Intelligent method and system for marine ranch resource assessment and management

Abstract

The invention discloses an intelligent method and system for marine ranch resource evaluation and management, and relates to the technical field of marine ranch resources, which comprises the steps of fusing various data sources such as satellite remote sensing, underwater acoustic detection and buoy sensors to carry out data acquisition, and carrying out data processing and evaluation by utilizing an evaluation model based on machine learning. And generating a management strategy by adopting an intelligent optimization algorithm, and performing display and interaction through a visual interface. The system comprises a data acquisition unit, a data processing unit, a management decision unit and a visual interaction unit. Through application of multi-source data fusion and the intelligent model, accurate assessment and dynamic management of marine ranching resources are realized, resource protection and utilization effects are improved, and sustainable development of the marine ranching is promoted.

Description

Technical Field

[0001] The present invention relates to the technical field of marine ranch resources, and in particular to an intelligent method and system for marine ranch resource assessment and management. Background Art

[0002] In the field of marine ranch resource assessment and management, relevant technologies have advanced significantly in recent years. Traditional methods primarily rely on single monitoring methods, such as manual surveys or simple instrumental measurements, to obtain partial environmental and biological information about marine ranches. Furthermore, early management systems were based on fixed rules and empirical formulas for resource assessment and management decisions, resulting in limited functionality and poor adaptability. With technological advances, improved methods have begun to emerge, integrating multiple monitoring technologies and incorporating data processing models. However, these methods are still in the process of gradual improvement and have yet to achieve comprehensive intelligent and dynamic management.

[0003] Existing technologies for marine ranch resource assessment and management have numerous shortcomings. On the one hand, data collection is insufficiently comprehensive and accurate, and most methods fail to effectively integrate multiple data sources such as satellite remote sensing, underwater acoustic detection, and buoy sensors. This results in limited information on marine ranches, which fails to accurately reflect the true state of the resources. On the other hand, resource assessment models are relatively simplistic, making it difficult to accurately analyze the complex interactions between biological populations and the dynamic response relationships between organisms and the environment, resulting in inaccurate and unreliable assessment results. Furthermore, management decisions lack flexibility and foresight, making it impossible to quickly generate dynamic management strategies that meet ecological and economic goals based on real-time assessment results, making it difficult to effectively address the complex changes in marine ranching and the long-term sustainable development needs. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a method for marine ranch resource assessment and management to solve the problems of existing marine ranch resource assessment and management methods, such as incomplete and inaccurate data collection, inaccurate resource assessment, and inability to effectively generate dynamic management strategies.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides an intelligent method for marine ranching resource assessment and management, characterized by comprising the following steps:

[0008] Data collection: Integrating multiple data sources such as satellite remote sensing, underwater acoustic detection, and buoy sensors to obtain real-time hydrological environment data and biological resource distribution data of the marine ranch, and transmitting this data to the data processing center;

[0009] Data processing and evaluation: The data processing center uses machine learning-based assessment models to analyze the interactions between biological populations and the responses of organisms to the environment, providing accurate assessments of marine ranch resources, including biomass estimation and biodiversity index calculations.

[0010] Management strategy generation: The management decision-making module uses intelligent optimization algorithms to automatically generate dynamic management strategies based on real-time assessment results, combined with ecological and economic goals, such as adjustments to fishing quotas and arrangements for ecological restoration measures.

[0011] Visual display and interaction: Through the visual interface of the management terminal, marine ranch resource data and management strategies are displayed to managers in the form of three-dimensional maps, dynamic charts, etc., and managers can make interactive simulation adjustments in the interface to predict the effectiveness of the strategy.

[0012] As a preferred embodiment of the method for marine ranch resource assessment and management described in the present invention, in which: in the data collection step, satellite remote sensing data includes information such as ocean surface temperature, chlorophyll concentration, current direction and speed; underwater acoustic detection data includes the distribution location, quantity and size information of underwater organisms such as fish and shellfish; buoy sensor data includes parameter information such as temperature, salinity, pH, dissolved oxygen, etc. of the ocean water body.

[0013] As a preferred solution of the method for marine ranch resource assessment and management described in the present invention, the machine learning-based assessment model is a deep learning neural network model trained with a large amount of historical marine ranch data, whose input is the collected multi-source data, and the output is the assessment indicators of marine ranch resources, including biological population prediction, biodiversity health status assessment, etc.

[0014] As a preferred solution of the method for marine ranch resource assessment and management described in the present invention, the intelligent optimization algorithm is specifically one of a simulated annealing algorithm, a genetic algorithm or an ant colony algorithm, which is used to search for the optimal management strategy parameter combination based on the assessment results while satisfying ecological constraints and economic goals.

