Time-varying aircraft circumnavigation path planning method based on weather avoidance zone prediction

Abstract

The present application is suitable for the field of aviation flight technology, and provides a time-varying aircraft circling path planning method based on meteorological avoidance area prediction, which comprises the following steps: obtaining the starting and ending points, time and historical meteorological radar chart data of an aircraft; extracting the edge features of the meteorological avoidance area and generating a convex polygon by using the Graham algorithm; fusing the grey prediction, Markov chain and Kalman filtering methods to predict the spatio-temporal evolution trend of the meteorological avoidance area; constructing a circling path planning model under the time-varying environment based on the predicted trend, taking the shortest path as the objective function, and imposing the minimum yaw angle, shortest distance and safety interval constraints; solving the model by using the improved scarab beetle optimization algorithm, improving the algorithm performance by introducing a dynamic inertia weight and an adaptive mutation mechanism, and outputting the optimal circling path. The present application realizes accurate prediction and efficient path planning of dynamic meteorological avoidance areas, and significantly improves the safety and economy of aircraft circling.

Description

Technical Field

[0001] This invention belongs to the field of aviation flight technology, and in particular relates to a time-varying aircraft detour path planning method based on weather avoidance zone prediction. Background Technology

[0002] In the field of aviation, severe convective weather is the primary factor affecting flight safety. It occurs, develops, and evolves rapidly, making accurate forecasting difficult. Severe convective weather causes the largest number of flight delays, cancellations, diversions, and detours, placing immense pressure on air traffic control operations, especially flow management decisions. The escalating global climate change is leading to more frequent extreme weather events, and the potential threat of meteorological disasters to air transport continues to expand. Against this backdrop, aviation meteorological users have an increasing demand for more refined, integrated, and decision-making support meteorological services, and there is an urgent need for reliable weather avoidance and route planning technologies to ensure flight safety and improve operational efficiency.

[0003] Existing weather avoidance zones are mostly generated based on meteorological forecast information at the same time, possessing high accuracy only within a short period. They cannot ensure spatial continuity between weather avoidance zones with adjacent times, resulting in weak robustness. Furthermore, the shape, size, location, and weather intensity of weather avoidance zones are all dynamically changing, and static delineation methods cannot accommodate this inherent attribute. Current technologies are all based on static weather avoidance zones for rerouting planning, failing to accurately predict changes in weather avoidance zones at different future times. They ignore the correlation between time-varying weather avoidance zones and rerouting paths, exhibiting a one-sidedness and making them difficult to apply to real-world scenarios.

[0004] Traditional route planning algorithms rely on fixed route models and static constraints. When faced with complex scenarios such as dynamic weather disturbances and airspace control changes, they suffer from low computational efficiency, limited optimization dimensions, and poor adaptability. Furthermore, in high-density airspace environments, traditional algorithms cannot adapt to dynamic adjustment needs, leading to persistently high flight delay rates. Simultaneously, route changes coordinated manually by air traffic controllers may not align with crew requirements, further impacting operational safety and economic efficiency. Summary of the Invention

[0005] The purpose of this invention is to provide a time-varying aircraft detour path planning method based on weather avoidance zone prediction, aiming to solve the technical problems existing in the prior art as identified in the background art.

[0006] This invention is implemented as follows: a time-varying aircraft detour path planning method based on weather avoidance zone prediction, comprising the following steps:

[0007] Step 1: Obtain the aircraft's origin and destination points, time, and historical weather radar data;

[0008] Step 2: Use the Graham algorithm to extract the edge features of the weather avoidance zone and generate a convex polygon of the flight hazard area;

[0009] Step 3: Combine grey prediction, Markov chain and Kalman filtering methods to predict the spatiotemporal evolution trend of the weather avoidance zone;

[0010] Step 4: Based on the spatiotemporal evolution trend obtained in Step 3, construct an aircraft detour path planning model under time-varying conditions, wherein the waypoint sequence of the detour path is the decision variable of the model; the model takes the shortest total length of the path formed by the path sequence as the objective function, and applies constraints including minimum yaw angle, shortest flight distance and safety separation.

[0011] Step 5: Solve the aircraft detour path planning model constructed in Step 4 using the improved dung beetle optimization algorithm, and output the detour path composed of the decision variables; the improved dung beetle optimization algorithm balances global search and local exploration by introducing dynamic inertial weights, and dynamically evaluates the mutation rate of individuals through an adaptive mutation mechanism.

