Real-time optimization management system based on flight track predicted arrival stand time

Abstract

The application relates to the field of flight arrival time prediction, and particularly discloses a real-time optimization management system for predicted arrival parking time based on a flight track, which comprises a residual time initial estimation module, a flight efficiency analysis module, a residual time correction module and an arrival time prediction module. The system analyzes the track, speed and energy data of the flight distance flown by the flight in the cruising stage, extracts flight efficiency factors from three dimensions of track directness, speed control and energy management, and weights and fuses the flight efficiency factors into a flight efficiency coefficient. The initial estimated value of the residual distance is corrected based on the flight efficiency coefficient, and finally, the predicted time length of the descent approach stage and the landing taxi stage is combined to real-time infer the flight arrival parking time. The application realizes dynamic and accurate prediction of the time length required for the residual distance of the flight cruising stage, effectively overcomes the prediction lag and deviation problem caused by the dependence on static planning or external factors, and improves the real-time performance and accuracy of the flight operation management.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of flight arrival time prediction, and relates to a real-time optimization management system for predicted arrival stand time based on flight track. BACKGROUND

[0002] Accurate prediction of flight arrival stand time is of great significance for improving airport operation efficiency, optimizing ground resource allocation, enhancing passenger service experience, and realizing fine management of air traffic.

[0003] In an air transportation network, the overall journey of any complex flight with stops or transfers can be decomposed into a series of consecutive point-to-point non-stop direct flight segments. These non-stop direct flight segments are the smallest, indivisible basic operation units that constitute the flight mission.

[0004] The flight profile of a non-stop direct flight can be divided into five main stages: ground, takeoff climb, cruise, descent approach, and landing taxi. Among them, the cruise stage is the longest in duration and is most significantly affected by external interference factors such as enroute weather and airspace flow control, making it the largest source of uncertainty in flight arrival stand time prediction. In contrast, the operation time of other stages is relatively stable, constrained by standardized operation procedures, and the deviation from the planned time is usually small.

[0005] Therefore, to build a high-precision flight arrival time prediction model, dynamic monitoring and prediction of the cruise stage should be the core focus.

[0006] Currently, existing flight arrival time prediction methods, such as training machine learning models based on historical data or relying on flight plans and external factors for multidimensional prediction, have the following shortcomings: (1) Most methods rely on complete data after flight or external environmental information, making it difficult to achieve real-time and dynamic prediction adjustment during flight.

[0007] (2) Lack of fine-grained and multi-dimensional real-time analysis and feedback mechanisms for changes in flight efficiency during the cruise stage.

[0008] (3) Failure to fully utilize the actual operating conditions reflected by the flown segments to continuously correct the remaining journey time, resulting in insufficient adaptability of the prediction results to real-time operating conditions and limited accuracy. SUMMARY

[0009] To address the above problems, the present application proposes a real-time optimization management system for predicted arrival stand time based on flight track, which realizes the function of flight arrival time prediction.

[0010] The technical scheme adopted by the present application to solve its technical problems is: the present application provides a real-time optimization management system for estimated arrival gate time based on flight track, comprising: a remaining time initial estimation module: according to the distance of the direct flight in the cruise stage and the actual flight time, the time required to complete the remaining distance is initially estimated, and the initial value of the cruise remaining time is obtained.

[0011] A flight efficiency analysis module: based on the flight data of the cruise stage, flight efficiency factors are extracted from three dimensions of track directness, speed control and energy management, and each flight efficiency factor is weighted and fused to obtain the flight efficiency coefficient of the flown distance.

[0012] A remaining time correction module: input the flight efficiency coefficient into the correlation model between the flight efficiency coefficient and the time prediction deviation, obtain the cruise remaining time compensation, and correct the initial value of the cruise remaining time to obtain the corrected cruise remaining time.

[0013] An arrival time prediction module: according to the historical flight data of the flight, the time required for the descent approach stage and the landing taxi stage after the cruise stage is predicted, the corrected cruise remaining time is combined to calculate the estimated time from the current time to the arrival of the aircraft into the gate, and the arrival time of the aircraft into the gate is inferred.

[0014] Compared with the prior art, the real-time optimization management system for estimated arrival gate time based on flight track has the following beneficial effects: 1. The present application estimates the time required for the remaining distance in the cruise stage by analyzing the flight data of the flown distance in the cruise stage, overcoming the limitations of traditional methods relying on static planning or historical average values.

