An icing warning method and system for low-altitude aircraft based on meteorological observations

Abstract

The present invention provides a method and system for icing warning of low-altitude aircraft based on meteorological observation. The method includes obtaining historical meteorological data, historical flight data, aircraft own data and icing conditions of the low-altitude aircraft in the same period; respectively establishing corresponding neural network models according to the division of icing components, and respectively training them to obtain corresponding target neural network models; obtaining various data collected by the low-altitude aircraft in real time, respectively inputting them into the corresponding target neural network models, and outputting corresponding icing thickness and icing type; judging whether the preset icing components reach the warning standard according to the output results of the target neural network models; if so, giving a first warning; if not, performing weighted processing according to the icing thickness and icing type of each icing component to obtain an evaluation value; judging whether the evaluation value is greater than a threshold; if so, giving a second warning. Specifically, by evaluating the icing situation of the aircraft, the flight risk can be effectively reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of icing warning for low-altitude aircraft, and particularly relates to a method and system for icing warning of low-altitude aircraft based on meteorological observation. Background Technique

[0002] Low-altitude aircraft refer to various aircraft flying within a relatively low altitude range from the ground. Low-altitude aircraft generally fly in the airspace below 1000 meters and have an icing risk within the cloud layer range of 0 - 3000 meters. Especially in the altitude range of 1500 meters to 3000 meters, it is a common high-incidence area of icing for low-altitude aircraft such as helicopters.

[0003] In order to reduce the risk of aircraft icing when manned low-altitude aircraft perform flight missions, predicted meteorological data is usually obtained, and a flight route is formulated based on the predicted meteorological data. However, this method is relatively dependent on the accuracy of meteorological prediction. It can be understood that the more accurate the predicted meteorological data is, the more reasonable the formulated flight route is, and the lower the probability of aircraft icing. In other words, if the meteorology suddenly changes and the pilot still flies according to the established route or takes countermeasures relatively late, there will be a greater potential safety hazard. Summary of the Invention

[0004] Based on this, an icing warning method and system for low-altitude aircraft based on meteorological observation are provided in the embodiments of the present invention, aiming to evaluate the icing situation of the aircraft in real time according to meteorological data so as to reduce flight risks.

[0005] The first aspect of the embodiments of the present invention provides an icing warning method for low-altitude aircraft based on meteorological observation, and the method includes:

[0006] Obtain the historical meteorological data, historical flight data, aircraft own data, and icing situation of the low-altitude aircraft in the same period. The historical meteorological data includes temperature, humidity, wind speed, wind direction, air pressure, precipitation type, and relative cloud position. The historical flight data includes flight altitude, flight speed, and flight time. The aircraft own data includes material and surface characteristics. The icing situation includes icing components, icing thickness, and icing type;

[0007] According to the division of icing components, corresponding neural network models are respectively established and trained respectively to obtain corresponding target neural network models. Specifically, the historical meteorological data, historical flight data, and aircraft own data are used as the input of the neural network model, and the icing thickness and icing type at a preset time interval are used as the output of the neural network model;

[0008] Obtain the meteorological data, flight data, and the data of the aircraft itself collected by the low-altitude aircraft in real time, and input them into the target neural network models corresponding to each icing component respectively to output the corresponding icing thickness and icing type;

[0009] According to the output result of the target neural network model, determine whether the preset icing component reaches the warning standard;

[0010] If it is determined that the preset icing component reaches the warning standard, issue a first warning;

[0011] If it is determined that the preset icing component does not reach the warning standard, perform weighted processing according to the icing thickness and icing type of each icing component to obtain an evaluation value;

[0012] Judge whether the evaluation value is greater than the threshold;

[0013] If it is determined that the evaluation value is greater than the threshold, issue a second warning.

[0014] Furthermore, the method further includes:

[0015] Obtain an initial planned path, and adjust the initial planned path according to the warning situation.

[0016] Furthermore, the steps of obtaining the initial planned path and adjusting the initial planned path according to the warning situation include:

[0017] Obtain a three-dimensional satellite cloud map of the current flight position of the low-altitude aircraft, and determine the target area in the three-dimensional satellite cloud map that is beneficial to alleviating icing;

[0018] When the warning situation is the first warning, plan a first flight path for the low-altitude aircraft to enter the first target area according to the first target area closest to the current flight position of the low-altitude aircraft, and adjust the initial planned path according to the first flight path;

[0019] When the warning situation is the second warning, correct the meteorological prediction data according to the three-dimensional satellite cloud map, generate a target three-dimensional satellite cloud map according to the corrected meteorological prediction data, plan a second flight path with the smallest deviation from the initial planned path for the low-altitude aircraft, and adjust the initial planned path according to the second flight path.

