A wave energy conversion enhancement method based on spatiotemporal convolutional neural network

Abstract

The present invention relates to the field of wave energy conversion technology, specifically a wave energy conversion enhancement method based on spatiotemporal convolutional neural network, comprising the following steps: receiving wave height, wave speed and frequency data collected by buoys and satellites, performing time series analysis on the data, calculating the wave energy intensity, analyzing the directional distribution of the wave according to the wave energy intensity calculation result, revealing the characteristics of the wave activity in the sea area, and generating a preliminary wave feature set. In the present invention, by integrating the spatiotemporal convolutional neural network with high-precision satellite and buoy data processing, the capture and conversion efficiency of wave energy is effectively improved, hierarchical feature extraction can be performed, wave energy patterns and periodic changes can be accurately analyzed, and the position of high-energy wave areas can be efficiently calibrated to ensure that the equipment in the high-energy wave area has higher energy capture potential and operating efficiency, dynamically adjust the equipment position and parameters, significantly improve the energy conversion efficiency, and enhance the sustainable utilization and economy of wave energy.

Description

Technical Field

[0001] The present invention relates to the technical field of wave energy conversion, and in particular to a wave energy conversion enhancement method based on a spatiotemporal convolutional neural network. Background Art

[0002] The field of wave energy conversion technology involves methods and devices for extracting energy from ocean waves and converting it into useful forms, such as electricity, using the kinetic energy or potential energy of the waves and converting it into electrical energy through various mechanical and electrical systems. Common devices include floating wave energy converters, progressive wave energy converters and oscillating water column converters, which are usually installed near the coast or in deep sea areas. They can effectively capture the movement of waves and drive generators through hydraulic or direct mechanical means. Wave energy conversion technology is a type of renewable energy technology that has the potential to not consume fuel, reduce environmental pollution and provide a stable energy supply.

[0003] Among them, wave energy conversion enhancement methods refer to strategies and technologies to improve existing wave energy conversion equipment and technologies to improve their energy conversion efficiency and economy, including optimizing equipment design, introducing advanced materials to reduce wear, and improving control systems and algorithms for energy capture rates. The purpose is to enable wave energy equipment to operate efficiently in a wider marine environment while reducing construction and maintenance costs. Through enhancement methods, ocean waves can be more effectively used as a continuous energy supply, which will help promote the popularization of renewable energy and the optimization of energy structure.

[0004] Although existing wave energy conversion technologies can capture wave energy in fixed areas, they usually face problems of inconsistent efficiency and high maintenance costs. Traditional equipment relies on static design and installation locations, which are difficult to adapt to rapidly changing ocean environments and wave conditions, resulting in low energy capture efficiency. In addition, the lack of real-time adjustment of position or parameters makes the maintenance cost of the equipment unchanged under low-energy wave conditions, increasing the economic burden of operation. Technical limitations affect the stability of energy output and limit the widespread application and development potential of wave energy as a renewable energy source. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and propose a wave energy conversion enhancement method based on spatiotemporal convolutional neural network.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: a wave energy conversion enhancement method based on spatiotemporal convolutional neural network, comprising the following steps:

[0007] S1: Receive wave height, wave speed and frequency data collected by buoys and satellites, perform time series analysis on the data, calculate wave energy intensity, analyze the directional distribution of waves based on the wave energy intensity calculation results, reveal the characteristics of wave activity in the sea area, and generate a preliminary wave feature set;

[0008] S2: Using the preliminary wave feature set, applying the spatiotemporal convolutional neural network, performing hierarchical feature extraction on the data, analyzing the wave energy pattern according to the feature extraction results, using the threshold to judge and calibrate the position of the high-energy wave area, and generating the wave energy feature mapping result;

[0009] S3: Analyze the spatial distribution of high-energy wave areas according to the wave energy characteristic mapping results, count the energy peaks, calculate the energy capture potential of the equipment at different positions according to the energy peak statistics, design a customized equipment deployment plan for each high-energy wave area, and output the equipment configuration optimization plan;

[0010] S4: According to the equipment configuration optimization plan, adjust the equipment position and operating parameters to match the capture of high-energy waves, monitor the equipment operating status and energy capture efficiency in real time, and generate a wave energy conversion enhancement report.

[0011] As a further solution of the present invention, the calculation steps of the fluctuation energy intensity are:

[0012] S111: Receive wave height, wave speed and frequency data from buoys and satellites, perform data aggregation, and generate a comprehensive data set;

[0013] S112: Performing time series analysis on the comprehensive data set, analyzing fluctuation patterns and periodicity, identifying key fluctuation indicators, and generating time series analysis results;

[0014] S113: Using the time series analysis results, adopt the formula:

[0015]

[0016] Calculate the wave energy intensity E and obtain the wave energy intensity result, where ρ represents the density of water and V represents the average value of the wave velocity.

