Marine fishery resource planning method based on big data

Abstract

The invention discloses a marine fishery resource planning method based on big data, and relates to the technical field of fishery resources, and the method comprises the steps: dividing an operation sea area into a plurality of grids, and obtaining the initial attribute of each grid; collecting a real-time data set of each fishing boat; calculating the ship admission priority of each grid to realize multi-ship collaborative path planning; the grids are combined into a plurality of block areas according to a primary selection rule, resource fishing indexes of the block areas are calculated based on initial attributes of the grids, path efficiency indexes are obtained according to the resource fishing indexes, and efficiency path planning is output; when the ship yaws in the block area, an optimal path plan is generated; and when the ship crosses the boundary of the block area to yaw, generating a migration path plan based on a resource fishing index and a path efficiency index, and triggering a dynamic compensation task redistribution mechanism to collaboratively optimize resource distribution at the same time.

Description

Technical Field

[0001] The present invention relates to the technical field of fishery resources, and in particular to a method for planning marine fishery resources based on big data. Background Art

[0002] Driven by global marine ecological sustainability goals, the field of fisheries resources has developed a multidisciplinary and integrated technical framework encompassing knowledge from diverse fields, including population dynamics, ecology, statistics, and oceanography. This framework aims to scientifically assess, monitor, and manage aquatic biological resources to address the challenge of overfishing. The core of this current technical framework includes: utilizing population assessment models to quantify resource dynamics; developing modern monitoring methods such as electronic monitoring, vessel monitoring systems, satellite remote sensing, acoustic surveys, and environmental DNA technology to enhance data acquisition capabilities; and promoting an ecosystem approach that integrates ecological interactions and environmental factors.

[0003] Chinese invention patent publication number CN115293658B provides a big data-based fishery resource planning method and system. This system uses ocean imagery and sonar data for resource analysis, real-time navigational information for production vessels, and deviation calculation and secondary route analysis. However, its applicability in multi-vessel collaborative and dynamic environments is limited by its support for single-vessel operations, its reliance on static area divisions resulting in insufficient dynamic response, its limited yaw correction range, its lack of a task reallocation mechanism, and its neglect of real-time load impacts in ecological assessments. Summary of the Invention

[0004] This application provides a big data marine fishery resource planning method to solve the problems of disordered resource competition, delayed dynamic environmental response, localized yaw correction range and static ecological assessment in multi-vessel operation scenarios in the existing technology. Through the dynamic grid coordination mechanism, multi-vessel signal orthogonal coding and yaw response strategy, it achieves the technical effect of improving the efficiency of multi-vessel collaborative operations and sustainable resource utilization.

[0005] This application provides a big data marine fishery resource planning method, including: S1: Divide the operating area into several grids and obtain the initial attributes of each grid. Collect real-time data sets from each fishing vessel, build a multidimensional spatial data model based on the initial attributes and real-time data sets, and generate a dynamic resource map. Based on the dynamic resource map, calculate the vessel access priority of each grid and implement multi-vessel collaborative path planning. S2: Combine the grids into several block areas according to the preliminary selection rules, calculate the resource capture index of the block area based on the initial attributes of the grid, obtain the path efficiency index based on the resource capture index, modify the collaborative path planning based on the path efficiency index, and output the efficient path planning; S3: When the ship deviates within the block area, it searches for grids in adjacent blocks with a resource fishing index greater than the current block area and a load rate less than 60%, and generates the optimal path planning; S4: When a ship deviates across the block area boundary, a migration path plan is generated based on the resource fishing index and path efficiency index, and a dynamic compensation task reallocation mechanism is triggered to collaboratively optimize resource distribution.

[0006] Furthermore, the initial attributes include: marine ecological diversity index, resource fishing index, maximum carrying capacity of ships and resource attenuation coefficient; The edge computing nodes are distributed around the operating sea area and are used to process ship data and resource information to form a fully covered distributed computing network.

[0007] Furthermore, the dynamic resource map is a multidimensional spatial data model dynamically generated by edge computing nodes, including: a basic attribute layer formed by edge nodes mapping initial attributes to spatial grid coordinates; a dynamic load layer formed by the real-time position, load value and sonar data of the ship obtained through orthogonal coding signals; and a spatiotemporal correction layer obtained based on environmental disturbances and resource trends.

[0008] Furthermore, the collaborative path planning includes the following steps: Based on the real-time load value, the ship access priority of each grid is calculated; based on the access priority, a basic path plan is generated through a path cost model; the ship speed is dynamically adjusted according to the basic path plan to ensure that the load value of the target grid is less than 80% upon arrival; and finally, a collaborative path planning instruction set is output, which includes the ship position sequence, arrival time window and communication frequency band allocation.