[0015] 2. The present application comprehensively evaluates flight efficiency from three dimensions of track directness, speed control and energy management, more comprehensively and accurately reflects the deviation between actual flight state and plan.

[0016] 3. The present application establishes a correlation model between the flight efficiency coefficient and the time prediction deviation, dynamically corrects the initial value of the cruise remaining time, and significantly improves the response ability to changes in real-time operation conditions.

[0017] 4. The present application analyzes the track, speed and energy profile by segmentation, identifies local flight characteristics, and further provides more reliable decision basis for real-time scheduling and resource allocation of flights. BRIEF DESCRIPTION OF DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the description of the embodiments. Obviously, the drawings described in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative effort.

[0019] Figure 1 The system module connection diagram of the present application.

[0020] Figure 2 The module workflow diagram of the flight efficiency analysis module of the present application.

[0021] Figure 3 The module workflow diagram of the remaining duration correction module of the present application. DETAILED DESCRIPTION

[0022] The technical solutions in the embodiments of the present application will be described clearly and completely in the following with reference to the drawings of the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, but not all the embodiments. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present application.

[0023] Please refer to Figure 1 The present application provides a real-time optimization management system of estimated arrival gate time based on flight track, which comprises a remaining duration initial estimation module, a flight efficiency analysis module, a remaining duration correction module and an arrival time prediction module.

[0024] The flight efficiency analysis module is connected with the remaining duration initial estimation module and the remaining duration correction module respectively, and the arrival time prediction module is connected with the remaining duration correction module.

[0025] The remaining duration initial estimation module is used to preliminarily estimate the duration required for completing the remaining distance according to the distance of the flight distance already flown in the cruise stage and the actual flight duration of the direct flight, so as to obtain the initial value of the cruise remaining duration.

[0026] Further, the specific working process of the remaining duration initial estimation module is as follows: the planned total distance and the planned total time of the direct flight without stop in the cruise stage are extracted from the pre-stored database, and the planned time per unit distance in the cruise stage is calculated .

[0027] The flight distance of the direct flight from the starting position of the cruise stage to the current position is obtained as the distance of the flight distance already flown in the cruise stage , and the actual flight duration corresponding to the flight distance already flown is obtained .

[0028] According to the distance of the flown range and the planned time of the unit distance , the planned flight time corresponding to the flown range is calculated wherein .

[0029] According to the track information of the flown range in the cruise phase, the position where the non-stop direct flight will arrive at the end of the cruise phase is predicted as the cruise phase end position.

[0030] The flight distance of the non-stop direct flight from the current position to the cruise phase end position is obtained as the distance of the cruise phase remaining range .

[0031] The initial estimate of the cruise phase remaining time is calculated based on the following formula : .

[0032] wherein, represents the deviation amount of the actual flight time per unit distance in the flown range from the planned flight time.

[0033] It should be noted that the non-stop direct flight refers to a flight that takes off from the departure airport, does not stop or land on the way, and directly flies to the destination airport. Any complex flight with stops or transfers can be decomposed into a series of continuous point-to-point non-stop direct flight segments. These non-stop direct flight segments are the smallest, indivisible basic operation units that constitute the flight task.

[0034] It should be noted that the flight profile of the non-stop direct flight can be divided into five main phases: ground, take-off climb, cruise, descent approach, and landing taxi. Among them, the ground phase includes the processes of pushing out, starting, and taxiing out; the take-off climb phase includes the processes of take-off, initial climb, and climb; the descent approach phase includes the processes of descent and approach; the landing taxi phase includes the processes of landing, taxiing in, and parking; and the cruise phase refers to the main phase of stable flight at the cruise altitude.

[0035] It should be noted that the planned total distance and the planned total time of the non-stop direct flight in the cruise phase are extracted from the pre-stored flight plan database. These planned data are inherent information that has been formulated and stored in the system database before the flight is executed, serving as the basis for calculating the planned time per unit distance.

[0036] It should be noted that the cruise phase start position refers to the position corresponding to the first time the non-stop direct flight in the take-off climb phase climbs to the preset cruise altitude layer.

[0037] It should be noted that, based on the historical track information of the flight range of the cruise phase, the cruise phase end position reached by the flight if the current flight trend is maintained is predicted through a track extrapolation algorithm such as fitting and extending according to the current heading, speed vector or flight trend, and the prediction point is determined as the cruise phase end position.

[0038] It should be noted that the cruise phase end position determined by the track prediction is a dynamic estimated value, which may not completely coincide with the planned end position set in the flight plan.