[0017] As a further solution of the present invention, the steps of obtaining the preliminary wave feature set are:

[0018] S121: Using the wave energy intensity result, perform wave direction distribution analysis using the formula:

[0019] E(θ)=∑(E·cos(θ-θ i ))

[0020] Calculate the total fluctuation energy value E(θ) in the target direction to generate the fluctuation directional distribution result, where E represents the fluctuation energy in the difference direction, θ is the fluctuation angle, and θ i is the angular difference in actual direction;

[0021] S122: According to the wave directional distribution result, identify the difference in wave characteristics between regions, reveal the behavior pattern of waves in different sea environments, and generate the result of sea wave activity characteristics;

[0022] S123: Using the results of the characteristics of the sea area wave activity, perform data clustering and integration to generate a preliminary wave feature set.

[0023] As a further solution of the present invention, the step of extracting the hierarchical features is:

[0024] S211: Input the preliminary wave feature set into the spatiotemporal convolutional neural network, perform time and space dimension feature extraction, and obtain preliminary results of time and space feature analysis;

[0025] S212: applying multi-layer convolution and pooling technology to the preliminary results of the temporal and spatial feature analysis, performing detection capabilities of local temporal and spatial features through the convolution layer, and reducing feature dimensions using the pooling layer to obtain deep feature extraction results;

[0026] S213: Input the deep feature extraction result into the fully connected layer, use a nonlinear activation function to capture complex nonlinear relationships, optimize parameters through network training, and obtain a hierarchical feature extraction result.

[0027] As a further solution of the present invention, the steps for obtaining the wave energy characteristic mapping result are:

[0028] S221: Based on the hierarchical feature extraction result, the formula is used:

[0029]

[0030] Calculate the energy density M of each wave event, evaluate the energy distribution in each area, and obtain the preliminary analysis results of wave energy distribution, where ρ is the density of water, g is the gravitational acceleration, and A is the wave height;

[0031] S222: setting an energy threshold, combining the preliminary analysis result of the wave energy distribution, evaluating the energy density of each wave event, marking the wave area exceeding the threshold as a high energy area by comparing with the set threshold, and generating a high energy area list;

[0032] S223: Based on the high-energy area list, the identified high-energy wave area is geographically calibrated, and the location is simultaneously marked on the ocean map to generate a wave energy feature mapping result.

[0033] As a further solution of the present invention, the statistical step of the energy peak is:

[0034] S311: Using the wave energy characteristic mapping result, extracting energy data of all regions and establishing wave energy distribution data;

[0035] S312: Based on the wave energy distribution data, perform spatial analysis, refine the energy level of each area, and generate an actual energy distribution map;

[0036] S313: Perform statistical analysis of energy peaks according to the actual energy distribution diagram, determine the maximum energy value of each identified high energy area, and obtain energy peak statistical results.

[0037] As a further solution of the present invention, the output step of the device configuration optimization solution is:

[0038] S321: Call the energy peak statistical result to extract the energy peak data of the high energy wave area, using the formula:

[0039]

[0040] Calculate the energy capture potential value P of each area c , and obtain the energy capture potential data, where E maxi is the energy peak of region i, A i is the effective capture area of ​​region i, d i is the distance from the device to the target area, and n represents the number of areas;

[0041] S322: Using the energy capture potential data, combined with the topography, depth and environmental conditions of the high-energy wave area, design energy capture equipment configuration for each area, match the energy capture potential requirements, and establish a customized equipment deployment plan;

[0042] S323: Verify and adjust the customized equipment deployment plan, analyze the energy capture capacity and cost efficiency of all configuration plans by simulating equipment operation performance and cost data, perform multi-index optimization, and obtain an equipment configuration optimization plan.

[0043] As a further solution of the present invention, the steps of obtaining the wave energy conversion enhancement report are:

[0044] S411: Based on the equipment configuration optimization plan, analyze the wave data, determine the optimal placement position of the equipment, and generate an equipment position adjustment plan;

[0045] S412: According to the equipment position adjustment plan, dynamically adjust the equipment operating parameters to match the wave dynamic conditions of the difference area, and obtain an operating parameter optimization plan;

[0046] S413: Apply the operation parameter optimization scheme to perform real-time monitoring, using the formula:

[0047]

[0048] Calculate the energy capture efficiency Ey and generate a wave energy conversion enhancement report; where CE represents the captured energy value and PE represents the potential energy value of the area.