[0020] Furthermore, the steps of generating a target three-dimensional satellite cloud map according to the corrected meteorological prediction data and planning a second flight path with the smallest deviation from the initial planned path for the low-altitude aircraft include:

[0021] Map the initial planned path to the target three-dimensional satellite cloud map, and remove the line segments corresponding to the flight paths at preset time intervals from the initial planned path to form a gap;

[0022] Obtain the target meteorological prediction data in the preset spatial range corresponding to the flight path at the preset interval, and input the target meteorological prediction data into the target neural network model corresponding to each icing component respectively to output the corresponding predicted icing thickness and predicted icing type;

[0023] Determine the target planned route according to the predicted icing thickness and predicted icing type;

[0024] Merge the line segments corresponding to the target planned route to the gap to obtain the second flight path.

[0025] Further, the step of determining the target planned route according to the predicted icing thickness and predicted icing type includes:

[0026] According to the importance degree of the icing components, as well as the predicted icing thickness and predicted icing type corresponding to the icing components, determine the second target area that meets the requirements in terms of the predicted icing thickness and predicted icing type and is within the preset spatial range;

[0027] In the second target area, determine the shortest flight route of the low-altitude aircraft as the target planned route.

[0028] Further, the neural network model is composed of an input layer, a hidden layer, and an output layer. The input layer contains a number of input nodes, and the number of input nodes is the same as the number of types of input data, which is used to receive different types of input data;

[0029] The hidden layer consists of five sub-layers. The first hidden sub-layer has 64 nodes, and the LeakyReLU function is used to extract features from the input data to find the key features related to the output result. The second hidden sub-layer and the third hidden sub-layer each have 128 nodes, and the Sigmoid function is used to further explore the features of the output data of the first hidden sub-layer and capture the internal connections between the nodes. The fourth hidden sub-layer and the fifth hidden sub-layer each contain 64 nodes, and with the help of the ReLU activation function, the data features output by the third hidden sub-layer are processed twice;

[0030] The output layer is provided with a number of output nodes, and the number of output nodes corresponds to the number of types of output data. The output layer uses a linear activation function to output the final prediction result.

[0031] Further, in the step of performing weighted processing on the icing thickness and icing type of each icing component to obtain an evaluation value, the expression of the weighted processing is:

[0032]

[0033] where P represents the evaluation value, denotes the weight coefficient of the k-th icing component, denotes the score value determined by the k-th icing component according to the icing thickness and icing type.

[0034] The second aspect of the embodiments of the present invention provides an icing warning system for low-altitude aircraft based on meteorological observations, which is used to implement the icing warning method for low-altitude aircraft based on meteorological observations described in the first aspect. The system includes:

[0035] A first acquisition module, configured to acquire the historical meteorological data, historical flight data, aircraft own data, and icing conditions of the low-altitude aircraft in the same period. The historical meteorological data includes temperature, humidity, wind speed, wind direction, air pressure, precipitation type, and relative position of clouds. The historical flight data includes flight altitude, flight speed, and flight time. The aircraft own data includes material and surface characteristics. The icing conditions include icing components, icing thickness, and icing type;

[0036] A model training module, configured to respectively establish corresponding neural network models according to the division of icing components, and respectively train them to obtain corresponding target neural network models. Specifically, the historical meteorological data, historical flight data, and aircraft own data are used as the input of the neural network model, and the icing thickness and icing type at a preset time interval are used as the output of the neural network model;

[0037] A second acquisition module, configured to acquire the meteorological data, flight data, and aircraft own data collected by the low-altitude aircraft in real time, and respectively input them into the target neural network models corresponding to each icing component, and output the corresponding icing thickness and icing type;

[0038] A first judgment module, configured to judge whether a preset icing component reaches the warning standard according to the output result of the target neural network model;

[0039] A first warning module, configured to perform a first warning if it is judged that the preset icing component reaches the warning standard;

[0040] A weighted processing module, configured to perform weighted processing according to the icing thickness and icing type of each icing component to obtain an evaluation value if it is judged that the preset icing component does not reach the warning standard;

[0041] A second judgment module, configured to judge whether the evaluation value is greater than a threshold;

[0042] A second warning module, configured to perform a second warning if it is judged that the evaluation value is greater than the threshold.

[0043] A third aspect of the embodiments of the present invention provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements the low-altitude aircraft icing warning method based on meteorological observations provided by the first aspect.

[0044] A fourth aspect of the embodiments of the present invention provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, it implements the low-altitude aircraft icing warning method based on meteorological observations provided by the first aspect.