[0039] It should be noted that in the analysis formula of the initial value of the cruise remaining time, the first term represents the remaining range corresponding to the planned flight time; the second term represents the predicted value of the flight deviation of the flown range to the remaining range flight time deviation. Wherein, is the flown range corresponding to the actual flight time, is the corresponding planned flight time, and the difference between the two reflects the flight time deviation of the flown range: if the difference is positive, it indicates that the actual flight time is greater than the planned flight time, and the flight is in a delay state; if it is zero, it indicates that the flight is on time; if it is negative, it indicates that the actual flight time is less than the planned flight time, and the flight is in an advance state. The deviation is converted to the remaining range according to the range proportion, that is, the predicted flight time deviation of the remaining range is obtained.

[0040] It should be noted that the flight deviation of the flown range is used to predict the flight time deviation of the remaining range, which is based on the fact that the actual operating conditions such as wind speed and air traffic reflected by the flown range are often persistent in the remaining range; by proportionally expanding the flight time deviation per unit distance of the flown range to the remaining range, dynamic correction of the flight plan can be achieved. This prediction method can effectively estimate the cumulative deviation that may occur in the remaining range, thereby providing real-time and reasonable reference for the advance estimation of the arrival time of the flight and the operation scheduling.

[0041] The flight efficiency analysis module is configured to extract flight efficiency factors from three dimensions of track directness, speed control and energy management based on the flight data of the flown range of the cruise phase, and to obtain a flight efficiency coefficient of the flown range by weighted fusion of the flight efficiency factors.

[0042] Further, the specific process of extracting flight efficiency factors from the track directness dimension in the flight efficiency analysis module is as follows: obtaining the actual track of the non-stop flight in the flown range of the cruise phase.

[0043] extracting a complete planned track of the direct flight from a database, and intercepting a track segment corresponding to the flown track as the planned track of the flown track.

[0044] comparing the actual track with the planned track to determine whether there is an intersection between them.

[0045] if there is no intersection, calculating a ratio of the length of the planned track to the length of the actual track, and taking the ratio as the flight efficiency factor based on track directness.

[0046] if there is an intersection, performing the following steps: dividing the actual track into several sub-segments according to the number of intersections.

[0047] for each sub-segment, comparing its length with the length of the corresponding sub-segment in the planned track: if the actual sub-segment length is greater than the planned corresponding length, marking the sub-segment as a detour segment.

[0048] if the actual sub-segment length is equal to the planned corresponding length, marking it as an equidistant segment.

[0049] if the actual sub-segment length is less than the planned corresponding length, marking it as a direct flight segment.

[0050] counting the cumulative length and the number of all direct flight segments.

[0051] calculating a first ratio of the cumulative length of the direct flight segments to the total length of the actual track and a second ratio of the number of the direct flight segments to the total number of the actual track sub-segments.

[0052] performing a weighted average of the first ratio and the second ratio, and taking the result as the flight efficiency factor based on track directness.

[0053] It should be noted that the flight efficiency factor is defined based on the directness of the track, and the greater the value of the factor, the smaller the deviation of the actual track from the planned track, or even better than the planned track, and the higher the flight efficiency.

[0054] It should be noted that during the cruise phase, if the track of the flown track of the aircraft shows high directness, it usually indicates that the current airspace environment, flow control and control strategy are conducive to achieving efficient direct flight. Based on the consistency or continuity of the track of the aircraft during the cruise phase, it can be reasonably inferred that the aircraft is more likely to maintain a high track directness in the remaining flight range; vice versa.

[0055] It should be noted that the flight efficiency factor based on the directness of the flight path is analyzed by using the above hierarchical judgment and segmented comparison method, which can more finely identify the local deviation characteristics of the actual flight path relative to the planned path, such as detouring, direct flying, etc., avoid the structural differences caused by the overall length comparison, and thus more accurately reflect the execution efficiency and flight path quality of the flight path, and improve the rationality and practicality of the evaluation results.

[0056] Further, the specific process of extracting the flight efficiency factor from the speed control dimension in the flight efficiency analysis module is: in the flown distance in the cruise phase, a plurality of continuous time windows are divided according to a preset rule.

[0057] Obtain the speed monitoring data of the direct flight in each time window, and calculate the corresponding actual average ground speed under each time window.

[0058] Extract the planned flight speed data of the direct flight in the cruise phase from the database, and determine the corresponding planned average ground speed of each time window.

[0059] For each time window, calculate the ratio of the actual average ground speed to the planned average ground speed.

