Method for predicting aircraft surface environment parameters

Abstract

The invention discloses an aircraft surface environmental parameter prediction method, and aims to improve flight safety and efficiency by monitoring and predicting aircraft surface environmental parameters in real time. The method comprises the following steps: acquiring environmental parameter data, such as temperature, humidity, air pressure and wind speed, near the surface of the aircraft in real time by using a sensor; the collected data is transmitted to a data processing module for preprocessing, including denoising and normalization processing, so that the data quality is improved; a prediction model based on a machine learning algorithm is constructed, the preprocessed data is used for training, and an environment parameter prediction model is generated; inputting environment parameters collected in real time into the prediction model, and calculating and predicting future environment parameters on the surface of the aircraft; and feeding back a prediction result to a flight control system so as to adjust a flight strategy in real time. By fusing multi-source data and an advanced machine learning technology, multi-dimensional environment parameters are accurately predicted, and more reliable decision support is provided for flight control of an aircraft.

Description

Technical Field

[0001] The present invention relates to the field of aircraft, and in particular to a method for predicting aircraft surface environment parameters. Background Art

[0002] In the field of modern aviation, the performance and flight safety of aircraft are affected by a variety of environmental factors, including temperature, humidity, air pressure, wind speed, etc. Changes in these environmental parameters directly affect the aerodynamic characteristics, power system and navigation accuracy of the aircraft, so real-time monitoring and prediction are particularly important.

[0003] At present, traditional environmental parameter prediction methods mostly rely on static meteorological models and historical meteorological data, and lack the full use of real-time data. These methods often cannot accurately reflect the environmental changes encountered by aircraft during actual flight, resulting in the flight control system not being able to respond in a timely manner when dealing with sudden climate events, which may affect flight safety.

[0004] In recent years, with the rapid development of sensor technology and data processing capabilities, prediction methods that combine real-time environmental data with machine learning algorithms have gradually attracted attention. These methods can more accurately predict environmental parameters, optimize flight paths, and improve flight safety and efficiency. However, the current related technologies still have the following deficiencies: Insufficient data fusion capabilities: Existing environmental parameter prediction systems usually lack an effective multi-source data fusion mechanism, and it is difficult to fully integrate historical data, real-time sensor data, and meteorological model data. Insufficient model generalization capabilities: Most existing machine learning models have limited generalization capabilities under different environmental conditions and are difficult to adapt to complex and changing flight environments. Insufficient real-time performance: Although some methods can provide predictions of environmental parameters, their real-time performance still needs to be improved and cannot meet the real-time requirements of flight control systems.

[0005] Therefore, there is an urgent need for a new environmental parameter prediction method that can combine multi-source data to improve the prediction accuracy and generalization ability of the model, thereby providing more reliable support for the safe and efficient operation of aircraft. Summary of the invention

[0006] The present invention provides a method for predicting aircraft surface environment parameters, which includes:

[0007] Use sensors to collect real-time data on environmental parameters near the aircraft surface, including temperature, humidity, air pressure, wind speed, solar radiation intensity, pollutant concentration, aerosol concentration, sound level, and other micro-meteorological parameters;

[0008] Transmitting the environmental parameter data to a data processing module for preprocessing, including denoising and normalization processing;

[0009] Build a prediction model based on machine learning algorithm, use preprocessed data for training, and generate an environmental parameter prediction model;

[0010] Inputting the environmental parameters collected in real time into the prediction model to calculate and predict the future environmental parameters of the aircraft surface;

[0011] The prediction results are fed back to the flight control system for real-time adjustment of flight strategies.

[0012] In the method for predicting the surface environmental parameters of an aircraft as described above, the sensors include a temperature sensor, a humidity sensor, an air pressure sensor, a wind speed sensor, a solar radiation sensor, a pollutant sensor, an aerosol sensor and a sound level sensor.

[0013] In the above-mentioned method for predicting the surface environment parameters of an aircraft, the denoising process in the preprocessing process uses a noise suppression function, which reduces the impact of noise on the data by adjusting the signal strength of the collected data, thereby improving the accuracy of the data.

