Offshore wind power-marine ranch collaborative development method

Abstract

The invention discloses an offshore wind power-marine ranch collaborative development method, and belongs to the field of comprehensive development and utilization of marine resources. Comprising the steps that ecological adaptation parameters of a net cage and a wind power pile foundation are determined by collecting ecological data of different water depths; a genetic algorithm is used for simulating the influence of pile foundation layout on hydrodynamic force, and the net cage space distribution proportion is optimized; combining water depth and tidal current characteristics to generate pile foundation and net cage differentiation design parameters; a modular parallel construction process is adopted, and a construction time cooperation plan is made; a digital twinning virtual interaction platform is constructed, and operation parameters are optimized in real time; deploying a sensor network, and obtaining pasture management real-time data; biological attachment and water quality change are analyzed, a pile foundation ecological coating is designed, and ecological restoration parameters are generated; the surplus wind power is used for driving a pasture water pump and oxygenation equipment, and energy and ecology collaborative power distribution is achieved. Multi-dimensional cooperation of wind power and the pasture in space, time, energy and ecology is achieved, and the resource utilization rate is increased.

Description

Technical Field

[0001] This invention belongs to the field of comprehensive development and utilization of marine resources, and in particular relates to a method for the collaborative development of offshore wind power and marine ranching. Background Technology

[0002] The synergistic development of offshore wind power and marine ranching is an important research area for achieving efficient utilization of marine resources and ecological environmental protection. Its core lies in combining renewable energy development with marine ecological aquaculture, balancing the dual goals of energy production and ecological restoration. This field is significant because it not only promotes clean energy development but also enhances marine fisheries production through ecological synergy, thus fostering sustainable marine economic development. However, existing methods have significant limitations in practical applications, particularly regarding the spatial coordination and operational efficiency of wind power facilities and ranching ecosystems. Traditional solutions often overlook the complex impacts of wind power facilities on ocean currents and biological habitats, resulting in a lack of targeted layout design, significant ecological disturbance during construction, and difficulty in achieving a dynamic balance between energy and ecological benefits.

[0003] Therefore, the core challenge of existing technologies lies first in the spatial adaptability of wind power facilities and ranch layout. Because the arrangement of wind turbine foundations and turbines alters local water flow and sunlight conditions, the distribution patterns of fish, algae, and other organisms in the ranch are disrupted, thus affecting aquaculture efficiency. For example, in a certain sea area, the dense arrangement of wind turbines slows water flow, resulting in insufficient water exchange within the ranch cages and limiting biological growth. Secondly, the spatial adaptability issue further exacerbates the time-sensitive conflict between construction and ecological protection. The parallel operation of rapidly installing wind turbine foundations and deploying ranch cages must be completed within a limited time window to minimize disturbance to the marine ecosystem. However, existing technologies lack effective collaborative planning, leading to extended construction cycles and increased ecological disturbance.

[0004] Therefore, how to optimize the spatial layout design of wind power and ranching, balance the impact of ecological elements such as water flow and sunlight, and formulate an efficient construction coordination plan to reduce disturbance to the marine ecosystem has become a key issue in realizing the coordinated development of offshore wind power and marine ranching. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention provides a method for the collaborative development of offshore wind power and marine ranching, comprising:

[0006] Acquire data on the ecological characteristics of marine ranches and generate ecological adaptation parameters;

[0007] Based on the ecological adaptation parameters, the impact of wind turbine pile foundation layout on hydrodynamics is simulated, and the space allocation ratio of the cage is calculated.

[0008] Based on the space allocation ratio of the net cages and combined with the regional water depth and tidal dynamics, the coordinated layout scheme of wind turbine pile foundations and net cages is optimized to obtain differentiated design parameters.

[0009] Based on the aforementioned differentiated design parameters, a modular parallel construction process is planned to generate a construction time coordination plan;

[0010] Based on the aforementioned construction time coordination plan, a digital twin virtual interaction platform is constructed to generate an optimized operation parameter scheme;

[0011] Based on the optimized operating parameter scheme, deploy a sensor network to acquire real-time data streams for intelligent ranch management;

[0012] Based on the real-time data stream, analyze the trends of biological attachment and water quality changes, and generate ecological restoration parameters;

[0013] Based on the ecological restoration parameters, the power support of surplus wind farm electricity for ranch equipment is simulated to generate a power distribution scheme that coordinates energy and ecology.

[0014] Preferably, the process of acquiring marine ranch ecological characteristic data and generating ecological adaptation parameters includes:

[0015] Data on water temperature, salinity, dissolved oxygen, and light intensity at different water depths were obtained to acquire a set of environmental parameters;

[0016] Based on the set of environmental parameters, a clustering algorithm is used to perform hierarchical clustering according to water depth to obtain the ecological characteristics of each water depth range;

[0017] Based on the aforementioned ecological characteristics, calculate the fish distribution characteristics;

[0018] Based on the aforementioned fish distribution characteristics, the correlation between algae distribution and fish distribution is predicted, and the distribution pattern of biological density is obtained.

[0019] Calculate the cage layout parameters based on the described biological density distribution pattern;

[0020] If the grid cage layout parameters overlap with the wind turbine pile foundation location, the ecological adaptation parameters are obtained by adjusting them through a spatial optimization algorithm.

[0021] Preferably, the process of simulating the impact of wind turbine foundation layout on hydrodynamics and calculating the cage space allocation ratio based on the ecological adaptation parameters includes:

[0022] A hydrodynamic model is constructed based on the ecological adaptation parameters to obtain the initial flow field distribution;

[0023] Based on the initial flow field distribution, a variety of layout schemes are generated using a genetic algorithm, and the changes in water flow velocity and wave height for each scheme are obtained.

[0024] Based on the changes in water flow velocity and wave height, calculate the set of cage space allocation ratios corresponding to each scheme;

[0025] Layout schemes with a proportion exceeding a preset threshold are removed to obtain an optimized scheme set;

[0026] The ecological impact factors are calculated based on the set of optimization schemes, and the optimal layout scheme is obtained after sorting them.

[0027] Based on the optimal layout scheme, the final cage space allocation ratio is output.

[0028] Preferably, the process of optimizing the coordinated layout scheme of wind turbine foundations and net cages based on the space allocation ratio of the net cages, combined with the regional water depth and tidal dynamic characteristics, to obtain differentiated design parameters includes:

[0029] Acquire regional water depth and tidal dynamics data to form a set of regional environmental characteristics;

[0030] Based on the set of regional environmental characteristics, the stress distribution of the pile foundation is calculated using finite element analysis to obtain the pile foundation structural parameters;

[0031] If the parameters of the pile foundation structure meet the stability threshold, then a preliminary collaborative layout scheme is generated by using a particle swarm optimization algorithm, taking into account the space allocation ratio of the cages.

[0032] Calculate the hydrodynamic influence coefficient based on the preliminary coordinated layout scheme;

[0033] If the hydrodynamic influence coefficient is lower than the preset threshold, the size and position of the cage are adjusted to obtain differentiated design parameters;

[0034] A genetic algorithm is used to iteratively adjust the spatial positions and output the final collaborative layout parameters and 3D layout model.

[0035] Preferably, the process of planning a modular parallel construction process and generating a construction time coordination plan based on the differentiated design parameters includes:

[0036] Obtain the design parameters of the pile foundation and the cage design parameters, extract the construction requirements and resource needs, and obtain the parameter difference analysis results;

[0037] Based on the parameter difference analysis results, a modular construction process model is constructed, and independent construction units are divided.

[0038] K-means clustering was used to classify the construction units, resulting in parallel operation groups;

[0039] Detect job conflicts in the parallel job groups, adjust group priorities, and obtain an optimized parallel job mode;

[0040] Based on the optimized parallel operation mode, a draft of the construction time coordination plan is generated.

[0041] The initial draft of the construction time coordination plan is iteratively optimized using the simulated annealing algorithm, and the final construction time coordination plan is output.