[0015] As a preferred solution of the method for marine ranch resource assessment and management described in the present invention, the visual interface has a data screening function, and managers can screen and view specific marine ranch resource data and corresponding management strategy recommendations based on conditions such as time, region, and biological species.

[0016] As a preferred solution of the method for marine ranch resource assessment and management described in the present invention, it also includes a data storage step, storing the collected raw data, processed assessment results and generated management strategies in a distributed database for subsequent data query, analysis and model optimization.

[0017] In a second aspect, the present invention provides a system for marine ranch resource assessment and management, comprising:

[0018] A data acquisition unit, which integrates satellite remote sensing equipment, underwater acoustic detection equipment, and buoy sensors to collect real-time hydrological and environmental data and biological resource distribution data of the marine ranch;

[0019] The data processing unit is connected to the data acquisition unit and has a built-in machine learning-based assessment model to analyze the interaction between biological populations and the laws of biological-environmental responses, thereby achieving an accurate assessment of marine ranch resources.

[0020] The management decision-making unit is connected to the data processing unit and uses intelligent optimization algorithms to generate dynamic management strategies based on the evaluation results;

[0021] The visualization interaction unit is connected to the management decision-making unit and is used to visualize marine ranch resource data and management strategies in the form of three-dimensional maps and dynamic charts, and supports managers to make interactive simulation adjustments to predict resource changes under different management strategies.

[0022] Thirdly, the satellite remote sensing equipment of the data acquisition unit has high-resolution imaging capabilities and multi-spectral monitoring functions, which can accurately obtain physical and bio-optical properties data of the ocean surface; the underwater acoustic detection equipment uses multi-beam sonar technology, which can achieve high-precision detection and classification of underwater organisms; the buoy sensor has adaptive data acquisition functions, which can automatically adjust the data acquisition frequency and parameter range according to changes in the ocean environment.

[0023] Fourthly, the data processing unit includes a data cleaning submodule, a feature extraction submodule and a model evaluation submodule; the data cleaning submodule is used to remove noise and outliers in the collected data; the feature extraction submodule extracts effective features related to marine ranch resource assessment from the cleaned data; the model evaluation submodule is used to monitor and evaluate the accuracy and reliability of the machine learning-based assessment model in real time, and trigger the model retraining process when the model performance deteriorates.

[0024] In the fifth aspect, the present invention provides that the visual interactive unit is equipped with a mobile terminal application, and managers can access the system anytime and anywhere through mobile devices such as mobile phones and tablets to view the marine ranch resource status and management strategies and perform remote interactive operations.

[0025] The beneficial effects of the present invention are as follows: by integrating multi-source data such as satellite remote sensing, underwater acoustic detection and buoy sensors, comprehensive and real-time monitoring of the hydrological environment and biological resource distribution of marine ranches is achieved, and data acquisition is richer and more accurate. The use of an assessment model based on machine learning can deeply analyze the complex relationships between biological populations and between organisms and the environment, making resource assessment more accurate and providing a reliable basis for scientific management. According to the assessment results, combined with ecological and economic goals, an intelligent optimization algorithm is used to generate dynamic management strategies, such as reasonably adjusting fishery catch limits and arranging ecological restoration measures to achieve the sustainable development of marine ranches. The visual interactive interface makes it convenient for managers to intuitively understand resource conditions and simulate the effects of management strategies, thereby improving decision-making efficiency. The data is stored in a distributed database, which facilitates subsequent query analysis and model optimization. Overall, the present invention improves the intelligence and dynamism of marine ranch resource assessment and management, enhances resource protection and utilization effects, and promotes the sustainable development of marine ranches. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0027] Figure 1 This is a flow chart of the method for marine ranch resource assessment and management in Example 1. DETAILED DESCRIPTION

[0028] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0029] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0030] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0031] Example 1, with reference to Figure 1, which is the first embodiment of the present invention, provides an intelligent method for marine ranch resource assessment and management, characterized by comprising the following steps:

[0032] Data collection: Integrating multiple data sources such as satellite remote sensing, underwater acoustic detection, and buoy sensors to obtain real-time hydrological environment data and biological resource distribution data of the marine ranch, and transmitting this data to the data processing center;

[0033] Data processing and evaluation: The data processing center uses machine learning-based assessment models to analyze the interactions between biological populations and the responses of organisms to the environment, providing accurate assessments of marine ranch resources, including biomass estimation and biodiversity index calculations.

[0034] Management strategy generation: The management decision-making module uses intelligent optimization algorithms to automatically generate dynamic management strategies based on real-time assessment results, combined with ecological and economic goals, such as adjustments to fishing quotas and arrangements for ecological restoration measures.