[0045] A low-altitude aircraft icing warning method and system provided by the embodiments of the present invention. The method obtains historical meteorological data, historical flight data, aircraft own data, and icing conditions of the low-altitude aircraft in the same period; according to the division of icing components, corresponding neural network models are established and trained respectively to obtain corresponding target neural network models; meteorological data, flight data, and aircraft own data collected by the low-altitude aircraft are obtained in real time and input into the target neural network models corresponding to each icing component respectively to output corresponding icing thickness and icing type; according to the output results of the target neural network models, it is judged whether the preset icing components reach the warning standard; if so, a first warning is carried out; if not, weighted processing is carried out according to the icing thickness and icing type of each icing component to obtain an evaluation value; it is judged whether the evaluation value is greater than a threshold; if it is judged that the evaluation value is greater than the threshold, a second warning is carried out. Specifically, by evaluating the icing situation of the aircraft, the flight risk can be effectively reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is a flowchart of the implementation of a low-altitude aircraft icing warning method provided by Embodiment 1 of the present invention;

[0047] Figure 2 It is a structural block diagram of a low-altitude aircraft icing warning system provided by Embodiment 2 of the present invention;

[0048] Figure 3 It is a structural block diagram of an electronic device provided by Embodiment 3 of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0049] To facilitate the understanding of the present invention, the present invention will be described more comprehensively below with reference to the relevant drawings. Several embodiments of the present invention are given in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, these embodiments are provided to make the disclosure of the present invention more thorough and comprehensive.

[0050] It should be noted that when an element is referred to as being "fixed to" another element, it can be directly on the other element or there can also be an intermediate element. When an element is considered to be "connected" to another element, it can be directly connected to the other element or there may be an intermediate element at the same time. The terms "vertical", "horizontal", "left", "right" and similar expressions used herein are for illustrative purposes only.

[0051] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which this invention belongs. The terms used herein in the specification of the present invention are for the purpose of describing specific embodiments only and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0052] Embodiment 1

[0053] According to an embodiment of the present invention, an embodiment of a method for icing warning of a low-altitude aircraft based on meteorological observation is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than here.

[0054] In this first embodiment, a method for icing warning of a low-altitude aircraft based on meteorological observation is provided, which can be used in an electronic device, such as a computer. Please refer to Figure 1 , Figure 1 , which shows an implementation flowchart of a method for icing warning of a low-altitude aircraft provided in the first embodiment of the present invention, specifically including steps S01 to S08.

[0055] Step S01, obtain the historical meteorological data, historical flight data, aircraft own data and icing conditions of the low-altitude aircraft in the same period.

[0056] Specifically, the historical meteorological data includes temperature, humidity, wind speed, wind direction, air pressure, precipitation type and relative position of clouds, the historical flight data includes flight altitude, flight speed and flight time, the aircraft own data includes material and surface characteristics, and the icing conditions include icing components, icing thickness and icing type. It should be noted that the above data all have different degrees of influence on the icing of the low-altitude aircraft.

[0057] Among them, for temperature, icing on the aircraft body surface usually occurs in the temperature range of 0°C to -20°C, and severe icing mostly occurs in the range of -2°C to -10°C; for humidity and air pressure, when the temperature dew point difference in the cloud is between 0°C and 5°C, icing is more likely to occur; for wind speed and wind direction, in combination with the flight speed of the aircraft, when the relative speed is relatively large, the collision frequency and energy between the aircraft and supercooled water droplets increase, and the supercooled water droplets are more likely to adhere to the aircraft surface and freeze quickly; for precipitation type, in freezing rain and drizzle weather, when the outside temperature is below 0°C, the supercooled raindrops are likely to freeze once they come into contact with the aircraft. In the environment of sleet, the coexistence of snowflakes and supercooled water droplets will also increase the possibility of aircraft icing; for the relative position of the cloud layer, when flying through the upper part of a cumulonimbus cloud or a cumulus congestus cloud, severe icing usually appears on the aircraft body surface; while when flying through a stratus cloud or a stratocumulus cloud, weak or moderate icing usually occurs; for flight altitude, the altitude range of 1500 meters to 3000 meters is a common high-incidence area of icing; for material and surface characteristics, if the aircraft surface is made of a metal material with good thermal conductivity, it is easier to rapidly reduce the surface temperature in a low-temperature environment, promoting the condensation and icing of water vapor. If the surface is rough and uneven, it will increase the attachment probability of supercooled water droplets and is also more likely to ice up.

[0058] In addition, according to the different functions and positions of the icing components, the icing effects also vary. For example, rotating components such as the rotor of a helicopter are more likely to ice up than fixed components due to the heat exchange with air during rotation and the complex airflow conditions. Prominent parts such as the engine air intake, airspeed tube, windshield, and antenna are also prone to icing because they are more likely to come into contact with supercooled water droplets.

[0059] The above data are all collected by various sensors carried on low-altitude aircraft. Exemplarily, the icing thickness and the relative position of the cloud layer, etc., can be obtained by installing visual sensors such as cameras on the aircraft to take images of the aircraft surface. Through image recognition technology, the icing areas in the images are analyzed and processed.

[0060] In some other embodiments of the present invention, the icing observation data of mountain weather stations and meteorological data such as wind, humidity, temperature, and pressure can also be used for reference.

