulation parameters based on the fishing score, habit score, environmental score, and ecological integrity, and perform resource regulation based on the resource regulation parameters.

[0059] In this embodiment, the generating of resource regulation parameters based on the fishing score, habit score, environmental score, and ecological integrity, and performing resource regulation based on the resource regulation parameters includes: Determine the regional regulation target and regulation direction based on the fishing score, habit score, environmental score, and ecological integrity; Match resource regulation parameters in the preset parameter rule base based on the regulation target and regulation direction, and perform resource regulation based on the resource regulation parameters.

[0060] In this embodiment, the integrity score is mapped to four ecological grades: excellent (≥0.8), good (0.6–0.8), medium (0.4–0.6), and poor (<0.4). The regulation targets and directions are respectively set as follows: for the "excellent" grade, it focuses on maintaining stability and fine-tuning; for the "good" grade, it focuses on appropriate proliferation and catch control; for the "medium" grade, it focuses on strengthening ecological restoration and environmental repair; for the "poor" grade, it focuses on halting fishing and large-scale habitat restoration.

[0061] In this embodiment, a preset resource regulation parameter rule base is established, which includes resource regulation parameters such as species type, geographical location, release amount, fishing intensity, and aeration frequency. For example: If a certain sub-region is rated as having a "medium" grade restoration target, then the following are matched from the rule base: the release density is set to 150% of the standard; 8 artificial fish reefs are placed per square kilometer; the aerator is started 2 times a day; the pH adjustment is maintained in the range of 7.8–8.2. If the sub-region is rated as "poor", then measures such as halting fishing, restoring 10 mu of seagrass beds, and sediment dredging once a month are matched.

[0062] In this embodiment, independent integrity score predictions are made for each node to precisely reflect the contributions and potential risks of different nodes in the ecosystem; then, the scores of each node are weighted and summed through preset or dynamically adjusted weights, achieving an effective aggregation of the global ecological integrity score. Taking into account both local node differences and overall network globality, it can not only provide detailed node-level health diagnoses but also output a clear global integrity indicator, providing a scientific basis for precise resource regulation and management decisions.

[0063] An embodiment of the present invention also provides a marine ranch resource regulation system, including: a data acquisition module, a screening module, a topology generation module, an evaluation module, and a resource regulation module; The data acquisition module is used to acquire multi-dimensional time-series data of the marine ranch, evaluate the scores of the multi-dimensional time-series data, and obtain the fishing score, habit score, and environment score at each moment; an endpoint set is constructed based on the fishing score, habit score, and environment score at each moment; The screening module is used to connect all endpoints in the endpoint set with a preset standard score as the base point to construct a three-dimensional geometric structure; and with the base point as the center of the sphere, a sphere structure is constructed based on a preset radius, and the endpoints within the sphere structure are screened as the candidate set; The topology generation module is used to connect any two endpoints in the candidate set to construct an initial ecological topology map; and the initial ecological topology map is corrected based on a preset distance threshold to obtain a target ecological topology map; The evaluation module is used to construct an ecological evaluation model based on a graph convolution model, and evaluate the ecological integrity of the marine ranch based on the target ecological topology map and the ecological evaluation model; The resource regulation module is used to generate resource regulation parameters based on the fishing score, habit score, environment score, and ecological integrity, and perform resource regulation based on the resource regulation parameters.

[0064] In this embodiment, the multi-dimensional time-series data includes historical fishing data, species habit data, and environmental information data; the data acquisition module is used to acquire the multi-dimensional time-series data of the marine ranch, evaluate the scores of the multi-dimensional time-series data, and obtain the fishing score, habit score, and environment score at each moment; constructing an endpoint set based on the fishing score, habit score, and environment score at each moment includes: Acquire the historical fishing data, species habit data, and environmental information data of the marine ranch; resample the historical fishing data, species habit data, and environmental information data based on a time-series model and a preset standard scale to obtain standard fishing data, standard habit data, and standard environmental data; Evaluate the scores of the standard catch data, standard habit data, and standard environment data based on the fuzzy comprehensive evaluation method to obtain the catch scores, habit scores, and environment scores at each moment; Construct a score vector corresponding to each moment based on the catch scores, habit scores, and environment at each moment, and construct an endpoint set based on the score vector.

