Flight Spatial Orientation Training Assistance System and Method
Through the flight space directional training assistance system, multi-dimensional meteorological models and real-time meteorological parameters are used to simulate flight missions and meteorological scenes, solving the problem that traditional training methods are difficult to simulate real scenes, and achieving improvement of pilot skills and enhanced response capabilities.
Patent Information
- Application Number
- CN202510122318.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2025-01-26
- Publication Date
- 2025-06-13
- Estimated Expiration
- 2045-01-26
AI Technical Summary
Traditional flight training methods are difficult to comprehensively and accurately simulate real scenes of flight missions and special flight missions at different stages, especially when simulating emergency landing, aerial refueling, low-altitude flight, formation flight, etc., the lack of a highly realistic training environment, resulting in the pilots' lack of sufficient response experience when actually performing missions.
It provides a flight space directional training assistance system, including flight mission simulation module, meteorological environment simulation module, meteorological environment parameter control module and training plan generation module. By establishing multi-dimensional meteorological models and combining real-time meteorological parameters, various meteorological scenarios are simulated, and personalized training plans are generated based on the historical training information of the flight personnel.
It has achieved a comprehensive, authentic and personalized simulated training environment for pilots, improved the professional skills of pilots and their ability to deal with various flight situations, and enhanced the authenticity and effectiveness of training.
Smart Images
Figure CN119580562B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of flight spatial orientation training. More specifically, the present invention relates to a flight spatial orientation training assistance system and method thereof. Background Art
[0002] With the rapid development of the aviation industry, flight safety has always been of utmost importance. Flight spatial orientation training is an important training to help pilots overcome spatial orientation obstacles and improve their understanding of themselves, the aircraft, and the flight environment during flight. The spiral ladder training is a common traditional method. By moving the body in a fixed roller to make the wheel rotate, it exercises the pilot's anti-vertigo ability, strength, courage, etc., and enhances their judgment ability of posture, position, and movement in space. However, there are safety hazards such as the risk of secondary injuries caused by the pilot letting go. Virtual reality technology training: Using this technology, various flight scenarios and complex environments can be simulated, such as simulating the low visibility environment of a space station, complex meteorological conditions of an aircraft, etc., providing pilots with a near-real visual experience of spatial movement, enhancing their immersion, and thus improving the training effect. Therefore, flight personnel not only need to master the operations of conventional flight tasks proficiently but also need to have the ability to handle various complex meteorological conditions and special flight situations. Traditional flight training methods have many limitations. It is often difficult to comprehensively and accurately simulate the real scenarios of different stages of flight tasks and special flight tasks. And when simulating special flight tasks such as emergency landing, in-air refueling, low-altitude flight, and formation flight, traditional training means are difficult to provide a highly realistic training environment, resulting in flight personnel lacking sufficient coping experience when actually performing tasks in the face of complex situations.
[0003] In summary, in order to improve the quality and effect of flight training and ensure flight safety, there is an urgent need for a flight spatial orientation training assistance system that can comprehensively simulate flight tasks and accurately simulate meteorological scenarios. Summary of the Invention
[0004] The present invention provides a flight spatial orientation training assistance system and method thereof, which can provide a comprehensive, real, and personalized simulation training environment for flight personnel, effectively improving the professional skills of flight personnel and their ability to handle various flight situations.
[0005] To achieve these and other advantages of the present invention, in a first aspect, the present invention provides a flight spatial orientation training assistance system, including:
[0006] A flight task simulation module, which is used to simulate flight tasks at different stages and special flight tasks;
[0007] A meteorological environment simulation module, which is connected to the flight mission simulation module and is used to simulate the meteorological scenario corresponding to the flight mission. The meteorological environment simulation module simulates the meteorological scenario by establishing meteorological models in multiple dimensions and combining the set meteorological parameters;
[0008] A meteorological environment parameter control module, which includes:
[0009] A time step control module, which inputs the meteorological parameter observation data collected in real time into the meteorological parameter setting module using a preset time step;
[0010] A specific meteorological event control module, which is used to transmit the meteorological parameters of the specific meteorological event area to the meteorological parameter setting module when the aircraft triggers a specific meteorological event, so that the meteorological environment simulation module updates the simulated meteorological scenario according to the meteorological parameters of the specific meteorological event area;
[0011] A training plan generation module, which generates a suitable training plan according to the historical training information of the flight crew. The training plan generation module is connected to the flight mission simulation module.
[0012] Preferably, it further includes an auxiliary function module, which provides the best operation steps and suggestions to the flight crew according to the current flight state and mission requirements of the flight crew.