[0061] Step S02, according to the division of the icing components, corresponding neural network models are respectively established and trained respectively to obtain the corresponding target neural network models.

[0062] Among them, historical meteorological data, historical flight data, and the data of the aircraft itself are used as the input of the neural network model, and the icing thickness and icing type at preset time intervals are used as the output of the neural network model. Before the data is input into the neural network model, data preprocessing can be performed, including missing value processing (interpolation method), outlier removal (based on physical constraints), feature engineering (Z-score normalization), and standardization (Z-score normalization), etc.

[0063] In this embodiment, the neural network model consists of an input layer, a hidden layer, and an output layer. The input layer contains a number of input nodes, and the number of input nodes is the same as the number of types of input data, which is used to receive different types of input data.

[0064] The hidden layer is composed of five sub-layers. The first hidden sub-layer has 64 nodes, and the LeakyReLU function is used to extract the features of the input data to find the key features related to the output result. The second hidden sub-layer and the third hidden sub-layer each have 128 nodes, and the Sigmoid function is used to further explore the features of the output data of the first hidden sub-layer and capture the internal connections between the nodes. The fourth hidden sub-layer and the fifth hidden sub-layer each contain 64 nodes, and with the help of the ReLU activation function, the data features output by the third hidden sub-layer are processed twice.

[0065] The output layer is provided with a number of output nodes, and the number of output nodes corresponds to the number of types of output data. The output layer uses a linear activation function to output the final prediction result.

[0066] Specifically, 80% of the dataset samples after random shuffling are used as training set samples, and the other 20% are used as test set samples. Among them, the neuron formulas of the hidden layer and the output layer can be expressed as:

[0067] The first hidden sub-layer:

[0068]

[0069] The second hidden sub-layer:

[0070]

[0071] The third hidden sub-layer:

[0072]

[0073] The fourth hidden sub-layer:

[0074]

[0075] The fifth hidden sub-layer:

[0076]

[0077] Output layer:

[0078]

[0079] where x j and y i represent the j-th node of the input layer and the i-th node of the output layer respectively, represents the weight connecting the j-th node of the layer and the i-th node of the output layer, represents the bias term of the i-th node of the layer. Exemplarily, represents the i-th neuron of the first hidden sub-layer, represents the i-th neuron of the second hidden sub-layer, represents the i-th neuron of the third hidden sub-layer, represents the i-th neuron of the fourth hidden sub-layer, represents the i-th neuron of the fifth hidden sub-layer. Similarly, represents the j-th neuron of the first hidden sub-layer, represents the j-th neuron of the second hidden sub-layer, represents the j-th neuron of the third hidden sub-layer, represents the j-th neuron of the fourth hidden sub-layer, represents the j-th neuron of the fifth hidden sub-layer. LeakyReLU, Sigmoid, and ReLU are activation functions.

[0080] Step S03: Real-time obtain the meteorological data, flight data, and aircraft own data collected by the low-altitude aircraft, and input them into the target neural network models corresponding to each icing component respectively, and output the corresponding icing thickness and icing type.

[0081] Step S04: According to the output result of the target neural network model, determine whether the preset icing component reaches the warning standard. If so, execute Step S05; if not, execute Step S06.

[0082] It can be understood that the preset icing components may include key aerodynamic components, power system-related components, and sensor components, etc. Exemplarily, icing on power system components such as the engine air intake and propeller will affect the normal operation of the engine, resulting in problems such as power decline and increased vibration. Once icing signs are found in these parts, even if it is only a small amount of icing, a warning should be issued in a timely manner.

[0083] Step S05: Then issue a first warning.

[0084] Among them, the pilot can be reminded of the possible icing by means of sound alarms and scrolling subtitles, and attention should be paid to taking safety precautions.

[0085] In step S06, weighted processing is performed according to the icing thickness and icing type of each icing component to obtain an evaluation value.

[0086] Specifically, the expression for weighted processing is:

[0087]

[0088] Among them, P represents the evaluation value, represents the weight coefficient of the kth icing component, represents the score value determined by the kth icing component according to the icing thickness and icing type. It should be noted that the weight coefficient of the more critical icing component is larger. In addition, a mapping relationship between the icing thickness and icing type of each icing component and the score value is established to determine the score value according to the mapping relationship.

[0089] In step S07, it is judged whether the evaluation value is greater than the threshold. If so, step S08 is executed.

[0090] In step S08, a second warning is given.