[0014] In the above-mentioned method for predicting the surface environmental parameters of an aircraft, the normalization process converts the environmental parameter data into a unified dimension so that data with different characteristics can be effectively compared and calculated.

[0015] A method for predicting aircraft surface environmental parameters as described above, wherein the constructed prediction model is a graph neural network-based model, which is used to process the multidimensional relationship of environmental parameter data to improve the accuracy of the prediction.

[0016] A method for predicting aircraft surface environment parameters as described above, wherein the preprocessed data set is divided into a training set, a validation set, and a test set in a ratio of 70%:15%:15% to ensure that the model has good generalization ability on unseen data and avoid overfitting.

[0017] As described above, a method for predicting the surface environmental parameters of an aircraft, wherein the output of the prediction model is the environmental parameter value within a future time window, including temperature, humidity, air pressure, wind speed, etc., to facilitate decision-making by the flight control system.

[0018] In the above-mentioned method for predicting the surface environmental parameters of an aircraft, the prediction results fed back to the flight control system can be dynamically adjusted according to real-time environmental changes to optimize the flight strategy.

[0019] The present invention also provides a computer storage medium, which includes: at least one memory and at least one processor;

[0020] A memory for storing one or more program instructions;

[0021] The processor is used for running one or more program instructions to execute a method for predicting aircraft surface environment parameters.

[0022] The beneficial effects achieved by the present invention are as follows: By comprehensively using data collection, processing and prediction models, the present invention can accurately and real-time predict the environmental parameters of the aircraft surface, thereby providing a decision-making basis for the aircraft control system and improving flight safety and performance reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0024] Figure 1 This is a flow chart of a method for predicting aircraft surface environment parameters provided in Example 1 of the present application. DETAILED DESCRIPTION

[0025] The following is a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0026] Embodiment 1

[0027] like Figure 1 As shown, the first embodiment of the present application provides a method for predicting aircraft surface environment parameters, comprising the following steps:

[0028] Step S10: Using sensors to collect environmental parameter data near the surface of the aircraft in real time;

[0029] In this step, environmental parameters near the aircraft surface are monitored in real time through a variety of sensors to ensure accurate and timely environmental data. These sensors include but are not limited to temperature sensors, humidity sensors, air pressure sensors, wind speed sensors, solar radiation sensors, pollutant concentration sensors, aerosol concentration sensors, and sound level sensors. The layout of each sensor should be optimized according to the design of the aircraft and the flight mode to ensure that interference and errors are minimized.

[0030] Define the environmental parameter vector E to be collected, which contains the following elements:

[0031] E=[T,H,P,W,S,C p ,C a ,L,M], where T represents temperature; H represents humidity; P represents air pressure; W represents wind speed; S represents solar radiation intensity; C p Indicates the pollutant concentration; C a represents aerosol concentration; L represents sound level; M represents other micrometeorological parameters (such as precipitation, visibility, etc.).

[0032] Environmental parameter data is collected periodically (e.g., every second) through sensors. The data collection process includes: initialization and calibration of sensors to ensure their accuracy and sensitivity; real-time recording and storage of data, using high-frequency sampling to capture instantaneous changes.

[0033] After data collection, preliminary preprocessing is required to improve the quality of the data, including: denoising, noise suppression of real-time collected data, and definition of a noise suppression function. Where E represents the current environmental parameter value, such as temperature, humidity, air pressure, etc. This value is the data input that needs to be noise suppressed; σ represents the standard deviation, which reflects the volatility of historical data; m represents the number of historical data, which is the number of samples collected within a certain time range; E mean It represents the average value of historical data, which is the value obtained by statistically analyzing the environmental parameters collected historically; It is the sum of squares of the deviations between the current environmental parameters and the average value of historical data. Through this calculation, the relationship between current data and historical data can be evaluated to help determine whether the current data is affected by noise. In order to better understand the impact of environmental changes, a comprehensive parameter is defined. Used to evaluate the impact of various environmental parameters: In this formula, n is the number of environmental parameters; α i is the weight coefficient of each environmental parameter; E i is the real-time value of the i-th environmental parameter; β i is a nonlinear weight, which indicates the contribution of different parameters to the comprehensive impact.