[0065] In this embodiment, the screening module is used to connect any two endpoints in the candidate set to construct an initial ecological topology map; and correct the initial ecological topology map based on a preset distance threshold to obtain a target ecological topology map, including: Connect any two endpoints in the candidate set to construct an initial ecological topology map; Calculate the endpoint distance between any two endpoints in the initial ecological topology map; Remove the connection between the two endpoints in the initial ecological topology map whose endpoint distance is greater than the preset distance threshold to obtain a target ecological topology map.

[0066] In this embodiment, the evaluation module is used to construct an ecological evaluation model based on the graph convolutional model, and evaluate the ecological integrity of the marine ranch based on the target ecological topology map and the ecological evaluation model, including: Construct an ecological evaluation model based on the graph convolutional model, where the ecological evaluation model includes a first graph convolutional module, a second graph convolutional module, and a spatial attention module; Perform node attribute encoding and edge attribute encoding on the target ecological topology map to obtain a node feature matrix, an adjacency matrix, and an edge tensor; Extract features from the node feature matrix and the adjacency matrix based on the first graph convolutional module to obtain local ecological features; Aggregate the local ecological features and the adjacency matrix based on the second graph convolutional module to obtain regional ecological features; Fuse the regional ecological features based on the spatial attention module and preset key nodes to obtain enhanced features; Generate a heterogeneous graph feature map based on the enhanced features and the edge tensor, and perform node prediction on the heterogeneous graph feature map to obtain an ecological integrity score.

[0067] In this embodiment, the evaluation module is used to generate a heterogeneous graph feature map based on the enhanced features, and perform node prediction on the heterogeneous graph feature map to obtain an ecological integrity score, including: Generate a heterogeneous graph feature map based on the enhanced features, perform node prediction on each node in the heterogeneous graph feature map to obtain the integrity score of each node; Weighted sum the integrity scores of each node to obtain the ecological integrity score.

[0068] In this embodiment, the resource regulation module is configured to generate resource regulation parameters based on the fishing score, habit score, environment score, and ecological integrity, and perform resource regulation based on the resource regulation parameters, including: Determine the regional regulation target and regulation direction based on the fishing score, habit score, environment score, and ecological integrity; Match the resource regulation parameters in the preset parameter rule library based on the regulation target and regulation direction, and perform resource regulation based on the resource regulation parameters.

[0069] In an embodiment of the present invention, a terminal device is further provided, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, the above-mentioned marine ranch resource regulation method is implemented.

[0070] In an embodiment of the present invention, a computer-readable storage medium is further provided. The computer-readable storage medium includes a stored computer program. When the computer program runs, it controls the device where the computer-readable storage medium is located to execute the above-mentioned marine ranch resource regulation method.

[0071] Exemplarily, the computer program can be divided into one or more modules. One or more modules are stored in the memory and executed by the processor to complete the present invention. One or more modules can be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program in the terminal device.

[0072] The terminal device can be a computing device such as a desktop computer, a notebook, a palm computer, and a cloud server. The terminal device may include, but is not limited to, a processor, a memory, and a display. Those skilled in the art can understand that the above components are only examples of the terminal device and do not constitute a limitation to the terminal device. It may include more or fewer components than those described, or combine some components, or different components. For example, the terminal device may further include input / output devices, network access devices, buses, etc.

[0073] The so-called processor may be a Central Processing Unit (CPU), or may also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The processor is the control center of the terminal device, and connects various parts of the entire terminal device through various interfaces and circuits.

