A wind farm prediction system based on multi-scale graph convolution and reinforcement learning

Abstract

The application discloses a wind field prediction system based on multi-scale graph convolution and reinforcement learning, which utilizes a multi-scale graph convolution network for dynamic modeling to comprehensively capture the spatio-temporal characteristics of the wind field. Through a reinforcement learning module, the model can optimize strategies in real time in a complex and dynamically changing wind field environment to improve the path planning and obstacle avoidance capabilities of the aircraft. The reinforcement learning algorithm dynamically adjusts the strategy based on the cumulative reward mechanism, enabling the agent to quickly adapt to changes in the wind field and ensuring flight safety. The application aims to improve the accuracy and real-time performance of wind field prediction by combining graph convolution neural networks and reinforcement learning optimization strategies in response to the dynamic changes in the wind field in complex meteorological environments. Through multi-scale graph convolution network modeling and analysis of the spatio-temporal characteristics of the wind field, and through reinforcement learning optimization of the prediction model, the system can provide more accurate and real-time wind field data support to ensure the flight safety and path optimization of the aircraft in complex meteorological conditions.

Description

Technical Field

[0001] The present invention belongs to the field of aircraft navigation technology, and in particular relates to a wind field prediction system based on multi-scale graph convolution and reinforcement learning. Background Art

[0002] Aircraft trajectory planning and flight safety assessment are closely linked to accurate wind field forecasting. Factors such as wind speed, direction, and turbulence directly impact an aircraft's flight stability, range accuracy, and fuel consumption. Sudden and extreme wind field fluctuations, particularly in adverse weather conditions, pose challenges to flight safety. Traditional wind field forecasting methods, such as numerical weather prediction (NWP) models, while capable of providing certain wind speed and direction forecasts, cannot meet the dynamic response requirements of aircraft during flight due to their limitations in small-scale and real-time updates.

[0003] Existing wind forecasting technologies primarily rely on static numerical forecast models, which typically operate at fixed temporal and spatial resolutions. These models struggle to capture rapidly changing local wind fields, particularly when predicting short-term, high-frequency variations in wind speed and direction. Furthermore, the time delay inherent in traditional models limits their ability to update in real time, preventing aircraft from rapidly adjusting their flight paths based on real-time wind data. Particularly in complex meteorological conditions, the predictive capabilities of traditional methods are insufficient to meet the demands of a dynamic environment.

[0004] Graph Convolutional Neural Networks (GCNs) have become a powerful tool for wind forecasting due to their ability to effectively process spatiotemporal graph data, capturing spatial dependencies between nodes and time series changes. Unlike traditional convolutional neural networks, GCNs are better able to handle the spatial and temporal complexity of wind data, particularly by establishing effective connections between multi-level data and data from aircraft at different altitudes. Through multi-scale modeling, GCNs can capture both global and local variations in wind fields, improving forecast accuracy and adaptability.

[0005] At the same time, reinforcement learning (RL), a technology capable of adaptive optimization in dynamic environments, can further enhance the real-time and robustness of wind forecasting. By optimizing decisions through interaction with the environment, RL can adjust forecasting strategies based on real-time feedback, enabling wind forecasting to dynamically respond to meteorological changes. Wind forecasting methods that combine graph convolutional networks and reinforcement learning can effectively address the shortcomings of traditional methods, providing highly accurate and real-time updated wind data support, significantly improving aircraft safety and the accuracy of range forecasts. Summary of the Invention

[0006] In view of this, the present application aims to overcome the above-mentioned problems in the prior art, and proposes a wind field prediction system based on multi-scale graph convolution and reinforcement learning, which aims to overcome the problems of insufficient precision, real-time performance and adaptability in existing wind field prediction technology, and provides efficient and accurate wind field prediction data support for aircraft.