[0091] Similarly, the pilot can be reminded of the possible icing by means of sound alarms and scrolling subtitles, and attention should be paid to taking safety precautions. In some other embodiments of the present invention, after the warning is given, the user can be more intelligently helped to plan the flight path to further reduce the flight risk. Specifically, the initial planned path is obtained and adjusted according to the warning situation. Among them, the initial planned path is the result of comprehensive consideration of meteorological data, flight distance, terrain, etc. It should be noted that a three-dimensional satellite cloud map of the flight position of the current low-altitude aircraft is obtained, and the target area beneficial to alleviating icing in the three-dimensional satellite cloud map is determined. Among them, the three-dimensional satellite cloud map is realized based on technologies such as volume rendering. The current CloudFantasy system already has this function. Since the three-dimensional satellite cloud map is rendered according to meteorological data, the meteorological data around the flight position of the current low-altitude aircraft can be obtained, and then the meteorological data is input into the target neural network, so that the target area beneficial to alleviating icing in the three-dimensional satellite cloud map can be determined;

[0092] When the warning situation is the first warning, it means that the meteorological change is rapid and the situation is relatively urgent, and rapid risk avoidance is required. Then, according to the first target area closest to the flight position of the current low-altitude aircraft, the first flight path for the low-altitude aircraft to enter the first target area is planned, and the initial planned path is adjusted according to the first flight path;

[0093] When the warning situation is the second warning, it indicates that there are potential risks, and the adaptive adjustment path is sufficient. Then, according to the three-dimensional satellite cloud map, the meteorological prediction data is corrected, and the target three-dimensional satellite cloud map is generated based on the corrected meteorological prediction data. The second flight path with the smallest deviation from the initial planned path of the low-altitude aircraft is planned, and the initial planned path is adjusted according to the second flight path. It can be understood that the obtained three-dimensional satellite cloud map is real-time, that is, the accuracy of the meteorological data is relatively high. The previously predicted meteorological data is corrected by the meteorological data with relatively high accuracy to obtain the corrected and relatively accurate target three-dimensional satellite cloud map.

[0094] Specifically, the initial planned path is mapped to the target three-dimensional satellite cloud map, and the line segments corresponding to the flight paths at preset time intervals are removed from the initial planned path to form a gap.

[0095] Obtain the target meteorological prediction data in the preset space range corresponding to the flight path at preset time intervals, and input the target meteorological prediction data into the target neural network model corresponding to each icing component respectively, and output the corresponding predicted icing thickness and predicted icing type. It can be understood that the preset space range refers to a space range such as a rectangle or a sphere obtained by expanding in height based on the line segment corresponding to the flight path at preset time intervals, and the target meteorological prediction data is obtained according to the target three-dimensional satellite cloud map.

[0096] According to the predicted icing thickness and predicted icing type, determine the target planned route. Among them, according to the importance level of the icing components, as well as the predicted icing thickness and predicted icing type corresponding to the icing components, determine the second target area that meets the requirements in the preset space range and has a predicted icing thickness and predicted icing type. Exemplarily, if there are 7 icing components a, b, c, d, e, f, g, where a, b, c are key icing components, and d, e, f, g are ordinary icing components, according to the importance levels of a, b, c, the position points that do not meet the requirements are sequentially removed within the preset space range. Not meeting the requirements can be understood as the icing type not meeting the requirements, the icing thickness exceeding the standard, etc. In the preset space range after removing the position points that do not meet the requirements, determine the position points where the four ordinary icing components d, e, f, g all meet the requirements, and merge these position points to form the second target area.

[0097] In the second target area, determine the shortest flight route of the low-altitude aircraft as the target planned route.

[0098] Merge the line segments corresponding to the target planned route to the gap to obtain the second flight path.

[0099] In summary, for the icing warning method of low-altitude aircraft based on meteorological observation in the above embodiments of the present invention, the method obtains historical meteorological data, historical flight data, aircraft own data, and icing conditions of the low-altitude aircraft in the same period; according to the division of icing components, corresponding neural network models are established and trained respectively to obtain corresponding target neural network models; meteorological data, flight data, and aircraft own data collected by the low-altitude aircraft are obtained in real time and input into the target neural network models corresponding to each icing component respectively, and the corresponding icing thickness and icing type are output; according to the output results of the target neural network models, it is judged whether the preset icing components reach the warning standard; if so, a first warning is given; if not, weighted processing is performed according to the icing thickness and icing type of each icing component to obtain an evaluation value; it is judged whether the evaluation value is greater than the threshold; if it is judged that the evaluation value is greater than the threshold, a second warning is given. Specifically, by evaluating the icing condition of the aircraft, the flight risk can be effectively reduced.

[0100] Embodiment 2

[0101] Please refer to Figure 2 , Figure 2 , which is a structural block diagram of an icing warning system for low-altitude aircraft based on meteorological observation provided by the second embodiment of the present invention. The icing warning system 200 for low-altitude aircraft based on meteorological observation is used to implement the above embodiments and preferred implementation manners, and those that have been described will not be repeated. As used hereinafter, the term "module" can be a combination of software and / or hardware that realizes a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware is also possible and contemplated.