[0034] Finally, wireless communication technology is used for real-time transmission to transmit the preprocessed data and comprehensive parameters to the data processing module for subsequent analysis and model training.

[0035] Step S20, preprocessing the environmental parameter data, including denoising and normalization processing;

[0036] In this step, the environmental parameter data transmitted to the data processing module needs to be preprocessed to ensure data quality and accuracy, laying the foundation for the subsequent construction of the prediction model. Data preprocessing mainly includes denoising, normalization, and feature selection.

[0037] Denoising: In order to eliminate the random noise in the sensor data, signal processing technology is used to denoise the data. The data is processed using a filter. Assume that the collected raw data is D raw =[d1,d2,…dn ] , denoised data D clean It can be expressed as: Among them, D clean (t) represents the denoised data value at time t; k is the window size, that is, the number of samples considered during the denoising process; d(tj) is the original data at time tj.

[0038] Normalization: To facilitate subsequent machine learning model processing, parameters in different ranges need to be normalized to the same range (usually [0,1]). The formula for normalization is as follows: Among them, E is the original data; E max and E min are the maximum and minimum values ​​of the parameter, respectively. After normalization, all parameters will be at the same level, which helps improve the efficiency and accuracy of model training.

[0039] Feature selection: Feature selection aims to select the environmental parameters that have the greatest impact on model performance. Correlation analysis is used to calculate the correlation coefficient r between each environmental parameter and the target variable. The formula is as follows: Among them, n represents the number of samples, the number of sample pairs involved in the calculation; and are the means of feature E and target variable T, respectively. Features with absolute values ​​of correlation coefficients higher than a certain threshold are selected and retained for model training.

[0040] In order to train and validate the model, the preprocessed data set needs to be split into training set, validation set and test set in a ratio of 70%:15%:15%. The segmentation process ensures that the model has good generalization ability on unseen data and avoids overfitting. After preprocessing, the cleaned, normalized and feature-selected data is stored in a database or file system for subsequent model training and real-time prediction. The data storage structure should facilitate fast retrieval and loading and support efficient data access.

[0041] Step S30: construct a prediction model based on a machine learning algorithm, use the preprocessed environmental parameter data for training, and generate an environmental parameter prediction model;

[0042] In this step, the preprocessed environmental parameter data is used to build a machine learning prediction model to achieve accurate prediction of the future environmental parameters of the aircraft surface. This step mainly includes model selection, model training, and model evaluation. According to the characteristics of the environmental parameters and the prediction target, an appropriate machine learning algorithm is selected. Considering the complexity of the environmental data and the relationship between the parameters, a graph neural network (GNN) is selected. GNN can effectively capture the topological structure and dependency relationship between the parameters.

[0043] The input of the model is expressed as: X = [E, A], where E is the environmental parameter matrix; A is the adjacency matrix, which represents the mutual influence between parameters. After the model is selected, the model is trained using the training set to optimize the model parameters to minimize the loss function. In order to improve the prediction accuracy of the model, an improved spatiotemporal loss function L is introduced sp , which not only takes into account the prediction error, but also integrates the correlation between parameters: Among them, y i is the true value; is the model prediction value; γ is a hyperparameter for adjusting the weight; corr(E i ,E j ) represents the environmental parameter E i and E j This loss function can not only effectively improve the prediction accuracy of the model, but also help the model better understand the interaction of environmental parameters. Optimization techniques such as back propagation algorithm and gradient descent method are used to adjust the model parameters. The basic formula for updating model parameters is: Among them, θ represents the parameters of the model; τ is the learning rate; is the gradient of the loss function. By continuously iterating this formula, the loss function is gradually reduced, thereby optimizing the model. A dynamic learning rate adjustment strategy is introduced to improve the training effect and accelerate convergence, and the adaptive learning rate formula is introduced: Here, τ0 is the initial learning rate; β is a tuning parameter, and t is the current number of iterations. This formula makes the learning rate higher in the early stages of training so that a better solution can be found quickly, and gradually decreases in the later stages of training to refine the learning process of the model, thereby preventing oscillation and overfitting.