[0007] To achieve the above purpose, the technical scheme of the present application is as follows:

[0008] A wind field prediction system based on multi-scale graph convolution and reinforcement learning, comprising a data input module, a graph convolution module, a reinforcement learning strategy optimization module and a cumulative reward calculation and result output module connected in sequence;

[0009] The data input module obtains aircraft communication addressing and reporting system data from the ground station and wind speed, wind direction, temperature and air pressure data obtained from the global forecast system, and integrates key data of the environmental wind field based on the space-time position of the aircraft;

[0010] The graph convolution module models the input wind field data in a multi-scale graph structure, extracts the spatial and space-time features of the wind field through a graph convolution network, and generates accurate wind field prediction results;

[0011] The reinforcement learning strategy optimization module uses the space-time features generated by the graph convolution module to optimize the prediction accuracy of the wind field prediction model through a reinforcement learning algorithm;

[0012] The cumulative reward calculation and result output module calculates the wind field prediction error according to the cumulative reward, outputs the optimized wind field prediction result, improves the accuracy and stability of the wind field prediction, and ensures the safety and flight efficiency of the aircraft in complex weather environments.

[0013] Further, the data input module specifically comprises:

[0014] A ground station data acquisition unit for real-time reception and decoding of ACARS data, capturing the position information of the aircraft at a specific time point, providing a spatial basis for subsequent space-time graph modeling;

[0015] A GFS data acquisition unit for real-time access to global forecast system data to collect wind speed, wind direction, temperature and air pressure data at a matched space-time resolution, combine them with the aircraft position information to generate space-time input data graphs, and normalize the processing to adapt to the input requirements of the graph convolution network.

[0016] Further, the graph convolution module specifically comprises:

[0017] Adjacency Matrix Construction Unit: Generates a graph-structured adjacency matrix of the wind farm based on the geographic location and time points of spatiotemporal data. It defines the weights between nodes based on the similarity of location and wind farm parameters, accurately capturing the spatial distribution and variation characteristics of the wind farm.

[0018] Multi-scale graph convolution unit: extracts wind field features at local and global scales through multi-scale graph convolution operations, adapts to the changing patterns of different scales in the wind field, and thus obtains multi-level wind field information;

[0019] Activation unit: Improves the feature representation capability after graph convolution through nonlinear activation functions, enhances the model's expression effect on complex wind field data, and better reflects the impact of high-variability areas in the wind field.

[0020] Furthermore, the reinforcement learning strategy optimization module includes:

[0021] State Estimation Unit: This unit uses the spatiotemporal graph embedding features generated by the graph convolution module as the current wind farm state input and ensures the accuracy of state estimation by selecting a state representation that adapts to the dynamic changes of the wind farm.

[0022] Policy Optimization Unit: Apply the policy gradient and Q-learning algorithms in reinforcement learning, use the current state and accumulated rewards to update the strategy, and further optimize the prediction accuracy of the wind farm prediction model.

[0023] Furthermore, the cumulative reward calculation and result output module includes:

[0024] Cumulative Reward Calculation Unit: This unit calculates cumulative rewards based on the model's prediction results, actual wind field measurement data, and flight safety parameters, covering both prediction accuracy and wind field stability, thereby providing effective feedback and adjustments to the wind field prediction strategy.

[0025] The result output unit generates and outputs optimized wind field prediction results, ensuring that the flight control system can obtain accurate wind field data in a timely manner, thereby improving the accuracy of navigation decisions and flight safety.

[0026] Furthermore, the implementation process of the wind field prediction system includes:

[0027] Step 1: The data input module collects and processes flight data and wind field data from ACARS and GFS to generate spatiotemporal map data input;

[0028] Step 2: Use the graph convolution module to process the input wind field map data, extract the multi-scale features of the wind field and generate spatiotemporal feature embedding;

[0029] Step 3: Dynamically adjust the wind farm prediction strategy based on the cumulative reward value through the reinforcement learning strategy optimization module to improve the accuracy of wind farm prediction;

[0030] Step 4, calculate and output the optimized wind field prediction results in the cumulative reward calculation and result output module, and provide real-time wind field prediction data support for the flight control system.

[0031] Compared with the prior art, the wind field prediction system based on multi-scale graph convolution and reinforcement learning has the following advantages:

[0032] The application combines graph convolution network with reinforcement learning technology, and innovatively proposes the overall architecture of multi-scale graph convolution wind field modeling and reinforcement learning strategy optimization, which has the following advantages:

[0033] High-precision wind field modeling: the wind field modeling method based on multi-scale graph convolution can capture spatial and temporal features of different scales, which significantly improves the wind field prediction accuracy;

[0034] Real-time path optimization: through the reinforcement learning algorithm, the strategy is optimized based on the graph convolution wind field features, and real-time and dynamic path planning scheme is provided for the aircraft;

[0035] Adaptability and robustness: the strategy optimization mechanism based on cumulative reward can actively adjust the path planning decision when the wind field changes, improving the safety and stability of the aircraft in complex wind field environment;

[0036] Aviation safety guarantee: the application can provide real-time wind field features and optimized path information, helping the aircraft to avoid high-risk areas in severe weather conditions, and effectively improving the overall safety of civil aviation flight.