[0009] Furthermore, the preliminary selection rule includes: calculating the value score of each grid according to the initial attributes of each grid, selecting grids with high value scores in order based on the value scores, and combining adjacent grids to form a block area; The resource fishing index of the block area is a comprehensive sea area efficiency evaluation value calculated in real time based on the initial attributes of the grid; The path efficiency indicators include: resource acquisition efficiency, time cost efficiency and ecological sustainability index.

[0010] Furthermore, the resource fishing index also includes: generating a regional priority ranking table based on resource abundance value assessment data, identifying the location and range of high-priority areas, screening out high-priority fishing areas, and avoiding low-priority fishing areas.

[0011] Furthermore, the efficiency path planning includes the following steps: Based on the path efficiency index, the expected comprehensive benefit of the ship from the current position to the target grid is calculated; according to the regional priority ranking table, the top three candidate paths with the highest benefit values ​​are screened; based on the current load value and the number of ships arriving at the same time, the predicted load rate of the target grid is calculated; based on the candidate paths, a target grid sequence is generated, and the recommended speed is dynamically determined according to the predicted load rate. In addition, a conflict-avoiding fishing time window is allocated based on the number of ships arriving at the same time; and the final efficiency path planning instruction set is output, which includes: target grid sequence, recommended speed and fishing time window.

[0012] Furthermore, the optimal path planning includes the following steps: Retrieve grids in the shared boundary; screen candidate grids with a resource fishing index greater than the current grid and a load rate less than 60%; calculate the path priority of the candidate grids, select the grid with the highest priority as the target grid, output its center position coordinates, and dynamically determine the recommended speed based on the predicted load rate of the target grid; generate an optimal path instruction set, including the target grid coordinates and the recommended speed.

[0013] Furthermore, the migration path planning includes: based on the dynamic resource map and path efficiency index, searching the entire sea area block area to determine the target with the highest resource fishing index; generating the target block area coordinates according to the block area spatial attributes of the target; calculating the collaborative migration time window based on the distance from the current position of the ship to the target and the recommended speed; allocating anti-interference communication frequency bands through the orthogonal coding mechanism; generating a migration instruction set to guide the collaborative migration of ships, and the migration instruction set includes: target block area coordinates, collaborative migration time window and communication frequency band allocation.

[0014] Furthermore, the dynamic compensation task reallocation mechanism includes: parsing the migration instruction set to obtain the coordinates of the target block area; dynamically allocating the original planned fishing task of the yawed ship to the ship with the lowest load rate or the closest distance in the target block area; calculating the compensation coefficient based on the real-time resource abundance value of the target block area, and increasing the fishing intensity of the assigned ship according to the coefficient; and optimizing the resource distribution in the entire sea area through the synergistic effect of migration path planning and compensation intensity improvement.

[0015] One or more technical solutions provided in this application have at least the following technical effects or advantages: By dividing the sea area into grids in real time and linking edge computing nodes, we can accurately perceive fish migration and environmental changes, and eliminate interference between sonar data from multiple ships; with the help of a cross-block collaborative migration mechanism, we can intelligently reallocate tasks and quickly compensate for resource gaps caused by yaw; at the same time, we can couple the load value and the ecological attenuation coefficient to dynamically balance fishing intensity. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a flow chart of a method for planning marine fishery resources based on big data in an embodiment of the present invention. DETAILED DESCRIPTION

[0017] To facilitate understanding of the present invention, the present application will be described more comprehensively below with reference to the relevant drawings; the drawings show preferred embodiments of the present invention, but the present invention can be implemented in many different forms and is not limited to the embodiments described herein; on the contrary, the purpose of providing these embodiments is to enable a more thorough and comprehensive understanding of the disclosed content of the present invention.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this invention pertains; the terms used herein in the specification of the present invention are for the purpose of describing specific embodiments only and are not intended to limit the present invention; the term "and / or" used herein includes any and all combinations of one or more of the associated listed items.

[0019] Example 1: Figure 1 As shown in Figure 1, a big data approach to marine fishery resource planning is presented.

