Data processing method, device and equipment applied to fishery management and storage medium

Abstract

This application relates to a data processing method, apparatus, equipment, and storage medium for fisheries management, comprising: acquiring pre-collected multi-source heterogeneous data; performing prediction processing on the multi-source heterogeneous data based on a preset fish school distribution prediction model to obtain corresponding fish school distribution prediction results; performing model construction processing based on the fish school distribution prediction results to obtain a corresponding heterogeneous fishing vessel multi-objective path planning model; acquiring a preset multi-objective cooperative search algorithm corresponding to the heterogeneous fishing vessel multi-objective path planning model; solving the heterogeneous fishing vessel multi-objective path planning model based on the multi-objective cooperative search algorithm to obtain a corresponding objective optimization solution set; decoding the objective optimization solution set to obtain the corresponding fishing vessel planning path; and outputting the fishing vessel planning path. This application effectively improves the processing efficiency of fishing vessel operations and reduces energy consumption and operating costs.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and in particular to data processing methods, apparatus, equipment and storage media applied to fisheries management. Background Technology

[0002] In the field of fisheries management, we are currently in a critical transition phase from traditional experience-driven to modern data-driven approaches. With the continuous depletion of global marine fishery resources and the decreasing quantity of catchable fish stocks, coupled with rising fishing costs, fisheries production faces significant challenges. Against this backdrop, building a precise and intelligent fisheries resource management system has become an inevitable trend for promoting the sustainable development of the fisheries industry. The rapid advancements in marine remote sensing technology, big data analytics, and artificial intelligence algorithms have brought new opportunities to fisheries resource management, prompting profound technological changes in areas such as fish distribution prediction and vessel path planning. Among these, vessel path planning technology, as a key link in improving fisheries operational efficiency and reducing costs, has attracted considerable attention for its development.

[0003] Early research on fishing vessel route planning largely focused on route optimization for single vessel types, primarily employing traditional planning algorithms. These algorithms can, to some extent, plan relatively reasonable routes for individual fishing vessels, improving their operational efficiency. However, in actual fisheries production, fishing fleets often exhibit significant heterogeneity. Different fishing vessels differ in multiple dimensions, including vessel type, tonnage, fishing gear type, and operational methods. This heterogeneity makes it difficult for traditional planning algorithms to comprehensively consider the characteristics and operational needs of all fishing vessels, thus failing to achieve optimal overall operational efficiency for the entire fishing fleet.

[0004] However, in actual operations, the lack of effective multi-vehicle collaborative path planning schemes often leads to fishing vessels navigating blindly and engaging in ineffective fishing. Some fishing vessels may operate for extended periods in areas with few or no fish due to unreasonable path planning, while other areas with fish are not developed in time; or multiple fishing vessels may concentrate in the same area, causing resource competition and waste. These situations not only result in low efficiency in fisheries operations but also significantly increase fuel consumption, further driving up operating costs and seriously affecting the economic benefits and sustainable development of fisheries production. Therefore, developing an efficient path planning technology that can achieve multi-vehicle collaborative optimization is of significant practical importance. Summary of the Invention

[0005] The purpose of this application is to provide a data processing method, apparatus, computer equipment, and storage medium for fisheries management, so as to solve the technical problems of low efficiency and high operating costs in existing fisheries operations.

[0006] Firstly, a data processing method for fisheries management is provided, including:

[0007] Acquire pre-collected multi-source heterogeneous data; wherein, the multi-source heterogeneous data includes at least remote sensing and monitoring data, fisheries data, and environmental data;

[0008] The multi-source heterogeneous data is processed based on a preset fish distribution prediction model to obtain the corresponding fish distribution prediction results.

[0009] Based on the fish distribution prediction results, a model construction process is performed to obtain the corresponding heterogeneous fishing vessel multi-objective path planning model.

[0010] Obtain a preset multi-objective cooperative search algorithm corresponding to the heterogeneous fishing vessel multi-objective path planning model;

[0011] The multi-objective cooperative search algorithm is used to solve the multi-objective path planning model of the heterogeneous fishing vessel to obtain the corresponding target optimization solution set.

[0012] The target optimization solution set is decoded to obtain the corresponding fishing boat planning path;

[0013] The planned route for the fishing vessel is output and processed.

[0014] Secondly, a data processing device for fisheries management is provided, comprising:

[0015] The first acquisition module is used to acquire pre-collected multi-source heterogeneous data; wherein, the multi-source heterogeneous data includes at least remote sensing and monitoring data, fishery data, and environmental data;

[0016] The prediction module is used to perform prediction processing on the multi-source heterogeneous data based on a preset fish distribution prediction model to obtain the corresponding fish distribution prediction results.

[0017] The construction module is used to perform model construction processing based on the fish distribution prediction results to obtain the corresponding heterogeneous fishing vessel multi-objective path planning model.

[0018] The second acquisition module is used to acquire a preset multi-objective cooperative search algorithm corresponding to the heterogeneous fishing vessel multi-objective path planning model.

[0019] The processing module is used to solve the heterogeneous fishing vessel multi-objective path planning model based on the multi-objective cooperative search algorithm to obtain the corresponding target optimization solution set;

[0020] The decoding module is used to decode the target optimization solution set to obtain the corresponding fishing boat planning path.

[0021] The output module is used to process the planned path of the fishing vessel.

[0022] Thirdly, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the data processing method applied to fisheries management described above.

[0023] Fourthly, a computer-readable storage medium is provided, which stores a computer program that, when executed by a processor, implements the steps of the data processing method described above for fisheries management.

[0024] The aforementioned data processing method, apparatus, computer equipment, and storage medium for fisheries management first acquire pre-collected multi-source heterogeneous data, including at least remote sensing and monitoring data, fisheries data, and environmental data. Then, based on a pre-defined fish school distribution prediction model, the multi-source heterogeneous data is processed to predict the fish school distribution, yielding corresponding prediction results. Next, based on the fish school distribution prediction results, a model is constructed to obtain a corresponding heterogeneous fishing vessel multi-objective path planning model. A pre-defined multi-objective cooperative search algorithm corresponding to the heterogeneous fishing vessel multi-objective path planning model is then acquired. Subsequently, the multi-objective cooperative search algorithm is used to solve the heterogeneous fishing vessel multi-objective path planning model, obtaining a corresponding optimized solution set. The optimized solution set is further decoded to obtain the corresponding fishing vessel planning path. Finally, the fishing vessel planning path is output. Based on the above automated processing flow, this application uses a fish swarm distribution prediction model to predict the distribution of multi-source heterogeneous data, then uses this prediction to construct a multi-objective path planning model for heterogeneous fishing vessels. A multi-objective collaborative search algorithm is then used to solve this model, yielding an optimized solution set. This optimized solution set is then decoded to obtain the planned fishing vessel path, which is finally output. Thus, by combining multi-source heterogeneous data with a fish swarm distribution prediction model, this application can achieve accurate short-term prediction of fish swarm distribution. Furthermore, by establishing a multi-objective path planning model tailored to the characteristics of heterogeneous fishing vessels and designing a multi-stage collaborative evolution multi-objective search algorithm to solve this model, the application can automatically and accurately optimize the global allocation of fishing vessel tasks and fishing routes. This effectively improves the efficiency and targeting of fishing operations, reduces energy consumption and operating costs, and ensures the coordination and efficiency of fishing vessel operation scheduling. Attached Figure Description

[0025] To more clearly illustrate the solutions in this application, the accompanying drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0026] Figure 1 This is an exemplary system architecture diagram to which this application can be applied;

[0027] Figure 2 This is a flowchart of an embodiment of a data processing method for fisheries management according to this application;

[0028] Figure 3 This is a schematic diagram of a structure of an embodiment of a data processing device for fisheries management according to this application;

[0029] Figure 4 This is a schematic diagram of the structure of one embodiment of the computer device according to this application. Detailed Implementation

[0030] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein in the specification of the application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application; the terms "comprising" and "having," and any variations thereof, in the specification, claims, and foregoing drawings of this application, are intended to cover non-exclusive inclusion. The terms "first," "second," etc., in the specification, claims, or foregoing drawings of this application are used to distinguish different objects, not to describe a particular order.