[0012] Preferably, in step 1, acquiring the aircraft's origin and destination points, time, and historical weather radar image data specifically includes acquiring the aircraft's original flight path. waypoints ,in and For origin and destination, each waypoint The coordinates are ( , Based on the aircraft's flight speed, calculate the time it takes to pass adjacent waypoints; obtain data from Doppler weather radar at different times. The collected radar echo image.

[0013] Preferably, in step 2, extracting the edge features of the weather avoidance zone using the Graham algorithm includes converting the weather radar image into a grayscale image and performing thresholding; performing edge detection to obtain the coordinates of edge points and the binary image boundary; generating a convex polygon with the bottom left edge point as the starting point; merging convex polygons that are less than 20km apart; and expanding the edge region of the convex polygon outward, by 5km during the day and 10km at night.

[0014] Preferably, in step 3, the method of combining grey prediction, Markov chain, and Kalman filtering to predict the spatiotemporal evolution trend includes: constructing a grey prediction model to process the initial coordinate sequence; constructing a Markov prediction model to correct the error of the grey prediction model; fusing the corrected prediction sequence with the actual radar echo map, and calculating the final prediction sequence through semi-adaptive Kalman filtering.

[0015] Preferably, in step 4, constructing an aircraft detour path planning model under time-varying conditions includes defining an objective function that minimizes the total path length, and imposing constraints on the decision variables such as segment length, weather avoidance zone, path length upper limit, and yaw angle.

[0016] Preferably, in step 5, the improved dung beetle optimization algorithm improves algorithm performance by introducing dynamic inertia weights and an adaptive mutation mechanism. The dynamic inertia weights decrease linearly during algorithm iteration, and the adaptive mutation probability is related to the individual fitness value.

[0017] The beneficial effects of this invention are:

[0018] This invention provides a time-varying aircraft detour path planning method and system based on weather avoidance zone prediction. It integrates methods such as grey prediction, Markov and Kalman filtering to predict the evolution trend, and constructs an aircraft detour path planning model under time-varying environment. It considers realistic factors such as minimum yaw angle, shortest flight distance, and safety interval to generate the shortest detour path.

[0019] Based on the combined algorithm to extrapolate the weather avoidance zone at different future times, and by integrating methods such as grey prediction, Markov and Kalman filtering to predict its evolution trend, an aircraft detour path planning model under time-varying conditions is constructed. Taking into account realistic factors such as minimum yaw angle, shortest flight distance, and safety interval, the shortest detour path is generated, providing technical support for aircraft detour path planning during thunderstorms. The embodiments of this invention can realize the function of automatically generating an executable detour path based on the future spatiotemporal evolution environment of the weather avoidance zone, providing technical support for dispatchers to conduct intelligent flight plan management in daily operations. Attached Figure Description

[0020] Figure 1 A flowchart of a time-varying aircraft detour path planning method based on weather avoidance zone prediction provided in an embodiment of the present invention;

[0021] Figure 2 A flowchart illustrating the implementation process of weather avoidance zone prediction based on a combined model, as provided in this embodiment of the invention.

[0022] Figure 3 A flowchart illustrating the implementation of the mathematical model for time-varying aircraft detour path planning provided in this embodiment of the invention;

[0023] Figure 4 This is a diagram illustrating the shortest flight segment restrictions;

[0024] Figure 5 A flowchart illustrating the implementation of the improved dung beetle optimization algorithm for solving time-varying aircraft bypass path planning models provided in this embodiment of the invention;

[0025] Figure 6 The present invention provides a structural block diagram of a time-varying aircraft detour path planning system based on weather avoidance zone prediction. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0027] like Figure 1 The diagram shows a flowchart of a time-varying aircraft detour path planning method based on weather avoidance zone prediction provided by an embodiment of the present invention. The method includes:

[0028] Step 1: Obtain the aircraft's origin and destination points and times, as well as historical weather radar data;

[0029] Obtain the aircraft's origin and destination points, scheduled times, and historical weather radar data. In practice, this requires reading the aircraft's original flight path. waypoints ,in and These are the departure point and the landing point, and each waypoint Having definite coordinates ( , According to the aircraft's flight speed Calculate the adjacent waypoints along its route. and Time and The following relation is satisfied: = Meanwhile, data from different time periods was obtained through a Doppler weather radar system. Radar echo map data.