[0020] S1: Divide the operating area into several grids and obtain the initial attributes of each grid. Collect real-time data sets from each fishing vessel, build a multidimensional spatial data model based on the initial attributes and real-time data sets, and generate a dynamic resource map. Based on the dynamic resource map, calculate the vessel access priority of each grid and implement multi-vessel collaborative path planning. The initial attributes include: marine ecological diversity index, resource fishing index, maximum carrying capacity of ships and resource attenuation coefficient; Specifically, a sea area with a radius of 10 km is divided into 200m×200m grids, and the initial attributes of each grid are marked. The initial attributes include: marine ecological diversity index, resource fishing index, maximum carrying capacity of ships and resource decay coefficient; the marine ecological diversity index includes but is not limited to: species composition abundance, habitat quality parameters and dynamic quantitative indicators for community stability assessment; the resource fishing index includes but is not limited to: resource abundance value, economic value coefficient, catchable coefficient and historical catch trend; the maximum carrying capacity of ships includes but is not limited to: grid area, single-vessel safe operating area, ship type adjustment coefficient and real-time load control benchmark; the resource decay coefficient includes but is not limited to: natural decay benchmark value, fishing intensity, environmental disturbance and resource recovery rate; The edge computing nodes are distributed around the operating sea area and are used to process ship data and resource information to form a fully covered distributed computing network: Specifically, the nodes are deployed in offshore areas 5-20 nautical miles from the coastline in the operating area, arranged in a hexagonal cellular topology. Each node is responsible for data processing in a 50-kilometer radius of the sea, and the overlapping coverage area of ​​adjacent nodes is ≥10 kilometers. The nodes are dually connected via submarine optical cables and satellite links to form a dual-ring redundant network. The nodes receive and decode orthogonal coded signals in real time to obtain the ship's position, speed, and sonar detection results. The dynamic resource map is a multidimensional spatial data model dynamically generated by edge computing nodes, including: a basic attribute layer formed by edge nodes mapping initial attributes to spatial grid coordinates; a dynamic load layer formed by the real-time position, load value and sonar data of ships obtained through orthogonal coding signals; and a spatiotemporal correction layer obtained based on environmental disturbances and resource trends.

[0021] Specifically, the edge computing node reads the initial attributes of each grid and maps them to the corresponding grid space coordinates. Each grid cell serves as a basic storage unit, providing baseline information on resource distribution, environmental carrying capacity limits, and ecological background. Edge computing nodes continuously receive signals from all ships covering their sea area of ​​responsibility. Using orthogonal coding technology, the nodes can simultaneously receive and accurately separate signals from different ships, extracting the ship's real-time position, ship load value, and sonar data from the decoded signals. The nodes summarize the load values ​​of all ships currently located in a certain grid and calculate the grid's real-time load rate based on the grid's maximum capacity. This reflects the distribution of ships in the operating area, resource consumption status, and the real-time status of local resources updated through sonar. The load value calculation formula is:

[0022] in, is the load value, n is the maximum number of ships, i is the current ship, is the vessel load factor, which is the weight of the resource consumption capacity of a single vessel. The coefficient for small fishing vessels is 1.0, the coefficient for medium-sized trawlers is 1.8, and the coefficient for large purse seine vessels is 2.5; is the maximum load reference value; is the base ship type coefficient, with a value of 1.0; Used to constrain overload conditions.

[0023] Edge nodes access real-time environmental data streams to analyze the impact of environmental changes on resource distribution. Combining historical resource data with current environmental changes, they use prediction models to predict possible short-term resource spatial trends. By incorporating dynamic environmental changes and resource migration trends into the model, future-oriented corrections are made to relatively static basic attributes and dynamic load information reflecting current conditions, giving resource maps predictive and adaptive capabilities. The basic attribute layer, dynamic load layer and spatiotemporal correction layer are integrated in a unified grid space framework to eventually generate a dynamic digital map covering the entire operation area with multi-dimensional attributes.

[0024] The collaborative path planning includes the following steps: calculating the ship access priority of each grid according to the real-time load value; generating a basic path plan based on the access priority through a path cost model; dynamically adjusting the ship speed according to the basic path plan to ensure that the load value of the target grid is less than 80% upon arrival; and finally outputting a collaborative path planning instruction set, which includes a ship position sequence, an arrival time window, and a communication frequency band allocation.

[0025] Specifically, the grid accessibility is dynamically evaluated and the ship access priority of each grid is calculated. The access priority is:

[0026] Among them, PJ is the access priority, RJ is the resource abundance value, which is obtained from the basic attribute layer of the resource map and has a value range of (0, 1]; EJ is the ecological diversity index (species abundance, habitat quality and historical data), with a value range of (0, 1]; A is the adjustment factor, which defaults to 1 and decreases when encountering environmental disturbances, such as 0.8 during wind and waves; Edge computing nodes update all grids every second And calculate PJ, and prioritize according to the obtained PJ: PJ>0.6 is high priority and can enter freely; 0.3≤PJ≤0.6 is medium priority and requires application for admission; PJ<0.3 is low priority and is prohibited from entering. Generate a grid admission priority matrix for path planning.

[0027] The current position of each ship is obtained based on the orthogonal code of each ship, and the path planning is carried out together with the resource map and the grid access priority matrix:

[0028] in, is the path cost, For distance, is the maximum distance of the current block area, is the maximum resource abundance value of the current block area, The maximum access priority for the operating sea area; starting from the current position, select the grid with the highest priority and resource abundance value greater than 0.7 in the neighborhood grid, conduct multi-vessel collaboration through edge nodes, detect conflicting paths, make adjustments, and output the basic path plan.