[0035] Visual display and interaction: Through the visual interface of the management terminal, marine ranch resource data and management strategies are displayed to managers in the form of three-dimensional maps, dynamic charts, etc., and managers can make interactive simulation adjustments in the interface to predict the effectiveness of the strategy.

[0036] It should be noted that, first, during the data collection phase, information is collected by integrating multiple data sources, including satellite remote sensing, underwater acoustic detection, and buoy sensors. Satellite remote sensing can obtain macro-environmental data such as sea surface temperature and chlorophyll concentration; underwater acoustic detection accurately detects the distribution, abundance, and size of underwater organisms; and buoy sensors provide real-time monitoring of key parameters such as ocean temperature, salinity, pH, and dissolved oxygen. This data is transmitted to the data processing center in real time, providing fundamental information for subsequent assessment and decision-making. During the data processing and evaluation phase, the center leverages machine learning-based assessment models to conduct in-depth analysis of the interactions between biological populations and the responses of organisms to the environment, thereby accurately assessing marine ranch resources. This includes key indicators such as biomass estimation and biodiversity index calculation. Next, the management strategy generation step begins. Based on the real-time assessment results and incorporating ecological and economic objectives, the management decision-making module applies intelligent optimization algorithms to automatically generate dynamic management strategies, such as adjustments to fishing quotas and the implementation of ecological restoration measures, to achieve scientific management of marine ranches. The last step is visualization and interaction. Through the visualization interface of the management terminal, the resource data and management strategies of the marine ranch are presented to managers using intuitive forms such as three-dimensional maps and dynamic charts. Managers can also make interactive simulation adjustments in the interface to predict the implementation effects of different strategies.

[0037] The data collection step integrates multiple advanced technologies to achieve comprehensive, multi-dimensional, and real-time data collection for marine ranches. This eliminates the incomplete information often associated with traditional single-source monitoring methods and lays a solid foundation for subsequent accurate assessments. The data processing and assessment step leverages cutting-edge machine learning techniques to delve deeper into the complex relationships between organisms and the environment, significantly improving the accuracy and reliability of resource assessments. The assessment results accurately reflect the actual resource status of marine ranches, providing a strong basis for scientific management. The management strategy generation step utilizes intelligent optimization algorithms based on the assessment results and multi-objective considerations to generate dynamic management strategies. Compared to traditional fixed management plans, these dynamic strategies are more adaptable to the complex dynamics of marine ranches, balancing ecological and economic interests for sustainable development. The visualization and interactive step provides managers with an intuitive and convenient platform, enabling them to quickly understand resource data and management strategies. They can also predict the effectiveness of strategy adjustments through simulations, thereby enhancing efficient and reliable decision-making. Overall, this advances the intelligentization of marine ranch management, improving management effectiveness and resource conservation and utilization.

[0038] Specifically, in the data collection step, satellite remote sensing data includes information such as ocean surface temperature, chlorophyll concentration, current direction and speed; underwater acoustic detection data includes the distribution location, quantity and size information of underwater organisms such as fish and shellfish; buoy sensor data includes parameter information such as temperature, salinity, pH, dissolved oxygen, etc. of ocean water bodies.

[0039] It should be noted that during the data collection process, satellite remote sensing data collected covers a wide range of information, including sea surface temperature, chlorophyll concentration, and current direction and speed. Specifically, satellite remote sensing uses specialized sensors to periodically scan the marine ranch area, capturing thermal infrared radiation from the ocean surface to measure temperature. Spectral analysis is also used to determine chlorophyll concentration, reflecting the distribution and growth of phytoplankton. Current direction and speed are determined using technologies such as synthetic aperture radar, analyzing characteristics such as the Doppler shift of surface echo signals. Underwater acoustic detection primarily utilizes sonar technology, transmitting acoustic signals underwater and receiving reflected signals to locate underwater organisms such as fish and shellfish. Signal processing analyzes echo intensity, frequency, and other characteristics to estimate the number and size of organisms. Buoy sensors are stationed at various locations within the marine ranch, providing real-time monitoring of water parameters such as temperature, salinity, pH, and dissolved oxygen. These sensors' built-in data acquisition modules sample data at preset intervals and transmit the data to a data processing center via wireless communication.

[0040] By specifying the specific content of satellite remote sensing data, we can more accurately grasp the macro-environmental characteristics of marine ranches. For example, sea surface temperature data helps understand the impact of phenomena such as hydrothermal circulation on biodistribution. Chlorophyll concentration is a key indicator for assessing primary productivity. Current direction and velocity data are important for predicting biodiversity migration and material transport. The refinement of underwater acoustic detection data allows for more accurate and comprehensive detection of underwater biological resources. It not only locates organisms but also quantifies their number and size, providing a detailed basis for their development, utilization, and conservation. The specific parameters of buoy sensor data directly reflect the physical and chemical properties of marine water, which are closely related to the survival and growth of organisms. The integration of these specific data collection elements further improves data accuracy and completeness, providing higher-quality input for subsequent resource assessment models, enhancing the credibility of assessment results, and ultimately making management strategies more relevant and effective, thereby strengthening the scientific nature and effectiveness of marine ranch resource management.