[0013] Preferably, the specific meteorological event control module includes:
[0014] A meteorological event recognition module, which learns from historical meteorological data and flight data based on deep learning methods, can real-time identify the regional meteorological parameters where the aircraft is located, and judge whether a specific meteorological event is triggered;
[0015] A meteorological parameter calculation module. If a specific meteorological event is triggered, the meteorological parameter calculation module calculates the changing meteorological parameters for the specific meteorological event and inputs them into the meteorological parameter setting module.
[0016] Preferably, the time step control module specifically includes:
[0017] A judgment module, which is used to judge the degree of change of meteorological parameters. If the meteorological parameters change greatly, the time step is shortened; if the meteorological parameters change smoothly, the time step is increased;
[0018] A time step adjustment module, which adjusts the time step according to the judgment result of the judgment module.
[0019] Preferably, the judgment module specifically includes:
[0020] A normalization module, which is used to normalize multiple different meteorological parameter data;
[0021] A weight determination module that determines the weight of each meteorological parameter according to the influence degree of different meteorological parameters on meteorological changes;
[0022] A comprehensive change index calculation module that calculates the comprehensive change index value through the comprehensive change index formula, specifically:
[0023] , where is the comprehensive change index value, is the wind speed weight, is the wind speed change amount, is the temperature weight, is the temperature change amount, is the air pressure weight, is the air pressure change amount;
[0024] A comparison module. If the comprehensive change index value C is greater than a predetermined first threshold, it is determined that the meteorological parameter changes greatly. If the comprehensive change index value C is less than a predetermined second threshold, it is determined that the meteorological parameter changes gently, and the second threshold is less than the first threshold.
[0025] Preferably, the meteorological parameter setting module fuses the meteorological parameter observation data with the simulation results of the meteorological model, and continuously adjusts the initial conditions of the meteorological model and the meteorological parameters to make the simulation results of the meteorological model close to the real meteorological situation.
[0026] Preferably, the multiple dimensions in the multiple-dimensional meteorological model include:
[0027] The spatial dimension, which includes the horizontal direction and the vertical direction;
[0028] The time dimension, which includes: short-term changes and long-term changes;
[0029] The physical process dimension, which includes: dynamic processes and thermodynamic processes;
[0030] Among them, in the meteorological model, there are also:
[0031] A raindrop simulation model that is used to simulate the generation, movement trajectory and collision effect of raindrops, and through the numerical solution of the Navier-Stokes equation, simulates the subtle changes of raindrops in a complex airflow field;
[0032] A strong wind simulation model that is used to simulate the wind speed, wind direction at different heights and positions, as well as the spatial distribution and changes of strong winds, and uses the large eddy simulation method to simulate the large-scale vortex structure in the wind field to capture the turbulent characteristics in strong winds.
[0033] Preferably, the meteorological event recognition module specifically includes:
[0034] A historical meteorological data acquisition module, which is used to acquire parameter information under various meteorological conditions, including temperature, humidity, air pressure, wind speed, wind direction, and cloud information in different regions, different seasons, and different weather types;
[0035] A flight data acquisition module, which includes the flight trajectory, altitude, speed, and attitude information of the aircraft under different meteorological conditions;
[0036] A convolutional neural network model, which extracts and reduces the dimensionality of image features in meteorological data through convolutional layers and pooling layers, takes the labeled meteorological data as input, the convolutional neural network model outputs predictions of meteorological events, and adjusts the weights of the convolutional kernels through the backpropagation algorithm to accurately identify meteorological events;
[0037] A recurrent neural network processing model, which is used to process the time series information in flight data to obtain features related to meteorological events in the flight data time series, so as to accurately identify the meteorological environment where the aircraft is located.
[0038] In a second aspect, the present invention also provides a flight spatial orientation training assistance method, which is applicable to the flight spatial orientation training assistance system, and includes the following steps:
[0039] Generate a suitable training plan according to the historical training information of the flight crew;
[0040] Determine the flight phases and special flight tasks to be simulated according to the training requirements of the training plan, and construct a flight scenario;
[0041] Construct a meteorological model: construct a meteorological model with multiple dimensions, including spatial dimension, time dimension, and physical process dimension;
[0042] Through the meteorological parameter setting module, combined with the real-time collected meteorological parameter observation data, set initial parameters for the meteorological model;
[0043] Use the established meteorological model and set meteorological parameters to simulate various meteorological scenarios;
[0044] Transmit the real-time collected meteorological parameter observation data to the meteorological parameter setting module according to the set time step;
[0045] Continuously monitor the flight state and environmental information of the aircraft to determine whether a specific meteorological event is triggered;
[0046] When a specific meteorological event is detected, obtain the meteorological parameters in the event area and transmit them to the meteorological parameter setting module, and the meteorological environment simulation module updates the simulated meteorological scenario according to the specific meteorological parameters.