[0042] Preferably, the process of constructing a digital twin virtual interaction platform and generating optimized operating parameter schemes based on the construction time coordination plan includes:

[0043] Obtain operational parameters from wind farms and ranches to create real-time operational datasets;

[0044] The real-time running dataset is cleaned and standardized to obtain a standardized running dataset;

[0045] Based on the standardized operational dataset, a digital twin virtual operational environment is constructed to obtain operational state simulation results;

[0046] Based on the simulation results of the operating status, the time conflict between equipment operation and pasture activities is analyzed, the construction time window is adjusted, and a collaborative scheduling mechanism is obtained.

[0047] The cooperative scheduling mechanism and the simulation results of the running status are loaded into the virtual interactive platform, and the optimal set of running parameters is calculated using the particle swarm optimization algorithm.

[0048] Based on the optimal set of operating parameters, key control parameters are extracted, and the final operating parameter scheme is output after iterative updates.

[0049] Preferably, the process of deploying a sensor network and acquiring real-time data streams for intelligent ranch management according to the optimized operating parameter scheme includes:

[0050] Acquire water quality monitoring data and biological growth data collected by sensor networks;

[0051] Time series analysis was used to determine the fluctuation trend of environmental parameters;

[0052] If the fluctuation trend exceeds the preset threshold, the abnormal state is classified using a support vector machine to obtain the abnormal environment parameters.

[0053] Based on the abnormal environmental parameters, a decision tree is used to predict the degree of impact on biological growth and obtain changes in growth status.

[0054] Based on the changes in growth status, the sensor sampling frequency is adjusted to acquire a high-frequency data stream;

[0055] If the high-frequency data stream continues to be abnormal, cluster analysis is used to divide the abnormal regions and obtain the environmental partitioning results.

[0056] Based on the environmental partitioning results, update the sensor network configuration and output a real-time optimized data stream.

[0057] Preferably, the process of analyzing the trends in biofouling and water quality changes based on the real-time data stream to generate ecological restoration parameters includes:

[0058] Real-time data streams are obtained through an intelligent ranch management platform to extract datasets of biological attachment distribution and water quality parameter changes.

[0059] Random forest algorithm was used to analyze time series characteristics and key influencing factors;

[0060] If the bio-attachment density exceeds the preset threshold, the parameters of the ecological coating on the surface of the wind turbine pile foundation are adjusted according to the key influencing factors to obtain the candidate coating formulation.

[0061] The candidate coating formulations are classified using a support vector machine to determine the preferred coating parameters;

[0062] Based on the preferred coating parameters, an ecological coating design scheme for the surface of wind turbine pile foundations is generated.

[0063] The coating effect is verified by monitoring environmental parameters, and iterative adjustments are made until the requirements are met, and the final ecological restoration parameters are output.

[0064] Preferably, the process of simulating the power support of surplus wind farm electricity for ranch equipment and generating a power allocation scheme that coordinates energy and ecology based on the ecological restoration parameters includes:

[0065] Real-time power output data of wind farms and power demand data of ranch water pump circulation and oxygenation equipment are obtained through digital twin models;

[0066] If the surplus power exceeds the water pump circulation demand, power will be allocated to the water pump circulation system first, and the remaining power share will be calculated.

[0067] Based on the remaining power share and the real-time demand of the oxygenation equipment, a linear programming algorithm is used to optimize the allocation ratio and obtain a power dispatching scheme for the oxygenation equipment.

[0068] The operational effect of the power dispatch scheme for the oxygenation equipment is simulated in a digital twin model to obtain ecosystem balance indicators;

[0069] If the ecosystem balance index is lower than the preset threshold, the linear programming parameters are adjusted and the allocation ratio is recalculated.

[0070] Time series analysis is used to predict future changes in electricity demand and generate dynamic power dispatch plans.

[0071] Verify the operational effectiveness of the dynamic power dispatch plan under different scenarios and output the final power allocation scheme.

[0072] Preferably, the differentiated design parameters include:

[0073] Structural dimensions, pile length, pile diameter, wall thickness, and corrosion protection level of wind turbine pile foundations;

[0074] The external dimensions, mesh size, floating material, and anchoring method of the cage;

[0075] Minimum safe distance, relative angle, and vertical spacing between wind turbine pile foundations and grid cages.

[0076] Compared with the prior art, the present invention has the following advantages and technical effects:

[0077] This invention discloses a solution to the operational challenges of marine ranching, where fish and algae distribution is influenced by water depth, tidal currents, and wind turbine foundation layout, as well as insufficient coordination between wind farm and ranch operation resources. It utilizes a genetic algorithm to simulate the dynamic impact of wind turbine foundation layout on water flow and waves, optimizing the allocation ratio of gabion space. Differentiated foundation and gabion layout parameters are designed based on water depth and tidal current characteristics. A modular construction process enables parallel operations, generating a coordinated construction time plan. A virtual interactive platform is constructed using a digital twin model to optimize operational parameters. A sensor network is deployed to acquire real-time water quality and biological growth data, analyzing trends in biological attachment and water quality changes. An ecological coating is designed to enhance biodiversity, and surplus power is simulated to support ranch equipment operation. Ultimately, an energy and ecological synergy power allocation scheme is generated. This invention, through digital twins and intelligent management, achieves deep integration of marine ranch ecological restoration and efficient wind power utilization, improving the comprehensive utilization efficiency of marine resources and the stability of the ecosystem. Attached Figure Description

[0078] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0079] Figure 1 This is a schematic diagram of the method flow according to an embodiment of the present invention. Detailed Implementation

[0080] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0081] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0082] like Figure 1As shown, this embodiment provides a method for the collaborative development of offshore wind power and marine ranching, including:

[0083] Acquire data on the ecological characteristics of marine ranches and generate ecological adaptation parameters;

[0084] Based on ecological adaptation parameters, the impact of wind turbine pile foundation layout on hydrodynamics is simulated, and the space allocation ratio of the cage is calculated.

[0085] Based on the space allocation ratio of the cages and combined with the regional water depth and tidal dynamics, the coordinated layout scheme of wind turbine pile foundations and cages is optimized to obtain differentiated design parameters.

[0086] Based on the differentiated design parameters, a modular parallel construction process is planned to generate a construction time coordination plan;

[0087] Based on the construction time coordination plan, a digital twin virtual interaction platform is built to generate optimized operation parameter schemes;

[0088] Based on the optimized operating parameter scheme, a sensor network is deployed to obtain real-time data streams for intelligent ranch management;

[0089] Based on real-time data streams, analyze the trends of biofouling and water quality changes to generate ecological restoration parameters;

[0090] Based on ecological restoration parameters, the power support of surplus wind farm electricity for ranch equipment is simulated to generate a power distribution scheme that coordinates energy and ecology.

[0091] Furthermore, the process of acquiring marine ranch ecological characteristic data and generating ecological adaptation parameters includes:

[0092] Data on water temperature, salinity, dissolved oxygen, and light intensity at different water depths were obtained to acquire a set of environmental parameters;

[0093] Based on the set of environmental parameters, a clustering algorithm was used to perform hierarchical clustering according to water depth to obtain the ecological characteristics of each water depth range;

[0094] Calculate the distribution characteristics of fish based on ecological features;

[0095] Based on the distribution characteristics of fish, the correlation between algae distribution and fish distribution is predicted, and the distribution pattern of biological density is obtained;

[0096] Calculate the cage layout parameters based on the distribution patterns of biological density;

[0097] If the layout parameters of the net cage overlap with the location of the wind turbine pile foundation, the ecological adaptation parameters are obtained by adjusting them through a spatial optimization algorithm.