[0074] The memory can be used to store computer programs and / or modules. The processor realizes various functions of the terminal device by running or executing the computer programs and / or modules stored in the memory, and by calling the data stored in the memory. The memory may mainly include a program storage area and a data storage area. Among them, the program storage area may store an operating system, application programs required for at least one function (such as a sound playback function, a text conversion function, etc.); the data storage area may store data created according to the use of the mobile phone (such as audio data, text message data, etc.). In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, at least one magnetic disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0075] Among them, when the module for regulating ocean ranch resources is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, to implement all or part of the processes in the above-mentioned embodiment methods of the present invention, it can also be completed by instructing relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above-mentioned various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electrical carrier signal, telecommunication signal, and software distribution medium, etc. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0076] The specific embodiments described above further elaborate on the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are only specific embodiments of the present invention and are not used to limit the protection scope of the present invention. In particular, it is pointed out that for those skilled in the art, any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for regulating and controlling marine ranch resources, characterized in that, Including: Obtain multi-dimensional time-series data of the marine ranch, conduct score evaluation on the multi-dimensional time-series data, and obtain the fishing score, habit score, and environmental score at each moment; Construct an endpoint set based on the fishing score, habit score, and environmental score at each moment; Taking a preset standard score as the base point, connect the base point and all endpoints in the endpoint set to construct a three-dimensional geometric structure; and taking the base point as the center of the sphere, construct a sphere structure based on a preset radius, and screen the endpoints within the sphere structure as the candidate set; Connect any two endpoints in the candidate set to construct an initial ecological topology map; And correct the initial ecological topology map based on a preset distance threshold to obtain the target ecological topology map; Construct an ecological evaluation model based on the graph convolutional model, and evaluate the ecological integrity of the marine ranch based on the target ecological topology map and the ecological evaluation model; Generate resource regulation parameters based on the fishing score, habit score, environmental score, and ecological integrity, and conduct resource regulation based on the resource regulation parameters.

2. The method for regulating and controlling ocean ranch resources according to claim 1, wherein The multi-dimensional time-series data includes historical fishing data, species habit data, and environmental information data; obtain the multi-dimensional time-series data of the marine ranch, conduct score evaluation on the multi-dimensional time-series data, and obtain the fishing score, habit score, and environmental score at each moment; Construct an endpoint set based on the fishing score, habit score, and environmental score at each moment, including: Obtain the historical fishing data, species habit data, and environmental information data of the marine ranch; resample the historical fishing data, species habit data, and environmental information data based on the time-series model and preset standard scale to obtain standard fishing data, standard habit data, and standard environmental data; Conduct score evaluation on the standard fishing data, standard habit data, and standard environmental data based on the fuzzy comprehensive evaluation method to obtain the fishing score, habit score, and environmental score at each moment; Construct a score vector corresponding to each moment based on the fishing score, habit score, and environmental score at each moment, and construct an endpoint set based on the score vector.

3. The method for regulating and controlling marine ranch resources according to claim 2, characterized in that, Connect any two endpoints in the candidate set to construct an initial ecological topology map; And correct the initial ecological topology map based on a preset distance threshold to obtain the target ecological topology map, including: Connect any two endpoints in the candidate set to construct an initial ecological topology map; Calculate the endpoint distance between any two endpoints in the initial ecological topology map; Delete the connection between the two endpoints in the initial ecological topology map whose endpoint distance is greater than the preset distance threshold to obtain the target ecological topology map.

4. The method for regulating and controlling ocean ranch resources according to claim 3, characterized in that, Construct an ecological evaluation model based on the graph convolutional model, and evaluate the ecological integrity of the marine ranch based on the target ecological topology map and the ecological evaluation model, including: Construct an ecological evaluation model based on the graph convolutional model, and the ecological evaluation model includes a first graph convolutional module, a second graph convolutional module, and a spatial attention module; Based on the target ecological topology map, obtain the node feature matrix adjacency matrix; Extract features from the node feature matrix and the adjacency matrix based on the first graph convolutional module to obtain local ecological features; Aggregate the local ecological features and the adjacency matrix based on the second graph convolution module to obtain regional ecological features; Fuse the regional ecological features based on the spatial attention module and the preset key nodes to obtain enhanced features; Generate a heterogeneous graph feature map based on the enhanced features, and perform node prediction on the heterogeneous graph feature map to obtain an ecological integrity score.

5. The method for regulating and controlling marine ranch resources according to claim 4, wherein, The generating a heterogeneous graph feature map based on the enhanced features, and performing node prediction on the heterogeneous graph feature map to obtain an ecological integrity score includes: Generate a heterogeneous graph feature map based on the enhanced features, perform node prediction on each node in the heterogeneous graph feature map to obtain the integrity scores of each node; Perform weighted summation on the integrity scores of each node to obtain an ecological integrity score.