[0031] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0032] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0033] like Figure 1As shown, system architecture 100 may include terminal device 101, network 102, and server 103. Terminal device 101 may be a laptop 1011, tablet 1012, or mobile phone 1013. Network 102 is used as a medium to provide a communication link between terminal device 101 and server 103. Network 102 may include various connection types, such as wired, wireless communication links, or fiber optic cables, etc.

[0034] Users can use terminal device 101 to interact with server 103 via network 102 to receive or send messages, etc. Various communication client applications can be installed on terminal device 101, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social media platform software, etc.

[0035] Terminal device 101 can be various electronic devices with a display screen and support web browsing. In addition to laptops 1011, tablets 1012, or mobile phones 1013, terminal device 101 can also be an e-book reader, an MP3 player (Moving Picture Experts Group Audio Layer III), an MP4 player (Moving Picture Experts Group Audio Layer IV), a laptop computer, and a desktop computer, etc.

[0036] Server 103 can be a server that provides various services, such as a backend server that provides support for the pages displayed on terminal device 101.

[0037] It should be noted that the data processing method for fisheries management provided in the embodiments of this application is generally executed by a server / terminal device, and correspondingly, the data processing device for fisheries management is generally located in the server / terminal device.

[0038] It should be understood that Figure 1 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0039] Continue to refer to Figure 2 A flowchart illustrating an embodiment of a data processing method for fisheries management according to this application is shown. The order of steps in the flowchart can be changed, and some steps can be omitted, depending on different needs. The data processing method for fisheries management includes the following steps:

[0040] Step S201: Acquire pre-collected multi-source heterogeneous data; wherein the multi-source heterogeneous data includes at least remote sensing and monitoring data, fishery data, and environmental data.

[0041] In this embodiment, the data processing method applied to fisheries management runs on electronic devices (e.g., Figure 1 The server / terminal device shown can acquire pre-collected multi-source heterogeneous data via wired or wireless connections. It should be noted that the aforementioned wireless connection methods may include, but are not limited to, 3G / 4G / 5G connections, WiFi connections, Bluetooth connections, WiMAX connections, Zigbee connections, UWB (ultra-wideband) connections, and other currently known or future-developed wireless connection methods.

[0042] The aforementioned remote sensing and monitoring data refers to: satellite remote sensing data (water surface temperature, water salinity, etc.) and surface buoy data, used to reflect the dynamic changes in the ecological environment of fishing grounds. The aforementioned fisheries data refers to: historical fishing locations, fish catch records, and Automatic Identification System (AIS) trajectory data, used to depict the spatiotemporal relationship between fish activity distribution and human fishing activities. The aforementioned environmental data refers to: real-time meteorological data (wind speed, air pressure, rainfall, etc.) and spatially constrained polygon data such as fishing ban areas and closed fishing areas, used to define the prediction area and effective fishing ground range.

[0043] In addition, data such as water surface temperature and salinity can be obtained through satellite remote sensing technology, typically presented in the form of images or numerical tables. Simultaneously, buoys are deployed on the water surface to collect relevant data in real time, such as water temperature and water quality; this buoy data can be transmitted wirelessly to a data center. Furthermore, historical fishing locations and fish catch records can be obtained from fisheries management departments or relevant fisheries databases. Automatic Identification System (AIS) trajectory data can be obtained by receiving AIS signals transmitted by vessels, which include information such as the vessel's position, speed, and heading. Real-time meteorological data, including wind speed, air pressure, and rainfall, can be obtained from meteorological monitoring stations. Spatial constraint polygon data such as fishing ban zones and closed fishing areas can be obtained from official documents of fisheries management departments or Geographic Information System (GIS) databases.

[0044] Step S202: Based on the preset fish distribution prediction model, the multi-source heterogeneous data is processed to predict the corresponding fish distribution prediction results.

[0045] In this embodiment, the specific implementation process of predicting the multi-source heterogeneous data based on the preset fish distribution prediction model to obtain the corresponding fish distribution prediction results will be further described in detail in subsequent specific embodiments of this application, and will not be elaborated on here.

[0046] Step S203: Based on the predicted fish distribution results, perform model construction processing to obtain the corresponding heterogeneous fishing vessel multi-objective path planning model.

[0047] In this embodiment, the construction process of the above-mentioned heterogeneous fishing vessel multi-objective path planning model includes: (1) Mathematical model. Based on the fish distribution prediction results contained in the fish distribution prediction model, including the location of fish appearance and the potential catch of fish, the heterogeneous fishing vessel multi-objective path problem is defined as a single fishing ground wharf having Various types of fishing boats, fishing boats are assembled into ,in Number the fishing boats, go Each node captures different schools of fish, and the customer set is... This can form an undirected graph. , Let be a set of nodes, where 0 represents a fishing ground or wharf. Number the nodes. Let be the set of edges, and let the length of each edge be expressed as... This indicates that the fishing boat departed from the node. To the node The driving distance.

[0048] The model notation definition includes: V: set of points , ; : Set of points , ; Ship type set , ; Fish school type set , ; : The set of the number of fishing boat types , ; Fishing boat type Maximum load capacity; :node For fish The amount captured; :node With nodes The Euclidean distance between them; : is a very large positive integer; Auxiliary variables; Decision variables ; Fishing boats at the node Load capacity at that time.

[0049] Furthermore, an algorithm for solving the multi-objective path optimization problem for heterogeneous fishing vessels is proposed. This algorithm comprehensively considers multiple optimization objectives such as the heterogeneous characteristics of fishing vessels, fishing revenue, and fuel consumption. It establishes a mathematical model with the dual objectives of minimizing the total number of fishing vessels used and minimizing the total delivery distance. Through population evolution combined with global perturbation and local search strategies, it achieves efficient solution and global optimization of fishing vessel path planning.

[0050] (2) Optimization objective:

[0051] 1. Minimize total delivery distance:

[0052]

[0053] 2. Minimize the total number of fishing boats in use:

[0054]

[0055] (3) Constraints:

[0056] 1. Each fishing boat departs from the fishing ground dock and returns after completing its catch:

[0057]

[0058] 2. Each fishing spot can only be caught once by one fishing boat:

[0059] ;

[0060]

[0061] 3. Once a fishing boat reaches a certain fishing spot, it must depart from that spot:

[0062]

[0063] 4. Fishing vessels are prohibited from making repeated visits to a single fishing spot:

[0064]

[0065] 5. The load carried by each fishing vessel after fishing at the fishing spot shall not exceed the maximum load capacity of the vessel:

[0066] ;

[0067]

[0068] 6. Each fishing vessel's delivery route does not include sub-loops:

[0069]

[0070] Establishing a multi-objective path planning model for heterogeneous fishing vessels is a crucial step in transforming fish distribution prediction results into actual fishing vessel path planning. By defining the problem, symbols, optimization objectives, and constraints, the complex path planning problem can be transformed into a mathematical model, which is beneficial for providing a foundation for the design and solution of subsequent multi-objective cooperative search algorithms (or heterogeneous fishing vessel multi-objective cooperative search algorithms).

[0071] Step S204: Obtain a preset multi-objective collaborative search algorithm corresponding to the heterogeneous fishing vessel multi-objective path planning model.

[0072] In this embodiment, the aforementioned heterogeneous fishing vessel multi-objective path planning model defines the type of fishing vessel, the distribution of fishing points, path constraints (such as sailing time, fuel consumption, and catch balance), and multiple objectives to be optimized (such as minimizing total sailing time, maximizing total catch, and balancing the catch of each fishing vessel). Correspondingly, the aforementioned multi-objective cooperative search algorithm (or heterogeneous fishing vessel multi-objective cooperative search algorithm) is a pre-constructed algorithm designed to find a set of solutions (i.e., the Pareto optimal solution set) that can simultaneously optimize multiple objectives, while satisfying the model constraints of the aforementioned heterogeneous fishing vessel multi-objective path planning model. Each step in this algorithm (such as initial solution generation, global perturbation, local search, non-dominated sorting, and elitist strategy) is aimed at searching for better solutions in the solution space to approximate the optimal solution set of the model.