[0098] Furthermore, this embodiment acquires ecological data at different water depths in the marine ranch. Water temperature, salinity, dissolved oxygen, and light intensity are collected using sensors to obtain a set of environmental parameters. For this set of environmental parameters, a K-means clustering algorithm is used to stratify the data by water depth, determining the ecological data characteristics within each depth range. If the fish distribution density exceeds a preset threshold in a certain depth range, a spatial analysis algorithm is used to calculate the fish distribution characteristics for that depth range. Based on the fish distribution characteristics, a random forest algorithm is used to predict the correlation between algae distribution and fish distribution, obtaining the biological density distribution pattern. Based on the biological density distribution pattern, candidate locations for the net cage layout are calculated, obtaining the net cage layout parameters. If the net cage layout parameters overlap with the wind turbine foundation locations, a spatial optimization algorithm is used to adjust the net cage layout parameters to determine the ecological adaptation parameters. Based on the ecological adaptation parameters, a joint layout scheme for the marine ranch net cages and wind turbine foundations is generated, obtaining the final deployment parameters.

[0099] Specifically, the acquisition and analysis of marine ranch ecological data is the core of achieving scientific management. For example, by deploying sensors at different water depths, environmental parameters such as water temperature, salinity, dissolved oxygen, and light intensity can be collected.

[0100] In one embodiment, it is assumed that a sensor node is deployed every 5 meters within a water depth range of 0-50 meters in a marine ranch. Data collection results might show that at a water depth of 10 meters, the average water temperature is 20.5℃, the salinity is 33‰, the dissolved oxygen is 6.5 mg / L, and the light intensity is 2000 lux. These data form an environmental parameter set, providing a basis for subsequent analysis. For this environmental parameter set, a K-means clustering algorithm is used to stratify the data by water depth to determine the ecological characteristics of each interval.

[0101] Specifically, the water depth of 0-50 meters is divided into 5 intervals, and the algorithm clusters the data based on the differences in parameters.

[0102] For example, the 0-10 meter depth range may form an active ecological zone due to strong light and high water temperature, while the 40-50 meter depth range exhibits low activity characteristics due to weak light and low dissolved oxygen. This stratification helps to accurately identify the ecological patterns at different water depths and improve management efficiency. In fish distribution density analysis, if the fish density in a certain water depth range, such as 10-15 meters, exceeds a preset threshold (e.g., 100 fish / cubic meter), the distribution characteristics are calculated using spatial analysis algorithms.

[0103] For example, this embodiment uses spatial autocorrelation analysis to determine whether fish are clustered. The results show that fish aggregation is high at water depths of 10-15 meters, and spatial hotspots are concentrated in a certain area. This analysis provides a basis for cage layout and avoids resource waste. Based on the fish distribution characteristics, a random forest algorithm is then used to predict the correlation between algae distribution and fish distribution.

[0104] For example, by inputting data on fish density, water temperature, and light intensity, the model predicts a positive correlation between algae density and fish density, with a correlation coefficient of 0.85. This indicates that fish aggregation areas are often accompanied by high algae density, providing a biological basis for cage site selection and optimizing aquaculture efficiency. Candidate cage layout locations are calculated based on the distribution patterns of biological density.

[0105] For example, in water depths of 10-15 meters, and considering areas with high fish and algae density, preliminary net cage layout parameters can be determined, such as one net cage per 100 square meters. However, if the net cage location overlaps with the wind turbine foundation, adjustments need to be made using spatial optimization algorithms.

[0106] For example, this embodiment uses a genetic algorithm, setting the objective function to maximize ecological adaptability and minimize spatial conflict, ultimately adjusting the net cage to an area 50 meters away from the pile foundation, and generating ecological adaptability parameters. Based on the ecological adaptability parameters, a joint layout scheme for marine ranch net cages and wind power pile foundations is generated.

[0107] For example, ten net cages can be deployed at a water depth of 10-15 meters, spaced 100 meters apart, avoiding the pile foundation area by 50 meters. The final deployment parameters ensure a balance between ecological benefits and engineering safety, enhancing the sustainable development capacity of the marine ranch. This combined layout scheme, through multi-dimensional data analysis and optimization, significantly improves resource utilization and economic benefits.

[0108] Furthermore, based on ecological adaptation parameters, the process of simulating the impact of wind turbine foundation layout on hydrodynamics and calculating the space allocation ratio of the net cages includes:

[0109] A hydrodynamic model is constructed based on ecological adaptation parameters to obtain the initial flow field distribution;

[0110] Based on the initial flow field distribution, a variety of layout schemes are generated using a genetic algorithm, and the changes in water flow velocity and wave height for each scheme are obtained.

[0111] Calculate the set of cage space allocation ratios corresponding to each scheme based on changes in water flow velocity and wave height;

[0112] Layout schemes with a proportion exceeding a preset threshold are removed to obtain an optimized scheme set;

[0113] The ecological impact factors are calculated based on the set of optimization schemes, and the optimal layout scheme is obtained after sorting them.

[0114] Based on the optimal layout scheme, output the final cage space allocation ratio.

[0115] Furthermore, this embodiment obtains the layout parameters of wind turbine pile foundations and gabion cages, extracts the initial boundary conditions for water flow and waves from the ecological adaptation parameters, constructs a hydrodynamic model, and obtains the initial flow field distribution. Based on the initial flow field distribution, a genetic algorithm is used to simulate the combination of wind turbine pile foundation locations and gabion cage distribution density, generating multiple layout schemes and obtaining the water flow velocity and wave height changes for each scheme. For the changes in water flow velocity and wave height, the gabion cage space allocation ratio in each layout scheme is calculated, obtaining a set of space allocation ratios. If any ratio in the set of space allocation ratios exceeds a preset threshold, the corresponding layout scheme is removed, resulting in an optimized set of layout schemes. From the optimized set of layout schemes, the hydrodynamic model is used to calculate the dynamic impact of water flow and waves for each scheme, obtaining the ecological impact factor for each scheme. Based on the ecological impact factor, a weighted evaluation method is used to rank the optimized set of layout schemes to obtain the optimal layout scheme. For the optimal layout scheme, the final gabion cage space allocation ratio is calculated to obtain the dynamic impact results of the wind turbine pile foundation layout on water flow and waves.

[0116] In one possible implementation, the parameters for wind turbine pile foundation layout and cage layout need to be obtained based on ecological adaptation parameters, with the initial boundary conditions of water flow and waves as the core.

[0117] For example, in this embodiment, sensors collect water flow velocity and wave height data at different water depths and in different areas within the marine ranch. Assuming a water flow velocity of 0.5 m / s and a wave height of 1.2 m in a certain area, initial boundary conditions are established. These conditions reflect the dynamic constraints of the marine environment on the layout scheme, helping to ensure the accuracy of subsequent models.

[0118] For example, when constructing a hydrodynamic model, initial flow field distribution is generated by using water flow velocity and wave height data, combined with environmental parameters such as water depth and tides. Assuming that the water flow velocity at a depth of 10 meters in a marine ranch is 0.8 m / s and at 20 meters is 0.4 m / s, the model can simulate the flow field distribution at different water depths. This distribution provides a foundation for subsequent optimization, ensuring that the layout scheme adapts to the marine environment.

[0119] In one embodiment, the genetic algorithm of this embodiment is used to simulate the layout combination of wind turbine pile foundations and gabions.

[0120] For example, setting up 10 wind turbine foundations and 50 net cages, the algorithm iterates 100 times to generate multiple layout schemes, each with different foundation spacing and net cage density. Suppose scheme A has a foundation spacing of 50 meters and a net cage density of 10 per hectare, while scheme B has a foundation spacing of 30 meters and a net cage density of 15 per hectare. The algorithm calculates the water flow velocity changes for each scheme, such as a 5% reduction in flow velocity in scheme A and a 10% reduction in scheme B, as well as wave height changes, and selects the scheme with the least impact on the flow field.

[0121] Specifically, the calculation of the cage space allocation ratio needs to take into account the dynamic effects of water flow and waves.