[0029] To prevent grid overload when a ship arrives, a predicted load value is calculated based on the real-time load change rate:

[0030] in, To predict the load value, is the estimated time of arrival (hours), is the ship speed; D is the decay rate (% / hour), , K is the disturbance factor, the value is -0.5, when encountering wind and waves, it is -2.0, is the resource recovery rate; if >80%, send a deceleration command or reselect the route. <60%, send acceleration command, if 60%≤ ≤80%, maintain the current speed; the edge node updates the prediction every 30 seconds and pushes adjustment instructions through orthogonal signals.

[0031] The final output is a collaborative path planning instruction set, which includes the ship position sequence, arrival time window and communication frequency band allocation.

[0032] The technical solutions in the above embodiments of the present application have at least the following technical effects or advantages: This application uses orthogonal coding signals to isolate ship sonar data to eliminate multi-ship interference, and combines the dynamic resource map and load value prediction model generated in real time by edge computing nodes to dynamically adjust the speed and access priority to achieve precise control of grid load rate. In the 200m grid division scenario of 10km sea area, the ship conflict rate is reduced by 92% and the operating efficiency is improved by 35%. At the same time, the resource attenuation coefficient is used to dynamically balance the fishing intensity, reducing the resource attenuation rate in ecologically sensitive areas by 60%, solving the problems of disordered resource competition among multiple ships, delayed environmental response and ecological static assessment.

[0033] Example 2: In Example 1, a dynamic resource map is generated by dividing the grid and deploying edge computing nodes, and multi-ship collaborative path planning is implemented based on orthogonal coding signals and load prediction models, which solves the problems of sonar interference and resource competition at the grid level. However, path optimization only focuses on a single grid, resulting in ships repeating operations between inefficient areas and insufficient global resource scheduling capabilities; and time cost and ecological sustainability indicators are not integrated, making it difficult to balance efficiency and ecological protection needs. This example further improves Example 1.

[0034] S2: Combine the grids into several block areas according to the preliminary selection rules, calculate the resource capture index of the block area based on the initial attributes of the grid, obtain the path efficiency index based on the resource capture index, modify the collaborative path planning based on the path efficiency index, and output the efficient path planning; The preliminary selection rules include: calculating the value score of each grid according to the initial attributes of each grid, selecting grids with high value scores in order based on the value scores, and combining adjacent grids to form a block area; Specifically, calculate the grid value:

[0035] in, is the grid value, is the grid row and column number, and All are normalized and the value range is [0,1]. For historical fishing trends, , is the maximum number of operations, is the number of operations, 0.6 and 0.4 are the corresponding weight coefficients; The grid with the maximum value is used as the seed. If there are multiple grids with the same maximum value, the row and column numbers are selected first ( ) The smallest grid is marked with the seed position, and the 8-direction grid is searched with the seed as the center to form a 3X3 matrix as the block area. If the neighboring grid in the boundary area is insufficient, it is marked as a reserved area and processed separately; The resource fishing index of the block area is a comprehensive sea area efficiency evaluation value calculated in real time based on the initial attributes of the grid; Specifically, adjacent 3×3 grids are combined into a block area, each of which covers a 600m×600m sea area. Based on the initial attributes of the grid, the resource fishing index of the block area is calculated:

[0036] in, is the resource fishing index of the block area, 9 is the number of grids in the block area composed of 3×3 grids, is the resource fishing index of the i-th grid, is the resource abundance value of the i-th grid, is the ecological diversity index of the i-th grid, 、 、 is the corresponding weight coefficient, the sum is 1, and the current , , , dominated by resource fishing index, is the dynamic attenuation factor of the i-th grid, and its calculation formula is: , is the resource attenuation coefficient of the i-th grid; The path efficiency indicators include: resource acquisition efficiency, time cost efficiency and ecological sustainability index.

[0037] Specifically, the resource acquisition efficiency is calculated based on the resource fishing index of the block area. The formula is:

[0038] Among them, RAE is resource acquisition efficiency, is the average time for ships to arrive at the block area (hours), is the resource capture efficiency coefficient, with a value of 0.7 for small fishing vessels, 0.85 for medium-sized trawlers, 1.0 for large purse seine vessels, and 0.3 for scientific research and monitoring vessels; Calculate the time cost efficiency, the formula is:

[0039] in, For time cost efficiency, is the economic value of fish stocks (yuan / ton), is the expected catch (tons), is the ship fuel efficiency (tons / nautical mile), is the current fuel price (yuan / ton), LD is the sailing distance of the target area (nautical miles), is the average cost per crew member (yuan / hour); Calculate the ecological sustainability index using the following formula:

[0040] in, is the ecological sustainability index, is the average resource attenuation coefficient of the block area, is the maximum attenuation threshold of the system, is the ecological correction factor, with a value range of [0.6,1], is the proportion of ecological protection area in the block area, is the total area of ​​the block; Calculate the path efficiency index based on resource acquisition efficiency, time cost efficiency and ecological sustainability index:

[0041] Among them, PE is the path efficiency index, 、 、 is the corresponding weight, the sum is 1, and the current =0.5, =0.3, =0.2, dominated by resource acquisition efficiency.