[0049] Compared with the prior art, the advantages and positive effects of the present invention are:

[0050] In the present invention, by integrating spatiotemporal convolutional neural networks with high-precision satellite and buoy data processing, the capture and conversion efficiency of wave energy is effectively improved, hierarchical feature extraction can be performed, wave energy patterns and periodic changes can be accurately analyzed, and the position of high-energy wave areas can be efficiently calibrated to ensure that the equipment in high-energy wave areas has higher energy capture potential and operating efficiency, dynamically adjust the equipment position and parameters, significantly improve the energy conversion efficiency, and enhance the sustainable utilization and economy of wave energy. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 It is a main step flow chart of the present invention;

[0052] Figure 2 is a flow chart for calculating the wave energy intensity of the present invention;

[0053] Figure 3 This is a flow chart for obtaining a preliminary wave feature set of the present invention;

[0054] Figure 4 It is a flowchart of extracting hierarchical features of the present invention;

[0055] Figure 5 This is a flow chart for obtaining the wave energy characteristic mapping results of the present invention;

[0056] Figure 6 It is a statistical flow chart of energy peak value of the present invention;

[0057] Figure 7 Output flow chart of the device configuration optimization scheme of the present invention;

[0058] Figure 8 This is a flow chart for obtaining the wave energy conversion enhancement report of the present invention. DETAILED DESCRIPTION

[0059] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0060] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0061] See also Figure 1 , a wave energy conversion enhancement method based on spatiotemporal convolutional neural network, comprising the following steps:

[0062] S1: Receive wave height, wave speed and frequency data collected by buoys and satellites, perform time series analysis on the data, calculate wave energy intensity, analyze the directional distribution of waves based on the wave energy intensity calculation results, reveal the characteristics of wave activity in the sea area, and generate a preliminary wave feature set;

[0063] S2: Based on the preliminary wave feature set, a spatiotemporal convolutional neural network is applied to extract hierarchical features of the data. The wave energy pattern is analyzed based on the feature extraction results. The location of the high-energy wave area is determined and calibrated using the threshold to generate the wave energy feature mapping results.

[0064] S3: Analyze the spatial distribution of high-energy wave areas based on the wave energy characteristic mapping results, count the energy peaks, calculate the energy capture potential of the equipment at different positions based on the energy peak statistics, design a customized equipment deployment plan for each high-energy wave area, and output the equipment configuration optimization plan;

[0065] S4: According to the equipment configuration optimization plan, adjust the equipment position and operating parameters to match the capture of high-energy waves, monitor the equipment operating status and energy capture efficiency in real time, and generate a wave energy conversion enhancement report.

[0066] The preliminary wave feature set includes energy indicators, directional indicators and activity characteristics; the wave energy feature mapping results include energy distribution diagrams, periodic characteristics and high-energy area calibration results; the equipment configuration optimization plan includes deployment records and energy capture strategies; the wave energy conversion enhancement report includes efficiency evaluation results, equipment status monitoring records and high-energy matching effects.

[0067] See also Figure 2 , the calculation steps of the wave energy intensity are:

[0068] S111: Receive wave height, wave speed and frequency data from buoys and satellites, perform data aggregation, and generate a comprehensive data set;

[0069] Receive wave height, wave speed and frequency data from buoys and satellites, and organize these data to form a comprehensive data set. The integration process includes verifying the format and range of the data, eliminating data that does not meet the specifications to reduce the impact of errors and biases, and performing data correction to adjust occasional errors and systematic biases in the data set. Time-stamp the data to ensure that each piece of data can be traced back to the actual collection time and location, which is crucial for subsequent time series analysis. Synchronous processing of data ensures that data collected from different devices and locations can be integrated together.

[0070] S112: Perform time series analysis on the comprehensive data set to analyze volatility patterns and periodicity, identify key volatility indicators, and generate time series analysis results;

[0071] Time series analysis is performed based on comprehensive data sets. Advanced statistical methods such as cycle decomposition and trend analysis are used to reveal the main components and seasonal patterns in the fluctuating data. The autoregressive moving average model (ARMA) is used, which includes determining the order of the model, estimating the model parameters, and performing model diagnosis to ensure that the selected model can effectively capture the dynamic characteristics of the data. In this process, information criteria such as the Akaike Information Criterion (AIC) are used to select the most appropriate model, thereby accurately describing the random fluctuations in the data and generating time series analysis results.