[0037] The application is not only suitable for path optimization of civil aircraft, but also can be popularized to navigation and obstacle avoidance application scenarios of unmanned aerial vehicles and other aircraft, providing an effective innovative solution for flight safety technology. BRIEF DESCRIPTION OF DRAWINGS

[0038] The accompanying drawings, which form a part of this application, are included to provide a further understanding of the application, and are incorporated in and constitute a part of this application. The schematic embodiments of the application and their descriptions are used to explain the application, and do not constitute an improper limitation on the application. In the drawings:

[0039] Figure 1 A structure diagram of a wind field prediction system based on multi-scale graph convolution and reinforcement learning of the application. DETAILED DESCRIPTION

[0040] It should be noted that the embodiments in the application and the features in the embodiments can be combined with each other without conflict.

[0041] In the description of the present application, it needs to be understood that the orientation or positional relationship indicated by the terms "center", "longitudinal", "transverse", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like is based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present application and simplifying the description, and does not indicate or imply that the device or element indicated must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present application. In addition, the terms "first", "second" and the like are only for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the technical features indicated. Therefore, the features defined with "first", "second" and the like can be explicitly or implicitly included one or more. In the description of the present application, unless otherwise specified, the meaning of "a plurality of" is two or more.

[0042] In the description of the present application, it needs to be understood that unless otherwise explicitly specified and limited, the terms "mounting", "connection", "connection" should be understood broadly, for example, it can be fixedly connected, or it can be detachably connected, or integrally connected; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium, and it can be the communication inside two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood through specific circumstances.

[0043] The present application will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

[0044] As Figure 1 shown, the present application provides a wind field prediction system based on multi-scale graph convolution and reinforcement learning, comprising a data input module, a graph convolution module, a reinforcement learning strategy optimization module and a cumulative reward calculation and result output module connected in sequence;

[0045] The data input module obtains aircraft communication addressing and reporting system data from the ground station and wind speed, wind direction, temperature and pressure data obtained from the global forecast system, and integrates key data of the environmental wind field based on the space-time position of the aircraft;

[0046] The graph convolution module models the input wind field data in a multi-scale graph structure, extracts the spatial and space-time features of the wind field through a graph convolution network, and generates an accurate wind field prediction result;

[0047] The reinforcement learning strategy optimization module uses the space-time features generated by the graph convolution module to optimize the prediction accuracy of the wind field prediction model through a reinforcement learning algorithm;

[0048] The cumulative reward calculation and result output module calculates the wind field prediction error according to the cumulative reward and outputs the optimized wind field prediction result, thereby improving the accuracy and stability of the wind field prediction and ensuring the safety and flight efficiency of the aircraft in complex meteorological environments.

[0049] Specifically, the data input module specifically includes:

[0050] The ground station data acquisition unit is used to receive and decode ACARS data in real time, capture the aircraft's position information at a specific point in time, and provide a spatial basis for subsequent space-time graph modeling;

[0051] The GFS data acquisition unit is used to access global forecast system data in real time, collect wind speed, wind direction, temperature and air pressure data with matching spatiotemporal resolution, combine it with aircraft position information to generate a spatiotemporal input data graph, and normalize it to meet the input requirements of the graph convolutional network.

[0052] Specifically, the graph convolution module includes:

[0053] Adjacency Matrix Construction Unit: Generates a graph-structured adjacency matrix of the wind farm based on the geographic location and time points of spatiotemporal data. It defines the weights between nodes based on the similarity of location and wind farm parameters, accurately capturing the spatial distribution and variation characteristics of the wind farm.

[0054] Multi-scale graph convolution unit: extracts wind field features at local and global scales through multi-scale graph convolution operations, adapts to the changing patterns of different scales in the wind field, and thus obtains multi-level wind field information;

[0055] Activation unit: Improves the feature representation capability after graph convolution through nonlinear activation functions, enhances the model's expression effect on complex wind field data, and better reflects the impact of high-variability areas in the wind field.