[0042] The resource fishing index also includes: generating a regional priority ranking table based on resource abundance value assessment data, identifying the location and range of high-priority areas, screening out high-priority fishing areas, and avoiding low-priority fishing areas.

[0043] Specifically, the resource fishing index for each block area Sort in descending order to generate the priority sequence of block areas in the entire sea area: , set the priority threshold:

[0044] in, is the priority threshold, is the priority coefficient, the value is 0.7, The maximum value of resource fishing index, filter Generate a high-priority area set H for the block area as the area priority sorting table; It is marked as a low priority area.

[0045] Mark the center point geographic coordinates for each high-priority block area , the calculation formula is:

[0046] in, , is the center coordinate of the i-th grid that makes up the block area.

[0047] When the ship path planning involves a low-priority block area, the edge node generates an avoidance instruction to replace all low-priority block areas in the ship's target path with the nearest high-priority block area in the H set.

[0048] The efficiency path planning includes the following steps: calculating the expected comprehensive benefit of a ship from its current position to a target grid based on a path efficiency index; screening the top three candidate paths with the highest benefit values ​​according to a regional priority ranking table; calculating the predicted load rate of the target grid based on the current load value and the number of ships arriving at the same time; generating a target grid sequence based on the candidate paths, dynamically determining a recommended speed based on the predicted load rate, and allocating a conflict-avoiding fishing time window based on the number of ships arriving at the same time; and outputting a final efficiency path planning instruction set, which includes: a target grid sequence, a recommended speed, and a fishing time window.

[0049] Specifically, the formula for calculating expected comprehensive income is:

[0050] in, is the expected comprehensive return, PE is the path efficiency indicator, is the sailing distance from the ship to the target block area, is the maximum diagonal distance of the operating sea area, is the ship type correction coefficient (fishing vessel: 1.0, scientific research vessel: 0.3, monitoring vessel: 0.1); the top three candidate paths with the highest benefit values ​​are selected as the optimal candidate paths.

[0051] According to the current load value and the number of ships arriving during the same period, the target grid predicted load rate is calculated using the following formula:

[0052] in, To predict the load factor, is the load value, is the maximum number of ships that the grid can carry, is the number of ships arriving during the same period, is the average load factor for other ships, which is 1.0 for small fishing vessels, 1.8 for medium-sized trawlers, 2.5 for large purse seine vessels, and 0.3 for scientific research and monitoring vessels.

[0053] Generate an ordered grid access sequence based on the selected optimal candidate path; determine the baseline speed Hs based on the ship type, and dynamically adjust the target grid's predicted load factor. When the predicted load factor is <60%, the recommended speed is Hs×1.2; when the predicted load factor is 60-80%, maintain the baseline speed; when the predicted load factor is >80%, the recommended speed is Hs×0.7; To resolve fishing conflicts caused by multiple vessels arriving at high-resource grids at the same time, a fishing time window is set, the UTC time format is unified, and a 15-minute buffer period is added to the estimated arrival time of the ship. The overlap of the time windows of ships in the same grid is detected in real time. When the overlap rate is greater than 30%, the starting time of the late-arriving ship is delayed by 15 minutes. The final output includes the target grid sequence, recommended speed, and fishing time window.

[0054] The technical solutions in the above embodiments of the present application have at least the following technical effects or advantages: This application dynamically aggregates 3×3 grids to form 600m×600m block areas, calculates the resource fishing index based on the dynamic weighting of resource abundance, ecological diversity and attenuation coefficient, generates a regional optimization ranking table to screen high-priority fishing areas, corrects the ship path based on the multi-objective path efficiency index, introduces the number of ships arriving at the same time to construct a predictive load rate model, and dynamically adjusts the speed and fishing time window simultaneously, systematically solving the local defects of grid-level planning and the problem of multi-vessel coordination conflicts.

[0055] Example 3: Example 2 predicts the load rate by the number of ships arriving at the same time and dynamically adjusts the speed and time window to achieve global resource scheduling optimization, but only supports global path replanning and cannot quickly respond to local sudden yaw. The yaw correction relies on the sorting of block areas in the entire sea area and does not utilize the local optimization potential of the current block area. This example further improves Example 2.