[0041] Specifically, the machine learning-based evaluation model is a deep learning neural network model trained with a large amount of historical marine ranch data. Its input is collected multi-source data, and its output is evaluation indicators of marine ranch resources, including biological population predictions, biodiversity health status assessments, etc.

[0042] It should be noted that the machine learning-based assessment model is a deep learning neural network model trained using a large amount of historical marine ranching data. During the model training phase, historical data covering multiple years of marine ranching, including hydrological and environmental data, biological resource monitoring data, and corresponding management activity records, is collected. This data is preprocessed, including removing missing values, outliers, and normalizing the data. The processed data is then fed into the neural network model. The network structure, such as the number of hidden layers and neurons per layer, is determined, and an appropriate activation function, such as the ReLU function, is selected to introduce nonlinearity. During training, the model's weight parameters are continuously adjusted using a backpropagation algorithm to minimize the error between predicted and actual values. Through multiple iterations of training, the model learns the complex mapping relationships between biodiversity succession, biodiversity changes, and multiple data sources within the marine ranching system. Ultimately, an assessment model is formed that accurately outputs key assessment indicators, including biodiversity population forecasts and biodiversity health assessments.

[0043] Compared to traditional statistical or simple machine learning models, deep learning neural network models trained on extensive historical data can more accurately capture the complex, nonlinear relationships between organisms and their environment. This model's learning from historical data gives it strong generalization capabilities, enabling more accurate predictions of population trends and biodiversity health, thus providing a highly precise quantitative tool for the dynamic assessment of marine ranch resources. This precise assessment helps to promptly identify anomalies in resource changes, providing a scientific and accurate basis for subsequent management decisions, avoiding management errors caused by assessment biases, and effectively ensuring the health of the marine ranch ecosystem and the sustainable use of fishery resources.

[0044] Specifically, the intelligent optimization algorithm is one of a simulated annealing algorithm, a genetic algorithm or an ant colony algorithm, which is used to search for the optimal combination of management strategy parameters under the conditions of meeting ecological constraints and economic goals based on the evaluation results.

[0045] It should be noted that in the management strategy generation step, intelligent optimization algorithms are the core technical means for dynamic management strategy formulation. When employing the simulated annealing algorithm, appropriate constraints and objective functions are first set based on the ecological and economic goals of the marine ranch. For example, the objective function is to maximize the economic benefits of the fishery while maintaining biodiversity at or above a certain threshold. A set of possible management strategy parameter combinations, such as fishing quotas and the area of ​​ecological restoration areas, is then initialized. Following the simulated annealing algorithm's iterative process, the objective function values ​​corresponding to the current parameter combination are calculated and evaluated. New parameter combinations are then accepted with a certain probability, while the temperature parameter is gradually lowered, controlling the algorithm's search range, until convergence conditions are met. Ultimately, the optimal management strategy parameter combination is obtained. Genetic algorithms and ant colony algorithms follow similar steps, respectively, mimicking the selection, crossover, and mutation operations of biological evolution or the pheromone renewal mechanism of ant foraging behavior. While satisfying ecological and economic constraints, they search for the management strategy parameter combination that optimizes the objective function.

[0046] Clarifying specific types of intelligent optimization algorithms, such as simulated annealing, genetic algorithms, or ant colony algorithms, ensures that the management strategy generation process has clear technical routes and operational methods. Each of these algorithms has unique advantages. For example, simulated annealing effectively avoids local optimal solutions, genetic algorithms excel at handling complex discrete optimization problems, and ant colony algorithms excel at path optimization. By applying these advanced optimization algorithms, it is possible to efficiently search for optimal solutions that meet ecological and economic goals among numerous possible management strategy parameter combinations. This improves the scientific and rational nature of management strategies, ensuring that marine ranching management can maximize economic benefits while maintaining ecological sustainability. It also enhances the convergence and stability of the algorithms, ensuring the efficiency and reliability of management strategy generation.

[0047] Specifically, the visualization interface has a data filtering function, and managers can filter and view specific marine ranch resource data and corresponding management strategy recommendations based on time, region, biological species and other conditions.