[0122] For example, if the water flow velocity in the area occupied by the net cages in a certain scheme drops to 0.3 m / s, which is lower than the preset threshold of 0.4 m / s, then that scheme is eliminated. In the optimized scheme set, assuming that the space allocation ratio of scheme C is 30% of the total area occupied by net cages and scheme D is 40%, both meet the threshold requirements. This screening ensures that the impact of the layout on the ecosystem is within a controllable range.

[0123] For example, this embodiment further analyzes the dynamic impact of the optimization scheme using a hydrodynamic model. It is assumed that scheme C has a lower ecological impact factor when the water flow velocity changes by 8% and the wave height changes by 5%, indicating less disturbance to fish and algae habitats. In contrast, scheme D may have a higher impact factor due to its higher cage density. Through weighted evaluation, considering water flow, wave, and ecological factors, scheme C is selected as the optimal layout scheme.

[0124] In one possible implementation, the cage space allocation ratio of the optimal layout scheme can be further refined.

[0125] For example, Option C ultimately determined that net cages would account for 28% of the total area, wind turbine foundations for 15%, and the remaining area as an ecological protection zone. This allocation balances aquaculture needs with environmental stability, reducing the risks of water flow obstruction and abnormal waves. The final results showed that the impact of the wind turbine foundation layout on water flow velocity was controlled within 10%, and wave height variation was less than 6%, providing a stable environmental foundation for the long-term operation of the marine ranch.

[0126] Furthermore, based on the space allocation ratio of the net cages and combined with the regional water depth and tidal dynamics, the process of optimizing the coordinated layout scheme of wind turbine foundations and net cages to obtain differentiated design parameters includes:

[0127] Acquire regional water depth and tidal dynamics data to form a set of regional environmental characteristics;

[0128] Based on the set of regional environmental characteristics, the stress distribution of the pile foundation is calculated using finite element analysis to obtain the pile foundation structural parameters;

[0129] If the pile foundation structural parameters meet the stability threshold, then a preliminary collaborative layout scheme is generated by using a particle swarm optimization algorithm, taking into account the cage space allocation ratio.

[0130] Calculate the hydrodynamic impact coefficient based on the preliminary coordinated layout plan;

[0131] If the hydrodynamic influence coefficient is lower than the preset threshold, the size and position of the cage are adjusted to obtain differentiated design parameters;

[0132] A genetic algorithm is used to iteratively adjust the spatial positions and output the final collaborative layout parameters and 3D layout model.

[0133] Furthermore, this embodiment acquires regional water depth data and tidal dynamic characteristic data. Using remote sensing technology and real-time monitoring equipment, high-resolution water depth distribution and tidal velocity vectors are extracted from the marine environmental database to obtain a set of regional environmental characteristics. Based on this set, the finite element method is used to calculate the stress distribution of the wind turbine pile foundation under different water depths and tidal conditions, determining the pile foundation structural parameters. If the pile foundation structural parameters meet a preset stability threshold, a preliminary collaborative layout scheme for the gabion and pile foundation is generated using a particle swarm optimization algorithm based on the gabion space allocation ratio and tidal dynamic characteristics, yielding initial layout parameters. For the initial layout parameters, the hydrodynamic impact and space utilization efficiency are analyzed. The flow field distribution of the gabion under tidal currents is calculated using numerical simulation methods, obtaining the hydrodynamic influence coefficient. If the hydrodynamic influence coefficient is lower than a preset threshold, the size and position of the gabion are adjusted according to structural stability requirements, generating differentiated design parameters to obtain an optimized layout scheme. Based on the optimized layout scheme and environmental adaptability requirements, a genetic algorithm is used to iteratively adjust the spatial positions of the gabion and pile foundation, obtaining the final collaborative layout parameters. By using the final collaborative layout parameters, differentiated design parameters for wind turbine pile foundations and net cages are generated, and a three-dimensional layout model is output.

[0134] For example, in this embodiment, when acquiring regional water depth data and tidal dynamics data, high-resolution water depth distribution is extracted from the marine environmental database by combining remote sensing technology with a multibeam echo sounder.

[0135] For example, in a near-shore wind farm area, the resolution can reach 0.5 meters, the measured water depth range is 10-30 meters, and the tidal velocity vector shows that the main tidal direction is northeast, with the velocity fluctuating between 0.2-1.5 m / s. This high-precision data provides a reliable foundation for subsequent analysis.

[0136] In one possible implementation, this embodiment uses the finite element analysis method to calculate the stress distribution of wind turbine pile foundations, and constructs a three-dimensional finite element model based on water depth and tidal flow data.

[0137] For example, with a pile diameter of 6 meters, a pile length of 40 meters, and a tidal current velocity of 1.0 m / s, the simulation results show that the pile base experiences stress concentration, with a maximum stress of approximately 200 MPa, meeting the stability threshold requirement of below 300 MPa. This method can accurately assess the structural safety of pile foundations in complex marine environments.

[0138] For example, when generating a coordinated layout scheme for net cages and pile foundations using the particle swarm optimization algorithm, the number of net cages is set to 20, with an initial space allocation ratio of 60% for net cage aquaculture and 40% for pile foundation support. After 100 algorithm iterations, the generated layout scheme shows that the net cages are evenly distributed in areas with low tidal current velocities, such as areas with a water depth of 20 meters and a tidal current velocity of 0.5 m / s, thereby reducing hydrodynamic impact. This layout can improve the stability of the net cages and optimize space utilization.

[0139] In one possible implementation, when analyzing the hydrodynamic effects, numerical simulation methods can be used to simulate the flow field around the cage using computational fluid dynamics software.

[0140] For example, with a cage size of 10m × 10m × 5m and a tidal current velocity of 0.8m / s, simulations show that the downstream eddy current intensity decreases by 10%, and the hydrodynamic influence coefficient is 0.6, which is below the threshold of 0.8. This analysis ensures that the cage layout has minimal impact on water flow and maintains ecological balance.

[0141] For example, when adjusting the size and location of the net cages, the size can be optimized to 8 meters × 8 meters × 4 meters based on the hydrodynamic influence coefficient and stability requirements. The net cages can then be moved to an area with a water depth of 15 meters, avoiding areas where the tidal current velocity exceeds 1.2 meters per second. This adjustment reduces the stress on the net cages and improves their long-term operational stability.

[0142] In one possible implementation, when iteratively adjusting the spatial positions of the gabion and pile foundation using a genetic algorithm, the objective can be set to minimize the hydrodynamic impact and maximize the space utilization efficiency.

[0143] For example, after 200 iterations, the cages are distributed in a ring around the pile foundation with a spacing of 5 meters, the hydrodynamic influence coefficient is reduced to 0.5, and the space utilization rate is increased to 70%. This scheme can balance ecological protection and economic benefits.

[0144] For example, when generating a 3D layout model, visualization software is used to convert the final collaborative layout parameters into a 3D model, which intuitively shows the spatial relationship between the cage and the pile foundation.

[0145] For example, the model shows that the cages are concentrated in shallower water areas, while the pile foundations are distributed in areas with gentler currents, resulting in a compact overall layout and a uniform flow field distribution. This type of model facilitates engineering implementation and subsequent optimization.

[0146] Furthermore, the process of planning a modular parallel construction process and generating a construction time coordination plan based on differentiated design parameters includes:

[0147] Obtain the design parameters of the pile foundation and the cage design parameters, extract the construction requirements and resource needs, and obtain the parameter difference analysis results;

[0148] Based on the results of parameter difference analysis, a modular construction process model is constructed, and independent construction units are divided.

[0149] K-means clustering was used to classify the construction units, resulting in parallel operation groups;

[0150] Detect job conflicts in parallel job groups, adjust group priorities, and obtain an optimized parallel job mode;

[0151] Based on the optimized parallel operation mode, a draft of the construction time coordination plan is generated.

[0152] The initial draft of the construction time coordination plan is iteratively optimized using the simulated annealing algorithm, and the final construction time coordination plan is output.