[0056] S3: When the ship deviates within the block area, it searches for grids in adjacent blocks with a resource fishing index greater than the current block area and a load rate less than 60%, and generates the optimal path planning; The optimal path planning includes the following steps: retrieving grids in the shared boundary; screening candidate grids with a resource fishing index greater than the current grid and a load rate less than 60 percent; calculating the path priority of the candidate grids, selecting the grid with the highest priority as the target grid, outputting the coordinates of its center position, and dynamically determining the recommended speed based on the predicted load rate of the target grid; and generating an optimal path instruction set, including the target grid coordinates and the recommended speed.

[0057] Specifically, when the ship's trajectory deviates from the target grid of the efficient path planning but does not cross the boundary of the current block area, the original task is not changed, only the path and speed are modified. In this way, only the boundary grid needs to be processed instead of the entire sea area, which can greatly reduce the amount of calculation. First, define the shared boundary:

[0058] in, For shared boundaries, , is the boundary grid between the current block area and the kth adjacent block area, It is a set of 6 adjacent block areas; only the boundary grids bordering the adjacent block areas are extracted to avoid traversal of the entire sea area grid.

[0059] Secondly, the grids with resource fishing index greater than the current grid and load rate less than 60% in the shared boundary grid are selected as candidate grids, and the path priority of the candidate grids is calculated:

[0060] in, is the path priority of the candidate grid, is the resource fishing index of the candidate grid, is the maximum value of resource fishing index in the current block area, is the real-time load value of the candidate grid, 0.6 and 0.4 are the corresponding weights; the grid with the highest path priority is selected from the candidate grid as the target grid:

[0061] in, is the target grid that is finally selected, argmax means finding the parameter that makes the function reach the maximum value, To traverse all grids G in the candidate grid set.

[0062] The speed is then dynamically adjusted based on the target grid load:

[0063] in, is the comprehensive load factor, is the number of other ships arriving during the same period, is the average load factor for other ships, which is 1.0 for small fishing vessels, 1.8 for medium-sized trawlers, 2.5 for large purse seine vessels, and 0.3 for scientific research and monitoring vessels. is the ship's load factor, is the maximum number of ships that the grid can support; when <60%, recommended speed ; 60%≤ When ≤80%, recommended speed ; >80%, recommended speed ;in, is the minimum economic speed of the ship, The maximum safe speed for a ship.

[0064] Finally, the optimal path instruction set is generated, including the target grid coordinates Center position and recommended speed . The technical solutions in the above embodiments of the present application have at least the following technical effects or advantages: This application accurately locates the shared boundary grids and only scans the grids at the intersection with the adjacent block areas, avoiding traversal of the entire sea area, significantly reducing the amount of calculation by 70% and achieving millisecond-level response. The system dynamically screens candidate grids that meet the dual conditions: the resource fishing index must be higher than the current grid and the real-time load rate must be lower than 60%, ensuring that the target area has both high resource value and low operating pressure. Based on the real-time status of the candidate grids, a priority calculation model is used to quantify the balance between resources and load, and the optimal target grid is selected. At the same time, a comprehensive load prediction model is introduced to integrate the load impact of ships arriving at the same time and dynamically adjust the speed: when When the speed is less than 60%, the speed is increased to 1.2Hs to improve efficiency. ... When ≤80%, maintain the reference speed Hs, When the speed is >80%, the speed is reduced to 0.7Hs to prevent overload. The solution finally generates the target grid coordinates. With real-time recommended speed The instruction set has achieved an 18% increase in ship operating efficiency and a 45% reduction in grid overload risk in actual measurements. It also reserved resource recovery space through load thresholds, reducing disturbances in ecologically sensitive areas by 30%, effectively solving the problem of dynamic resource optimization and ecological sustainability coordination in local yaw scenarios.

[0065] Example 4: Example 3 aims at the scenario where the ship yawing in the block area, and achieves rapid response through the local optimization mechanism. However, it can only handle the yaw within the block area and cannot handle the yaw outside the block area. This embodiment further improves Example 3.

[0066] S4: When a ship deviates across the block area boundary, a migration path plan is generated based on the resource fishing index and path efficiency index, and a dynamic compensation task reallocation mechanism is triggered to collaboratively optimize resource distribution.

[0067] The migration path planning includes: based on a dynamic resource map and path efficiency indicators, searching the entire sea area block area to determine the target with the highest resource fishing index; generating the target block area coordinates according to the block area spatial attributes of the target; calculating the collaborative migration time window based on the distance from the current position of the ship to the target and the recommended speed; allocating anti-interference communication frequency bands through an orthogonal coding mechanism; and generating a migration instruction set to guide the collaborative migration of ships. The migration instruction set includes: target block area coordinates, collaborative migration time window and communication frequency band allocation.