[0102] Specifically, the icing warning system 200 for low-altitude aircraft based on meteorological observation includes: a first acquisition module 21, a model training module 22, a second acquisition module 23, a first judgment module 24, a first warning module 25, a weighted processing module 26, a second judgment module 27, and a second warning module 28, where:

[0103] The first acquisition module 21 is used to acquire historical meteorological data, historical flight data, aircraft own data, and icing conditions of the low-altitude aircraft in the same period. The historical meteorological data includes temperature, humidity, wind speed, wind direction, air pressure, precipitation type, and relative position of cloud layers. The historical flight data includes flight altitude, flight speed, and flight time. The aircraft own data includes material and surface characteristics. The icing conditions include icing components, icing thickness, and icing type;

[0104] The model training module 22 is used to respectively establish corresponding neural network models according to the division of icing components, and train them respectively to obtain corresponding target neural network models. Specifically, historical meteorological data, historical flight data, and aircraft own data are used as the input of the neural network model, and the icing thickness and icing type at preset time intervals are used as the output of the neural network model. The neural network model consists of an input layer, a hidden layer, and an output layer. The input layer contains several input nodes, and the number of input nodes is the same as the number of input data types, which is used to receive different types of input data;

[0105] The hidden layer consists of five sub-layers. The first hidden sub-layer has 64 nodes, and the LeakyReLU function is used to extract features from the input data to find key features related to the output result. The second and third hidden sub-layers each have 128 nodes, and the Sigmoid function is used to further explore the features of the output data of the first hidden sub-layer and capture the internal connections between nodes. The fourth and fifth hidden sub-layers each contain 64 nodes, and with the help of the ReLU activation function, the data features output by the third hidden sub-layer are processed twice;

[0106] The output layer is provided with several output nodes, and the number of output nodes corresponds to the number of output data types. The output layer uses a linear activation function to output the final prediction result;

[0107] The second acquisition module 23 is used to acquire in real time the meteorological data, flight data, and aircraft own data collected by the low-altitude aircraft, and input them into the target neural network models corresponding to the respective icing components, and output the corresponding icing thickness and icing type;

[0108] The first judgment module 24 is used to judge whether the preset icing component reaches the warning standard according to the output result of the target neural network model;

[0109] The first warning module 25 is used to issue a first warning if it is judged that the preset icing component reaches the warning standard;

[0110] The weighted processing module 26 is used to perform weighted processing according to the icing thickness and icing type of each icing component to obtain an evaluation value if it is judged that the preset icing component does not reach the warning standard. The expression of the weighted processing is:

[0111]

[0112] where P represents the evaluation value, represents the weight coefficient of the kth icing component, represents the score value determined by the kth icing component according to the icing thickness and icing type;

[0113] The second judgment module 27 is used to judge whether the evaluation value is greater than the threshold value;

[0114] The second warning module 28 is used to give a second warning if it is judged that the evaluation value is greater than the threshold value.

[0115] Furthermore, in some alternative embodiments of the present invention, the low-altitude aircraft icing warning system 200 based on meteorological observation further includes:

[0116] An adjustment module is used to obtain an initial planned path and adjust the initial planned path according to the warning situation.

[0117] Furthermore, in some alternative embodiments of the present invention, the adjustment module includes:

[0118] An acquisition unit is used to obtain a three-dimensional satellite cloud map of the current flight position of the low-altitude aircraft and determine a target area in the three-dimensional satellite cloud map that is beneficial to alleviating icing;

[0119] The first adjustment unit is used to, when the warning situation is the first warning, plan a first flight path for the low-altitude aircraft to enter the first target area according to the first target area closest to the current flight position of the low-altitude aircraft, and adjust the initial planned path according to the first flight path;

[0120] The second adjustment unit is used to, when the warning situation is the second warning, correct the meteorological prediction data according to the three-dimensional satellite cloud map, generate a target three-dimensional satellite cloud map according to the corrected meteorological prediction data, plan a second flight path with the smallest deviation from the initial planned path for the low-altitude aircraft, and adjust the initial planned path according to the second flight path.

[0121] Furthermore, in some alternative embodiments of the present invention, the second adjustment unit includes:

[0122] A mapping sub-unit is used to map the initial planned path to the target three-dimensional satellite cloud map and remove the line segment corresponding to the flight path at the preset time interval from the initial planned path to form a gap;

[0123] An acquisition sub-unit is used to obtain target meteorological prediction data in a preset spatial range corresponding to the flight path at the preset time interval, input the target meteorological prediction data into the target neural network model corresponding to each icing component, and output the corresponding predicted icing thickness and predicted icing type;

[0124] A determining subunit, configured to determine a target planned route according to a predicted icing thickness and a predicted icing type. Specifically, according to the importance degree of an icing component, and the predicted icing thickness and the predicted icing type corresponding to the icing component, a second target area that meets the requirements for the predicted icing thickness and the predicted icing type and is within the preset space range is determined;

[0125] In the second target area, the shortest flight route of the low-altitude aircraft is determined as the target planned route;

[0126] A merging subunit, configured to merge the line segments corresponding to the target planned route to the gap to obtain the second flight path.