[0153] Furthermore, this embodiment obtains the design parameters of the pile foundation and the cage mesh, extracts their respective construction requirements and resource needs, and obtains the parameter difference analysis results. Based on the parameter difference analysis results, a modular construction process model is constructed to determine the independent construction units for pile foundation installation and cage mesh placement. The K-means clustering algorithm is used to classify the modular construction units, and parallel operation groups are obtained based on construction time and resource requirements. For the parallel operation groups, operation conflicts between pile foundation installation and cage mesh placement are detected. If there is time overlap or resource competition, the group priority is adjusted to obtain an optimized parallel operation mode. A time resource allocation scheme is extracted from the optimized parallel operation mode to generate a draft of the construction time coordination plan. The draft of the construction time coordination plan is iteratively optimized using a simulated annealing algorithm, adjusting the operation sequence and time allocation to obtain the final construction time coordination plan. If there are operation delays in the final construction time coordination plan, the parameter differences are re-analyzed, the modular construction process model is updated, and the above steps are repeated to obtain a construction time coordination plan that meets the requirements.

[0154] For example, in this embodiment, when obtaining the design parameters of the pile foundation and the cage design parameters, the diameter, length, material strength and other parameters of the pile foundation are extracted by combining on-site survey and historical data, as well as the size, mesh density and anchoring method of the cage.

[0155] Specifically, the pile foundation design parameters include reinforced concrete piles with a diameter of 2 meters and a length of 30 meters, requiring a compressive strength of 50 MPa; the gabion design parameters are a 10m × 10m × 5m cubic structure with a mesh diameter of 5 cm, and the anchor points must withstand a tensile force of 10 tons. Regarding construction requirements, the pile foundation requires a hydraulic pile driver, with a construction cycle of approximately 3 days per pile; the gabion requires underwater hoisting equipment, with a deployment cycle of approximately 1 day per gabion. In terms of resource requirements, pile foundation construction requires 50 tons of steel reinforcement and 200 cubic meters of concrete, while the gabion requires 500 square meters of high-strength polyethylene mesh. Parameter difference analysis reveals that pile foundation construction has a high dependence on heavy equipment, while gabion deployment requires high precision in underwater operations.

[0156] In one possible implementation, this embodiment constructs a modular construction process model, dividing pile foundation installation into three modules: foundation preparation, pile hoisting, and pile driving and fixing; and gabion placement into three modules: gabion assembly, underwater positioning, and anchoring connection. Each module defines an independent construction unit; for example, the pile foundation driving unit requires one pile driver and five workers, while the gabion anchoring unit requires two divers and one work vessel. This modular design facilitates standardized operation and improves construction efficiency.

[0157] For example, in this embodiment, when using the K-means clustering algorithm to classify construction units, based on construction time and resource requirements, pile foundation installation units are divided into high resource-intensive types (such as pile driving) and low resource-intensive types (such as foundation preparation), and gabion placement units are divided into high-precision types (such as anchoring) and conventional types (such as assembly). Assuming there are 10 pile foundations and 20 gabions, the clustering result may group 5 pile foundation driving and 10 gabion anchoring into a high-priority parallel operation group, optimizing resource allocation.

[0158] In one possible implementation, this embodiment detects operational conflicts by analyzing the schedule and identifying potential conflicts between pile driving and gabion anchoring due to sharing an underwater work vessel. When adjusting priorities, pile driving can be prioritized, while gabion anchoring can be postponed by two days, thus generating an optimized parallel operation mode. This mode ensures that equipment is not overloaded and improves construction smoothness.

[0159] Specifically, when extracting the time resource allocation plan in this embodiment, a draft plan can be generated: 2 piles per day for 5 days; 4 gabion cages per day for 5 days. When optimizing using the simulated annealing algorithm, it can be adjusted to 3 piles per day for the first 3 days and 1 pile per day for the next 2 days; 2 gabion cages per day for the first 2 days and 6 gabion cages per day for the next 3 days. This adjustment balances equipment utilization and shortens the overall construction period.

[0160] For example, if the final plan is delayed, such as the cage deployment being postponed by one day due to strong ocean currents, the parameter differences can be re-analyzed, the modular model updated, additional backup divers added, or the anchoring time extended. This iterative optimization ensures that the construction plan adapts to the dynamic environment and improves the stability of project execution.

[0161] Furthermore, the differentiated design parameters in this embodiment include:

[0162] Structural dimensions, pile length, pile diameter, wall thickness, and corrosion protection level of wind turbine pile foundations;

[0163] The external dimensions, mesh size, floating material, and anchoring method of the cage;

[0164] The minimum safe distance, relative angle, vertical spacing, and dynamic adjustment range of each parameter between the wind turbine pile foundation and the net cage as water depth, tidal current velocity, and wave height change.

[0165] Furthermore, based on the construction time coordination plan, the process of building a digital twin virtual interaction platform and generating optimized operation parameter schemes includes:

[0166] Obtain operational parameters from wind farms and ranches to create real-time operational datasets;

[0167] The real-time running dataset is cleaned and standardized to obtain a standardized running dataset.

[0168] Based on the standardized operational dataset, a digital twin virtual operational environment is constructed to obtain operational status simulation results;

[0169] Based on the simulation results of the operational status, the time conflict between equipment operation and pasture activities was analyzed, the construction time window was adjusted, and a collaborative scheduling mechanism was obtained.

[0170] The collaborative scheduling mechanism and the simulation results of the running status are loaded into the virtual interactive platform, and the optimal set of running parameters is calculated using the particle swarm optimization algorithm.

[0171] Key control parameters are extracted based on the optimal set of operating parameters, and the final operating parameter scheme is output after iterative updates.

[0172] Furthermore, this embodiment acquires the operating parameters of the wind farm and the ranch. Through sensors and a data acquisition system, wind speed, power output, and equipment status are obtained from the wind farm, and livestock distribution and pasture environmental data are obtained from the ranch, resulting in a real-time operating dataset. Through data integration and processing, the real-time operating dataset is cleaned and standardized. A preprocessing algorithm is used to remove noise and outliers. If the data integrity is below a preset threshold, missing data is supplemented using interpolation methods to obtain a standardized operating dataset. Using a digital twin model, based on the standardized operating dataset, a virtual operating environment for the wind farm and the ranch is constructed to simulate equipment operating status and livestock activity, obtaining operating status simulation results. Based on the operating status simulation results and the construction schedule, time conflicts between wind farm equipment operation and pasture activities are analyzed. If time overlap exists, the construction time window is adjusted to obtain a collaborative scheduling mechanism. Through a virtual interactive platform, the collaborative scheduling mechanism and operating status simulation results are loaded to generate an interactive operating scenario. Using a particle swarm optimization algorithm, optimal operating parameters are calculated for wind speed, power output, and livestock distribution, resulting in an optimized operating parameter set. Key control parameters are extracted from the optimized operating parameters set, and the operating mode of wind farm equipment and ranch management strategy are adjusted in real time. If the operating efficiency after adjustment is lower than the preset threshold, the parameter optimization algorithm is iteratively updated to obtain the final operating parameter scheme.

[0173] Furthermore, based on the optimized operating parameter scheme, the process of deploying a sensor network to acquire real-time data streams for intelligent ranch management includes:

[0174] Acquire water quality monitoring data and biological growth data collected by sensor networks;

[0175] Time series analysis was used to determine the fluctuation trend of environmental parameters;

[0176] If the fluctuation trend exceeds the preset threshold, the abnormal state is classified using a support vector machine to obtain the abnormal environment parameters.

[0177] Based on abnormal environmental parameters, decision trees are used to predict the degree of impact on biological growth and obtain changes in growth status.

[0178] Adjust the sensor sampling frequency according to changes in growth status to acquire high-frequency data streams;

[0179] If the high-frequency data stream continues to be abnormal, cluster analysis is used to divide the abnormal regions and obtain the environmental partitioning results.

[0180] Based on the environmental zoning results, update the sensor network configuration and output a real-time optimized data stream.