[0047] The present invention has at least the following beneficial effects:
[0048] First, it can simulate flight missions at different stages and special flight missions, providing diverse training support for flight personnel, offering comprehensive and highly targeted training scenarios, and helping to improve the ability of flight personnel to handle various flight situations.
[0049] Second, by establishing a multi-dimensional meteorological model and simulating meteorological scenarios in combination with meteorological parameters set, it can accurately simulate meteorological scenarios corresponding to flight missions. This enables flight personnel to encounter various meteorological conditions that may be encountered in actual flight during training, enhancing their familiarity and response ability to flight operations in different meteorological environments, and improving the authenticity and effectiveness of training.
[0050] Third, by using a preset time step to input real-time collected meteorological parameter observation data, it ensures the timeliness and accuracy of the data used by the meteorological environment simulation module, making it closer to the meteorological changes in actual flight, and further improving the authenticity of training. When the aircraft triggers a specific meteorological event, it can timely transmit the meteorological parameters of a specific area to the meteorological parameter setting module to achieve dynamic update of the meteorological scenario. This enables flight personnel to obtain a real-time adjusted meteorological environment when facing sudden meteorological events during training, enhancing their response training to emergencies and improving their emergency handling ability in actual flight.
[0051] Fourth, generating a suitable training plan based on the historical training information of flight personnel realizes personalized customization of the training plan. According to the actual level and training experience of flight personnel, more targeted training tasks are arranged, improving the training effect and promoting the improvement of the ability of flight personnel.
[0052] Other advantages, objectives, and features of the present invention will be partially reflected by the following description and partially understood by those skilled in the art through the research and practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS
[0053] Figure 1 It is a schematic diagram of the relationship structure of the flight spatial orientation training assistance system described in the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS
[0054] The following further describes the present invention in detail with reference to the drawings, so that those skilled in the art can implement it according to the description in the specification.
[0055] It should be understood that terms such as "having", "comprising", and "including" used herein do not exclude the presence or addition of one or more other elements or their combinations.
[0056] As Figure 1 shown, the present invention provides a flight spatial orientation training assistance system, including:
[0057] A flight mission simulation module, which is used to simulate flight missions at different stages and special flight missions;
[0058] A meteorological environment simulation module, which is connected to the flight mission simulation module and is used to simulate the meteorological scenario corresponding to the flight mission. The meteorological environment simulation module simulates the meteorological scenario by establishing meteorological models in multiple dimensions and combining the set meteorological parameters;
[0059] A meteorological environment parameter control module, which includes:
[0060] A time step control module, which inputs the meteorological parameter observation data collected in real time into the meteorological parameter setting module by using a preset time step;
[0061] A specific meteorological event control module, which is used to transmit the meteorological parameters of the specific meteorological event area to the meteorological parameter setting module when the aircraft triggers a specific meteorological event, so that the meteorological environment simulation module updates the simulated meteorological scenario according to the meteorological parameters of the specific meteorological event area;
[0062] A training plan generation module, which generates a suitable training plan according to the historical training information of the flight crew. The training plan generation module is connected to the flight mission simulation module.
[0063] In the above embodiment, the flight mission simulation module can simulate flight missions at different stages, such as takeoff stage simulation, cruise stage simulation, and landing stage simulation. Special flight missions, such as emergency situation simulation, for example, engine failure, bird strike event, emergency landing, in-air refueling, etc. The flight environment simulation mainly includes meteorological environment simulation, geographical environment simulation, and time environment simulation. The meteorological environment simulation module is used to simulate the meteorological scenario corresponding to the flight mission. The meteorological scenario, such as sunny, cloudy, overcast, heavy rain, thunderstorm, strong wind, icing conditions, etc. And the meteorological parameter setting module is used to set the parameter data when simulating the meteorological scenario, such as temperature, pressure, humidity, wind speed and direction, cloud amount and cloud height, etc. For the geographical environment simulation and time environment simulation, this embodiment does not make specific limitations. For example, the geographical environment simulation module is used to simulate different geographical environments, such as plain areas, mountainous areas, ocean areas, plateau areas, etc.; the time environment simulation module is used to simulate the time environment, such as day, night; or different seasonal times, such as spring, summer, autumn, and winter. This embodiment focuses on the improvement of the meteorological environment simulation.