[0073] Specifically, the algorithm flow of the proposed heterogeneous fishing vessel multi-objective cooperative search algorithm is as follows:

[0074] Step 1: Algorithm Initialization. Set the main algorithm parameters, including population size. and maximum number of iterations An initial population solution set is generated based on information about the fishing vessel type and fishing location. Step 2: Global Perturbation Operation. The potential high-quality solution space is explored using a global perturbation operator, generating a new population solution set. This enhances the algorithm's global search capability and avoids getting trapped in local optima. Step 3: Local search optimization. This involves optimizing the generated... Perform a local search optimization operation to generate an optimized population solution set. Step 4: Non-dominant sorting and elite preservation. and merged into The biobjective function values ​​were calculated and non-dominated rankings were performed. An elite retention strategy was adopted based on individual rank and crowding distance, selecting the top performers. Individuals as the next generation of the population Step 5: Termination Condition Check. Determine if the maximum number of iterations or the convergence condition has been reached. If not, return to Step 2; if so, output the Pareto optimal solution set and decode the individual solutions into fishing boat paths.

[0075] Step S205: Solve the heterogeneous fishing vessel multi-objective path planning model based on the multi-objective cooperative search algorithm to obtain the corresponding target optimization solution set.

[0076] In this embodiment, the specific implementation process of solving the heterogeneous fishing vessel multi-objective path planning model based on the multi-objective cooperative search algorithm to obtain the corresponding target optimization solution set will be further described in detail in subsequent specific embodiments of this application, and will not be elaborated on here.

[0077] Step S206: Decode the target optimization solution set to obtain the corresponding fishing boat planning path.

[0078] In this embodiment, the individual codes corresponding to the aforementioned target optimization solution set, i.e., the Pareto optimal solution, can be decoded into actual fishing vessel fishing paths, i.e., fishing vessel planning paths. The individual codes typically use a certain encoding method (such as binary encoding, real number encoding, permutation encoding, etc.) to digitally represent the fishing vessel scheduling scheme. The decoding process involves converting these codes back into specific, operable fishing vessel action sequences, such as each fishing vessel's departure port, the order of visiting fishing points, and arrival and departure times.

[0079] Step S207: Output the planned path for the fishing boat.

[0080] In this embodiment, the generated fishing vessel planning path can be sent to the corresponding decision-maker via email, text message, or interface display, thereby completing the output processing of the aforementioned fishing vessel planning path. Furthermore, the decision-maker can directly select the most suitable planning path from the above-mentioned fishing vessel planning paths as the final fishing vessel collaborative scheduling scheme based on their experience, preferences, and actual needs.

[0081] This application first acquires pre-collected multi-source heterogeneous data; wherein, the multi-source heterogeneous data includes at least remote sensing and monitoring data, fisheries data, and environmental data; then, based on a preset fish school distribution prediction model, the multi-source heterogeneous data is processed for prediction to obtain the corresponding fish school distribution prediction results; subsequently, based on the fish school distribution prediction results, a model construction process is performed to obtain the corresponding heterogeneous fishing vessel multi-objective path planning model; and a preset multi-objective cooperative search algorithm corresponding to the heterogeneous fishing vessel multi-objective path planning model is acquired; subsequently, based on the multi-objective cooperative search algorithm, the heterogeneous fishing vessel multi-objective path planning model is solved to obtain the corresponding target optimization solution set; further, the target optimization solution set is decoded to obtain the corresponding fishing vessel planning path; finally, the fishing vessel planning path is output. Based on the above automated processing flow, this application uses a fish swarm distribution prediction model to predict the distribution of multi-source heterogeneous data, then uses this prediction to construct a multi-objective path planning model for heterogeneous fishing vessels. A multi-objective collaborative search algorithm is then used to solve this model, yielding an optimized solution set. This optimized solution set is then decoded to obtain the planned fishing vessel path, which is finally output. Thus, by combining multi-source heterogeneous data with a fish swarm distribution prediction model, this application can achieve accurate short-term prediction of fish swarm distribution. Furthermore, by establishing a multi-objective path planning model tailored to the characteristics of heterogeneous fishing vessels and designing a multi-stage collaborative evolution multi-objective search algorithm to solve this model, the application can automatically and accurately optimize the global allocation of fishing vessel tasks and fishing routes. This effectively improves the efficiency and targeting of fishing operations, reduces energy consumption and operating costs, and ensures the coordination and efficiency of fishing vessel operation scheduling.

[0082] In some optional implementations, the fish swarm distribution prediction model includes an input layer, a feature extraction layer, a temporal modeling layer, and an output layer; step S202 includes the following steps:

[0083] The multi-source heterogeneous data is input into the fish distribution prediction model. Based on the input layer, the multi-source heterogeneous data is preprocessed and features are constructed to obtain the corresponding processed data.

[0084] In this embodiment, the specific implementation process of performing data preprocessing and feature construction on the multi-source heterogeneous data based on the input layer to obtain the corresponding processed data will be further described in detail in subsequent specific embodiments of this application, and will not be elaborated on here.

[0085] Based on the feature extraction layer, features are extracted from the processed data to obtain the corresponding first feature data.

[0086] In this embodiment, the aforementioned feature extraction layer (CNN layer) can be a CNN module, specifically employing a one-dimensional convolutional neural network (CNN) to extract features from the spatiotemporal data cube. The CNN layer captures local correlations in the time series or spatial dimensions of the data through a local receptive field mechanism, thereby effectively extracting potential feature patterns of the fishing vessel's fishing environment and removing noise and unstable components. The convolutional layer extracts local patterns (such as temperature gradients and flow velocity trends); the pooling layer achieves feature dimensionality reduction and noise suppression; finally, a fusion feature mapping matrix of multi-source factors is formed, which is the aforementioned first feature data.

[0087] The feature data is processed based on the time-series modeling layer to obtain the corresponding second feature data.

[0088] In this embodiment, the aforementioned temporal modeling layer can specifically be a BiLSTM module. Temporal modeling is performed by inputting the high-dimensional features extracted by the CNN into a Bidirectional Long Short-Term Memory (BiLSTM) network. By learning temporal dependency information simultaneously from both forward (past → future) and backward (future → past) directions, BiLSTM can capture the bidirectional dynamic features of the fish school distribution in the time dimension and output the corresponding second feature data.

[0089] The internal calculation formula of BiLSTM is as follows: Each LSTM unit in each direction contains an input gate, a forget gate, an output gate, and a candidate memory unit, and its state update formula is as follows:

[0090]

[0091] in, Enter the current time; Hidden state; Memory unit; Sigmoid activation function; Hyperbolic tangent activation function.

[0092] The forward and backward computations of BiLSTM are represented as follows:

[0093] ;

[0094]

[0095] The final hidden state is the result of concatenating the previous and next directions:

[0096]

[0097] This structure can capture both historical trends and future impacts, enabling the model to better identify periodic and abrupt patterns in fish migration.

[0098] Based on the output layer, the first feature data and the second feature data are predicted to obtain the corresponding prediction results.

[0099] In this embodiment, the output layer is the part of the entire model that generates the final prediction result. It receives data after CNN feature extraction and BiLSTM temporal modeling, and outputs the prediction results of the target fishing grounds for the next 24-48 hours based on the task and objectives set by the fish distribution prediction model. Specifically, the output layer provides the probability distribution of fish appearance locations, reflecting the likelihood of fish appearing in different geographical locations within a specific future time period; it also provides the potential catch for each predicted area, helping fishermen understand which areas are likely to yield higher catches; and it provides fish density levels, visually demonstrating the density of fish in different areas. Furthermore, the output layer overlays these prediction results with polygons representing prohibited fishing areas to generate a map of the operable fishing grounds, providing fishermen with clear operational guidance. For example, the output layer may use a series of calculations and transformations to map the BiLSTM-processed features into specific probability values, catch amounts, or density levels, and present the operable fishing grounds graphically, thus completing the entire process from input data to actual prediction results.

[0100] The prediction results are used as the prediction results for the fish population distribution.