[0181] Furthermore, this embodiment acquires water quality monitoring data and biological growth data from the deployed sensor network, and uses time series analysis to determine the fluctuation trend of environmental parameters. If the fluctuation trend exceeds a preset threshold, an abnormal state is classified using a support vector machine algorithm to obtain abnormal environmental parameters. Based on the abnormal environmental parameters, a decision tree algorithm is used to predict the degree of impact on biological growth and determine changes in growth status. The sampling frequency of the sensor network is adjusted based on these changes in growth status to acquire high-frequency data streams. If the high-frequency data stream shows continuous anomalies, cluster analysis is used to divide the abnormal regions, obtaining environmental zoning results. Based on the environmental zoning results, an optimized operating parameter scheme is generated to determine the ranch management adjustment strategy. The sensor network configuration is updated by adjusting the strategy to acquire real-time optimized data streams.

[0182] For example, in the scenario of monitoring water quality and biological growth in a ranch, the deployment of sensor networks is a core component. Sensor networks typically include water quality sensors and biological growth monitoring equipment. The former monitors parameters such as pH, dissolved oxygen, and turbidity, while the latter records data such as the weight and activity frequency of farmed organisms.

[0183] In one embodiment, a water quality sensor is deployed every 100 meters in the pasture waters to record hourly pH fluctuations, assuming a normal range of 6.5-8.5. If a sensor detects a pH value of 9.0, exceeding the threshold, an anomaly analysis is triggered. This deployment method ensures comprehensive data coverage, facilitating subsequent analysis of the impact of environmental changes on biological growth.

[0184] Specifically, time series analysis is used to capture the fluctuation trends of environmental parameters. By analyzing pH and dissolved oxygen data over seven consecutive days, the system can identify periodic fluctuations or anomalous changes.

[0185] For example, if the dissolved oxygen level in a pasture's waters remains below 5 mg / L at night, it indicates a potential pollution source. Time series analysis uses a sliding window to calculate the mean and standard deviation, determining whether fluctuations exceed a preset threshold, such as a 10% rate of change. This method can quickly pinpoint the time points of environmental anomalies, providing data support for subsequent classification.

[0186] In one embodiment, a support vector machine algorithm is used to classify abnormal states. Assuming that pH, dissolved oxygen, and turbidity are used as feature inputs in the water quality data, the algorithm classifies the state into normal, slightly abnormal, and severely abnormal.

[0187] For example, a combination of pH 9.0 and dissolved oxygen 4 mg / L is classified as a severe anomaly. Support Vector Machines (SVMs) construct hyperplanes to distinguish different states, making them suitable for high-dimensional data processing. This ensures the accuracy of anomaly classification and provides a reliable basis for subsequent predictions.

[0188] For example, decision tree algorithms predict the extent to which marine organism growth is affected. Based on abnormal environmental parameters, such as a pH of 9.0, the decision tree analyzes whether the decrease in the weight gain rate of farmed organisms exceeds 5%. Through training with historical data, the algorithm found that when abnormal pH values ​​persisted for more than 3 days, the average daily weight gain rate of farmed organisms decreased to 0.2 kg, far below the normal 0.5 kg. Decision trees, through hierarchical judgment, clearly demonstrate the causal relationship between the environment and growth, facilitating rapid decision-making for farm managers.

[0189] Specifically, adjusting the sampling frequency of the sensor network is key to dynamic response. If predictions indicate that growth is affected, the sensors are adjusted from sampling once per hour to once every 15 minutes to acquire a high-frequency data stream.

[0190] For example, if dissolved oxygen levels in a body of water remain below 4 mg / L for three consecutive hours, high-frequency sampling is triggered to capture more precise fluctuation details. This adjustment improves data resolution and provides a richer data foundation for identifying anomalous areas.

[0191] In one embodiment, cluster analysis is used to segment abnormal regions. Based on high-frequency data streams, the K-means clustering algorithm divides the pasture waters into normal and abnormal areas.

[0192] For example, the northwest corner of a ranch's waterway was designated as an abnormal area due to dissolved oxygen levels consistently below 4 mg / L. Cluster analysis automatically identifies the boundaries of this abnormal area based on the similarity of data points, facilitating targeted management.

[0193] For example, the optimized operating parameter scheme is generated based on the environmental zoning results. For abnormal areas, the system recommends adding water purification equipment. For instance, deploying temporary aeration equipment in the northwest corner of the abnormal area; this scheme reduces the negative impact of the environment on biological growth through zoning management.

[0194] Specifically, updating the sensor network configuration ensures real-time optimization of the data flow. The adjusted configuration may include increasing sensor density or optimizing data transmission frequency.

[0195] For example, an additional sensor was added every 50 meters in the abnormal zone, and the data transmission interval was reduced from 10 seconds to 5 seconds. This optimization ensures real-time data and improves the responsiveness of ranch management strategies.

[0196] Furthermore, the process of analyzing biofouling and water quality change trends based on real-time data streams to generate ecological restoration parameters includes:

[0197] Real-time data streams are obtained through an intelligent ranch management platform to extract datasets of biological attachment distribution and water quality parameter changes.

[0198] Random forest algorithm was used to analyze time series characteristics and key influencing factors;

[0199] If the bio-attachment density exceeds the preset threshold, the parameters of the ecological coating on the surface of the wind turbine pile foundation are adjusted according to the key influencing factors to obtain the candidate coating formulation.

[0200] Support vector machine is used to classify candidate coating formulations and determine the optimal coating parameters;

[0201] Based on the optimized coating parameters, a design scheme for an ecological coating on the surface of wind turbine pile foundations is generated.

[0202] The coating effect is verified by monitoring environmental parameters, and iterative adjustments are made until the requirements are met, and the final ecological restoration parameters are output.

[0203] Furthermore, this embodiment obtains real-time data streams from an intelligent ranch management platform, extracts datasets of biofouling distribution and water quality parameter changes, and determines the initial environmental state. The random forest algorithm is used to analyze the trends in biofouling distribution and water quality parameter changes, obtaining time-series features and key influencing factors. If the time-series features show that the biofouling density exceeds a preset threshold, the parameters of the ecological coating on the wind turbine pile foundation surface are adjusted based on the key influencing factors to generate candidate coating formulations. A support vector machine algorithm is used to classify the candidate coating formulations, assess the biodiversity enhancement effect, and determine the optimal coating parameters. Based on the optimal coating parameters, an ecological coating design scheme for the wind turbine pile foundation surface is generated, outputting the optimized surface ecological configuration. The environmental parameter monitoring module verifies the impact of the optimized configuration on biofouling distribution and water quality parameters, obtaining updated real-time data streams. If the updated real-time data stream shows that the biodiversity enhancement effect does not reach the preset threshold, the ecological coating parameters are iteratively adjusted, and the above steps are repeated to obtain the final ecological restoration scheme.

[0204] For example, an intelligent ranch management platform acquires real-time data streams through a sensor network, collecting data on biofouling distribution and water quality parameters from the surface of wind turbine foundations. Biofouling distribution mainly refers to the density and types of algae, shellfish, etc., attached to the foundation surface, while water quality parameters include dissolved oxygen, pH, and turbidity. Determining the initial environmental state requires integrating this data to generate a baseline reflecting ecological balance.

[0205] For example, monitoring of wind turbine pile foundations in a certain sea area showed that the biofouling density was 2,000 shellfish individuals per square meter, the dissolved oxygen was 6.5 mg / L, the pH value was 7.8, and the turbidity was 10 NTU, indicating that the environment was in a stable state.

[0206] In one possible implementation, this embodiment uses a random forest algorithm to analyze the time-series characteristics of biofouling distribution and water quality parameters. The random forest constructs multiple decision trees to extract key influencing factors, such as dissolved oxygen and pH, on the significant impact of shellfish attachment density. Hypothetical analysis revealed that shellfish attachment density significantly decreases when dissolved oxygen is below 5.5 mg / L. Through time-series analysis, the trend of attachment density changes over the next week can be predicted, assisting ranch managers in early intervention.