[0072] S113: Using the time series analysis results, the formula is:

[0073]

[0074] Calculate the wave energy intensity E and obtain the wave energy intensity result, where ρ represents the density of water and V represents the average value of the wave velocity;

[0075] ρ is the density of water, usually taken as 1000 kg / m 3 , V is the average value of the wave velocity, the measured wave velocity is 5m / s, then the wave energy can be calculated as:

[0076]

[0077] The results show that the wave energy intensity is 12,500 joules, reflecting the kinetic energy level of the fluctuation at the moment of measurement, which is of great significance for analyzing the energy distribution and propagation characteristics of the fluctuation.

[0078] See also Figure 3, the steps to obtain the preliminary wave feature set are:

[0079] S121: Using the wave energy intensity results, the wave direction distribution analysis is performed using the formula:

[0080] E(θ)=∑(E·cos(θ-θ i ))

[0081] Calculate the total fluctuation energy value E(θ) in the target direction to generate the fluctuation directional distribution result, where E represents the fluctuation energy in the difference direction, θ is the fluctuation angle, and θ i is the angular difference in actual direction;

[0082] There is a direction θ=90° (eastward), and there are three wave energy components E1=200J, E2=150J, E3=100J, and the corresponding direction angle differences are θ1=80°, θ2=90°, θ3=100°.

[0083] The calculation process is as follows:

[0084]

[0085] E(θ)≈200×0.9848+150×1+100×0.9848

[0086] E(θ)=196.96+150+98.48

[0087] E(θ)=445.44J

[0088] The results show that the total wave energy in the 90-degree direction is 445.44 joules, indicating that the main energy is concentrated in the east direction, providing data support for further analysis of the wave propagation characteristics in this direction.

[0089] S122: Based on the wave directional distribution results, identify the differences in wave characteristics between regions, reveal the behavior patterns of waves in different sea environments, and generate the results of sea wave activity characteristics;

[0090] Using the results of the directional distribution of fluctuations, energy density mapping technology is used to analyze the main characteristics of the fluctuation activities in the sea area. By comparing and analyzing the different fluctuation areas, the process includes mapping the energy results in each direction into two-dimensional or three-dimensional space, and using color coding or heat maps to visualize the energy distribution. By collecting the energy values ​​in the fluctuation data and classifying and counting the data according to direction, each category will be represented by a different color according to the size of the energy value, and then a heat map will be created on the geographic information system (GIS) platform. By observing the color changes and concentrated areas in the heat map, the high-energy areas in the sea area and their relative intensity to the surrounding fluctuations can be intuitively identified to determine which directions of fluctuations are beyond the scope of conventional activities, thereby helping to understand the behavior patterns and energy transfer paths of the fluctuations and generate the results of the characteristics of the sea area fluctuation activities.

[0091] S123: Using the results of the characteristics of the sea wave activity, perform data clustering and integration to generate a preliminary wave feature set;

[0092] Based on the results of the characteristics of sea wave activity, including integrated wave height, wave speed, wave direction and wave frequency, the process uses data clustering methods to integrate these characteristics. Cluster analysis groups data according to the physical characteristics of waves to identify patterns of similar wave behaviors. The clustering results will be displayed in the form of charts and graphs to provide intuitive wave behavior analysis. The process first involves data preprocessing, including normalizing the wave data to eliminate the influence of different measurement scales. Then the K-means clustering algorithm is applied to analyze the processed data, and the appropriate number of clusters is selected to best represent the wave patterns in the data. Each cluster will represent a unique wave characteristic, such as average wave height and wave frequency. Finally, the clustering information will be compiled into a comprehensive wave feature set for use in marine engineering design and safety management of marine activities, providing decision makers with key marine environmental information.

[0093] See also Figure 4 , the steps of extracting hierarchical features are:

[0094] S211: Input the preliminary wave feature set into the spatiotemporal convolutional neural network, perform time and space dimension feature extraction, and obtain preliminary results of time and space feature analysis;

[0095] The preliminary wave feature set is input into the spatiotemporal convolutional neural network. This network structure is specially designed to process time series and space-related data. The temporal and spatial patterns in the wave data are identified and extracted through deep learning of the layer-by-layer network to enhance the model's ability to predict the dynamic changes of waves. The process involves the use of a large number of neurons and layers to simulate the multidimensionality of wave data and perform feature extraction in the time and space dimensions, including the use of convolutional layers to capture the relationship between adjacent data in space and the use of recursive layers to identify the time-dependent relationship between data. Through data processing, the accuracy and efficiency of the prediction model are improved, and preliminary results of the temporal and spatial feature analysis are obtained.