[0101] This application inputs the multi-source heterogeneous data into the fish school distribution prediction model. Based on the input layer, it performs data preprocessing and feature construction on the multi-source heterogeneous data to obtain corresponding processed data. Then, based on the feature extraction layer, it extracts features from the processed data to obtain corresponding first feature data. Next, based on the temporal modeling layer, it processes the feature data to obtain corresponding second feature data. Subsequently, based on the output layer, it performs prediction processing on the first and second feature data to obtain corresponding prediction results. Finally, it uses the prediction results as the fish school distribution prediction results. Based on this processing flow, this application, through the use of a multi-objective path planning model for heterogeneous fishing vessels, can fully utilize the information from multi-source heterogeneous data to accurately predict the distribution of fish schools. By using the input layer to construct features to provide input data for the model, the feature extraction layer and the temporal modeling layer respectively model the features and temporal relationships of the data, and the output layer finally generates accurate prediction results, improving the generation efficiency and accuracy of fish school distribution prediction results and providing a scientific basis for fishing vessel path planning.

[0102] In some optional implementations of this embodiment, the step of performing data preprocessing and feature construction on the multi-source heterogeneous data based on the input layer to obtain corresponding processed data includes the following steps:

[0103] The multi-source heterogeneous data is filled with missing values ​​based on the input layer to obtain the corresponding first processed data.

[0104] In this embodiment, the collected multi-source heterogeneous data can be processed uniformly. Let the target dataset be: { , } = {historical fishing locations, historical fish catches}, with multiple influencing factors including: { , , ... The dataset includes {water surface temperature, water salinity, meteorological parameters, etc.}. To ensure temporal continuity and data comparability, linear interpolation can be used to fill in missing values ​​in the collected multi-source heterogeneous data. For time series data, linear interpolation is performed based on data values ​​from adjacent time points; for spatial data, interpolation can be performed based on data values ​​from the surrounding area.

[0105] The first processed data is normalized to obtain the corresponding second processed data.

[0106] In this embodiment, the normalization process mentioned above refers to normalizing data with different dimensions so that their range is uniformly within [0, 1]. Specifically, this can be achieved using the formula... Normalization was performed, where, This represents the value of a certain factor at a specific moment. These are the minimum and maximum observed values ​​for this factor, respectively.

[0107] The second processed data is processed by feature construction based on a preset sliding window mechanism to obtain the corresponding third processed data.

[0108] In this embodiment, a sliding window mechanism is introduced to construct training samples in order to capture the temporal evolution of fish school distribution. Let the window length be... Step size is Then the model input samples are continuous Composition of multi-source influencing factors at each time step: Output corresponding to the future Step-by-step prediction target: [ , This approach can enhance time stationarity, smooth out abrupt changes, and improve the model's ability to capture short-term fluctuations and long-term trends.

[0109] The third processed data is used as the processed data.

[0110] This application performs missing value completion processing on the multi-source heterogeneous data based on the input layer to obtain corresponding first processed data; then, it normalizes the first processed data to obtain corresponding second processed data; subsequently, it performs feature construction processing on the second processed data based on a preset sliding window mechanism to obtain corresponding third processed data; finally, it uses the third processed data as the final processed data. Based on the above processing flow, this application, by performing missing value completion, normalization, and feature construction processing on the collected multi-source heterogeneous data, can eliminate noise and missing values ​​in the data, improve the quality and usability of the data, and thus provide accurate input data for subsequent fish swarm distribution prediction models.

[0111] In some alternative implementations, step S205 includes the following steps:

[0112] Set the algorithm parameters corresponding to the multi-objective collaborative search algorithm, and generate an initial population solution set based on the fishing vessel type and fishing point information in the heterogeneous fishing vessel multi-objective path planning model; wherein, the algorithm parameters include at least the population size and the maximum number of iterations.

[0113] In this embodiment, the main parameters (algorithm parameters) of the above-mentioned multi-objective cooperative search algorithm can be set according to actual processing needs, including population size. and maximum number of iterations The specific implementation process of generating the initial population solution set based on the fishing vessel type and fishing point information in the heterogeneous fishing vessel multi-objective path planning model will be further described in detail in subsequent specific embodiments of this application, and will not be elaborated on here.

[0114] A global perturbation operation is performed on the initial population solution set to generate a new first population solution set.

[0115] In this embodiment, a global perturbation operator can be used to explore the potential high-quality solution space and generate a new population solution set, thereby enhancing the global search capability of the aforementioned multi-objective cooperative search algorithm and avoiding getting trapped in local optima. The specific implementation process of performing a global perturbation operation on the initial population solution set to generate a new first-type population solution set will be described in further detail in subsequent embodiments of this application, and will not be elaborated upon here.

[0116] Based on a preset local search optimization strategy, a local search optimization operation is performed on the first type of group solution set to generate an optimized second type of group solution set.

[0117] In this embodiment, to further improve the accuracy and stability of the path solution, this application designs three local neighborhood operations to achieve deep optimization of the path, including: 1. Single task point relocation operation; 2. Whole path task reallocation operation; 3. Path internal structure reversal adjustment operation. The local search optimization strategy applies the three local neighborhood operations to the population solution set obtained after the global perturbation operation, and then integrates the results of these operations to obtain the optimized population solution set. Furthermore, the specific implementation process of performing local search optimization operations on the first population solution set based on the preset local search optimization strategy to generate the optimized second population solution set will be further described in detail in subsequent specific embodiments of this application, and will not be elaborated upon here.

[0118] The initial population solution set and the second population solution set are subjected to non-dominated sorting and elite retention processing to obtain a third population solution set that matches the population size.

[0119] In this embodiment, a combined population, i.e., a third population set, can be obtained by merging the initial population set with the locally optimized second population set. Then, the bi-objective function value for each individual is calculated, and a non-dominated ranking method is used to classify individual ranks. Finally, based on the individual rank and crowding distance, an elite retention strategy is used to select the top-ranked individuals. (Population size) 100 high-quality individuals, which are used to form the initial population for the next generation. This ensures the diversity of solutions and global convergence performance.

[0120] Determine whether the current iteration count has reached the maximum iteration count.

[0121] In this embodiment, if it is detected that the current iteration count has not reached the maximum iteration count, the solution set of the third population is used as the initial population for the next generation. Then, return to the above global perturbation operation and continue to execute the algorithm iterative processing.

[0122] If so, stop the iteration, obtain the specified optimal solution set from the third group solution set, and use the specified optimal solution set as the target optimization solution set.

[0123] In this embodiment, when the current iteration count is detected to have reached the maximum iteration count, the multi-objective cooperative search algorithm terminates and outputs the Pareto optimal solution set (i.e., the specified optimal solution set), and uses the Pareto optimal solution set as the final target optimization solution set.

[0124] The Pareto optimal solution set is defined as follows: In multi-objective optimization problems, there are usually multiple conflicting objective functions (for example, in the cooperative scheduling problem of fishing vessels, multiple objectives such as fishing efficiency, fuel consumption, and sailing time may be considered simultaneously). The Pareto optimal solution set refers to a set of solutions where, for any given solution, no other solution is better than it in all objectives, or worse than it in at least one objective. In other words, in the Pareto optimal solution set, it is impossible to obtain a better solution by improving one objective without sacrificing others. These solutions provide decision-makers with multiple options for weighing trade-offs among multiple objectives. Therefore, the aforementioned Pareto optimal solution set (objective optimization solution set) is a set containing multiple solutions, each representing a cooperative scheduling scheme for fishing vessels. These schemes achieve a relatively optimal balance across multiple objectives (such as fishing efficiency, fuel consumption, and sailing time), and no other scheme can improve all objectives without compromising at least one objective.

[0125] This application sets algorithm parameters corresponding to the multi-objective collaborative search algorithm and generates an initial population solution set based on the fishing vessel type and fishing point information in the heterogeneous fishing vessel multi-objective path planning model. The algorithm parameters include at least the population size and the maximum number of iterations. Then, a global perturbation operation is performed on the initial population solution set to generate a new first population solution set. Subsequently, a local search optimization operation is performed on the first population solution set based on a preset local search optimization strategy to generate an optimized second population solution set. Next, non-dominated sorting and elite retention processing are performed on the initial population solution set and the second population solution set to obtain a third population solution set matching the population size. It is further determined whether the current number of iterations has reached the maximum number of iterations. If so, iteration is stopped, and a specified optimal solution set is obtained from the third population solution set, which is then used as the target optimized solution set. Based on the above processing flow, this application continuously optimizes individuals in the population through steps such as algorithm initialization, initial solution generation, global perturbation operation, local search optimization, non-dominated sorting, and elite strategy, gradually approaching the optimal solution and improving the accuracy and adaptability of the generated target optimization solution set. Furthermore, the final output target optimization solution set can provide multiple feasible solutions for fishing vessel path planning, allowing decision-makers to select the optimal solution based on actual needs.