[0044] After training, the performance of the model is evaluated using stability feedback evaluation. The prediction results of multiple time periods are collected and the consistency of their feedback is calculated. Define the stability function S: in, is the predicted value at the current moment; y i-1is the true value at the previous moment; N is the number of observation time periods. This formula measures the relative changes between adjacent predictions and reflects the stability of the model. After the model training is completed, the trained model is saved to the database or file system to ensure its reusability and version traceability.

[0045] Step S40: input the environmental parameters collected in real time into the environmental parameter prediction model to calculate and predict the future environmental parameters of the aircraft surface.

[0046] In this step, the environmental parameters collected in real time will be passed as input data to the previously constructed prediction model to calculate and predict the future environmental parameters of the aircraft surface.

[0047] Integrate the environmental parameters collected by real-time sensors into an input vector X input , to suit the input requirements of the model;

[0048] X input =[T current ,H current ,P current ,W current ,S current ,C current ,A current ,L current ,M current ]

[0049] Among them, T current Indicates temperature; H current Indicates humidity; P current Indicates air pressure; W current Indicates wind speed; S current Indicates the intensity of solar radiation; C current Indicates the pollutant concentration; A current Indicates aerosol concentration; L current Indicates the sound level; M current Represents other micro-meteorological parameters. The formatted environmental parameter vector X input The trained prediction model is passed in. This model is based on learning from historical data and can capture the complex relationship between environmental parameters. The input data is processed by a machine learning algorithm to generate a prediction of the future environmental state. The model will generate the predicted values ​​of future environmental parameters:

[0050]

[0051] The output of the model (predicted future environmental parameters) is integrated into a data package for subsequent processing. The prediction results are fed back to the flight control system in real time through the data channel to support dynamic adjustment of flight strategies. The system will monitor the difference between the actual environmental parameters and the predicted parameters, and use this information to continuously optimize the model and improve the accuracy of future predictions.

[0052] Step S50: Feedback the prediction result to the flight control system for real-time adjustment of the flight strategy.

[0053] At this stage, the generated prediction results of future environmental parameters will be transmitted to the flight control system to support real-time decision-making and flight strategy adjustment.

[0054] Integrate the predicted environmental parameters into a structured data packet for easy processing by the flight control system. Use a reliable communication protocol (such as TCP / IP, UDP, etc.) to send the formatted prediction results to the flight control system in real time. Ensure the timeliness and integrity of data transmission so that the flight control system can respond quickly. During the transmission process, the flight control system should perform data verification to ensure that the received prediction results are valid and complete. After receiving the prediction results, the flight control system parses and processes them, and comprehensively evaluates the predicted environmental parameters through the following formula The actual environmental parameters E actual The relationship between: Where ΔE is the overall error estimate of the environmental parameters; n is the number of predicted parameters (such as temperature, humidity, etc.); is the i-th predicted future environmental parameter; is the ith actual environmental parameter; α is a constant that controls the attenuation rate and is used to emphasize the importance of larger errors. Set an acceptable error threshold. If ΔE exceeds this threshold, it indicates that the prediction may not be accurate enough and the strategy needs to be adjusted. For example, if the wind speed and wind direction forecasts differ significantly from the actual data, the heading may need to be adjusted. Use the current wind speed and heading information to recalculate the optimal heading to reduce wind resistance and improve flight efficiency. If the actual air pressure and temperature data differ significantly from the predicted values, it may affect flight performance and safety. According to the actual meteorological conditions and aircraft performance characteristics, adjust the flight altitude to find a more suitable aerodynamic environment and improve flight stability.

[0055] Continuously monitor actual environmental parameters and compare them with predicted values ​​to ensure timely detection of new changes. Feedback the adjusted strategy effects to the model for optimization in future forecasts to improve the accuracy and adaptability of the model.

[0056] Corresponding to the above embodiment, an embodiment of the present invention provides a computer storage medium, including: at least one memory and at least one processor;

[0057] The memory is used to store one or more program instructions;

[0058] The processor is used for running one or more program instructions to execute a method for predicting aircraft surface environment parameters.