[0207] For example, if time-series data shows that the biofilm density exceeds 3,000 individuals per square meter, triggering a preset threshold, the parameters of the ecological coating on the wind turbine foundation surface need to be adjusted. Ecological coatings are typically coatings containing specific chemical components designed to regulate biofilm adhesion. Based on key influencing factors, such as dissolved oxygen, candidate coating formulations can increase the proportion of nanomaterials that promote oxygen adsorption, for example, increasing the titanium dioxide content to 15%, to enhance surface oxygen concentration and reduce the risk of excessive adhesion.

[0208] Specifically, this embodiment uses a support vector machine algorithm to classify candidate coating formulations and evaluate their effect on enhancing biodiversity. The classification criterion is whether the coating can balance the ratio of shellfish and algae.

[0209] For example, a certain formulation reduced the shellfish attachment density to 2,500 individuals per square meter while increasing algae coverage to 30%, indicating improved biodiversity. The optimal coating parameters were thus determined to be a formulation with 15% titanium dioxide and 60% silicon-based materials, resulting in an eco-friendly coating design.

[0210] In one possible implementation, this embodiment can also verify the optimization effect through an environmental parameter monitoring module. Monitoring revealed that after applying the new coating, dissolved oxygen increased to 6.8 mg / L, shellfish attachment density stabilized at 2200 individuals per square meter, and the biodiversity index increased from 0.65 to 0.75. If the real-time data stream shows that the biodiversity enhancement effect does not meet expectations, for example, if the index is below 0.8, the coating parameters are iteratively adjusted, such as increasing the silicon-based material to 65%, and the optimization process is repeated until the ecological restoration goal is achieved.

[0211] For example, after iterative adjustments, the ecological restoration scheme can significantly improve the ecological compatibility of wind turbine foundations, maintain water quality stability, and promote the sustainable development of ranches. This method, through multi-dimensional data analysis and formula optimization, ensures the efficiency and eco-friendliness of ranch management.

[0212] Furthermore, based on ecological restoration parameters, the process of simulating the power support of surplus wind farm electricity for ranch equipment and generating a power distribution scheme that coordinates energy and ecology includes:

[0213] Real-time power output data of wind farms and power demand data of ranch water pump circulation and oxygenation equipment are obtained through digital twin models;

[0214] If the surplus power exceeds the water pump circulation demand, power will be allocated to the water pump circulation system first, and the remaining power share will be calculated.

[0215] Based on the remaining power share and the real-time demand of the oxygenation equipment, a linear programming algorithm is used to optimize the allocation ratio and obtain a power dispatching scheme for the oxygenation equipment.

[0216] The operational effects of the power dispatch scheme for oxygenation equipment are simulated in a digital twin model to obtain ecosystem balance indicators;

[0217] If the ecosystem balance index is lower than the preset threshold, the linear programming parameters are adjusted and the allocation ratio is recalculated.

[0218] Time series analysis is used to predict future changes in electricity demand and generate dynamic power dispatch plans.

[0219] Verify the effectiveness of the dynamic power dispatch plan under different scenarios and output the final power allocation scheme.

[0220] Furthermore, this embodiment uses a digital twin model to acquire real-time power output data from the wind farm and power demand data from the pasture's water pump circulation and aeration equipment, constructing a virtual simulation environment to determine the initial state of power allocation. If the surplus power data output by the digital twin model is higher than the power demand of the pasture's water pump circulation, power is preferentially allocated to the water pump circulation system, and the remaining power is calculated to obtain the share of power available for the aeration equipment. Based on the real-time demand of the aeration equipment operation and ecological restoration parameters, a linear programming algorithm is used to optimize the allocation ratio of the remaining power, determining the power dispatch scheme for the aeration equipment operation. The optimized power allocation scheme is simulated in the virtual environment using the digital twin model to obtain ecosystem balance indicators and power dispatch efficiency data, judging the stability of the allocation scheme. If the ecosystem balance indicator in the simulation results is lower than a preset threshold, the parameters of the linear programming algorithm are adjusted, the power allocation ratio is recalculated, and a new power allocation scheme is obtained. Based on the adjusted power allocation scheme, a time series analysis algorithm is used to predict the trend of power demand changes over a future period, generating a dynamic power dispatch plan. The dynamic power dispatch plan is verified using a digital twin model under different scenarios to obtain the final power allocation scheme that coordinates energy and biodiversity.

[0221] For example, this embodiment applies a digital twin model to the collaborative management of a wind farm and a ranch. A virtual simulation environment can be constructed through real-time data streams to accurately reflect the wind farm's power output and the ranch's equipment needs. Assume the wind farm outputs 1000kW of power at a certain moment, the ranch's water pump circulation system requires 400kW, and the aeration equipment requires 300kW. The digital twin model collects wind speed, power generation efficiency, and water pump operating status data from sensors to generate a virtual environment and calculate the initial power allocation. The model shows a surplus of 600kW of power. After prioritizing the water pump's needs, the remaining power can be used for the aeration equipment to ensure stable water quality.

[0222] In one possible implementation, a linear programming algorithm optimizes the allocation of remaining power. Assume the aeration equipment has two modes: high-efficiency and low-efficiency. The high-efficiency mode requires 250kW and offers better biodiversity enhancement; the low-efficiency mode requires 150kW but has limited effectiveness. The algorithm calculates the optimal allocation ratio based on ecological restoration parameters such as dissolved oxygen levels and fish activity data.

[0223] For example, the model predicts that the high-efficiency mode can increase dissolved oxygen levels to 8 mg / L, meeting the ecological threshold, while the low-efficiency mode only reaches 6 mg / L. The preferred solution is to allocate 250 kW to the high-efficiency mode, with the remaining 350 kW reserved or used for other equipment.

[0224] Specifically, the digital twin model simulates the operational effects of the optimization scheme to obtain ecosystem balance indicators, such as dissolved oxygen and biomass density. Assuming the simulation shows stable dissolved oxygen and a 10% reduction in biomass density under high-efficiency mode, this indicates effective power allocation. If the indicators fail to meet the standards, such as dissolved oxygen below 7 mg / L, the algorithm weights are adjusted, the power of the oxygenation equipment is increased to 300 kW, and the simulation is repeated for verification. This process ensures the coordinated optimization of ecology and energy.

[0225] For example, this embodiment uses time series analysis to predict the electricity demand trend for the next 24 hours. Based on historical data, the model finds that nighttime wind speeds decrease, power output drops to 800kW, water pump demand remains unchanged, and oxygenation demand increases to 350kW. The algorithm generates a dynamic scheduling plan, prioritizing water pumps and ensuring that oxygenation equipment operates in high-efficiency mode at night. This plan is validated using a digital twin model, simulating different wind speed scenarios to ensure dissolved oxygen levels remain stable above 7.5mg / L, thereby improving biodiversity indicators.

[0226] In one possible implementation, the final power allocation scheme incorporates verification results from multiple scenarios.

[0227] For example, in strong wind scenarios, with ample power, the aeration equipment operates efficiently around the clock; in weak wind scenarios, water pumps are prioritized, and the aeration equipment operates intermittently. The solution continuously optimizes power allocation through real-time data feedback, ensuring the effectiveness of pasture ecological restoration and improving biodiversity and energy efficiency.

[0228] The above are merely preferred embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for the collaborative development of offshore wind power and marine ranching, characterized in that, include: Acquire data on the ecological characteristics of marine ranches and generate ecological adaptation parameters; Based on the ecological adaptation parameters, the impact of wind turbine pile foundation layout on hydrodynamics is simulated, and the space allocation ratio of the cage is calculated. Based on the space allocation ratio of the net cages and combined with the regional water depth and tidal dynamics, the coordinated layout scheme of wind turbine pile foundations and net cages is optimized to obtain differentiated design parameters. Based on the aforementioned differentiated design parameters, a modular parallel construction process is planned to generate a construction time coordination plan; Based on the aforementioned construction time coordination plan, a digital twin virtual interaction platform is constructed to generate an optimized operation parameter scheme; Based on the optimized operating parameter scheme, deploy a sensor network to acquire real-time data streams for intelligent ranch management; Based on the real-time data stream, analyze the trends of biological attachment and water quality changes, and generate ecological restoration parameters; Based on the ecological restoration parameters, the power support of surplus wind farm electricity for ranch equipment is simulated to generate a power distribution scheme that coordinates energy and ecology.