[0126] In some optional implementations, generating the initial population solution set based on the fishing vessel type and fishing point information in the heterogeneous fishing vessel multi-objective path planning model includes the following steps:

[0127] Obtain information on fishing vessel type and fishing point from the heterogeneous fishing vessel multi-objective path planning model.

[0128] In this embodiment, when constructing a multi-objective path planning model for heterogeneous fishing vessels, it is first necessary to define the type of fishing vessel. These vessel types may be categorized based on characteristics such as vessel size, speed, fishing capacity, and equipment configuration. For example, some fishing vessels may be more suitable for deep-sea fishing, while others are more suitable for near-shore operations. This type information is determined during the model construction phase and serves as the basis for subsequent algorithm design. Furthermore, the selection of fishing points is also an important optimization objective when constructing the multi-objective path planning model for heterogeneous fishing vessels. Through model solving, the optimal distribution of fishing points under given vessel types and task requirements can be found. This fishing point information obtained through model optimization can also be used as part of the initial solution set.

[0129] During the construction of the multi-objective path planning model for heterogeneous fishing vessels, the required fishing vessel type and fishing point information can be recorded simultaneously.

[0130] Obtain the preset multi-segment encoding method based on fishing vessel type.

[0131] In this embodiment, the above-mentioned multi-segment encoding method based on fishing vessel type refers to a multi-segment encoding method designed to adapt to the heterogeneous characteristics of fishing vessels, using fishing vessels as the basic unit to encode the path of fishing points. The specific steps are as follows: 1. Initialize the fishing vessel path set R to be empty; 2. Each fishing vessel departs from the fishing ground dock (number 0) and adds the dock number to the path r1; 3. Randomly select a fishing point and add it to the path. If the constraints such as capacity and operation are met, the fishing point is added to the path and removed from the candidate set; 4. If there is no feasible fishing point on the current path, the fishing vessel returns to the dock (number 0); 5. Repeat steps 2 to 4 until all fishing points are assigned; 6. Combine all fishing vessel paths into path set R, thus forming the initial decoding structure.

[0132] Based on the fishing vessel type and the fishing point information, the path encoding process is performed using the multi-segment encoding method based on the fishing vessel type to obtain the corresponding initial decoding structure.

[0133] In this embodiment, based on the above-mentioned processing steps of the multi-segment encoding method based on fishing vessel type, the fishing point can be path encoded using the fishing vessel as the basic unit to obtain the corresponding initial decoding structure, which is then used as the corresponding initial population solution set.

[0134] The initial solution encoding structure is used as the initial population solution set.

[0135] This application obtains information on fishing vessel type and fishing point from the heterogeneous fishing vessel multi-objective path planning model; then, it obtains a preset multi-segment encoding method based on fishing vessel type; subsequently, based on the fishing vessel type and fishing point information, it uses the multi-segment encoding method based on fishing vessel type to perform path encoding processing to obtain the corresponding initial decoding structure; subsequently, the initial decoding structure is used as the initial population solution set. Based on the above processing flow, this application, by using the multi-segment encoding method based on fishing vessel type and performing path encoding processing according to the fishing vessel type and fishing point information in the heterogeneous fishing vessel multi-objective path planning model, can effectively improve the quality and distribution diversity of the initial population solution set while ensuring the feasibility of constraints.

[0136] In some optional implementations of this embodiment, the step of performing a global perturbation operation on the initial population solution set to generate a new first population solution set includes the following steps:

[0137] Get a randomly generated number within a preset numerical range.

[0138] In this embodiment, the aforementioned numerical range is specifically [0,1]. When the multi-objective cooperative search algorithm is executed, a random number 'a' within the range [0,1] is randomly generated.

[0139] Obtain two preset global perturbation methods.

[0140] In this embodiment, to enhance the global search capability of the aforementioned multi-objective cooperative search algorithm and prevent it from getting trapped in local optima, this application designs two global perturbation operators (corresponding to two global perturbation methods), including: 1. Cross-path task redistribution operation: Randomly select two different fishing boat paths, select one fishing point from each path, remove it from its original path, and randomly insert it into any position on the other path to achieve cross-boat task redistribution. 2. Cross-path task swap adjustment operation: Randomly select two fishing boat paths, select one fishing point from each path, remove the two fishing points, and swap them, inserting them into their original positions on the other path to achieve path swapping of fishing points.

[0141] Select the target global perturbation method corresponding to the random number from the global perturbation methods.

[0142] In this embodiment, the random number 'a' can be compared with a preset threshold (0.5). When 'a' < the preset threshold (0.5), a cross-path task reallocation operation is performed; when 'a' ≥ the preset threshold (0.5), a cross-path task swap adjustment operation is performed.

[0143] Based on the target global perturbation method, a global perturbation operation is performed on the initial population solution set to generate the corresponding specified population solution set.

[0144] In this embodiment, the initial population solution set can be subjected to a global perturbation operation based on a global perturbation method selected by random numbers, and the new specified population solution set generated after the global perturbation operation is completed can be used as the corresponding first type of population solution set.

[0145] The specified population solution set is used as the first population solution set.

[0146] This application obtains randomly generated random numbers within a preset numerical range and two preset global perturbation methods. Then, it selects a target global perturbation method corresponding to the random number from the global perturbation methods. Subsequently, it performs a global perturbation operation on the initial population solution set based on the target global perturbation method to generate a corresponding specified population solution set. This specified population solution set is then used as the first population solution set. Based on the above processing flow, this application effectively enhances the global search capability of multi-objective cooperative search algorithms by selecting a target global perturbation method from preset global perturbation methods based on generated random numbers within a preset numerical range, then performs a global perturbation operation on the initial population solution set based on the use of the target global perturbation method, and uses the generated new specified population solution set as the corresponding first population solution set. This prevents getting trapped in local optima, guides the algorithm to explore other potential regions by disrupting the structure of existing solutions, and thus effectively enhances the diversity of solutions, improving the intelligence and richness of the generated first population solution set.

[0147] In some optional implementations of this embodiment, the step of performing a local search optimization operation on the first type of group solution set based on a preset local search optimization strategy to generate an optimized second type of group solution set includes the following steps:

[0148] Perform a single-task-point relocation operation on the first population solution set to obtain the corresponding first target population solution set.

[0149] In this embodiment, the single-task point relocation operation includes: selecting two paths from the fishing vessel path set, removing a fishing point from one path, and inserting it into the optimal position on the other path (i.e., the position that minimizes the change in the total path distance). If the new path satisfies the constraints and reduces the total travel distance, the path is updated.

[0150] Perform a whole-path task reallocation operation on the first target population solution set to obtain the corresponding second target population solution set.

[0151] In this embodiment, the above-mentioned whole-path task reallocation operation includes: selecting a path from the fishing vessel path set and removing it, then inserting the fishing points in that path one by one into the optimal positions of other fishing vessel paths. If the new solution satisfies the constraints and improves on any objective function, then the solution is updated.

[0152] Perform a path internal structure reversal adjustment operation on the first target population solution set to obtain the corresponding third target population solution set.

[0153] In this embodiment, the above-mentioned path internal structure reversal and adjustment operation includes, for the internal structure of a single fishing boat path, selecting two edges, reversing the path segment between the two edges, and reconnecting the path structure. If the new path satisfies the constraints and reduces the travel distance, the path is updated.

[0154] The first target population solution set, the second target population solution set, and the third target population solution set are integrated to obtain the corresponding integrated population solution set.

[0155] In this embodiment, after completing the three local neighborhood operations, the results of all operations are integrated to form an optimized population (i.e., an integrated population solution set). The integration process may include: merging the better solutions (feasible solutions, i.e., solutions that satisfy the constraints and improve on at least one objective function) generated by each operation into a set, and using the resulting integrated population solution set as the corresponding second population solution set.

[0156] The integrated population solution set is used as the second population solution set.