[0048] It should be noted that the visual interface is designed with a data filtering function module, which allows managers to set different filtering conditions through the operation interface. For example, managers can select a specific time range in the filter bar of the interface, such as a certain year or a certain season, and specify the area of ​​interest, such as a specific bay or a certain depth range of the marine ranch. They can also select specific biological species, such as a certain economic fish or rare shellfish. After setting these conditions, the system background will quickly query and filter the stored marine ranch resource data and management strategy recommendations according to the filtering conditions, filter out the data that meets the conditions, and display it in the form of a clear list or chart on the visual interface, which is convenient for managers to view and analyze. At the same time, the system also supports managers to further refine the filtering results, such as sorting by data size or performing comparative analysis on data from different regions.

[0049] The data filtering function of the visual interface gives managers greater autonomy and flexibility, allowing them to quickly locate the resource data and management strategies of interest based on actual work needs. When faced with massive amounts of marine ranch data, this function avoids the tedious process of managers browsing all the data one by one, greatly saving time and energy. By accurately filtering data under specific conditions, managers can conduct targeted analysis more efficiently, such as comparing changes in biological resources in the same area in different seasons, or evaluating the impact of a specific management strategy on a specific biological species, thereby providing strong support for the formulation of more targeted and timely management measures, enhancing the scientific nature and adaptability of management decisions, and improving the refinement and work efficiency of marine ranch management.

[0050] Specifically, it also includes a data storage step, storing the collected raw data, processed evaluation results and generated management strategies in a distributed database for subsequent data query, analysis and model optimization.

[0051] It should be noted that during the data storage step, the collected raw data, the evaluation results processed by the data processing unit, and the management policies generated by the management decision module are stored in a distributed database. This distributed database consists of multiple storage nodes interconnected by a high-speed network to enable distributed data storage and parallel processing. When writing data, the system distributes different types of data to different storage nodes based on a preset data distribution strategy. For example, satellite remote sensing data is stored on nodes with large-capacity storage devices, while underwater acoustic detection data is stored on nodes with high-speed read and write performance. Furthermore, each storage node regularly backs up data and employs redundant storage mechanisms to ensure data security and reliability. During data query and analysis, the system coordinates parallel data retrieval across storage nodes based on user requests through a distributed computing framework, rapidly returning query results to support subsequent data query, analysis, and model optimization operations.

[0052] By using a distributed database to store data, we can efficiently cope with the large amount of data, diverse data types, and rapid data growth characteristics of marine ranches. The distributed storage architecture not only increases data storage capacity, but also accelerates data reading and writing speeds through parallel processing capabilities, ensuring the system's performance and responsiveness when processing massive amounts of data. The redundant data backup mechanism effectively guarantees data security and prevents data loss due to single points of failure. At the same time, it facilitates subsequent data query and analysis functions, allowing researchers to fully utilize accumulated data for in-depth data mining and model optimization. For example, by analyzing historical data to improve the accuracy of evaluation models, or to evaluate the long-term effects of different management strategies, it provides solid data support for the long-term planning and management of marine ranches, promoting the continuous optimization of management decisions and the refined management of marine ranch resources.

[0053] Specifically, an intelligent system for marine ranch resource assessment and management is characterized by including:

[0054] A data acquisition unit, which integrates satellite remote sensing equipment, underwater acoustic detection equipment, and buoy sensors to collect real-time hydrological and environmental data and biological resource distribution data of the marine ranch;

[0055] The data processing unit is connected to the data acquisition unit and has a built-in machine learning-based assessment model to analyze the interaction between biological populations and the laws of biological-environmental responses, thereby achieving an accurate assessment of marine ranch resources.

[0056] The management decision-making unit is connected to the data processing unit and uses intelligent optimization algorithms to generate dynamic management strategies based on the evaluation results;

[0057] The visualization interaction unit is connected to the management decision-making unit and is used to visualize marine ranch resource data and management strategies in the form of three-dimensional maps, dynamic charts, etc., and supports managers to make interactive simulation adjustments to predict resource changes under different management strategies.

[0058] It should be noted that this patent also proposes an intelligent system for marine ranch resource assessment and management, corresponding to the aforementioned intelligent method. This system consists of four main units, interconnected via high-speed communication interfaces. The data acquisition unit is equipped with satellite remote sensing equipment, underwater acoustic detection equipment, and buoy sensors. These devices collect real-time marine ranch hydrological and biological resource distribution data according to a preset sampling frequency and strategy, and transmit this data to the data processing unit via wireless or wired communication links. The data processing unit incorporates a machine learning-based assessment model. Upon receiving data from the data acquisition unit, it first preprocesses the data, such as removing noise and filling in missing values. The processed data is then input into the assessment model, which then runs the model and outputs assessment results for marine ranch resources, including key indicators such as biomass estimates and biodiversity indices. The management decision-making unit receives the assessment results from the data processing unit in real time and invokes an intelligent optimization algorithm module. Based on preset ecological and economic objective functions, it calculates and generates dynamic management strategies, such as recommendations for adjusting fishing quotas and planning specific implementation plans for ecological restoration measures. These strategies are then sent to the visualization and interaction unit. After receiving the management strategy, the visualization interaction unit converts it into visualization forms such as three-dimensional maps and dynamic charts through its graphics processing module, and displays them on the interface of the management terminal. It also receives simulation adjustment instructions input by managers through interactive devices, updates the visualization display content in real time according to the instructions, and predicts resource changes under different management strategies.