[0096] S212: Apply multi-layer convolution and pooling techniques to the preliminary results of temporal and spatial feature analysis, use the convolution layer to detect local temporal and spatial features, and use the pooling layer to reduce the feature dimension to obtain deep feature extraction results;

[0097] Multi-layer convolution and pooling techniques are applied to the preliminary results. By configuring different convolution kernel sizes and step sizes, different levels of features of the wave data are systematically extracted. The features include wave height, frequency and other related attributes. The superposition of multiple convolutional layers enables the network to deeply explore the complex patterns within the wave data, while the pooling layer reduces the amount of calculation by reducing the feature dimension and improves the generalization ability of the model. The pooling operation is performed by selecting the maximum value or average value in a specific area, which not only retains the most significant features but also suppresses noise, and finally obtains a deep feature extraction result containing the core characteristics of the wave.

[0098] S213: Input the deep feature extraction results into the fully connected layer, use the nonlinear activation function to capture the complex nonlinear relationship, optimize the parameters through network training, and obtain the hierarchical feature extraction results;

[0099] The deep feature extraction results are input into the fully connected layer, which involves the superposition of multiple fully connected layers. Each layer fully connects and calculates the output of the previous layer. Nonlinear activation functions such as ReLU are used to enhance the model's ability to capture complex nonlinear relationships. The ReLU function helps the model remove inactive neurons during forward propagation and reduce the risk of overfitting. The output of the fully connected layer optimizes the neural network weights through the gradient descent method, adjusts the contribution of each neuron, and optimizes the parameters through network training. The entire network continuously adjusts its internal structure through iterative learning to adapt to the characteristic distribution of wave data, and finally obtains a hierarchical feature extraction result that can represent wave behavior.

[0100] See also Figure 5 , the steps to obtain the wave energy characteristic mapping results are:

[0101] S221: Based on the hierarchical feature extraction results, the formula is:

[0102]

[0103] Calculate the energy density M of each wave event, evaluate the energy distribution in each area, and obtain the preliminary analysis results of wave energy distribution, where ρ is the density of water, g is the gravitational acceleration, and A is the wave height;

[0104] ρ is the density of water (1000kg / m^3, the density of water under standard conditions), g is the acceleration due to gravity (9.8m / s^2, the standard acceleration due to gravity on the surface of the earth), and A is the wave height (measured as 2 meters). Plug in the values ​​and calculate:

[0105]

[0106] The results show that the energy stored in every cubic meter of seawater is 19,600 Joules, which is calculated through the wave height data obtained through actual monitoring and reflects the kinetic energy density of the waves.

[0107] S222: setting an energy threshold, combining the preliminary analysis results of wave energy distribution, evaluating the energy density of each wave event, marking the wave area exceeding the threshold as a high energy area by comparing with the set threshold, and generating a high energy area list;

[0108] After setting the energy threshold, the energy density of each wave event is evaluated. By comparing the energy density of each wave event with the set threshold, the wave areas that exceed this threshold are identified. This process involves several key steps: First, the energy threshold is determined based on historical data. It is often set by statistically analyzing the distribution characteristics of historical wave energy data. For example, the upper 90% of the energy value is selected as the threshold to ensure that only the strongest wave events are identified as high-energy areas; then the energy data of the current wave event is processed and compared with the threshold; in addition, the process also uses graphical tools to visualize the energy distribution, helping analysts to intuitively see which areas exceed the energy threshold, thereby more effectively identifying high-energy areas; finally, a detailed list of all high-energy areas is generated, which will be marked and used in subsequent research or practical applications, such as the design and layout optimization of marine structures.

[0109] S223: Based on the high-energy area list, the identified high-energy wave area is geographically calibrated, and the location is simultaneously marked on the ocean map to generate a wave energy feature mapping result;

[0110] The identified high-energy wave areas are precisely calibrated using Geographic Information System (GIS) technology, and their locations are marked in detail on ocean maps. First, the data of the identified high-energy wave areas are connected to the geographic coordinate system. Usually, the data collected by the wave monitoring equipment needs to be synchronized with the GPS data to ensure that the geographic location of each data point is accurate. These geographic coordinates are then converted into visual markers on the map. Each marker represents the specific location of a high-energy area. The markers can be different colors or shapes to indicate different energy levels. In addition, the process also includes further refinement of the map, such as adding additional information such as wave frequency and duration to enhance the map's information content and practicality. The resulting wave energy feature mapping results not only provide visual information, but can also be used by marine scientists, engineers and policymakers to make more informed decisions, such as designing more robust marine structures near high-energy areas or developing relevant safety warning systems.