[0157] This application obtains a first target population solution set by performing a single-task-point relocation operation on the first population solution set; simultaneously, it performs a whole-path task reallocation operation on the first target population solution set to obtain a second target population solution set; and it performs a path internal structure reversal adjustment operation on the first target population solution set to obtain a third target population solution set. Then, it integrates the first, second, and third target population solution sets to obtain an integrated population solution set; subsequently, it uses this integrated population solution set as the second population solution set. Based on the above processing flow, this application effectively increases the diversity of the population and avoids premature convergence to a local optimum by performing single-task-point relocation, whole-path task reallocation, and path internal structure reversal adjustment operations on the first population solution set, and by integrating the results of multiple local neighborhood operations. Furthermore, each local neighborhood operation has its unique optimization effect. Integrating the results of these operations allows for the comprehensive utilization of their advantages, achieving more comprehensive path optimization, and thus effectively improving the accuracy and comprehensiveness of the generated second population solution set.

[0158] In some optional implementations of this embodiment, the system may further include intelligent processing of the Pareto optimal solution set (i.e., the target optimal solution set), including:

[0159] 1. Visualization. Objective Space Visualization: Each solution in the Pareto optimal solution set is visualized in a multi-dimensional space composed of various objective functions. For example, if there are two objective functions (such as fishing efficiency and fuel consumption), each solution can be represented as a point on a two-dimensional plane, with the horizontal axis representing fishing efficiency and the vertical axis representing fuel consumption. In this way, decision-makers can intuitively see the performance of different solutions on various objectives and their relative positions, thus better understanding the trade-offs between different options. Decision Variable Space Visualization: If the problem involves multiple decision variables (such as the number of fishing boats, the initial position of each boat, etc.), these variables can also be visualized in relation to the objective functions to help decision-makers understand the impact of different decision variables on the objectives.

[0160] 2. Solution Screening and Ranking. Screening based on preference information: Decision-makers may prioritize different objectives based on their preferences. For example, they might focus more on fishing efficiency and less on fuel consumption. Weights can be assigned to each objective function based on these preferences, and then a comprehensive score (e.g., a weighted sum) can be calculated for each Pareto optimal solution. Solutions can then be ranked according to their scores. This prioritizes solutions that better align with the decision-maker's preferences, making selection easier. Screening based on specific rules: In addition to preference weights, specific rules can be established to screen solutions. For example, an upper limit can be set for fuel consumption, retaining only Pareto optimal solutions that meet this limit; or a lower limit can be set for fishing efficiency, excluding solutions with excessively low fishing efficiency. These rules further narrow down the range of solutions, making it easier for decision-makers to make a decision.

[0161] 3. Sensitivity Analysis. Analyzing the impact of changes in objective function weights: Change the weights of each objective function and observe the changes in the Pareto optimal solution set. This helps decision-makers understand the relative importance of different objectives and their impact on the final solution. For example, if a slight increase in the weight of fishing efficiency causes a significant change in the Pareto optimal solution set, it indicates that fishing efficiency has a greater impact on the solution; conversely, if the change is small, it suggests that other objectives may have a more critical impact on the solution. Analyzing the impact of changes in decision variables: Study how changes in decision variables affect the Pareto optimal solution set. For example, change the number of fishing boats and observe the changes in the distribution of the solution set in the objective space. This helps decision-makers understand the impact of different decision variables on the scheduling scheme, providing more comprehensive information for practical decision-making.

[0162] 4. Decision Making. Direct Selection: Decision-makers can directly select the most suitable solution from the Pareto optimal solution set as the final fishing vessel coordinated scheduling scheme based on their experience, preferences, and actual needs. Combination and Improvement: In some cases, decision-makers may not be satisfied with the existing Pareto optimal solution and can try combining multiple solutions or further improving the solution. For example, combining the better parts of two solutions to form a new scheduling scheme; or performing local optimization based on a certain solution to obtain better target performance.

[0163] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0164] Furthermore, this application proposes a method for predicting fish swarm distribution and optimizing multi-objective paths for heterogeneous fishing vessels based on multi-source data fusion. By combining multi-source data fusion with a deep learning model, it achieves accurate short-term prediction of fish swarm distribution. Based on this, a multi-objective path optimization model considering the characteristics of heterogeneous fishing vessels is established, and a multi-stage co-evolutionary algorithm is designed to achieve global optimization of fishing vessel task allocation and fishing paths, thereby effectively improving the overall economic efficiency of the fleet and reducing fuel consumption and fishing costs. The key technical points of this application include:

[0165] 1. A fish school distribution prediction method based on multi-source heterogeneous data fusion. This method establishes a unified data fusion system by fusing multi-source heterogeneous data, including satellite remote sensing data (water surface temperature, salinity, etc.) and surface buoy data, providing high-precision input for fish school distribution prediction. A multi-feature spatiotemporal prediction model based on CNN-BiLSTM is employed to achieve dynamic modeling and future trend prediction of fish school distribution, capturing the nonlinear temporal variation patterns of fishing grounds.

[0166] 2. Multi-objective path optimization model for heterogeneous fishing vessels based on prediction results. A mathematical model for path planning of heterogeneous fishing vessels is constructed, considering constraints such as vessel load and fish location. The model aims to minimize both total delivery distance and the number of fishing vessels used, balancing fishing efficiency and operating costs to improve the overall operational efficiency of the fleet.

[0167] 3. Design and Optimization of Multi-Objective Cooperative Search Algorithm. A multi-objective cooperative search algorithm combining global perturbation and local search mechanisms is proposed. Operations such as task redistribution, path swapping, and structure reversal are used to improve search depth and solution diversity. Non-dominated sorting and elite preservation strategies are employed to achieve stable convergence of the Pareto solution set, obtaining high-quality multi-objective optimization results.

[0168] 4. Inter-model collaboration mechanism. The fish school distribution prediction results are integrated with the path planning model to achieve an integrated decision-making process from "fish school prediction → operation planning → fleet scheduling." This improves fishing efficiency, reduces energy consumption, and promotes the sustainable use of fishery resources.

[0169] In addition, this application has the following technical effects:

[0170] This application achieves scientific assessment and rational utilization of fishing ground resources through accurate prediction of the distribution areas of target fish species, effectively improving the targeting and capture efficiency of fishing vessel operations. Optimizing fishing areas based on the prediction results reduces aimless navigation and ineffective fishing activities, lowering energy consumption and operating costs. Furthermore, this application comprehensively considers the balance between the number of fishing vessels and the navigation distance during route optimization, improving the overall utilization rate of fishing resources and ensuring the coordination and efficiency of operational scheduling. Through these technical means, this application not only improves the economic benefits of fishing operations but also promotes the sustainable development of fishery resources and the protection of the ecological environment. Therefore, this application has high application value and promotional significance in the field of intelligent fishing and fleet scheduling optimization.

[0171] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by instructing related hardware through computer-readable instructions. These computer-readable instructions can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. The aforementioned storage medium can be a non-volatile storage medium such as a magnetic disk, optical disk, or read-only memory (ROM), or random access memory (RAM).

[0172] It should be understood that although the steps in the flowcharts of the accompanying figures are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the accompanying figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0173] Further reference Figure 3 As a response to the above Figure 2The implementation of the method shown in this application provides an embodiment of a data processing device for fisheries management, which is similar to... Figure 2 Corresponding to the method embodiments shown, this device can be specifically applied to various electronic devices.

[0174] like Figure 3 As shown, the data processing device 300 for fisheries management described in this embodiment includes: a first acquisition module 301, a prediction module 302, a construction module 303, a second acquisition module 304, a processing module 305, a decoding module 306, and an output module 307. Wherein:

[0175] The first acquisition module 301 is used to acquire pre-collected multi-source heterogeneous data; wherein, the multi-source heterogeneous data includes at least remote sensing and monitoring data, fishery data, and environmental data;

[0176] The prediction module 302 is used to perform prediction processing on the multi-source heterogeneous data based on a preset fish distribution prediction model to obtain the corresponding fish distribution prediction result.

[0177] The construction module 303 is used to perform model construction processing based on the fish distribution prediction results to obtain the corresponding heterogeneous fishing vessel multi-objective path planning model.

[0178] The second acquisition module 304 is used to acquire a preset multi-objective cooperative search algorithm corresponding to the heterogeneous fishing vessel multi-objective path planning model.