2. The method according to claim 1, characterized in that, The process of acquiring marine ranch ecological characteristic data and generating ecological adaptation parameters includes: Data on water temperature, salinity, dissolved oxygen, and light intensity at different water depths were obtained to acquire a set of environmental parameters; Based on the set of environmental parameters, a clustering algorithm is used to perform hierarchical clustering according to water depth to obtain the ecological characteristics of each water depth range; Based on the aforementioned ecological characteristics, calculate the fish distribution characteristics; Based on the aforementioned fish distribution characteristics, the correlation between algae distribution and fish distribution is predicted, and the distribution pattern of biological density is obtained. Calculate the cage layout parameters based on the described biological density distribution pattern; If the grid cage layout parameters overlap with the wind turbine pile foundation location, the ecological adaptation parameters are obtained by adjusting them through a spatial optimization algorithm.

3. The method according to claim 1, characterized in that, Based on the aforementioned ecological adaptation parameters, the process of simulating the impact of wind turbine foundation layout on hydrodynamics and calculating the space allocation ratio of the net cages includes: A hydrodynamic model is constructed based on the ecological adaptation parameters to obtain the initial flow field distribution; Based on the initial flow field distribution, a variety of layout schemes are generated using a genetic algorithm, and the changes in water flow velocity and wave height for each scheme are obtained. Based on the changes in water flow velocity and wave height, calculate the set of cage space allocation ratios corresponding to each scheme; Layout schemes with a proportion exceeding a preset threshold are removed to obtain an optimized scheme set. The ecological impact factors are calculated based on the set of optimization schemes, and the optimal layout scheme is obtained after sorting them. Based on the optimal layout scheme, the final cage space allocation ratio is output.

4. The method according to claim 1, characterized in that, Based on the aforementioned cage space allocation ratio, and combined with the regional water depth and tidal dynamics characteristics, the process of optimizing the coordinated layout scheme of wind turbine foundations and cages to obtain differentiated design parameters includes: Acquire regional water depth and tidal dynamics data to form a set of regional environmental characteristics; Based on the set of regional environmental characteristics, the stress distribution of the pile foundation is calculated using finite element analysis to obtain the pile foundation structural parameters; If the pile foundation structural parameters meet the stability threshold, then a preliminary collaborative layout scheme is generated by using a particle swarm optimization algorithm, taking into account the cage space allocation ratio. Calculate the hydrodynamic influence coefficient based on the preliminary coordinated layout scheme; If the hydrodynamic influence coefficient is lower than the preset threshold, the size and position of the cage are adjusted to obtain differentiated design parameters; A genetic algorithm is used to iteratively adjust the spatial positions and output the final collaborative layout parameters and 3D layout model.

5. The method according to claim 1, characterized in that, Based on the aforementioned differentiated design parameters, the process of planning a modular parallel construction process and generating a construction time coordination plan includes: Obtain the design parameters of the pile foundation and the cage design parameters, extract the construction requirements and resource needs, and obtain the parameter difference analysis results; Based on the parameter difference analysis results, a modular construction process model is constructed, and independent construction units are divided. K-means clustering was used to classify the construction units, resulting in parallel operation groups; Detect job conflicts in the parallel job groups, adjust group priorities, and obtain an optimized parallel job mode; Based on the optimized parallel operation mode, a draft of the construction time coordination plan is generated. The initial draft of the construction time coordination plan is iteratively optimized using the simulated annealing algorithm, and the final construction time coordination plan is output.

6. The method according to claim 1, characterized in that, The process of constructing a digital twin virtual interaction platform and generating optimized operating parameter schemes based on the aforementioned construction time coordination plan includes: Obtain operational parameters from wind farms and ranches to create real-time operational datasets; The real-time running dataset is cleaned and standardized to obtain a standardized running dataset; Based on the standardized operational dataset, a digital twin virtual operational environment is constructed to obtain operational state simulation results; Based on the simulation results of the operating status, the time conflict between equipment operation and pasture activities is analyzed, the construction time window is adjusted, and a collaborative scheduling mechanism is obtained. The cooperative scheduling mechanism and the simulation results of the running status are loaded into the virtual interactive platform, and the optimal set of running parameters is calculated using the particle swarm optimization algorithm. Based on the optimal set of operating parameters, key control parameters are extracted, and the final operating parameter scheme is output after iterative updates.

7. The method according to claim 1, characterized in that, The process of deploying a sensor network and acquiring real-time data streams for intelligent ranch management based on the optimized operating parameter scheme includes: Acquire water quality monitoring data and biological growth data collected by sensor networks; Time series analysis was used to determine the fluctuation trends of environmental parameters; If the fluctuation trend exceeds the preset threshold, the abnormal state is classified using a support vector machine to obtain the abnormal environment parameters. Based on the abnormal environmental parameters, a decision tree is used to predict the degree of impact on biological growth and obtain changes in growth status. Based on the changes in growth status, the sensor sampling frequency is adjusted to acquire a high-frequency data stream; If the high-frequency data stream continues to be abnormal, cluster analysis is used to divide the abnormal regions and obtain the environmental partitioning results. Based on the environmental partitioning results, update the sensor network configuration and output a real-time optimized data stream.

8. The method according to claim 1, characterized in that, The process of analyzing biofouling and water quality change trends based on the real-time data stream and generating ecological restoration parameters includes: Real-time data streams are obtained through an intelligent ranch management platform to extract datasets of biological attachment distribution and water quality parameter changes. Random forest algorithm was used to analyze time series characteristics and key influencing factors; If the bio-attachment density exceeds the preset threshold, the parameters of the ecological coating on the surface of the wind turbine pile foundation are adjusted according to the key influencing factors to obtain the candidate coating formulation. The candidate coating formulations are classified using a support vector machine to determine the preferred coating parameters; Based on the preferred coating parameters, an ecological coating design scheme for the surface of wind turbine pile foundations is generated. The coating effect is verified by monitoring environmental parameters, and iterative adjustments are made until the requirements are met, and the final ecological restoration parameters are output.

9. The method according to claim 1, characterized in that, Based on the aforementioned ecological restoration parameters, the process of simulating the power support of surplus wind farm electricity for ranch equipment and generating a power allocation scheme that coordinates energy and ecology includes: Real-time power output data of wind farms and power demand data of ranch water pump circulation and oxygenation equipment are obtained through digital twin models; If the surplus power exceeds the water pump circulation demand, power will be allocated to the water pump circulation system first, and the remaining power share will be calculated. Based on the remaining power share and the real-time demand of the oxygenation equipment, a linear programming algorithm is used to optimize the allocation ratio and obtain a power dispatching scheme for the oxygenation equipment. The operational effect of the power dispatch scheme for the oxygenation equipment is simulated in a digital twin model to obtain ecosystem balance indicators; If the ecosystem balance index is lower than the preset threshold, the linear programming parameters are adjusted and the allocation ratio is recalculated. Time series analysis is used to predict future changes in electricity demand and generate dynamic power dispatch plans. Verify the operational effectiveness of the dynamic power dispatch plan under different scenarios and output the final power allocation scheme.

10. The method according to claim 1, characterized in that, The differentiated design parameters include: Structural dimensions, pile length, pile diameter, wall thickness, and corrosion protection level of wind turbine pile foundations; The external dimensions, mesh size, floating material, and anchoring method of the cage; Minimum safe distance, relative angle, and vertical spacing between wind turbine pile foundations and grid cages.

Images