[0111] See also Figure 6 , the statistical steps of energy peak are:

[0112] S311: Using the wave energy characteristic mapping result, extracting energy data of all regions and establishing wave energy distribution data;

[0113] A detailed analysis is conducted based on the results of wave energy feature mapping. First, the wave energy data of each area is extracted from the data set, including basic information such as sea depth, wave height, frequency, etc. Then, data analysis techniques such as peak analysis are used to identify characteristic points with higher energy in each area. This process requires normalization of the data to eliminate the deviation between different measuring devices. The normalized data can more truly reflect the actual distribution of wave energy. Then, based on the normalized data, representative high-energy areas are determined through cluster analysis. The final wave energy distribution data provides a basis for subsequent steps.

[0114] S312: Based on the wave energy distribution data, perform spatial analysis, refine the energy level of each area, and generate an actual energy distribution map;

[0115] After obtaining the wave energy distribution data, the data is applied to the geographic information system (GIS) for more detailed spatial analysis. The spatial analysis tools of GIS software, such as spatial interpolation methods and surface analysis techniques, are mainly used to visualize the wave energy data. Through this processing, the energy intensity of different regions can be intuitively displayed in geographic space, which facilitates the identification of energy-concentrated areas. These areas are usually the best locations for wave energy development. Through GIS analysis, specific coordinates can be accurately determined, which facilitates subsequent resource development and utilization planning. The actual energy distribution map generated provides decision support for the deployment of wave energy equipment.

[0116] S313: performing statistical analysis of energy peaks according to the actual energy distribution diagram, determining the maximum energy value of each identified high energy area, and obtaining energy peak statistical results;

[0117] After obtaining the specific energy distribution diagram, the data is statistically analyzed for energy peaks and peak recognition is performed. The highest energy points in the data can be automatically identified and the points marked and recorded. Statistical analysis is not only about finding the maximum value, but also includes analyzing the distribution characteristics of these peaks, such as spatial continuity, frequency distribution, etc. The analysis results are crucial to understanding the spatial distribution law of wave energy and ultimately obtaining energy peak statistical results.

[0118] See also Figure 7 , the output steps of the equipment configuration optimization solution are:

[0119] S321: Call the energy peak statistical results to extract the energy peak data of the high energy wave area, using the formula:

[0120]

[0121] Calculate the energy capture potential value P of each area c , and obtain the energy capture potential data, where E maxi is the energy peak of region i, A i is the effective capture area of ​​region i, d i is the distance from the device to the target area, and n represents the number of areas;

[0122] Collect the following data: the energy peak E in region 1 max1 =500kW / m^2, effective capture area A1=2000m^2, distance d1=50m; Substitute the values ​​into the formula:

[0123]

[0124] By calculating similar capture potentials for all areas and summing them up, the final energy capture potential result is obtained. The results show that the energy capture potential value of area 1 is 20,000 kW, which provides an optimization basis for subsequent equipment deployment.

[0125] S322: Using the energy capture potential data, combined with the topography, depth and environmental conditions of the high-energy wave area, design the energy capture equipment configuration for each area, match the energy capture potential requirements, and establish a customized equipment deployment plan;

[0126] Using the energy capture potential data and combining it with the environmental conditions of each area, the actual characteristics of the sea area are analyzed in detail, including the wave direction distribution, seabed topography and water depth. First, according to the wave direction distribution map and the performance parameters of the energy capture equipment, the type of equipment suitable for each high-energy area is determined. For example, a fixed energy capture device is selected in areas with stable wave direction, and a floating energy capture device is selected in multi-directional wave areas. Then, the optimal equipment installation angle is calculated through statistics on wave height and frequency, and the main frequency characteristics of the wave direction are used to adjust the equipment's entry direction into the water to maximize energy capture. Finally, based on the regional energy capture potential and the equipment's energy capture efficiency, the optimal equipment installation density is calculated. By matching the equipment capture range with the regional area, the number and distribution plan of equipment in each area are determined, and finally an adaptive equipment deployment plan is generated.

[0127] S323: Verify and adjust the customized equipment deployment plan, analyze the energy capture capacity and cost efficiency of all configuration plans by simulating equipment operation efficiency and cost data, perform multi-index optimization, and obtain the equipment configuration optimization plan;

[0128] The customized equipment deployment plan is simulated and verified. First, the wave energy distribution, wave frequency and wave direction data of each equipment operation area are collected. The energy capture efficiency of the equipment in these environments is analyzed to verify its adaptability. Then, the long-term operation cost of the equipment is calculated in detail, including equipment procurement cost, installation cost and daily maintenance cost. At the same time, the long-term energy output of each equipment is evaluated. According to the simulation results of equipment operation, the unit energy cost index (total cost divided by energy output) is used to evaluate the equipment performance. On this basis, the cost-effectiveness ratio is improved by adjusting the equipment installation angle and optimizing the number or density of equipment. At the same time, the deployment effect of each equipment is comprehensively analyzed in combination with regional characteristics. Finally, a deployment plan with high energy capture efficiency and strong economy is selected, and it is optimized into the final equipment configuration optimization plan.