[0179] Processing module 305 is used to solve the heterogeneous fishing vessel multi-objective path planning model based on the multi-objective cooperative search algorithm to obtain the corresponding target optimization solution set;

[0180] Decoding module 306 is used to decode the target optimization solution set to obtain the corresponding fishing boat planning path;

[0181] The output module 307 is used to output the planned path of the fishing vessel.

[0182] In this embodiment, the operations performed by the above-mentioned modules or units correspond one-to-one with the steps of the data processing method for fisheries management in the aforementioned implementation method, and will not be repeated here.

[0183] In some optional implementations of this embodiment, the fish swarm distribution prediction model includes an input layer, a feature extraction layer, a temporal modeling layer, and an output layer; the prediction module 302 includes:

[0184] The first processing submodule is used to input the multi-source heterogeneous data into the fish distribution prediction model, and perform data preprocessing and feature construction on the multi-source heterogeneous data based on the input layer to obtain the corresponding processed data.

[0185] An extraction submodule is used to extract features from the processed data based on the feature extraction layer to obtain corresponding first feature data;

[0186] The second processing submodule is used to process the feature data based on the time series modeling layer to obtain the corresponding second feature data;

[0187] The prediction submodule is used to perform prediction processing on the first feature data and the second feature data based on the output layer to obtain the corresponding prediction result;

[0188] The first determining submodule is used to use the prediction result as the fish population distribution prediction result.

[0189] In some optional implementations of this embodiment, the first processing submodule includes:

[0190] The first processing unit is used to perform missing value completion processing on the multi-source heterogeneous data based on the input layer to obtain the corresponding first processed data.

[0191] The second processing unit is used to normalize the first processed data to obtain the corresponding second processed data.

[0192] The third processing unit is used to perform feature construction processing on the second processing data based on a preset sliding window mechanism to obtain the corresponding third processing data.

[0193] The first determining unit is used to use the third processed data as the processed data.

[0194] In some optional implementations of this embodiment, the processing module 305 includes:

[0195] A generation submodule is used to set the algorithm parameters corresponding to the multi-objective collaborative search algorithm, and to generate an initial population solution set based on the fishing vessel type and fishing point information in the heterogeneous fishing vessel multi-objective path planning model; wherein, the algorithm parameters include at least the population size and the maximum number of iterations;

[0196] The first operation submodule is used to perform a global perturbation operation on the initial population solution set to generate a new first type of population solution set;

[0197] The second operation submodule is used to perform local search optimization operation on the first type of group solution set based on a preset local search optimization strategy to generate an optimized second type of group solution set.

[0198] The third processing submodule is used to perform non-dominated sorting and elite retention processing on the initial population solution set and the second population solution set to obtain a third population solution set that matches the population size.

[0199] The judgment submodule is used to determine whether the current iteration count has reached the maximum iteration count;

[0200] The second determining submodule is used to stop the iteration if the condition is met, obtain a specified optimal solution set from the third group solution set, and use the specified optimal solution set as the target optimized solution set.

[0201] In some optional implementations of this embodiment, the generation submodule includes:

[0202] The first acquisition unit is used to acquire information on fishing vessel type and fishing point in the heterogeneous fishing vessel multi-objective path planning model.

[0203] The second acquisition unit is used to acquire a preset multi-segment encoding method based on fishing vessel type;

[0204] The encoding unit is used to perform path encoding processing based on the fishing vessel type and the fishing point information using the multi-segment encoding method based on the fishing vessel type to obtain the corresponding initial decoding structure.

[0205] The second determining unit is used to use the initial solution encoding structure as the initial population solution set.

[0206] In some optional implementations of this embodiment, the first operation submodule includes:

[0207] The third acquisition unit is used to acquire randomly generated random numbers within a preset numerical range;

[0208] The fourth acquisition unit is used to acquire two preset global perturbation methods;

[0209] A filtering unit is used to filter out the target global perturbation mode corresponding to the random number from the global perturbation modes;

[0210] The first operation unit is used to perform a global perturbation operation on the initial population solution set based on the target global perturbation method to generate a corresponding specified population solution set;

[0211] The third determining unit is used to take the specified population solution set as the first population solution set.

[0212] In some optional implementations of this embodiment, the second operation submodule includes:

[0213] The second operation unit is used to perform a single-task-point relocation operation on the first population solution set to obtain the corresponding first target population solution set.

[0214] The third operation unit is used to perform a whole path task reallocation operation on the first target population solution set to obtain the corresponding second target population solution set.

[0215] The fourth operation unit is used to perform a path internal structure reversal adjustment operation on the first target population solution set to obtain the corresponding third target population solution set;

[0216] The fifth operation unit is used to integrate the first target population solution set, the second target population solution set, and the third target population solution set to obtain the corresponding integrated population solution set.

[0217] The fourth determining unit is used to use the integrated population solution set as the second population solution set.

[0218] To address the aforementioned technical problems, embodiments of this application also provide a computer device. Please refer to [link / reference needed]. Figure 4 , Figure 4 This is a basic structural block diagram of the computer device in this embodiment.

[0219] The computer device 4 includes a memory 41, a processor 42, and a network interface 43 that are interconnected via a system bus. It should be noted that only the computer device 4 with components 41-43 is shown in the figure; however, it should be understood that it is not required to implement all the shown components, and more or fewer components can be implemented alternatively. Those skilled in the art will understand that the computer device described here is a device capable of automatically performing numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes, but is not limited to, microprocessors, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.

[0220] The computer device can be a desktop computer, laptop, handheld computer, or cloud server, etc. The computer device can interact with the user via a keyboard, mouse, remote control, touchpad, or voice control.

[0221] The memory 41 includes at least one type of readable storage medium, including flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 41 may be an internal storage unit of the computer device 4, such as the hard disk or memory of the computer device 4. In other embodiments, the memory 41 may also be an external storage device of the computer device 4, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the computer device 4. Of course, the memory 41 may also include both the internal storage unit and its external storage device of the computer device 4. In this embodiment, the memory 41 is typically used to store the operating system and various application software installed on the computer device 4, such as computer-readable instructions for data processing methods applied to fisheries management. In addition, the memory 41 can also be used to temporarily store various types of data that have been output or will be output.

[0222] In some embodiments, the processor 42 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chip. The processor 42 is typically used to control the overall operation of the computer device 4. In this embodiment, the processor 42 is used to execute computer-readable instructions stored in the memory 41 or to process data, for example, to execute computer-readable instructions for the data processing method applied to fisheries management.

[0223] The network interface 43 may include a wireless network interface or a wired network interface, which is typically used to establish communication connections between the computer device 4 and other electronic devices.

[0224] This application also provides another embodiment, namely, providing a computer-readable storage medium storing computer-readable instructions that can be executed by at least one processor to cause the at least one processor to perform the steps of the data processing method applied to fisheries management as described above.

[0225] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0226] Obviously, the embodiments described above are only some embodiments of this application, not all embodiments. The accompanying drawings show preferred embodiments of this application, but do not limit the patent scope of this application. This application can be implemented in many different forms; rather, the purpose of providing these embodiments is to provide a more thorough and comprehensive understanding of the disclosure of this application. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing specific embodiments, or make equivalent substitutions for some of the technical features. Any equivalent structures made using the content of this application's specification and drawings, directly or indirectly applied to other related technical fields, are similarly within the scope of patent protection of this application.