[0030] Step 2: Use the Graham algorithm to extract the edge features of the weather avoidance zone;

[0031] The acquired weather radar images were converted into grayscale images, and thresholding was performed using the Otsu method to enhance the image contour features.

[0032] Edge detection is performed using the Canny operator to obtain the coordinates of edge points and the binary image boundary.

[0033] Find the lowest and leftmost point in the image as the starting point, and sort all edge points according to their polar angle relative to the starting point;

[0034] The Graham algorithm is applied to generate convex polygons representing hazardous flight zones. Based on this, in accordance with aviation safety regulations, convex polygons with a distance of less than 20km are merged, and the edges of the merged convex polygons are expanded. The expansion width is 5km during the day and 10km at night to ensure that aircraft maintain a safe distance from hazardous weather areas.

[0035] Step 3: Combine grey prediction, Markov and Kalman filtering to predict the spatiotemporal evolution trend of the weather avoidance zone;

[0036] Specifically, such as Figure 2 The diagram shown illustrates an implementation flowchart of a weather avoidance zone prediction process based on a combined model, according to an embodiment of the present invention. The main processes and steps include:

[0037] 1. Acquire multiple weather radar images at different times;

[0038] 2. Extract the edge features of the weather avoidance zone from the weather radar images at each time point;

[0039] 3. Based on historical time periods The changes in weather avoidance zones have inherent limitations in traditional methods such as grey prediction models, Markov prediction, and Kalman filtering. Combining these methods to extrapolate and predict their changes in future periods is crucial. The spatiotemporal evolution trend:

[0040] Among them, the edge features of the weather avoidance zone are extracted as follows:

[0041] 2.1 Take a weather radar image at a certain moment and convert it into a grayscale image, then perform thresholding to enhance the contour features;

[0042] 2.2 Perform edge detection and obtain edge point coordinates to obtain the boundary of the binary image;

[0043] 2.3 Find the bottom leftmost point as the starting point, sort all points according to the polar angle relative to the starting point, and finally obtain the point set and the convex polygon of the danger zone.

[0044] 2.4 When two adjacent hazardous weather areas are less than 20km apart, aircraft are prohibited from crossing the thunderstorm area. Therefore, it is necessary to merge two convex hull areas that are less than 20km apart. Thunderstorm detours must maintain a safe distance radius of 5km during the day and 10km at night, which requires expanding the edge area.

[0045] The projection predicts its future period The spatiotemporal evolution trends include:

[0046] 3.1 The input of the Markov prediction model is the error sequence of the numerical values ​​directly predicted by the grey model, and the output is the prediction of the grey model residuals;

[0047] 3.2 The input to the grey prediction model is the initial coordinate sequence after transformation. Axial direction sequence and The axial direction sequence is output as the initial prediction sequence after correction by the combined Markov prediction model.

[0048] 3.3 The initial predicted sequence is fused with the actual radar echo map as the input of the semi-adaptive Kalman filter to calculate the final predicted sequence of the weather avoidance area.

[0049] Step 4: Based on the spatiotemporal evolution trend, construct an aircraft detour path planning model under time-varying environment, taking into account realistic factors such as minimum yaw angle, shortest flight distance, and safety interval, and generate the shortest detour path.

[0050] like Figure 3 The diagram shown is a flowchart illustrating the implementation of a mathematical model for time-varying aircraft detour path planning according to an embodiment of the present invention:

[0051] (1) Problem Description

[0052] Based on weather radar data and combined forecasting models, it is possible to extract the data for a specific airspace at different times. of Weather avoidance zone ( An aircraft's original path was affected by thunderstorms, requiring it to travel at a different speed. To avoid the weather avoidance zone, the total number of flights involved is... waypoints ,in and These are the origin and destination points, each waypoint. The coordinates are ( , Aircraft passing through adjacent waypoints. and The times are respectively and ,satisfy = Obviously, in the planning of rerouting routes, practical factors such as the single turning angle should not exceed 90 degrees, the shortest flight distance, and the avoidance of weather avoidance zones must be considered.

[0053] Without loss of generality, the model makes the following main assumptions: it does not consider the impact of factors such as handover between ATC control sectors on rerouting routes; it maintains constant aircraft speed; and it ignores interference from other aircraft, changes in altitude, and pilot errors. Based on the spatiotemporal evolution trend of the weather avoidance zone, a mathematical model is established to address this problem, aiming to generate the shortest possible rerouting route.