[0064] In the above embodiments, the meteorological environment simulation module simulates meteorological scenarios by establishing meteorological models in multiple dimensions and combining the set meteorological parameters. The meteorological models in multiple dimensions include the spatial dimension, the time dimension, and the physical process dimension, realizing an all-round and three-dimensional simulation of meteorological scenarios. In the spatial dimension, the grid division in the horizontal direction and the layering in the vertical direction can accurately describe the distribution of meteorological parameters in different regions and at different heights. Exemplarily, a high-resolution grid can capture the urban heat island effect. In the time dimension, both short-term weather changes from several hours to several days and long-term meteorological trends from several years to several decades are simulated. It can predict short-term rainfall brought by the passage of a cold front and analyze the changes in the frequency and intensity of extreme weather events under global climate change. The physical process dimension comprehensively considers dynamics and thermodynamics and can accurately simulate the formation of wind, precipitation mechanisms, and the impact of atmospheric chemical reactions on meteorology. According to different flight missions, corresponding meteorological parameters are set. For example, in a flight mission in the tropical region, parameters related to high temperature, high humidity, and possible tropical cyclones are set to simulate the real meteorological environment in this region, enabling pilots to feel the actual situation of flying under specific meteorological conditions. The time step control module inputs the meteorological parameter observation data (such as satellite cloud images, ground meteorological station data, or upper-air sounding data, etc.) collected in real time using a preset time step into the meteorological parameter setting module to ensure that the meteorological model can obtain the latest data in a timely manner. When meteorological conditions change, such as the rapid movement of a cold front, the data is updated according to the set time step, making the simulated meteorological scenario fit the actual meteorological changes. Flight personnel can experience the dynamic evolution of meteorological conditions in real time during training, rather than a static simulation based on outdated data, enhancing the authenticity and timeliness of training. When the aircraft triggers a specific meteorological event, the specific meteorological event control module quickly transmits the meteorological parameters in the event area to the meteorological parameter setting module to update the meteorological scenario. Exemplarily, when the aircraft enters a cumulonimbus cloud area, meteorological parameters such as strong convection, lightning, and drastic temperature changes in the cumulonimbus cloud area are immediately captured to update the simulated scenario. This enables flight personnel to truly face and respond to sudden meteorological events during training, improving their emergency handling ability under complex meteorological conditions.
[0065] The meteorological environment simulation module is connected to the flight mission simulation module and can simulate corresponding meteorological scenarios according to different flight missions. Exemplarily, when simulating the training of a low-altitude flight mission in mountainous areas, meteorological conditions such as complex valley winds and terrain waves in mountainous areas can be simulated; for an over-ocean flight mission, variable meteorological conditions over the ocean, such as tropical cyclones and severe convective weather, can be simulated. This meteorological simulation combined with flight missions provides personalized training for pilots, enhances their adaptability to different flight environments and missions, and comprehensively improves flight skills and professional qualities.
[0066] In one specific embodiment, it further includes an auxiliary function module, which provides the best operation steps and suggestions to the flight crew according to the current flight status and mission requirements of the flight crew.
[0067] In the above embodiment, during the training process, the auxiliary function module can provide real-time operation guidance. For example, when the flight crew performs a certain operation, the auxiliary function module will give the best operation steps and suggestions according to the current flight status and mission requirements. In addition, for the wrong behaviors of the flight crew, it can also be corrected in time. Through voice prompts, pop-up warnings, etc., the flight crew is informed of the places where the operation is wrong and a correct operation demonstration is provided.
[0068] In one specific embodiment, the specific meteorological event control module includes:
[0069] A meteorological event recognition module, which learns from historical meteorological data and flight data based on deep learning methods, can real-time identify the regional meteorological parameters where the aircraft is located, and determine whether a specific meteorological event is triggered;
[0070] A meteorological parameter calculation module. If a specific meteorological event is triggered, the meteorological parameter calculation module calculates the changing meteorological parameters for the specific meteorological event and inputs them into the meteorological parameter setting module.
[0071] Specifically, the meteorological event recognition module specifically includes:
[0072] A historical meteorological data collection module, which is used to collect parameter information under various meteorological conditions, including temperature, humidity, air pressure, wind speed, wind direction, cloud information in different regions, different seasons, and different weather types;
[0073] A flight data collection module, which includes the flight trajectory, altitude, speed, and attitude information of the aircraft under different meteorological conditions;
[0074] A convolutional neural network model, which extracts and reduces the dimensionality of the image features in the meteorological data through convolutional layers and pooling layers. Using the labeled meteorological data as input, the convolutional neural network model outputs the prediction of the meteorological event and adjusts the weights of the convolutional kernels through the backpropagation algorithm to accurately identify the meteorological event;
[0075] A recurrent neural network processing model, which is used to process the time series information in the flight data to obtain the features related to the meteorological event in the flight data time series, so as to accurately identify the meteorological environment where the aircraft is located.