Claims

1. A data processing method applied to fisheries management, characterized in that, Includes the following steps: Acquire pre-collected multi-source heterogeneous data; wherein, the multi-source heterogeneous data includes at least remote sensing and monitoring data, fisheries data, and environmental data; The multi-source heterogeneous data is processed based on a preset fish distribution prediction model to obtain the corresponding fish distribution prediction results. Based on the fish distribution prediction results, a model construction process is performed to obtain the corresponding heterogeneous fishing vessel multi-objective path planning model. Obtain a preset multi-objective cooperative search algorithm corresponding to the heterogeneous fishing vessel multi-objective path planning model; The multi-objective cooperative search algorithm is used to solve the multi-objective path planning model of the heterogeneous fishing vessel to obtain the corresponding target optimization solution set. The target optimization solution set is decoded to obtain the corresponding fishing boat planning path; The planned route for the fishing vessel is output and processed. The step of solving the multi-objective path planning model of the heterogeneous fishing vessels based on the multi-objective cooperative search algorithm to obtain the corresponding target optimization solution set specifically includes: Set the algorithm parameters corresponding to the multi-objective collaborative search algorithm, and generate an initial population solution set based on the fishing vessel type and fishing point information in the heterogeneous fishing vessel multi-objective path planning model; wherein, the algorithm parameters include at least the population size and the maximum number of iterations; A global perturbation operation is performed on the initial population solution set to generate a new first type of population solution set; Based on a preset local search optimization strategy, a local search optimization operation is performed on the first type of group solution set to generate an optimized second type of group solution set. The initial population solution set and the second population solution set are subjected to non-dominated sorting and elite retention processing to obtain a third population solution set that matches the population size. Determine whether the current iteration count has reached the maximum iteration count; If so, stop the iteration, obtain the specified optimal solution set from the third group solution set, and use the specified optimal solution set as the target optimization solution set; The step of performing a local search optimization operation on the first type of group solution set based on a preset local search optimization strategy to generate an optimized second type of group solution set specifically includes: Perform a single-task-point relocation operation on the first population solution set to obtain the corresponding first target population solution set; Perform a whole-path task reallocation operation on the first target population solution set to obtain the corresponding second target population solution set; Perform a path internal structure reversal adjustment operation on the first target population solution set to obtain the corresponding third target population solution set; The first target population solution set, the second target population solution set, and the third target population solution set are integrated to obtain the corresponding integrated population solution set. The integrated population solution set is used as the second population solution set; The single-task point relocation operation includes: selecting two paths from the fishing vessel path set, removing one fishing point from one path, and inserting it into the optimal position of the other path, i.e., the position that minimizes the change in the total path distance. If the new path satisfies the constraints and reduces the total travel distance, then the path is updated. The whole-path task reassignment operation includes: selecting one path from the fishing vessel path set and removing it, inserting the fishing points in that path one by one into the optimal positions of other fishing vessel paths. If the new solution satisfies the constraints and improves on any objective function, then the solution is updated. The path internal structure reversal adjustment operation includes: for the internal structure of a single fishing vessel path, selecting two edges, reversing the path segments between the two edges, and reconnecting the path structure. If the new path satisfies the constraints and reduces the travel distance, then the path is updated.

2. The data processing method for fisheries management according to claim 1, characterized in that, The fish distribution prediction model includes an input layer, a feature extraction layer, a temporal modeling layer, and an output layer; the step of performing prediction processing on the multi-source heterogeneous data based on the preset fish distribution prediction model to obtain the corresponding fish distribution prediction results specifically includes: The multi-source heterogeneous data is input into the fish distribution prediction model. Based on the input layer, the multi-source heterogeneous data is preprocessed and feature constructed to obtain the corresponding processed data. Based on the feature extraction layer, feature extraction is performed on the processed data to obtain the corresponding first feature data; The feature data is processed based on the time-series modeling layer to obtain the corresponding second feature data; Based on the output layer, the first feature data and the second feature data are predicted to obtain the corresponding prediction results; The prediction results are used as the prediction results for the fish population distribution.

3. The data processing method for fisheries management according to claim 2, characterized in that, The step of performing data preprocessing and feature construction on the multi-source heterogeneous data based on the input layer to obtain corresponding processed data specifically includes: Based on the input layer, missing value completion processing is performed on the multi-source heterogeneous data to obtain the corresponding first processed data; The first processed data is normalized to obtain the corresponding second processed data; The second processed data is processed by feature construction based on a preset sliding window mechanism to obtain the corresponding third processed data; The third processed data is used as the processed data.

4. The data processing method for fisheries management according to claim 1, characterized in that, The step of generating an initial population solution set based on the fishing vessel type and fishing point information in the heterogeneous fishing vessel multi-objective path planning model specifically includes: Obtain the fishing vessel type and fishing point information from the heterogeneous fishing vessel multi-objective path planning model; Obtain the preset multi-segment encoding method based on fishing vessel type; Based on the fishing vessel type and the fishing point information, the path encoding process is performed using the multi-segment encoding method based on the fishing vessel type to obtain the corresponding initial decoding structure. The initial solution encoding structure is used as the initial population solution set.

5. The data processing method for fisheries management according to claim 1, characterized in that, The step of performing a global perturbation operation on the initial population solution set to generate a new first type of population solution set specifically includes: Get a randomly generated number within a preset numerical range; Obtain two preset global perturbation methods; Select the target global perturbation method corresponding to the random number from the global perturbation methods; Based on the target global perturbation method, a global perturbation operation is performed on the initial population solution set to generate the corresponding specified population solution set; The specified population solution set is used as the first population solution set.

6. A data processing device for fisheries management, characterized in that, include: The first acquisition module is used to acquire pre-collected multi-source heterogeneous data; wherein, the multi-source heterogeneous data includes at least remote sensing and monitoring data, fishery data, and environmental data; The prediction module is used to perform prediction processing on the multi-source heterogeneous data based on a preset fish distribution prediction model to obtain the corresponding fish distribution prediction results. The construction module is used to perform model construction processing based on the fish distribution prediction results to obtain the corresponding heterogeneous fishing vessel multi-objective path planning model. The second acquisition module is used to acquire a preset multi-objective cooperative search algorithm corresponding to the heterogeneous fishing vessel multi-objective path planning model. The processing module is used to solve the heterogeneous fishing vessel multi-objective path planning model based on the multi-objective cooperative search algorithm to obtain the corresponding target optimization solution set; The decoding module is used to decode the target optimization solution set to obtain the corresponding fishing boat planning path. The output module is used to process the planned path of the fishing vessel. The processing module includes: A generation submodule is used to set the algorithm parameters corresponding to the multi-objective collaborative search algorithm, and to generate an initial population solution set based on the fishing vessel type and fishing point information in the heterogeneous fishing vessel multi-objective path planning model; wherein, the algorithm parameters include at least the population size and the maximum number of iterations; The first operation submodule is used to perform a global perturbation operation on the initial population solution set to generate a new first type of population solution set; The second operation submodule is used to perform local search optimization operation on the first type of group solution set based on a preset local search optimization strategy to generate an optimized second type of group solution set. The third processing submodule is used to perform non-dominated sorting and elite retention processing on the initial population solution set and the second population solution set to obtain a third population solution set that matches the population size. The judgment submodule is used to determine whether the current iteration count has reached the maximum iteration count; The second determining submodule is used to stop the iteration if the condition is met, and to obtain a specified optimal solution set from the third group solution set, and to use the specified optimal solution set as the target optimization solution set. The second operation submodule includes: The second operation unit is used to perform a single-task-point relocation operation on the first population solution set to obtain the corresponding first target population solution set. The third operation unit is used to perform a whole path task reallocation operation on the first target population solution set to obtain the corresponding second target population solution set. The fourth operation unit is used to perform a path internal structure reversal adjustment operation on the first target population solution set to obtain the corresponding third target population solution set; The fifth operation unit is used to integrate the first target population solution set, the second target population solution set, and the third target population solution set to obtain the corresponding integrated population solution set. The fourth determining unit is used to use the integrated population solution set as the second population solution set; The single-task point relocation operation includes: selecting two paths from the fishing vessel path set, removing one fishing point from one path, and inserting it into the optimal position of the other path, i.e., the position that minimizes the change in the total path distance. If the new path satisfies the constraints and reduces the total travel distance, then the path is updated. The whole-path task reassignment operation includes: selecting one path from the fishing vessel path set and removing it, inserting the fishing points in that path one by one into the optimal positions of other fishing vessel paths. If the new solution satisfies the constraints and improves on any objective function, then the solution is updated. The path internal structure reversal adjustment operation includes: for the internal structure of a single fishing vessel path, selecting two edges, reversing the path segments between the two edges, and reconnecting the path structure. If the new path satisfies the constraints and reduces the travel distance, then the path is updated.

7. A computer device, characterized in that, The system includes a memory and a processor, wherein the memory stores computer-readable instructions, and the processor executes the computer-readable instructions to implement the steps of the data processing method for fisheries management as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-readable instructions, which, when executed by a processor, implement the steps of the data processing method for fisheries management as described in any one of claims 1 to 5.

Images