[0056] Specifically, the reinforcement learning strategy optimization module includes:

[0057] State Estimation Unit: This unit uses the spatiotemporal graph embedding features generated by the graph convolution module as the current wind farm state input and ensures the accuracy of state estimation by selecting a state representation that adapts to the dynamic changes of the wind farm.

[0058] Policy Optimization Unit: Apply the policy gradient and Q-learning algorithms in reinforcement learning, use the current state and accumulated rewards to update the strategy, and further optimize the prediction accuracy of the wind farm prediction model.

[0059] Specifically, the cumulative reward calculation and result output module includes:

[0060] The cumulative reward calculation unit calculates the cumulative reward based on the prediction result of the model, the actual wind field measurement data and the flight safety parameter, covering the prediction accuracy and the wind field stability, so as to effectively feedback and adjust the wind field prediction strategy;

[0061] The result output unit generates and outputs the optimized wind field prediction result, ensures that the flight control system can obtain accurate wind field data in time, and improves the accuracy of navigation decision and flight safety.

[0062] The implementation process of the wind field prediction system of the present application comprises:

[0063] Step 1, collecting and processing flight data and wind field data from ACARS and GFS through the data input module to generate spatiotemporal graph data input;

[0064] Step 1.1: decoding the ACARS data through the ARINC 618 protocol to obtain the data of flight height, speed, heading, and surrounding wind speed, temperature and air pressure, and obtaining the data of wind speed, wind direction, temperature and air pressure at the corresponding time from the GFS meteorological database;

[0065] Step 1.2: the values deviating from the normal range in the data collected in step 1 are detected by using the anomaly value detection method of adversarial training to generate adversarial samples to enhance the robustness of the model to abnormal data. This method is especially suitable for meteorological data such as wind speed data with strong time series and spatiality. By generating and identifying adversarial samples through adversarial training, the model can learn more detailed features of abnormal data, thereby improving the effect of anomaly value detection in wind speed prediction;

[0066] Step 1.3: for the repair of missing parts and abnormal values in the data collected in steps 1 and 1.1, linear interpolation is used for filling, and the missing values are linearly calculated according to the values of the previous and next two time points, is the estimation of the missing value, t i is the time of the missing value, y i+1 and y i-1 are the values of the previous and next time points, and the formula is:

[0067]

[0068] Step 1.4: standardizing the data processed in step 1.3, and taking the processed data as the data set for model training;

[0069] Step 1.5: constructing the wind field prediction training set and test set from the obtained data, inputting the test set into the graph convolution reinforcement learning model, and taking the training set as D train , the test set as D text , and each node x i(t) a wind field feature vector containing wind speed, wind direction, temperature, air pressure, the feature vector of each node is x i (t) = [v i (t), θ i (t), T i (t), p i (t)] T , v i (t) represents the wind speed of node i at time t, θ i (t) represents the wind direction of node i at time t, T i (t) represents the temperature of node i at time t, and p i (t) represents the air pressure of node i at time t.

[0070] For time t, the node feature matrix of the graph is represented as X t :

[0071]

[0072] where N represents the number of nodes, represents the position in the wind field, D represents the feature dimension, and represents the number of observation features contained by each node.

[0073] Step 2, use the graph convolution module to process the input wind field graph data, extract the multi-scale features of the wind field and generate the spatio-temporal feature embedding;

[0074] Step 2.1: The spatial structure of the wind field depends on the adjacency matrix A t , which reflects the spatial connectivity between the measurement points in the wind field and determines which nodes have information propagation. In the spatio-temporal graph data, each element A ij , of the adjacency matrix represents the connection strength between nodes i and j, and the similarity of node features is calculated to quantify the correlation between nodes to construct A t :

[0075]

[0076] In this way, the adjacency matrix A t can reflect the similarity between nodes based on input data such as wind speed, wind direction, temperature and air pressure.

[0077] Step 2.2: In order to prevent the node degree from having too great an impact on feature propagation, the invention uses a standardized adjacency matrix to normalize the adjacency matrix. The degree matrix D t The formula is:

[0078]

[0079] element a ijRepresents the relationship weight between nodes i and j. This weight represents the spatial correlation or similarity between wind field nodes at different measurement points. If the wind speed, wind direction, temperature, air pressure and other data of nodes i and j are highly similar or affect each other's wind field characteristics, then a ij The value will be higher.