[0076] In the above embodiments, the core components of the convolutional neural network model (CNN) are the convolutional layer and the pooling layer. The convolutional layer performs convolutional operations on meteorological data (such as satellite cloud images) by designing convolutional kernels of different sizes and weights. Exemplarily, a 3×3 convolutional kernel scans the satellite cloud image pixel by pixel, multiplies the pixel values in the local area by the weights of the convolutional kernel and sums them to obtain a new feature value. This process can extract local features in the satellite cloud image, such as the edges and textures of clouds. Stacking multiple convolutional layers with different convolutional kernels can extract richer and more complex image features. The pooling layer is used to reduce the data dimension. For example, the point with the largest pixel value is selected within a 2×2 area, which not only retains the main features but also reduces the data volume and improves the computational efficiency. For the collected historical meteorological data, it must be processed before it can be used. Exemplarily, for the satellite cloud images in meteorological data, they are first converted into a format suitable for CNN processing, generally a three-dimensional tensor, which includes the height, width, and number of channels of the image. The labeled satellite cloud image data is divided into a training set, a validation set, and a test set according to a certain ratio. During the training process, the training set data is continuously input into the convolutional neural network model, and the convolutional layer and the pooling layer sequentially perform feature extraction and dimensionality reduction processing on the data. For example, when inputting a 100×100 pixel satellite cloud image, after being processed by several convolutional layers and pooling layers, the data may be compressed into a 10×10 feature map, which contains the most important feature information in the satellite cloud image. The labeled meteorological data (such as cloud images marked as cumulonimbus clouds, fair weather clouds, etc.) is used as input, and the convolutional neural network model outputs the prediction results of meteorological events. Through the backpropagation algorithm, the error between the prediction result and the true label is backpropagated to each layer of the network to adjust the weights and bias values of the convolutional kernels. For example, if the convolutional neural network model mispredicts a cumulonimbus cloud as a fair weather cloud, the backpropagation algorithm will calculate how to adjust the convolutional kernel weights so that a more accurate prediction can be made when encountering a similar cumulonimbus cloud image next time. After multiple iterative trainings, the convolutional neural network model gradually learns the feature patterns of different meteorological events in the satellite cloud image, so as to accurately identify meteorological events. The characteristic of the recurrent neural network processing model (RNN) is that there are recurrent connections between its hidden layers, which can process time series data. At each time step, it receives the input at the current moment and the output of the hidden layer at the previous moment, and calculates the output of the hidden layer at the current moment through a non-linear activation function, enabling the RNN to remember the previous information and thus process the time series information in the flight data. For flight data, such as the sequence data of the height, speed, attitude, etc. of an aircraft changing over time, it is sorted into a format suitable for RNN input in chronological order. It is also divided into a training set, a validation set, and a test set. During the training process, the training set data is input into the RNN model step by step according to the time steps.For example, the aircraft records flight data every minute within 10 minutes, and the data of these 10 time steps are sequentially input into the RNN. The input of each time step is calculated in combination with the output of the hidden layer at the previous moment. Similar to the CNN, it is trained by the backpropagation algorithm, but in the RNN, the backpropagation is carried out by unfolding in time steps, that is, starting from the error at the last time step, gradually backpropagating to each time step to adjust the weights and biases of the network in each time step. For example, during the training process, if the model makes a wrong prediction about the flight state of the aircraft when encountering a thunderstorm at a certain moment, the backpropagation algorithm will trace back along the time steps and adjust the parameters of the network in each time step, so that the RNN model can better handle the relationship between meteorological events and flight states in this time series. After a large number of trainings, the RNN can learn the features related to meteorological events in the flight data time series, so as to accurately identify the meteorological environment where the aircraft is located. First, use the CNN to extract image features from meteorological data such as satellite cloud images to obtain a feature vector, and then input this feature vector together with the time series of flight data into the RNN. In this way, the extraction ability of the CNN for image features and the processing ability of the RNN for time series data can be fully utilized to improve the accuracy of meteorological event recognition.
[0077] In one specific embodiment, the time step control module specifically includes:
[0078] A judgment module, which is used to judge the degree of change of meteorological parameters. If the meteorological parameters change greatly, the time step is shortened; if the meteorological parameters change gently, the time step is increased;
[0079] A time step adjustment module, which adjusts the time step according to the judgment result of the judgment module.