[0059] Corresponding to the above-mentioned embodiment, an embodiment of the present invention provides a computer-readable storage medium, wherein the computer storage medium contains one or more program instructions, and the one or more program instructions are used by a processor to execute a method for predicting aircraft surface environment parameters.

[0060] The embodiment disclosed in the present invention provides a computer-readable storage medium, in which computer program instructions are stored. When the computer program instructions are executed on a computer, the computer executes the above-mentioned method for predicting the surface environment parameters of an aircraft.

[0061] In the embodiment of the present invention, the processor may be an integrated circuit chip having the ability to process signals. The processor may be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0062] The methods, steps and logic block diagrams disclosed in the embodiments of the present invention can be implemented or executed. The general processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in the embodiments of the present invention can be directly embodied as a hardware decoding processor for execution, or can be executed by a combination of hardware and software modules in the decoding processor. The software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The processor reads the information in the storage medium and completes the steps of the above method in combination with its hardware.

[0063] The storage medium may be a memory, which may be, for example, a volatile memory or a nonvolatile memory, or may include both volatile and nonvolatile memory.

[0064] Among them, the non-volatile memory can be a read-only memory (ROM), a programmable ROM (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory.

[0065] The volatile memory may be a random access memory (RAM) which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate synchronous DRAM (DDRSDRAM), enhanced synchronous DRAM (ESDRAM), synchronous link DRAM (SLDRAM), and direct RAM bus DRAM (DRRAM).

[0066] The storage media described in the embodiments of the present invention are intended to include, but are not limited to, these and any other suitable types of memory.

[0067] Those skilled in the art will appreciate that in one or more of the above examples, the functions described in the present invention can be implemented using a combination of hardware and software. When software is used, the corresponding functions can be stored in a computer-readable medium or transmitted as one or more instructions or codes on a computer-readable medium. Computer-readable media include computer storage media and communication media, wherein communication media include any media that facilitates the transmission of computer programs from one place to another. Storage media can be any available media that can be accessed by general or special-purpose computers.

[0068] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made on the basis of the technical solution of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for predicting aircraft surface environment parameters, characterized in that: include: Use sensors to collect environmental parameter data near the surface of the aircraft in real time; Preprocessing the environmental parameter data; Construct a prediction model based on machine learning algorithm, use the preprocessed environmental parameter data for training, and generate an environmental parameter prediction model; Inputting the environmental parameters collected in real time into the prediction model to calculate and predict the future environmental parameters of the aircraft surface; The prediction results are fed back to the flight control system for real-time adjustment of flight strategies.

2. The method for predicting aircraft surface environment parameters according to claim 1, characterized in that: The sensors include temperature sensors, humidity sensors, air pressure sensors, wind speed sensors, solar radiation sensors, pollutant sensors, aerosol sensors and sound level sensors.

3. A method for predicting aircraft surface environment parameters as claimed in claim 2, characterized in that: The denoising process in the preprocessing process uses a noise suppression function, which reduces the impact of noise on the data and improves the accuracy of the data by adjusting the signal strength of the collected data.

4. The method for predicting aircraft surface environment parameters according to claim 3, characterized in that: The normalization process converts the environmental parameter data into a unified dimension so that data with different characteristics can be effectively compared and calculated.

5. The method for predicting aircraft surface environment parameters according to claim 4, characterized in that: The constructed prediction model is a graph neural network-based model, which is used to process the multidimensional relationship of environmental parameter data to improve the accuracy of prediction.

6. A method for predicting aircraft surface environment parameters as claimed in claim 5, characterized in that: The output of the prediction model is the environmental parameter values ​​within the future time window, including temperature, humidity, air pressure, and wind speed, which facilitates the flight control system to make decisions.

7. A method for predicting aircraft surface environment parameters as claimed in claim 6, characterized in that: The prediction results fed back to the flight control system can be dynamically adjusted according to real-time environmental changes to optimize the flight strategy.

8. A computer storage medium, characterized in that: include: at least one memory and at least one processor; A memory for storing one or more program instructions; A processor, used to run one or more program instructions to execute a method for predicting aircraft surface environment parameters as described in any one of claims 1-7.

Images