[0060] The ratio corresponding to all time windows is accumulated, and the arithmetic mean value is calculated, which is taken as the flight efficiency factor based on speed control.

[0061] It should be noted that the flight efficiency factor is defined based on the precision of speed control, and the larger the factor value, the higher the degree of coincidence between the actual flight speed and the planned flight speed, or even faster than the planned flight speed, and the flight efficiency is correspondingly higher, and the remaining distance time is likely to be shorter than the plan.

[0062] It should be noted that in the cruise phase, if the speed control performance of the aircraft in the flown distance shows high stability and compliance, it usually indicates that the current meteorological conditions, aircraft performance and automation system are conducive to achieving efficient cruising. Based on the inertia or persistence of the speed control of the aircraft in the cruise phase, it can be reasonably inferred that the aircraft has a high probability of maintaining good speed control performance in the remaining distance; vice versa.

[0063] It should be noted that the flight efficiency factor based on speed control is analyzed by comparing the actual average ground speed with the planned average ground speed in the time window, which can more finely capture the fluctuations and overall trends of the speed execution in each period of actual flight, avoid the local speed deviation or control characteristics caused by the comparison of the overall speed average, and thus more truly reflect the execution effect and control accuracy of the speed strategy, and improve the accuracy and guidance value of the evaluation results.

[0064] Further, the specific process of extracting flight efficiency factors from the energy management dimension in the flight efficiency analysis module is as follows: in the flown distance of the cruise stage, a plurality of monitoring time points are selected according to the equal time interval rule.

[0065] From the flight data of the straight flight in the flown distance of the cruise stage, the altitude and true airspeed corresponding to each monitoring time point are extracted, the energy altitude of each monitoring time point is calculated based on the energy altitude formula, and the energy altitude change trend curve of the flown distance of the cruise stage is plotted according to the time sequence.

[0066] The cruise stage is sequentially divided into a gradient rising process, a stable cruising process and a slow descending process.

[0067] According to the energy altitude change trend curve, the curve segments corresponding to each flight process are intercepted respectively.

[0068] Each of the curve segments is linearly fitted to obtain the fitting straight line slope as the energy altitude change slope of the process.

[0069] The energy altitude change slopes corresponding to each flight process are compared with the expected energy altitude change slopes of the corresponding processes pre-stored in the database, and the slope coincidence degrees of each process are calculated respectively.

[0070] The arithmetic mean of the slope coincidence degrees of all flight processes is calculated, and the mean value is taken as the flight efficiency factor based on energy management.

[0071] It should be noted that the energy altitude formula is wherein represents the energy altitude, represents the altitude, represents the true airspeed, represents the gravitational acceleration.

[0072] It should be noted that, in order to plot the energy altitude change trend curve of the flown distance of the cruise stage, the specific steps are as follows: a coordinate system is established with time as the horizontal axis and energy altitude as the vertical axis, the corresponding data points are marked in the coordinate system according to the energy altitude values of each monitoring time point, and the energy altitude change trend curve of the flown distance of the cruise stage is fitted and plotted based on the mathematical modeling method.

[0073] It should be noted that the following two implementation manners can be adopted to intercept the curve section corresponding to each flight process from the energy altitude change trend curve: the first manner is based on the morphological characteristics of the energy altitude change trend curve itself: the gradient rising process corresponds to the curve section with slow rising trend of energy altitude, the steady cruising process corresponds to the curve section with basically stable energy altitude, and the slow descending process corresponds to the curve section with steady descending trend of energy altitude. By identifying the trend characteristics of each stage in the curve, manual or automatic division of the curve is realized.

[0074] The second manner is based on the actual track information corresponding to each flight process: first, the start and end positions of the gradient rising, steady cruising and slow descending processes on the track are determined, and then the time points corresponding to the arrival of the aircraft at each position are obtained, and the energy altitude change trend curve is segmented and intercepted according to these time points.

[0075] It should be noted that the specific steps of calculating the slope fitness degree include: comparing the energy altitude change slope corresponding to a flight process with the pre-stored expected energy altitude change slope corresponding to the process in the database, and calculating the relative deviation between the two.

[0076] The relative deviation is input into a pre-set slope fitness degree evaluation model, and the slope fitness degree of the flight process is calculated, wherein the evaluation model is a linear negative correlation function with the relative deviation as the independent variable and the slope fitness degree as the dependent variable, that is, the larger the relative deviation, the lower the corresponding slope fitness degree.