[0054] (2) Constraints

[0055] like Figure 4 As shown, based on the turning radius of the aircraft as it flies over the tangent point ( ) and flight speed ( ), turning slope ( Turning rate () The relationship between them is shown in the following formula:

[0056] ;

[0057] ;

[0058] According to ICAO document 8168, the shortest segment length between adjacent waypoints should generally not be less than [a certain length]. To ensure that the aircraft has sufficient time to complete its course adjustment, data from a typical commercial airliner is used for calculations. =250 sections =25°, at this time and =3.6 kilometers.

[0059] ;

[0060] For safe flight, any waypoint on any detour route. At its arrival time It should be in accordance with the weather avoidance zone during this period. Maintain a safe distance.

[0061] ;

[0062] When an aircraft detours to avoid a weather avoidance zone, fuel constraints will be taken into account, and its detour route will not be more than 1.3 times longer than the original route.

[0063] ;

[0064] in: It is the length of the original path.

[0065] Two adjacent flight segments and The headings are respectively and Then it is based on the heading difference waypoints The angle calculation is as follows:

[0066] ;

[0067] by segment course For example, its calculation method is shown below.

[0068] ;

[0069] in: .

[0070] According to relevant regulations of my country's civil aviation, when choosing a rerouting route, the turning angle should not exceed 90° at a time, and the smaller the turning angle should be, as shown below.

[0071] ;

[0072] (3) Objective function

[0073] During thunderstorm detours, while satisfying practical factors such as minimum yaw angle and avoiding weather avoidance zones, we aim to minimize the detour path within time-varying weather avoidance zones, as shown below:

[0074] ;

[0075] Step 5: Based on the aircraft detour path planning model and spatiotemporal evolution trend, the improved dung beetle optimization algorithm is used to solve the problem. Dynamic inertial weight is introduced to balance global search and local exploration. The mutation rate of individuals is dynamically evaluated through an adaptive mutation mechanism, and the generated detour path is output.

[0076] like Figure 5 The diagram shown illustrates the implementation flowchart of an improved dung beetle optimization algorithm for solving time-varying aircraft bypass path planning models, provided by an embodiment of the present invention. DBO originates from biological behavior and is a swarm intelligence algorithm that simulates the rolling, dancing, foraging, stealing, and reproductive behaviors of dung beetle populations. The dung beetle population is divided into rolling dung beetles, reproductive dung beetles, foraging dung beetles, and stealing dung beetles. Rolling dung beetles are responsible for global exploration; reproductive dung beetles perform dynamic local exploration; foraging dung beetles are responsible for local fine-grained search; and stealing dung beetles are responsible for the overall population competition mechanism. The main processes and steps include:

[0077] (1) Initialize the population

[0078] The position of an individual in a dung beetle population corresponds to a candidate solution to an optimization problem. The population is randomly initialized. Afterwards, any individual dung beetle We explore better solutions through continuous competition among organisms.

[0079] ;

[0080] in: This represents the number of dung beetles in the population.

[0081] (2) Global search

[0082] In the During the nth iteration, the 1st A dung beetle relies on bright celestial bodies such as the sun or moon for navigation while rolling the ball, maintaining a straight line throughout the process. The update formula is as follows:

[0083] ;

[0084] in: This represents the natural coefficient, whose main function is to control the randomness of direction; b is the path deflection coefficient; It is a constant value; This represents the worst position within the dung beetle population, used to expand the search area; It is used to simulate changes in light intensity in nature.

[0085] When dung beetles encounter obstacles, they adjust their angle... The position update formula for repositioning via dancing is as follows:

[0086] ;

[0087] Obviously, if At this time, the dung beetle's position will not be updated. The above formula simulates the process by which the dung beetle continuously rolls to find a new path after receiving environmental information such as light and wind direction.

[0088] (3) Local search

[0089] In nature, dung balls are not only used as food by dung beetles, but also as oviposition sites for females to reproduce. Female dung beetles lay their eggs within a dynamically shrinking, safe area. Inspired by this behavior, the boundaries of the female dung beetle's oviposition area are:

[0090] ;

[0091] in: and These are the upper and lower boundaries of the spawning area, respectively; This is the current local optimal position; This represents the maximum number of iterations. This is a dynamic factor that determines the size of the spawning zone boundary. In the first... During the nth iteration, the 1st The location of the brooding feces ball as follows:

[0092] ;

[0093] in: and It is a random vector.