[0080] Specifically, the judgment module specifically includes:
[0081] A normalization module, which is used to normalize multiple different meteorological parameter data;
[0082] A weight determination module, which determines the weight of each meteorological parameter according to the influence degree of different meteorological parameters on meteorological changes;
[0083] A comprehensive change index calculation module, which calculates the comprehensive change index value through the comprehensive change index formula, specifically:
[0084] , where is the comprehensive change index value, is the wind speed weight, is the wind speed change amount, is the temperature weight, is the temperature change amount, is the air pressure weight, Air pressure change;
[0085] The comparison module determines that the meteorological parameter changes greatly if the comprehensive change index value C is greater than a predetermined first threshold value, and determines that the meteorological parameter changes slowly if the comprehensive change index value C is less than a predetermined second threshold value, and the second threshold value is less than the first threshold value.
[0086] In the above embodiment, the normalization module is used to normalize multiple different meteorological parameter data. For example, for temperature T, the minimum value is set to T min , the maximum value is T max , the normalized temperature is , so that all meteorological parameters are on the same scale. Different meteorological parameters have different degrees of influence on meteorological changes. For example, in severe convective weather, changes in wind speed and vertical wind shear may have a greater impact on flight safety than relative humidity. The weight of each meteorological parameter is determined according to the degree of influence of different meteorological parameters on meteorological changes. Since meteorological parameters do not exist in isolation, but affect each other. For example, rising temperature may lead to increased evaporation of water vapor, which in turn affects humidity and air pressure. Exemplarily, a machine learning algorithm can be used to learn the association pattern between parameters. For example, in a neural network model, through a large amount of data training, the model automatically learns the complex relationship between parameters such as temperature, humidity, and air pressure. According to the normalized parameter values and weights, a comprehensive change index is calculated. The larger the comprehensive change index, the more drastic the comprehensive change of the meteorological parameters. If the comprehensive change index value C is greater than the predetermined first threshold, it is determined that the meteorological parameter changes greatly. If the comprehensive change index value C is less than the predetermined second threshold, it is determined that the meteorological parameter changes slowly, and the second threshold is less than the first threshold. If the comprehensive change index value C is less than the predetermined first threshold value and greater than the predetermined second threshold value, the time step is not changed. The first threshold value and the second threshold value can be specifically limited according to the actual situation, and this embodiment does not make specific limitations. In the meteorological environment simulation, a smaller time step can be used, set at the millisecond level, to ensure that changes in meteorological parameters can be reflected in the meteorological scene in real time.
[0087] In one specific implementation manner, the meteorological parameter setting module integrates the meteorological parameter observation data with the simulation results of the meteorological model, and continuously adjusts the initial conditions and meteorological parameters of the meteorological model to make the simulation results of the meteorological model close to the actual meteorological conditions.
[0088] In one specific embodiment, the multiple dimensions of the multiple-dimensional meteorological model include:
[0089] spatial dimensions, which include horizontal and vertical directions;
[0090] The time dimension, which includes: short-term changes and long-term changes;
[0091] The physical process dimension, which includes: dynamic processes and thermodynamic processes;
[0092] Among them, in the meteorological model, there are also set:
[0093] A raindrop simulation model, which is used to simulate the generation, movement trajectory and collision effect of raindrops, and through the numerical solution of the Navier-Stokes equation, simulate the subtle changes of raindrops in a complex airflow field;
[0094] A strong wind simulation model, which is used to simulate the wind speed, wind direction at different heights and positions, as well as the spatial distribution and changes of strong winds, and uses the large eddy simulation method to simulate the large-scale vortex structure in the wind field to capture the turbulent characteristics in strong winds.
[0095] In the above embodiment, the raindrop simulation model can set the size distribution and density of raindrops according to different rainfall intensities. For example, in heavy rain, the raindrops are larger and the density is higher. At the same time, it also considers the influence of the falling speed of raindrops in the air by gravity, air resistance and wind force, and calculates the movement trajectory of raindrops in real time, so that it presents a real falling effect in the simulation scene. When simulating raindrops, in addition to considering gravity, air resistance and wind force, more complex physical models are introduced. Such as considering the influence of the surface tension and deformation of raindrops on their movement trajectory. And through the numerical solution of the Navier-Stokes equation, simulate the subtle changes of raindrops in a complex airflow field, making the simulation of raindrops more realistic. The strong wind simulation model can use the principles of fluid mechanics to establish a three-dimensional wind field model, simulate the spatial distribution and changes of strong winds by calculating the wind speed and wind direction at different heights and positions. And use the large eddy simulation method to simulate the large-scale vortex structure in the wind field to capture the turbulent characteristics in strong winds, thereby improving the simulation accuracy of the influence of the flight attitude of the aircraft in strong winds.