[0080] Step 2.3: Perform normalization to obtain a standardized adjacency matrix

[0081]

[0082] Step 2.4: Aggregate the features of each node through the adjacency matrix A t And the normalized adjacency matrix Get the feature matrix H t : The feature matrix of the kth layer is

[0083]

[0084] is the output feature matrix of the k-1th layer, initially k = 0, W (k) is the weight matrix of the kth layer, which is used to perform linear transformation on the features, and σ is the ReLU activation function, which is used to capture the nonlinear characteristic changes of the wind field.

[0085] Step 3: Dynamically adjust the wind farm prediction strategy based on the cumulative reward value through the reinforcement learning strategy optimization module to improve the accuracy of wind farm prediction; the goal of the reinforcement learning strategy optimization module is to continuously optimize the strategy π(s t ,a t |θ) to maximize the accuracy of wind field prediction. Specifically including:

[0086] Step 3.1: State space s t Represents the wind field characteristics at the current time t, which includes the wind speed, wind direction, temperature and air pressure of the node. This information is extracted and encoded into a state vector through the graph convolutional network. The state expression of the wind field is:

[0087]

[0088] Step 3.2: Policy function π(s t ,a t |θ) is learned through neural network, according to the current state of the wind field s t Output action probability distribution:

[0089]

[0090] Where Q(st ,a t ; θ) is the action value function, which means taking action a under the current wind field state t The corresponding value, Q(s t ,a′;θ) represents the current state s t The expected cumulative reward that can be obtained by taking an action a′.

[0091] Step 3.3: In order to evaluate the contribution of a specific action to the cumulative reward, we introduce the advantage function A(s t ,a t ), which is expressed as:

[0092] A(s t ,a t )=Q(s t ,a t )-V(s t );

[0093] Among them, Q(s t ,a t ) means in state s t Select action a t The expected reward after , thus helping the model to identify the optimal decision in a specific wind farm scenario, the value function V(s t ) is the estimated wind field state s t The expected cumulative reward of is expressed as:

[0094]

[0095] E π [] represents the expectation of a random variable under a given policy π. k Represents the discount weight multiplied by the reward after time step k. t+k Represents the immediate reward obtained at time step t+k.

[0096] Step 3.4: To further enhance the ability of policy optimization, we combine the Q-learning method with the value function. Q-learning calculates the value Q(s) of each state-action pair. t ,a t ) to evaluate the effect of taking a certain action. The update formula of the Q-function is:

[0097]

[0098] where r t+1 is the reward at time t+1, γ is the discount factor, and this update mechanism enables the model to gradually optimize the wind field forecast. α is the learning rate, which controls the update speed of new information.

[0099] Step 3.5: Update the policy by the policy gradient method, so that the selected action a t At the current state s t , the expected reward is maximized:

[0100]

[0101] E π [] represents the expectation of a random variable under a given policy π.

[0102] The update formula for θ is:

[0103]

[0104] where η is the learning rate of the policy, and the policy parameters are updated to maximize the cumulative reward in the wind field changes, improving the accuracy of wind field prediction.

[0105] Step 3.6: Use the cumulative reward J(θ) as the evaluation index to reflect the effectiveness of the model's strategy under complex wind field conditions: the expression of J(θ) is:

[0106]

[0107] By maximizing the objective function J(θ), we can find the optimal policy π(s t , a t |θ), thereby improving the accuracy of wind field prediction.

[0108] Step 4: Calculate and output the optimized wind field prediction results in the cumulative reward calculation and result output module, providing real-time wind field prediction data support for the flight control system.

[0109] Step 4.1: After the graph convolution operation and reinforcement learning strategy optimization, the final model outputs the prediction results of the wind field The prediction value is output by the graph convolution network H L , which is mapped through the fully connected layer, and the prediction result is expressed as:

[0110]

[0111] W f is the weight matrix of the fully connected layer, responsible for mapping the output of the graph convolution network to the prediction result space, and b f is the bias term. The model achieves high-precision prediction of dynamic wind field changes by continuously optimizing the reinforcement learning strategy.

[0112] Step 4.2: The loss function uses Mean Squared Error (MSE) to measure the gap between the predicted value and the true value, improving the accuracy of prediction.

[0113]

[0114] Where T is the total number of time steps, is the predicted output of the model at time step t, Y t is the corresponding real wind field state.