[0094] The foraging behavior of dung beetles mimics the behavior of mature dung beetles that venture outside the brooding area. Their foraging range is as follows:

[0095] ;

[0096] in: and These represent the upper and lower boundaries of the dung beetle's foraging area; This is the globally optimal solution. According to the definition of the foraging area, in the... During the nth iteration, the 1st The locations of the dung beetle larvae have been updated as follows:

[0097] ;

[0098] in: and It is a random vector. Dung beetles sometimes engage in competitive behavior after obtaining food, or they may steal from each other. The position update for simulating dung beetles snatching dung balls from other individuals is as follows:

[0099] ;

[0100] in: represents a constant; g represents a single-row vector; The current global optimum is considered the optimal food source in the dung beetle population, leading to competition and theft primarily occurring in its vicinity. Therefore, by simulating the theft and competition mechanisms of dung beetles, DBO not only possesses strong development performance but also improves convergence speed.

[0101] (4) Termination conditions

[0102] To balance the convergence accuracy and speed of DBO, the algorithm's termination conditions are: (1) reaching the maximum number of iterations; (2) the change in the optimal fitness value over several consecutive iterations is lower than a preset threshold.

[0103] To address the issue of standard DBO easily getting trapped in local optima, a dynamic inertial weight is introduced to balance global search and local exploration. An adaptive mutation mechanism is used to dynamically evaluate the mutation rate of individuals, thereby improving the convergence accuracy and speed of the algorithm.

[0104] (1) Optimization of dynamic inertia weight

[0105] To balance global and local exploration, dung beetle thieves and dung beetles cooperate and mutually constrain each other. In standard DBO, dung beetles tend towards the current global optimum in the early stages of the algorithm, easily getting trapped in local optima. Therefore, in the... Dynamic inertia weights were added to the rolling dung beetle behavior in the next iteration. This enhances the transition from global to local, as shown below:

[0106] ;

[0107] in: and These are the upper and lower bounds of the weights. Compared to fixed weights, dynamic inertia weights can favor global search in the early stages of the algorithm and focus on local exploration in the later stages, thus avoiding the algorithm getting trapped in local optima.

[0108] (2) Adaptive mutation mechanism

[0109] In standard DBO, a mutation mechanism is introduced to explore better solutions. A fixed mutation rate means that the fitness of an individual is independent of the mutation probability, which leads to slow convergence in the early stages of the algorithm and wasted computational resources in the later stages. Therefore, an adaptive mutation mechanism is introduced to ensure that the mutation rate is positively correlated with the fitness of the individual, as shown below.

[0110] ;

[0111] in: Indicates the first The mutation probability of a dung beetle; It is a constant. and They represent the first A dung beetle And the best dung beetle The current fitness value; This represents the largest fitness difference in the population.

[0112] like Figure 6 The diagram shown illustrates a structural block diagram of a time-varying aircraft detour path planning system based on weather avoidance zone prediction, provided as an embodiment of the present invention. The system includes a data preparation module, a prediction module, a model solving module, and a visualization module, specifically comprising:

[0113] The data preparation module involves real data such as the aircraft's origin and destination points and scheduled times, and the storage of weather forecast time series diagrams;

[0114] The prediction module integrates methods such as grey prediction, Markov and Kalman filtering, based on historical time periods. Weather avoidance zone changes are predicted and projected for future periods. The spatiotemporal evolution trend;

[0115] The model solving module takes into account the aircraft's origin and destination points and planned time, as well as realistic factors such as minimum yaw angle, shortest flight distance, and safety interval, and calls the model and algorithm interface to generate the shortest detour path;

[0116] The visualization module displays the spatiotemporal evolution of the weather avoidance zone from the prediction module and the rerouting path from the model solving module using GIS.

[0117] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0118] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

[0119] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A time-varying aircraft detour path planning method based on weather avoidance zone prediction, characterized in that, The method includes: Step 1: Obtain the aircraft's origin and destination points and times, as well as historical weather radar data; Step 2: Use the Graham algorithm to extract the edge features of the weather avoidance zone; Step 3: Combine grey prediction, Markov and Kalman filtering to predict the spatiotemporal evolution trend of the weather avoidance zone; Step 4: Based on the spatiotemporal evolution trend, construct an aircraft detour path planning model under time-varying environment, taking into account realistic factors such as minimum yaw angle, shortest flight distance, and safety interval, and generate the shortest detour path. Step 5: Based on the aircraft detour path planning model and spatiotemporal evolution trend, the improved dung beetle optimization algorithm is used to solve the problem. Dynamic inertial weight is introduced to balance global search and local exploration. The mutation rate of individuals is dynamically evaluated through an adaptive mutation mechanism, and the generated detour path is output.