[0068] Specifically, when a ship deviates from the target grid of the efficiency path planning and crosses the boundary of the current block area, the original task is transferred to other ships and the ship receives the new task. , combined with the path efficiency index PE, confirm the migration target block area:

[0069] in, For the migration target block area, is the resource fishing index of target block area k, is the path efficiency index of the ship from the current position to the block area k, is the sailing distance from the current position of the ship to the center of block area k, 1 is the denominator protection item to prevent the denominator from being 0, is the distance attenuation factor, ranging from [0.01, 0.1], , is the average distance to the operating sea area, is a set of block regions, To select the parameter that maximizes the function value from the set.

[0070] get The central geographic coordinates of , the calculation formula is:

[0071] in, , It is composed The 9 grid center coordinates.

[0072] According to the current position of the ship distance and recommended speed Calculate the co-migration time window:

[0073] in, is the start time of the migration task, is the current UTC time, The migration task end time. The default value is 0.5 hours.

[0074] Assign orthogonal communication frequency bands to migrating ships:

[0075] in, For orthogonal communication frequency bands, is the reference frequency band, is the frequency band spacing, is the total number of co-migrating ships.

[0076] The migration instruction set is jointly constructed according to the target block area coordinates, collaborative migration time window and communication frequency band allocation.

[0077] The dynamic compensation task reallocation mechanism includes: parsing the migration instruction set to obtain the coordinates of the target block area; dynamically allocating the original planned fishing task of the yawed ship to the ship with the lowest load rate or the closest distance in the target block area; calculating the compensation coefficient based on the real-time resource abundance value of the target block area, and increasing the fishing intensity of the assigned ship according to the coefficient; and optimizing the resource distribution in the entire sea area through the synergistic effect of migration path planning and compensation intensity improvement.

[0078] Specifically, the edge node parses the migration instruction set and extracts the coordinates of the center of the target block area and its boundary range to obtain the target block area Real-time resource data, including resource abundance values, current ship distribution and remaining mission carrying capacity; Determine the basic fishing efficiency based on the type of the yawed vessel:

[0079] in, is the standardized task volume to be transferred (tons per hour), is the ship type efficiency coefficient (1.0 for fishing vessels and 0.3 for scientific research vessels), is the length of the originally planned fishing time window (in hours).

[0080] Select the ship with the lowest load rate and the ship closest to the original target position of the deviating ship in the target block area, and calculate the comprehensive priority formula:

[0081] in, For the comprehensive priority, is the sailing distance from the ship to the target block area, is the maximum diagonal distance of the operating sea area, 0.7 and 0.3 are the corresponding weight coefficients, and the load factor is dominant. Assigned to the highest priority ship.

[0082] Calculate the compensation coefficient using the formula:

[0083] in, is the compensation coefficient, is the real-time resource abundance value of the target block area, is the resource depletion threshold, which is 0.3. is the resource saturation threshold, with a value of 1.0. It is a dynamic adjustment factor, which is 0.5 in normal conditions and 0.2 in stormy weather.

[0084] Migration solves the problem of resource location after ships cross regions, and compensates for resource gaps caused by yaw through task transfer and intensity adjustment. Through synergy, load balancing and efficient resource utilization are achieved, thus globally optimizing resource distribution and utilization efficiency across the entire sea area.

[0085] The technical solutions in the above embodiments of the present application have at least the following technical effects or advantages: This application adopts the dual technologies of cross-block collaborative migration mechanism and dynamic compensation task reallocation. The edge computing node parses the target block area coordinates of the migration instruction set in real time, accurately guides the ship to migrate to the target area with the highest resource fishing index, and dynamically transfers the original planned fishing task of the deviating ship to other ships with the highest comprehensive priority in the target area. The fishing intensity is dynamically increased according to the compensation coefficient based on the real-time resource abundance value. Through the synergistic effect of migration and compensation, load balancing is achieved, with a 92% reduction in ship conflict rate and a 40% increase in target area resource utilization. The coupled resource attenuation coefficient constraint reduces the disturbance in ecologically sensitive areas by 60%. Finally, the 35% increase in the operating efficiency of the entire sea area fills the blind spot of the existing cross-regional response technology, and achieves a global optimization closed loop of efficient resource utilization and ecological sustainability.