[0127] Embodiment III

[0128] On the other hand, the present invention further provides an electronic device. Please refer to Figure 3 , which shows the electronic device in Embodiment III of the present invention, including a memory 20, a processor 10, and a computer program 30 stored on the memory and executable on the processor. When the processor 10 executes the computer program 30, the method for warning of icing of a low-altitude aircraft based on meteorological observation as described above is implemented.

[0129] Among them, in some embodiments, the processor 10 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chips, and is used to run the program code stored in the memory 20 or process data, such as executing an access restriction program, etc.

[0130] Among them, the memory 20 includes at least one type of readable storage medium. The readable storage medium includes flash memory, a hard disk, a multimedia card, a card-type memory (for example, an SD or DX memory, etc.), a magnetic memory, a magnetic disk, an optical disk, etc. The memory 20 may be an internal storage unit of the electronic device in some embodiments, such as the hard disk of the electronic device. The memory 20 may also be an external storage device of the electronic device in other embodiments, such as a plug-in hard disk equipped on the electronic device, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. Further, the memory 20 may also include both an internal storage unit and an external storage device of the electronic device. The memory 20 can not only be used to store application software and various types of data of the electronic device, but also be used to temporarily store data that has been output or will be output.

[0131] It should be noted that Figure 3The structures shown do not constitute a limitation on the electronic device. In other embodiments, the electronic device may include fewer or more components than shown, or combine certain components, or have a different component arrangement.

[0132] An embodiment of the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the method for icing warning of a low-altitude aircraft based on meteorological observations as described above.

[0133] Those skilled in the art can understand that the logic and / or steps represented in the flowchart or described in other ways herein, for example, can be considered as a definite sequence list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other systems that can fetch instructions from the instruction execution system, apparatus, or device and execute the instructions), or in combination with these instruction execution systems, apparatus, or devices. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.

[0134] More specific examples (non-exhaustive list) of computer-readable media include the following: an electrical connection part (electronic device) having one or more wirings, a portable computer disk cartridge (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpretation, or other appropriate processing as necessary, and then stored in a computer memory.

[0135] It should be understood that various parts of the present invention can be implemented by hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, any one or a combination of the following techniques well known in the art can be used: discrete logic circuits having logic gate circuits for implementing logical functions on data signals, application-specific integrated circuits having appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0136] In the description of this specification, the description referring to terms such as "one embodiment", "some embodiments", "example", "specific example", or "some examples", etc. means that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples.

[0137] The above embodiments only express several implementation manners of the present invention, and their descriptions are relatively specific and detailed, but should not be construed as limiting the scope of the patent of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several modifications and improvements can still be made, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the appended claims.

Claims

1. A low-altitude aircraft icing early warning method based on meteorological observation, characterized in that: The method comprises: Obtain historical meteorological data, historical flight data, aircraft data and ice accumulation conditions of the low-altitude aircraft in the same period, wherein the historical meteorological data includes temperature, humidity, wind speed, wind direction, air pressure, precipitation type and relative position of clouds, the historical flight data includes flight altitude, flight speed and flight time, the aircraft data includes material and surface characteristics, and the ice accumulation conditions include ice accumulation components, ice thickness and ice accumulation type; According to the division of ice accumulation components, corresponding neural network models are established respectively, and training is performed respectively to obtain corresponding target neural network models. Specifically, historical meteorological data, historical flight data and aircraft data are used as inputs of the neural network model, and ice accumulation thickness and ice accumulation type at preset time intervals are used as outputs of the neural network model. Acquire meteorological data, flight data and aircraft data collected by low-altitude aircraft in real time, input them into the target neural network model corresponding to each icing component, and output the corresponding ice thickness and ice type; According to the output results of the target neural network model, it is determined whether the preset ice accumulation components meet the warning standard; If it is determined that the preset ice accumulation component reaches the warning standard, a first warning is issued; If it is determined that the preset ice accumulation component does not meet the warning standard, weighted processing is performed according to the ice accumulation thickness and ice accumulation type of each ice accumulation component to obtain an evaluation value; Determine whether the evaluation value is greater than a threshold; If it is determined that the evaluation value is greater than the threshold, a second warning is issued; Obtaining an initial planned path, and adjusting the initial planned path according to the warning situation. Specifically, obtaining a three-dimensional satellite cloud map of the current flight position of the low-altitude aircraft, and determining a target area in the three-dimensional satellite cloud map that is conducive to alleviating icing; When the warning situation is the first warning, a first flight path for the low-altitude aircraft to enter the first target area is planned according to a first target area closest to the current flight position of the low-altitude aircraft, and the initial planned path is adjusted according to the first flight path; When the warning situation is the second warning, the meteorological forecast data is corrected according to the three-dimensional satellite cloud map, and a target three-dimensional satellite cloud map is generated according to the corrected meteorological forecast data, a second flight path of the low-altitude aircraft that deviates from the initial planned path to the minimum is planned, and the initial planned path is adjusted according to the second flight path. Specifically, the initial planned path is mapped to the target three-dimensional satellite cloud map, and the line segment corresponding to the flight path at the preset interval time is removed from the initial planned path to form a gap; Obtaining target meteorological forecast data in a preset spatial range corresponding to the flight path at the preset time interval, inputting the target meteorological forecast data into a target neural network model corresponding to each icing component, and outputting the corresponding predicted icing thickness and predicted icing type; Determine the target planning route based on the predicted ice thickness and predicted ice type; The line segments corresponding to the target planned route are merged into the gap to obtain the second flight path.