[0115] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A wind field prediction system based on multi-scale graph convolution and reinforcement learning, characterized by: It includes a data input module, a graph convolution module, a reinforcement learning strategy optimization module, and a cumulative reward calculation and result output module connected in sequence; The data input module obtains aircraft communication addressing and reporting system data from the ground station and wind speed, wind direction, temperature and air pressure data from the global forecast system, and integrates key data of the environmental wind field based on the aircraft's spatiotemporal position; The graph convolution module performs multi-scale graph structure modeling on the input wind field data, extracts the spatial and spatiotemporal features of the wind field through the graph convolution network, and generates accurate wind field prediction results; The reinforcement learning strategy optimization module uses the spatiotemporal features generated by the graph convolution module to optimize the prediction accuracy of the wind field prediction model through a reinforcement learning algorithm; The cumulative reward calculation and result output module calculates the wind field prediction error according to the cumulative reward and outputs the optimized wind field prediction result, thereby improving the accuracy and stability of the wind field prediction and ensuring the safety and flight efficiency of the aircraft in complex meteorological environments.

2. The wind field prediction system based on multi-scale graph convolution and reinforcement learning according to claim 1, characterized in that: The data input module specifically includes: The ground station data acquisition unit is used to receive and decode ACARS data in real time, capture the aircraft's position information at a specific point in time, and provide a spatial basis for subsequent space-time graph modeling; The GFS data acquisition unit is used to access global forecast system data in real time, collect wind speed, wind direction, temperature and air pressure data with matching spatiotemporal resolution, combine it with aircraft position information to generate a spatiotemporal input data graph, and normalize it to meet the input requirements of the graph convolutional network.

3. The wind field prediction system based on multi-scale graph convolution and reinforcement learning according to claim 1, characterized in that: The graph convolution module specifically includes: Adjacency Matrix Construction Unit: Generates a graph-structured adjacency matrix of the wind farm based on the geographic location and time points of spatiotemporal data. It defines the weights between nodes based on the similarity of location and wind farm parameters, accurately capturing the spatial distribution and variation characteristics of the wind farm. Multi-scale graph convolution unit: extracts wind field features at local and global scales through multi-scale graph convolution operations, adapts to the changing patterns of different scales in the wind field, and thus obtains multi-level wind field information; Activation unit: Improves the feature representation capability after graph convolution through nonlinear activation functions, enhances the model's expression effect on complex wind field data, and better reflects the impact of high-variability areas in the wind field.

4. The wind field prediction system based on multi-scale graph convolution and reinforcement learning according to claim 1, characterized in that: The reinforcement learning strategy optimization module includes: State Estimation Unit: This unit uses the spatiotemporal graph embedding features generated by the graph convolution module as the current wind farm state input and ensures the accuracy of state estimation by selecting a state representation that adapts to the dynamic changes of the wind farm. Policy Optimization Unit: Apply the policy gradient and Q-learning algorithms in reinforcement learning, use the current state and accumulated rewards to update the strategy, and further optimize the prediction accuracy of the wind farm prediction model.

5. The wind field prediction system based on multi-scale graph convolution and reinforcement learning according to claim 1, characterized in that: The cumulative reward calculation and result output module includes: Cumulative Reward Calculation Unit: This unit calculates cumulative rewards based on the model's prediction results, actual wind field measurement data, and flight safety parameters, covering both prediction accuracy and wind field stability, thereby providing effective feedback and adjustments to the wind field prediction strategy. The result output unit generates and outputs optimized wind field prediction results, ensuring that the flight control system can obtain accurate wind field data in a timely manner, improving the accuracy of navigation decisions and flight safety.

6. The wind field prediction system based on multi-scale graph convolution and reinforcement learning according to claim 1, characterized in that: The implementation process of the wind field prediction system includes: Step 1: The data input module collects and processes flight data and wind field data from ACARS and GFS to generate spatiotemporal map data input; Step 2: Use the graph convolution module to process the input wind field map data, extract the multi-scale features of the wind field and generate spatiotemporal feature embedding; Step 3: Dynamically adjust the wind farm prediction strategy based on the cumulative reward value through the reinforcement learning strategy optimization module to improve the accuracy of wind farm prediction; Step 4: Calculate and output the optimized wind field prediction results in the cumulative reward calculation and result output module to provide real-time wind field prediction data support for the flight control system.

Images