[0129] See also Figure 8 , the steps to obtain the wave energy conversion enhancement report are:

[0130] S411: Based on the equipment configuration optimization plan, analyze the wave data, determine the optimal placement position of the equipment, and generate an equipment position adjustment plan;

[0131] Based on the equipment configuration optimization plan, the best equipment placement location is determined through GIS system and ocean wave model analysis. First, the GIS system accurately measures the geographic information to generate terrain and geographic location data sets. Then, combined with the wave energy distribution data output by the ocean wave model, the energy fusion technology is used to determine the area with maximized energy. This process involves big data analysis technology and the spatial data processing capabilities of the geographic information system. By calculating the average energy value of each area, comparing the energy values ​​of each area, and finding the location of the maximum value, the equipment placement location is finally determined to provide the optimal solution and generate an equipment location adjustment plan.

[0132] S412: According to the equipment position adjustment plan, dynamically adjust the equipment operating parameters to match the wave dynamic conditions of the difference area, and obtain an operating parameter optimization plan;

[0133] Fine-tuning of the equipment's operating parameters is based on control theory and feedback mechanisms. First, the control system collects data on the equipment's current operating status and wave dynamic conditions. Then, the algorithm analyzes this data to identify the need for parameter adjustment. The feedback mechanism dynamically adjusts operating parameters based on data changes to ensure that the equipment maintains optimal performance under different ocean conditions. Through continuous testing and adjustment, each parameter is precisely controlled to ensure maximum energy capture efficiency and obtain an optimized operating parameter solution.

[0134] S413: Apply the operation parameter optimization scheme to perform real-time monitoring using the formula:

[0135]

[0136] Calculate the energy capture efficiency Ey and generate a wave energy conversion enhancement report; where CE represents the captured energy value and PE represents the potential energy value of the area;

[0137] CE represents the energy value captured by the device. It is measured by the real-time monitoring system and the energy captured in a specific time is 1500 Joules. PE represents the potential energy value of the waves in the area predicted by the wave model and calculated as 1800 Joules. Substitute the values ​​into the formula for calculation:

[0138]

[0139] The results show that under the current device configuration and operation, the energy conversion efficiency is 83.33%. This efficiency level reflects the performance of the device in actual operation. Such a high conversion efficiency further proves the effectiveness of the operating parameter adjustment scheme, which means efficient energy conversion and optimization of subsequent operational adjustments.

[0140] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A wave energy conversion enhancement method based on spatiotemporal convolutional neural network, characterized in that: The following steps are involved: Receive wave height, wave speed and frequency data collected by buoys and satellites, perform time series analysis on the data, calculate wave energy intensity, analyze the directional distribution of waves based on the wave energy intensity calculation results, reveal the characteristics of sea wave activity, and generate a preliminary wave feature set; Using the preliminary wave feature set, applying a spatiotemporal convolutional neural network, hierarchical feature extraction is performed on the data, wave energy patterns are analyzed based on the feature extraction results, and the location of high-energy wave areas is determined and calibrated using thresholds to generate wave energy feature mapping results; According to the wave energy characteristic mapping results, the spatial distribution of high-energy wave areas is analyzed, the energy peaks are counted, and the energy capture potential of the equipment at different positions is calculated according to the energy peak statistics. A customized equipment deployment plan is designed for each high-energy wave area, and an equipment configuration optimization plan is output; According to the equipment configuration optimization scheme, the equipment position and operating parameters are adjusted to match the capture of high-energy waves, the equipment operating status and energy capture efficiency are monitored in real time, and a wave energy conversion enhancement report is generated.

2. The wave energy conversion enhancement method based on spatiotemporal convolutional neural network according to claim 1 is characterized in that: The calculation steps of the fluctuation energy intensity are: Receives wave height, wave speed and frequency data from buoys and satellites and performs data aggregation to produce a comprehensive data set; performing time series analysis on the comprehensive data set to analyze volatility patterns and periodicity, identify key volatility indicators, and generate time series analysis results; Using the time series analysis results, the formula is adopted: Calculate the wave energy intensity E and obtain the wave energy intensity result, where ρ represents the density of water and V represents the average value of the wave velocity.