[0059] The construction of a complete intelligent system organically integrates the various functional modules for marine ranch resource assessment and management, achieving full automation and intelligentization of the entire process, from data collection to processing and evaluation, and from decision-making to visualization. The coordinated operation of various system units ensures the real-time and accuracy of data, the timely and reliable evaluation results, and the scientific and dynamic nature of management strategies. Compared with decentralized, manual operation methods, this system greatly improves the efficiency and scientific nature of marine ranch management, reduces labor costs and management risks, and provides comprehensive and efficient technical support and management tools for the long-term sustainable development of marine ranches, promoting the modernization and intelligentization of marine ranch management.

[0060] Specifically, the satellite remote sensing equipment of the data acquisition unit has high-resolution imaging capabilities and multi-spectral monitoring functions, which can accurately obtain physical and bio-optical property data of the ocean surface; the underwater acoustic detection equipment uses multi-beam sonar technology, which can achieve high-precision detection and classification of underwater organisms; the buoy sensor has adaptive data acquisition functions, which can automatically adjust the data acquisition frequency and parameter range according to changes in the ocean environment.

[0061] It should be noted that the data acquisition unit of the satellite remote sensing equipment is equipped with high-resolution imaging sensors and multispectral monitors. Using advanced optical systems and highly sensitive detectors, high-resolution imaging sensors can capture images of the ocean surface with sub-meter spatial resolution, clearly revealing subtle features on the surface, such as small algal blooms or the distribution of aquaculture facilities. Multispectral monitors, with detection capabilities covering multiple wavelengths from visible light to infrared, measure the reflectivity of electromagnetic radiation at different wavelengths to accurately capture data on the physical and bio-optical properties of the ocean surface, enabling applications such as distinguishing different types of phytoplankton or monitoring the distribution of oil pollution on the surface. Underwater acoustic detection equipment utilizes multi-beam sonar technology. This technology transmits a fan-shaped acoustic beam and receives return signals from different directions. Using signal processing algorithms, it calculates the precise position, shape, and distance of underwater objects, enabling high-precision detection and classification of underwater life, such as distinguishing different species of fish or identifying obstacles such as submarine reefs. The buoy sensor has adaptive data collection capabilities and an environmental perception module installed inside it, which can monitor changes in marine environmental parameters in real time. When it detects that the parameter changes exceed the preset threshold, such as a sharp rise in water temperature in a short period of time or a sudden drop in dissolved oxygen concentration, the sensor will automatically adjust the data collection frequency, switching from normal mode to encrypted sampling mode to record the environmental changes in more detail and expand the parameter monitoring range to ensure that data changes under extreme environmental conditions can be captured.

[0062] The high-resolution imaging and multispectral monitoring capabilities of satellite remote sensing equipment provide richer and more accurate ocean surface data in terms of spatial detail and physical properties. This provides highly accurate environmental information for the meticulous management of marine ranches, such as pinpointing algal blooms for timely remediation measures or clearly understanding the layout of aquaculture facilities to optimize aquaculture planning. The multibeam sonar technology of underwater acoustic detection equipment significantly improves the accuracy and classification of underwater organism detection, enabling accurate identification and location of different underwater organisms. This provides a deeper understanding of biological community structure and reliable data support for the rational development, utilization, and protection of biological resources. The adaptive data collection capabilities of buoy sensors enhance sensitivity to changes in the marine environment. When unusual environmental changes occur, sampling strategies can be adjusted promptly to ensure critical data is not missed. This improves the integrity and representativeness of data collection, enabling the system to more accurately capture dynamic changes in the marine ranch environment, providing a powerful guarantee for timely response to environmental emergencies and optimizing management strategies.

[0063] Specifically, the data processing unit includes a data cleaning submodule, a feature extraction submodule and a model evaluation submodule; the data cleaning submodule is used to remove noise and outliers in the collected data; the feature extraction submodule extracts effective features related to marine ranch resource assessment from the cleaned data; the model evaluation submodule is used to monitor and evaluate the accuracy and reliability of the machine learning-based assessment model in real time, and trigger the model retraining process when the model performance deteriorates.