[0096] Another embodiment of the present invention also provides a flight space orientation training assistance method, which is applicable to the flight space orientation training assistance system described above, and includes the following steps:
[0097] S1. Generate a suitable training plan according to the historical training information of the flight crew.
[0098] Among them, analyze the historical training information of the flight crew, combine the skill level, training objectives and flight mission requirements of the flight crew, generate a personalized training plan, and clarify the training content, sequence, intensity and time arrangement, etc.
[0099] S2. Determine the flight phases and special flight missions to be simulated according to the training requirements of the training plan, and construct a flight scene.
[0100] Among them, according to the training requirements, clarify the flight phases to be simulated, such as takeoff, cruise, landing, etc., as well as special flight tasks, such as emergency landing, in-flight refueling, etc. According to the selected task type, construct the corresponding flight environment, including selecting a suitable airport, setting the flight route, and simulating different geographical environments (plain, mountainous area, ocean, etc.).
[0101] S3. Construct a meteorological model: Construct a meteorological model with multiple dimensions, including spatial dimension, time dimension, and physical process dimension.
[0102] S4. Through the meteorological parameter setting module, combine the meteorological parameter observation data collected in real time to set the initial parameters for the meteorological model.
[0103] S5. Utilize the established meteorological model and the set meteorological parameters to simulate various meteorological scenarios.
[0104] Among them, utilize the established meteorological model and the set parameters to simulate various meteorological scenarios, such as sunny day, rainy day, thunderstorm, icing, etc.
[0105] S6. According to the set time step, transmit the meteorological parameter observation data collected in real time to the meteorological parameter setting module.
[0106] Among them, preset a suitable time step to facilitate controlling the input frequency of the meteorological parameter observation data. According to the set time step, transmit the meteorological parameter observation data collected in real time to the meteorological parameter setting module to ensure that the meteorological model can obtain the latest data in a timely manner.
[0107] S7. Continuously monitor the flight state and environmental information of the aircraft to determine whether a specific meteorological event is triggered, such as entering a thunderstorm area, encountering strong wind shear, etc.
[0108] S8. When a specific meteorological event is detected, obtain the meteorological parameters of the event area and transmit them to the meteorological parameter setting module. The meteorological environment simulation module updates the simulated meteorological scenario according to the specific meteorological parameters to present a more realistic sudden meteorological situation for the pilot.
[0109] In the above embodiments, it is possible to provide a comprehensive, realistic and personalized simulation training environment for flight personnel, effectively improving the professional skills of flight personnel and their ability to handle various flight situations.
[0110] The number of devices and the processing scale described here are used to simplify the description of the present invention. The application, modification and variation of the present invention are obvious to those skilled in the art.
[0111] Although the embodiments of the present invention have been disclosed as above, they are not limited to the applications listed in the specification and embodiments. It can be fully applied to various fields suitable for the present invention. For those skilled in the art, additional modifications can be easily achieved. Therefore, without departing from the general concept defined by the claims and the scope of equivalents, the present invention is not limited to specific details and the examples shown and described herein.
Claims
1. Flight space orientation training auxiliary system, characterized in that: include: Flight mission simulation module, which is used to simulate flight missions at different stages and special flight missions; A meteorological environment simulation module, which is connected to the flight mission simulation module and is used to simulate the meteorological scene corresponding to the flight mission. The meteorological environment simulation module simulates the meteorological scene by establishing a meteorological model of multiple dimensions and combining the set meteorological parameters; Meteorological environment parameter control module, which includes: A time step control module, which uses a preset time step to input the meteorological parameter observation data collected in real time into the meteorological parameter setting module; A specific meteorological event control module, which is used to transmit the meteorological parameters of the specific meteorological event area to the meteorological parameter setting module when the aircraft triggers a specific meteorological event, so that the meteorological environment simulation module updates the simulated meteorological scene according to the meteorological parameters of the specific meteorological event area, wherein the specific meteorological events include: meteorological events in tropical areas with high temperature, high humidity and tropical cyclones, meteorological events in cumulonimbus areas with severe convection, lightning and drastic temperature changes, thunderstorm area events, and strong wind shear events; A training program generation module, which generates a suitable training program according to the historical training information of the flight personnel, and the training program generation module is connected to the flight mission simulation module; Wherein, the time step control module specifically includes: A judgment module is used to judge the degree of change of meteorological parameters. If the meteorological parameters change greatly, the time step is shortened; if the meteorological parameters change slowly, the time step is increased; A time step adjustment module, which adjusts the time step according to the judgment result of the judgment module; The judgment module specifically includes: A normalization module, which is used to normalize multiple different meteorological parameter data; A weight determination module determines the weight of each meteorological parameter according to the degree of influence of different meteorological parameters on meteorological changes; The comprehensive change index calculation module calculates the comprehensive change index value through the comprehensive change index formula, which is specifically: ,in, is the comprehensive change index value, is the wind speed weight, is the wind speed change, is the temperature weight, is the temperature change, is the air pressure weight, Air pressure change; The comparison module determines that the meteorological parameter changes greatly if the comprehensive change index value C is greater than a predetermined first threshold value, and determines that the meteorological parameter changes slowly if the comprehensive change index value C is less than a predetermined second threshold value, and the second threshold value is less than the first threshold value.