3. The wave energy conversion enhancement method based on spatiotemporal convolutional neural network according to claim 2 is characterized in that: The steps for obtaining the preliminary wave feature set are: The fluctuation energy intensity results are used to perform fluctuation directional distribution analysis using the formula: E(θ)=∑(E·cos(θ-θ) i )) Calculate the total fluctuation energy value E(θ) in the target direction to generate the fluctuation directional distribution result, where E represents the fluctuation energy in the difference direction, θ is the fluctuation angle, and θ i is the angular difference in actual direction; According to the wave directional distribution results, the differences in wave characteristics between regions are identified, the behavior patterns of waves in different sea environments are revealed, and the results of sea wave activity characteristics are generated; Using the results of the characteristics of the sea area wave activity, data clustering and integration are performed to generate a preliminary wave feature set.

4. The wave energy conversion enhancement method based on spatiotemporal convolutional neural network according to claim 3 is characterized in that: The steps of extracting the hierarchical features are: Input the preliminary wave feature set into the spatiotemporal convolutional neural network, perform time and space dimension feature extraction, and obtain preliminary results of time and space feature analysis; Apply multi-layer convolution and pooling technology to the preliminary results of the temporal and spatial feature analysis, perform local temporal and spatial feature detection through the convolution layer, and use the pooling layer to reduce the feature dimension to obtain deep feature extraction results; The deep feature extraction result is input into the fully connected layer, a nonlinear activation function is used to capture complex nonlinear relationships, and parameters are optimized through network training to obtain hierarchical feature extraction results.

5. The wave energy conversion enhancement method based on spatiotemporal convolutional neural network according to claim 4 is characterized in that: The steps for obtaining the wave energy characteristic mapping result are: Based on the hierarchical feature extraction results, the formula is adopted: Calculate the energy density M of each wave event, evaluate the energy distribution in each area, and obtain the preliminary analysis results of wave energy distribution, where ρ is the density of water, g is the gravitational acceleration, and A is the wave height; An energy threshold is set, and the energy density of each wave event is evaluated in combination with the preliminary analysis result of the wave energy distribution. By comparing with the set threshold, the wave area exceeding the threshold is marked as a high energy area, and a high energy area list is generated; Based on the high-energy area list, the identified high-energy wave areas are geographically calibrated, and the locations are simultaneously marked on the ocean map to generate wave energy feature mapping results.

6. The wave energy conversion enhancement method based on spatiotemporal convolutional neural network according to claim 5 is characterized in that: The statistical steps of the energy peak are: Utilizing the wave energy characteristic mapping results, extracting energy data of all regions and establishing wave energy distribution data; Based on the wave energy distribution data, spatial analysis is performed to refine the energy level of each area and generate an actual energy distribution map; According to the actual energy distribution diagram, a statistical analysis of the energy peak is performed to determine the maximum energy value of each identified high energy area to obtain energy peak statistical results.

7. The wave energy conversion enhancement method based on spatiotemporal convolutional neural network according to claim 6 is characterized in that: The output steps of the device configuration optimization solution are: The energy peak statistical results are called to extract the energy peak data of the high energy wave area, using the formula: Calculate the energy capture potential value P of each area c , and obtain the energy capture potential data, where E maxi is the energy peak of region i, A i is the effective capture area of ​​region i, d i is the distance from the device to the target area, and n represents the number of areas; Using the energy capture potential data, combined with the topography, depth and environmental conditions of the high-energy wave area, design energy capture equipment configuration for each area, match the energy capture potential requirements, and establish a customized equipment deployment plan; The customized equipment deployment plan is verified and adjusted, and the energy capture capacity and cost efficiency of all configuration plans are analyzed by simulating equipment operation performance and cost data, and multi-index optimization is performed to obtain an equipment configuration optimization plan.

8. The wave energy conversion enhancement method based on spatiotemporal convolutional neural network according to claim 7 is characterized in that: The steps for obtaining the wave energy conversion enhancement report are: Based on the equipment configuration optimization plan, analyze the wave data, determine the optimal placement position of the equipment, and generate an equipment position adjustment plan; According to the equipment position adjustment plan, the equipment operating parameters are dynamically adjusted to match the wave dynamic conditions of the difference area, and the operating parameter optimization plan is obtained; Apply the operating parameter optimization scheme to perform real-time monitoring using the formula: Calculate the energy capture efficiency Ey and generate a wave energy conversion enhancement report; where CE represents the captured energy value and PE represents the potential energy value of the area.

Images