[0064] It should be noted that the data processing unit incorporates a data cleaning submodule, a feature extraction submodule, and a model evaluation submodule to ensure data processing accuracy and model reliability. After receiving raw data from the data acquisition unit, the data cleaning submodule first uses statistical analysis methods to identify and remove noise. For example, it calculates the mean and standard deviation of the data to eliminate outliers that fall outside a reasonable range. Furthermore, for data records with missing values, a machine learning-based interpolation algorithm is used to supplement them to ensure data integrity. Based on the cleaned data, the feature extraction submodule applies mathematical methods such as principal component analysis to extract effective features relevant to marine ranch resource assessment. For example, it extracts key water temperature gradient features that influence biological distribution from hydrological data and core species diversity indicators from biological resource data. This converts high-dimensional data into low-dimensional feature vectors to improve data processing efficiency. After each model evaluation run, the model evaluation submodule quantitatively assesses the accuracy and reliability of the model by calculating error metrics such as mean squared error and mean absolute error between the predicted and actual values. When the evaluation indicators show that the model performance drops below the preset threshold, the model retraining process will be automatically triggered, the latest data samples will be retrieved from the database, and the model parameters will be readjusted to restore the model's evaluation accuracy.

[0065] By setting up a data cleaning submodule, noise and outliers in the data are effectively removed, missing data is supplemented, and the quality of data input into the evaluation model is ensured. This prevents low-quality data from interfering with the evaluation results, thereby improving the accuracy of the evaluation. The feature extraction submodule can extract key features from massive amounts of data, reducing the data dimension. This not only speeds up data processing but also highlights information closely related to resource evaluation, enabling the evaluation model to more efficiently focus on important features for calculations, further improving evaluation efficiency and the reliability of results. The real-time monitoring function of the model evaluation submodule ensures the continued effectiveness of the evaluation model. If the model's performance degrades, timely retraining and adjustment ensure that the system can always output accurate evaluation results, providing a stable and reliable basis for management decisions and enhancing the stability and credibility of the entire intelligent system.

[0066] Specifically, the visual interaction unit is equipped with a mobile terminal application, and managers can access the system anytime and anywhere through mobile devices such as mobile phones and tablets to view the marine ranch resource status and management strategies and perform remote interactive operations.

[0067] It should be noted that the visualization interaction unit is equipped with a mobile terminal application. This application establishes a connection with the communication interface of the system backend, allowing managers to access the system anytime and anywhere using mobile devices such as mobile phones and tablets. The application's interface design uses responsive layout technology, which can automatically adjust the display mode of interface elements according to the screen size of the mobile device to ensure a good user experience. When using it, managers only need to log in to the system through authentication to view the resource data and management strategies of the marine ranch on their mobile devices, such as viewing a heat map of fish distribution in a certain area or a chart of ecological restoration progress. At the same time, the application supports touch interaction operations. Managers can adjust the visualization content and view detailed information through simple gestures such as sliding, clicking, and zooming. They can also enter simulation adjustment instructions. The system will provide real-time feedback on the predicted results based on the instructions. Managers can remotely monitor and participate in the management decision-making process through mobile devices.

[0068] The mobile terminal application breaks down geographical restrictions, enabling managers to access the system anytime, anywhere via mobile devices, keeping abreast of the latest marine ranch resource status and management strategy implementation, significantly improving the flexibility and timeliness of management work. The responsive interface design and convenient touch-screen interaction simplify operational complexity for managers, improving system usability and enabling them to quickly get started and efficiently utilize system functions. Interactive simulation adjustments and predictive analysis on mobile devices enable managers to more easily optimize decisions and respond to emergencies promptly, enhancing the timeliness and scientific nature of management decisions and further improving the intelligence and efficiency of marine ranch management.

[0069] In summary, the present invention achieves comprehensive, real-time monitoring of the hydrological environment and biological resource distribution of marine ranches by integrating multi-source data such as satellite remote sensing, underwater acoustic detection, and buoy sensors, resulting in richer and more accurate data acquisition. Utilizing an assessment model based on machine learning, it is possible to deeply analyze the complex relationships between biological populations and between organisms and the environment, making resource assessments more accurate and providing a reliable basis for scientific management. Based on the assessment results and combined with ecological and economic goals, an intelligent optimization algorithm is used to generate dynamic management strategies, such as rationally adjusting fishery catch limits and arranging ecological restoration measures, to achieve the sustainable development of marine ranches. The visual interactive interface allows managers to intuitively understand resource conditions and simulate the effects of management strategies, thereby improving decision-making efficiency. Data is stored in a distributed database, facilitating subsequent query analysis and model optimization. Overall, the present invention improves the intelligence and dynamism of marine ranch resource assessment and management, enhances resource protection and utilization, and promotes the sustainable development of marine ranches.