2. The method according to claim 1, characterized in that, The acquisition of the aircraft's origin and destination points and times, as well as historical weather radar image data, specifically includes: When an aircraft is avoiding a weather avoidance zone, read all waypoints. waypoints ,in and For origin and destination, each waypoint The coordinates are ( , ); Aircraft passing through adjacent waypoints and The times are respectively and And satisfy = ; Based on data collected by Doppler weather radar, information was obtained at different times. Radar echo map.

3. The method according to claim 1, characterized in that, Step 2 involves using the Graham algorithm to extract the edge features of the weather avoidance zone. Based on Doppler weather radar data, the Graham algorithm is used to extract areas with a reflectivity greater than 41 dBZ from the radar echo map as flight hazard zones, specifically including: The weather radar image was converted into a grayscale image and then thresholded. Perform edge detection and obtain edge point coordinates to obtain the boundary of the binary image. Take the bottom left point as the starting point and sort all points according to the polar angle relative to the starting point to obtain the point set and the convex polygon of the danger area. Convex polygons that are less than 20km apart are merged, and the edge regions of the convex polygons are extended outward, with an extension width of 5km during the day and 10km at night.

4. The method according to claim 1, characterized in that, Step 3, which involves fusing grey prediction, Markov and Kalman filtering to predict the spatiotemporal evolution trend of the weather avoidance zone, specifically includes: A grey prediction model is constructed, which takes the x-axis direction sequence and y-axis direction sequence obtained by transforming the initial coordinate sequence of the meteorological avoidance zone as input and outputs the initial prediction sequence. A Markov prediction model is constructed, and the error sequence of the output value of the grey prediction model is used as input to output the prediction of the grey model residuals and correct the initial prediction sequence. The predicted sequence, corrected by the Markov model, is fused with the actual radar echo image and used as input for a semi-adaptive Kalman filter to calculate the final predicted sequence for the weather avoidance zone.

5. The method according to claim 1, characterized in that, Step 4, which describes constructing an aircraft detour path planning model under time-varying conditions, specifically includes: Define the objective function as minimizing the total length of the detour path: ； Constraints are imposed on the waypoint sequence of the bypass route, including: Segment length constraint: The distance between adjacent waypoints shall not be less than the aircraft's turning radius R. ； The turning radius R is based on the aircraft's flight speed. Turning slope and turning rate calculate: ； Weather avoidance zone constraints, any waypoint Arrival time Weather avoidance zone during this period intersect: ； The path length is subject to an upper limit constraint: the total length of the detour route cannot exceed 1.3 times the original path length L. ； Yaw angle constraints, waypoints The corner No more than 90°: ； Wherein, the turning angle Difference in heading between adjacent segments calculate: ； The heading The calculation method is as follows: ； in, .

6. The method according to claim 1, characterized in that, Step 5, which involves using an improved dung beetle optimization algorithm, includes: Introducing dynamic inertia weights Its update formula is: ； in, This represents the current iteration number. The maximum number of iterations, and These are the minimum and maximum values ​​of the weights, respectively. Introducing an adaptive mutation mechanism, the first The mutation probability of an individual dung beetle The calculation formula is: ； in, It is a constant. For the first Individual dung beetle fitness value, The best dung beetle individual currently available fitness value, This represents the maximum fitness difference in the population.

7. A time-varying aircraft detour path planning platform based on weather avoidance zone prediction, characterized in that, include: The data preparation module is used to acquire and store aircraft origin and destination points, scheduled times, and weather forecast time series data; The prediction module is used to integrate grey prediction, Markov chain and Kalman filtering methods to extrapolate and predict the spatiotemporal evolution trend of meteorological avoidance zones in the future based on the changes in meteorological avoidance zones in historical periods. The model solving module is used to construct and solve the detour path planning model under time-varying conditions based on the origin and destination points and planned time of the aircraft, taking into account realistic factors such as minimum yaw angle, shortest flight distance and safety interval, to generate the shortest detour path; The visualization module is used to display the spatiotemporal evolution information of the weather avoidance zone obtained by the prediction module and the rerouting path obtained by the model solving module in GIS.

Images