2. The low-altitude aircraft icing early warning method based on meteorological observation according to claim 1 is characterized in that: The step of determining the target planning route according to the predicted ice accumulation thickness and the predicted ice accumulation type comprises: According to the importance of the ice accretion component, and the predicted ice accretion thickness and predicted ice accretion type corresponding to the ice accretion component, determine a second target area within the preset spatial range where the predicted ice accretion thickness and the predicted ice accretion type meet the requirements; In the second target area, the shortest flight route of the low-altitude aircraft is determined as the target planning route.

3. The low-altitude aircraft icing early warning method based on meteorological observation according to claim 2 is characterized in that: The neural network model is composed of an input layer, a hidden layer and an output layer. The input layer includes a number of input nodes, the number of which is the same as the number of input data types, and is used to receive different types of input data; The hidden layer consists of five sublayers. The first hidden sublayer has 64 nodes. The LeakyReLU function is used to extract the features of the input data to find the key features related to the output results. The second and third hidden sublayers each have 128 nodes. The Sigmoid function is used to further explore the features of the output data of the first hidden sublayer and capture the internal connections between the nodes. The fourth and fifth hidden sublayers each contain 64 nodes. The ReLU activation function is used to perform secondary processing on the data features output by the third hidden sublayer. The output layer is provided with a number of output nodes, the number of output nodes corresponds to the number of output data types, and the output layer adopts a linear activation function to output the final prediction result.

4. The low-altitude aircraft icing early warning method based on meteorological observation according to claim 3 is characterized in that: In the step of performing weighted processing according to the ice thickness and ice type of each ice accretion component to obtain an evaluation value, the expression for weighted processing is: Among them, P represents the evaluation value, represents the weight coefficient of the kth ice accretion component, It represents the score value of the kth ice accretion component determined according to the ice accretion thickness and ice accretion type.

5. A low-altitude aircraft icing warning system based on meteorological observation, characterized in that: Used to implement the low-altitude aircraft icing warning method based on meteorological observation as described in any one of claims 1 to 4, the system includes: The first acquisition module is used to acquire historical meteorological data, historical flight data, aircraft data and ice accumulation conditions of the low-altitude aircraft in the same period, wherein the historical meteorological data includes temperature, humidity, wind speed, wind direction, air pressure, precipitation type and relative position of clouds, the historical flight data includes flight altitude, flight speed and flight time, the aircraft data includes material and surface characteristics, and the ice accumulation conditions include ice accumulation components, ice accumulation thickness and ice accumulation type; A model training module is used to establish corresponding neural network models according to the division of ice-accumulated parts, and train them respectively to obtain corresponding target neural network models. Specifically, historical meteorological data, historical flight data and aircraft data are used as inputs of the neural network model, and ice thickness and ice type at preset time intervals are used as outputs of the neural network model. The second acquisition module is used to acquire the meteorological data, flight data and aircraft data collected by the low-altitude aircraft in real time, input them into the target neural network model corresponding to each ice accumulation component, and output the corresponding ice accumulation thickness and ice accumulation type; A first judgment module is used to judge whether the preset ice accumulation component reaches the warning standard according to the output result of the target neural network model; A first warning module, configured to issue a first warning if it is determined that a preset ice accumulation component has reached a warning standard; A weighted processing module, for performing weighted processing according to the ice thickness and ice type of each ice accretion component to obtain an evaluation value if it is determined that the preset ice accretion component does not meet the warning standard; A second judgment module, used to judge whether the evaluation value is greater than a threshold; The second warning module is used to issue a second warning if it is determined that the evaluation value is greater than a threshold.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the low-altitude aircraft icing warning method based on meteorological observation as described in any one of claims 1 to 4 is implemented.

7. An electronic device, characterized in that: The invention comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the low-altitude aircraft icing warning method based on meteorological observation as described in any one of claims 1 to 4 is implemented.

Images