[0070] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. An intelligent method for marine ranching resource assessment and management, characterized by: The following steps are involved: Data collection: Integrating multiple data sources such as satellite remote sensing, underwater acoustic detection, and buoy sensors to obtain real-time hydrological environment data and biological resource distribution data of the marine ranch, and transmitting this data to the data processing center; Data processing and evaluation: The data processing center uses machine learning-based assessment models to analyze the interactions between biological populations and the responses of organisms to the environment, providing accurate assessments of marine ranch resources, including biomass estimation and biodiversity index calculations. Management strategy generation: The management decision-making module uses intelligent optimization algorithms to automatically generate dynamic management strategies based on real-time assessment results, combined with ecological and economic goals, such as adjustments to fishing quotas and arrangements for ecological restoration measures. Visual display and interaction: Through the visual interface of the management terminal, marine ranch resource data and management strategies are displayed to managers in the form of three-dimensional maps, dynamic charts, etc., and managers can make interactive simulation adjustments in the interface to predict the effectiveness of the strategy.

2. The intelligent method for marine ranching resource assessment and management according to claim 1, characterized in that: In the data collection step, satellite remote sensing data includes information such as ocean surface temperature, chlorophyll concentration, current direction and speed; underwater acoustic detection data includes the distribution location, quantity and size information of underwater organisms such as fish and shellfish; buoy sensor data includes parameter information such as temperature, salinity, pH, dissolved oxygen, etc. of ocean water bodies.

3. The intelligent method for marine ranching resource assessment and management according to claim 1, characterized in that: The machine learning-based evaluation model is a deep learning neural network model trained with a large amount of historical marine ranch data. Its input is collected multi-source data, and its output is evaluation indicators of marine ranch resources, including biological population predictions, biodiversity health status assessments, etc.

4. The intelligent method for marine ranching resource assessment and management according to claim 1, characterized in that: The intelligent optimization algorithm is specifically one of a simulated annealing algorithm, a genetic algorithm or an ant colony algorithm, which is used to search for the optimal management strategy parameter combination under the conditions of meeting ecological constraints and economic goals based on the evaluation results.

5. The intelligent method for marine ranching resource assessment and management according to claim 1, characterized in that: The visualization interface has a data filtering function, and managers can filter and view specific marine ranch resource data and corresponding management strategy recommendations based on time, region, biological species and other conditions.

6. The intelligent method for marine ranching resource assessment and management according to claim 1, characterized in that: It also includes a data storage step, which stores the collected raw data, processed evaluation results, and generated management strategies in a distributed database for subsequent data query, analysis, and model optimization.

7. An intelligent system for marine ranch resource assessment and management, characterized by: include: A data acquisition unit, which integrates satellite remote sensing equipment, underwater acoustic detection equipment, and buoy sensors to collect real-time hydrological and environmental data and biological resource distribution data of the marine ranch; The data processing unit is connected to the data acquisition unit and has a built-in machine learning-based assessment model to analyze the interaction between biological populations and the laws of biological-environmental responses, thereby achieving an accurate assessment of marine ranch resources. The management decision-making unit is connected to the data processing unit and uses intelligent optimization algorithms to generate dynamic management strategies based on the evaluation results; The visualization interaction unit is connected to the management decision-making unit and is used to visualize marine ranch resource data and management strategies in the form of three-dimensional maps, dynamic charts, etc., and supports managers to make interactive simulation adjustments to predict resource changes under different management strategies.

8. The intelligent system for marine ranching resource assessment and management according to claim 7, characterized in that: The satellite remote sensing equipment of the data acquisition unit has high-resolution imaging capabilities and multi-spectral monitoring functions, and can accurately obtain physical and bio-optical property data of the ocean surface; the underwater acoustic detection equipment uses multi-beam sonar technology, which can achieve high-precision detection and classification of underwater organisms; the buoy sensor has adaptive data acquisition functions, which can automatically adjust the data acquisition frequency and parameter range according to changes in the ocean environment.

9. The intelligent system for marine ranching resource assessment and management according to claim 7, characterized in that: The data processing unit includes a data cleaning submodule, a feature extraction submodule and a model evaluation submodule; the data cleaning submodule is used to remove noise and outliers in the collected data; The feature extraction submodule extracts effective features related to marine ranch resource assessment from the cleaned data; The model evaluation submodule is used to monitor and evaluate the accuracy and reliability of the machine learning-based evaluation model in real time, and trigger the model retraining process when the model performance degrades.

10. The intelligent system for marine ranching resource assessment and management according to claim 7, characterized in that: The visual interaction unit is equipped with a mobile terminal application, and managers can access the system anytime and anywhere through mobile devices such as mobile phones and tablets to view the marine ranch resource status and management strategies and perform remote interactive operations.

Images