2. The flight space orientation training auxiliary system according to claim 1, characterized in that: It also includes an auxiliary function module, which provides the flight crew with the best operating procedures and suggestions based on the flight crew's current flight status and mission requirements.
3. The flight space orientation training auxiliary system according to claim 1, characterized in that: The specific meteorological event control module includes: The meteorological event recognition module uses deep learning methods to learn historical meteorological data and flight data, and can identify the meteorological parameters of the area where the aircraft is located in real time and determine whether a specific meteorological event is triggered; The meteorological parameter calculation module calculates the changing meteorological parameters for the specific meteorological event if a specific meteorological event is triggered, and inputs the calculated meteorological parameters into the meteorological parameter setting module.
4. The flight space orientation training auxiliary system according to claim 1, characterized in that: The meteorological parameter setting module integrates the meteorological parameter observation data with the simulation results of the meteorological model, and continuously adjusts the initial conditions and meteorological parameters of the meteorological model to make the simulation results of the meteorological model close to the actual meteorological situation.
5. The flight space orientation training auxiliary system according to claim 1, characterized in that: The multiple dimensions of the multi-dimensional meteorological model include: spatial dimensions, which include horizontal and vertical directions; The time dimension, which includes: short-term changes and long-term changes; The physical process dimension, which includes: dynamic process and thermodynamic process; Among them, in the meteorological model, it is also provided with: Raindrop simulation model, which is used to simulate the generation, movement trajectory and collision effect of raindrops, and simulate the subtle changes of raindrops in complex airflow fields through numerical solution of Navier-Stokes equations; The strong wind simulation model is used to simulate the spatial distribution and changes of wind speed, wind direction and strong wind at different heights and locations, and use the large eddy simulation method to simulate the large-scale vortex structure in the wind field to capture the turbulent characteristics in strong winds.
6. The flight space orientation training auxiliary system according to claim 3, characterized in that: The meteorological event identification module specifically includes: Historical meteorological data collection module, which is used to collect parameter information under various meteorological conditions, including temperature, humidity, air pressure, wind speed, wind direction, and cloud information in different regions, seasons, and weather types; Flight data acquisition module, which includes the flight trajectory, altitude, speed, and attitude information of the aircraft under different meteorological conditions; A convolutional neural network model extracts and reduces the dimensionality of image features in meteorological data through convolutional layers and pooling layers, takes the labeled meteorological data as input, outputs predictions of meteorological events, and adjusts the weights of the convolution kernels through a back-propagation algorithm to accurately identify meteorological events; A recurrent neural network processing model is used to process the time series information in the flight data to obtain the features related to meteorological events in the flight data time series, so as to accurately identify the meteorological environment in which the aircraft is located.
7. A flight space orientation training assistance method, applicable to the flight space orientation training assistance system according to any one of claims 1 to 6, characterized in that: The following steps are involved: Generate appropriate training plans based on the flight crew's historical training information; According to the training requirements of the training program, determine the flight phases and special flight tasks to be simulated and construct flight scenarios; Build meteorological models: Build meteorological models in multiple dimensions, including spatial dimension, temporal dimension, and physical process dimension; By using the meteorological parameter setting module, combined with the meteorological parameter observation data collected in real time, initial parameters are set for the meteorological model; Use the established meteorological models and set meteorological parameters to simulate various meteorological scenarios; According to the set time step, the meteorological parameter observation data collected in real time is transmitted to the meteorological parameter setting module; Continuously monitor the aircraft's flight status and environmental information to determine whether specific meteorological events are triggered; When a specific meteorological event is detected, the meteorological parameters of the event area are obtained and transmitted to the meteorological parameter setting module, and the meteorological environment simulation module updates the simulated meteorological scene according to the specific meteorological parameters.
Citation Information
Patent Citations
Simulator system for simulating weather
US9583020B1