[0086] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Various modifications and variations are readily apparent to those skilled in the art. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A method for marine fishery resource planning based on big data, characterized in that: include: S1: Divide the operating sea area into several grids and obtain the initial attributes of each grid; Collect real-time data sets from each fishing vessel, build a multidimensional spatial data model based on initial attributes and real-time data sets, and generate a dynamic resource map; Based on the dynamic resource map, the ship access priority of each grid is calculated to realize multi-vessel collaborative path planning; the initial attributes include: marine ecological diversity index, resource fishing index, maximum number of ships carrying capacity and resource attenuation coefficient; S2: Combine the grids into several block areas according to the preliminary selection rules, calculate the resource capture index of the block area based on the initial attributes of the grid, obtain the path efficiency index based on the resource capture index, modify the collaborative path planning based on the path efficiency index, and output the efficient path planning; S3: When the ship deviates within the block area, it searches for grids in adjacent blocks with a resource fishing index greater than the current block area and a load rate less than 60%, and generates the optimal path planning; S4: When a ship deviates from its course across the block boundary, a migration path plan is generated based on the resource fishing index and path efficiency index, and a dynamic compensation task reallocation mechanism is triggered to collaboratively optimize resource distribution. The migration path planning includes: based on a dynamic resource map and path efficiency indicators, searching the entire sea area block area to determine the target with the highest resource fishing index; generating the target block area coordinates according to the block area spatial attributes of the target; calculating the collaborative migration time window based on the distance from the current position of the ship to the target and the recommended speed; allocating anti-interference communication frequency bands through an orthogonal coding mechanism; and generating a migration instruction set to guide the collaborative migration of ships. The migration instruction set includes: target block area coordinates, collaborative migration time window and communication frequency band allocation.

2. A method for marine fishery resource planning based on big data as claimed in claim 1, characterized in that: The dynamic resource map is a multidimensional spatial data model dynamically generated by edge computing nodes, including: a basic attribute layer formed by edge nodes mapping initial attributes to spatial grid coordinates; a dynamic load layer formed by real-time ship positions, load values, and sonar data obtained through orthogonal coding signals; and a spatiotemporal correction layer obtained based on environmental disturbances and resource trends. The edge computing nodes are distributed around the operating sea area and are used to process ship data and resource information to form a fully covered distributed computing network.

3. A method for marine fishery resource planning based on big data as claimed in claim 1, characterized in that: The collaborative path planning includes the following steps: Based on the real-time load value, the ship access priority of each grid is calculated; based on the access priority, a basic path plan is generated through a path cost model; the ship speed is dynamically adjusted according to the basic path plan to ensure that the load value of the target grid is less than 80% upon arrival; and finally, a collaborative path planning instruction set is output, which includes the ship position sequence, arrival time window and communication frequency band allocation.

4. A method for marine fishery resource planning based on big data as claimed in claim 1, characterized in that: The preliminary selection rules include: calculating the value score of each grid according to the initial attributes of each grid, selecting grids with high value scores in order based on the value scores, and combining adjacent grids to form a block area; The resource fishing index of the block area is a comprehensive sea area efficiency evaluation value calculated in real time based on the initial attributes of the grid; The path efficiency indicators include: resource acquisition efficiency, time cost efficiency and ecological sustainability index.

5. A method for marine fishery resource planning based on big data as claimed in claim 4, characterized in that: The resource fishing index also includes: generating a regional priority ranking table based on resource abundance value assessment data, identifying the location and range of high-priority areas, screening out high-priority fishing areas, and avoiding low-priority fishing areas.

6. A method for marine fishery resource planning based on big data as claimed in claim 1, characterized in that: The efficiency path planning includes the following steps: Based on the path efficiency index, the expected comprehensive benefit of the ship from the current position to the target grid is calculated; according to the regional priority ranking table, the top three candidate paths with the highest benefit values ​​are screened; based on the current load value and the number of ships arriving at the same time, the predicted load rate of the target grid is calculated; based on the candidate paths, a target grid sequence is generated, and the recommended speed is dynamically determined according to the predicted load rate. In addition, a conflict-avoiding fishing time window is allocated based on the number of ships arriving at the same time; and the final efficiency path planning instruction set is output, which includes: target grid sequence, recommended speed and fishing time window.

7. A method for marine fishery resource planning based on big data as claimed in claim 1, characterized in that: The optimal path planning includes the following steps: Retrieve grids in the shared boundary; screen candidate grids with a resource fishing index greater than the current grid and a load rate less than 60%; calculate the path priority of the candidate grids, select the grid with the highest priority as the target grid, output its center position coordinates, and dynamically determine the recommended speed based on the predicted load rate of the target grid; generate an optimal path instruction set, including the target grid coordinates and the recommended speed.

8. A method for marine fishery resource planning based on big data as claimed in claim 1, characterized in that: The dynamic compensation task reallocation mechanism includes: parsing the migration instruction set to obtain the coordinates of the target block area; dynamically allocating the original planned fishing task of the yawed ship to the ship with the lowest load rate or the closest distance in the target block area; calculating the compensation coefficient based on the real-time resource abundance value of the target block area, and increasing the fishing intensity of the assigned ship according to the coefficient; and optimizing the resource distribution in the entire sea area through the synergistic effect of migration path planning and compensation intensity improvement.

Images