6. The method for regulating and controlling marine ranch resources according to claim 5, characterized in that, The generating resource regulation parameters based on the fishing score, habit score, environment score, and ecological integrity, and performing resource regulation based on the resource regulation parameters includes: Determine the regional regulation target and regulation direction based on the fishing score, habit score, environment score, and ecological integrity; Match resource regulation parameters in a preset parameter rule library based on the regulation target and regulation direction, and perform resource regulation based on the resource regulation parameters.

7. A marine ranch resource regulation system, characterized in that, Includes: A data acquisition module, a screening module, a topology generation module, an evaluation module, and a resource regulation module; The data acquisition module is used to acquire multi-dimensional time-series data of the marine ranch, perform score evaluation on the multi-dimensional time-series data to obtain the fishing score, habit score, and environment score at each moment; Construct an endpoint set based on the fishing score, habit score, and environment score at each moment; The screening module is used to connect all endpoints in the endpoint set with a preset standard score as the base point to construct a three-dimensional geometric structure; and use the base point as the center of the sphere, construct a sphere structure based on a preset radius, and screen the endpoints within the sphere structure as a candidate set; The topology generation module is used to connect any two endpoints in the candidate set to construct an initial ecological topology graph; And correct the initial ecological topology graph based on a preset distance threshold to obtain a target ecological topology graph; The evaluation module is used to construct an ecological evaluation model based on a graph convolution model, and evaluate the ecological integrity of the marine ranch based on the target ecological topology graph and the ecological evaluation model; The resource regulation module is used to generate resource regulation parameters based on the fishing score, habit score, environment score, and ecological integrity, and perform resource regulation based on the resource regulation parameters.

8. The marine ranch resource regulation system according to claim 7, characterized in that, The multi-dimensional time-series data includes historical fishing data, species habit data, and environmental information data; the data acquisition module is used to acquire multi-dimensional time-series data of the marine ranch, perform score evaluation on the multi-dimensional time-series data to obtain the fishing score, habit score, and environment score at each moment; Constructing an endpoint set based on the fishing score, habit score, and environment score at each moment includes: Obtain the historical catch data, species habit data, and environmental information data of the marine ranch; resample the historical catch data, species habit data, and environmental information data based on the time series model and the preset standard scale to obtain the standard catch data, standard habit data, and standard environmental data; Evaluate the scores of the standard catch data, standard habit data, and standard environmental data based on the fuzzy comprehensive evaluation method to obtain the catch score, habit score, and environmental score at each moment; Construct a score vector corresponding to each moment based on the catch score, habit score, and environment at each moment, and construct an endpoint set based on the score vector.

9. The marine ranch resource regulation system according to claim 8, characterized in that The screening module is used to connect any two endpoints in the candidate set to construct an initial ecological topology map; And correct the initial ecological topology map based on the preset distance threshold to obtain the target ecological topology map, including: Connect any two endpoints in the candidate set to construct an initial ecological topology map; Calculate the endpoint distance between any two endpoints in the initial ecological topology map; Remove the connection between the two endpoints in the initial ecological topology map whose endpoint distance is greater than the preset distance threshold to obtain the target ecological topology map.

10. The marine ranch resource regulation system according to claim 8, characterized in that, The evaluation module is used to construct an ecological evaluation model based on the graph convolutional model, and evaluate the ecological integrity of the marine ranch based on the target ecological topology map and the ecological evaluation model, including: Construct an ecological evaluation model based on the graph convolutional model, and the ecological evaluation model includes a first graph convolutional module, a second graph convolutional module, and a spatial attention module; Based on the target ecological topology map, obtain the node feature matrix adjacency matrix; Extract features from the node feature matrix and the adjacency matrix based on the first graph convolutional module to obtain local ecological features; Aggregate the local ecological features and the adjacency matrix based on the second graph convolutional module to obtain regional ecological features; Fuse the regional ecological features based on the spatial attention module and the preset key nodes to obtain enhanced features; Generate a heterogeneous graph feature map based on the enhanced features, and perform node prediction on the heterogeneous graph feature map to obtain an ecological integrity score.

Images