[0077] It should be noted that the energy altitude comprehensively reflects the kinetic energy and potential energy state of the aircraft, and is the core physical quantity for representing the overall energy management of the aircraft. One of the core targets of high-efficiency cruising is to make the aircraft fly as close as possible to the pre-set optimal energy altitude profile, so as to maximize the fuel efficiency or optimize the flight time under given constraints. By quantifying the degree of fitness between the actual energy altitude change and the expected profile, the accuracy and economy of energy utilization in the flight process can be effectively evaluated.

[0078] It should be noted that during the cruising stage, if the energy management of the aircraft for the flight distance shows high stability and conformity, it usually indicates that the current flight profile planning, thrust management and external conditions are conducive to achieving high-efficiency cruising. Based on the inertia or persistence of the energy state change of the aircraft during the cruising stage, it can be reasonably inferred that the aircraft is more likely to maintain good energy management performance in the remaining flight distance; vice versa.

[0079] Further, refer to Figure 2As shown, the specific process for obtaining the flight efficiency coefficient of the flown distance in the flight efficiency analysis module is: extracting the pre-set weight of the flight efficiency factor based on the track directness, speed control and energy management from the database.

[0080] Linearly weighting and fusing the numerical value of each flight efficiency factor with its corresponding weight to obtain the flight efficiency coefficient of the flown distance.

[0081] It should be noted that the weight of each flight efficiency factor can be directly set according to industry experience or calculated based on limited test data. For example, first, collect the statistical frequency of the influence of each dimension of track directness, speed control and energy management on flight efficiency in historical cruise data, then calculate the correlation coefficient of each dimension efficiency factor and the final cruise efficiency index, determine the contribution of each factor by regression analysis, and finally normalize the contribution to convert it into the corresponding weight and ensure that the sum of all weights is 1.

[0082] It should be noted that the flight efficiency during the cruise phase is the result of the joint action of track path optimization, speed execution accuracy and energy utilization efficiency, and a single dimension cannot fully reflect the overall flight performance. Through multi-dimensional fusion evaluation, the economy of the spatial path, the stability of the time dimension and the scientificity of the energy management can be considered at the same time, so as to more systematically and accurately represent the comprehensive deviation degree between the actual flight and the ideal state. This way can avoid the one-sidedness of a single index, provide a more reliable and robust comprehensive efficiency basis for the remaining flight time prediction and flight strategy optimization, and improve the overall judgment and decision quality of the management system.

[0083] In this embodiment, the flight efficiency is comprehensively evaluated from the track directness, speed control and energy management, which more comprehensively and accurately reflects the deviation between the actual flight state and the plan.

[0084] In this embodiment, the flight efficiency is comprehensively evaluated from the track directness, speed control and energy management, which more comprehensively and accurately reflects the deviation between the actual flight state and the plan.

[0085] In this embodiment, the flight efficiency is comprehensively evaluated from the track directness, speed control and energy management, which more comprehensively and accurately reflects the deviation between the actual flight state and the plan.

[0086] The remaining time correction module is configured to input the flight efficiency coefficient into an association model between the flight efficiency coefficient and the time prediction deviation, obtain a cruise remaining time compensation amount, and correct the initial estimate of the cruise remaining time to obtain a corrected cruise remaining time.

[0087] Further, referring to Figure 3As shown, the specific working process of the remaining time correction module is: according to the historical flight data of the direct flight, a correlation model between the flight efficiency coefficient and the time prediction deviation is established.

[0088] The flight efficiency coefficient of the current flown distance is input into the correlation model to obtain a corresponding cruise remaining time compensation amount.

[0089] The initial value of the cruise remaining time is algebraically added to the compensation amount to obtain the corrected cruise remaining time.

[0090] It should be noted that the track directness, speed control accuracy and energy management level shown in the flown distance jointly reflect the actual running state and environmental influence of the current flight, and these efficiency characteristics are usually persistent in the cruise stage. By establishing a correlation model between the flight efficiency and the time prediction deviation, the current efficiency performance can be quantified as a compensation amount for the remaining distance. Its effect is to overcome the limitations of pure reliance on plan parameter prediction, dynamically correct the estimation deviation caused by real-time efficiency fluctuations, thereby significantly improving the accuracy of remaining time prediction and adaptability to actual running conditions, and providing a more reliable time benchmark for subsequent flight management and decision-making.

[0091] In the embodiment, the application estimates the time required for the remaining distance in the cruise stage in real time by analyzing the flight data of the flown distance in the cruise stage, overcoming the limitations of traditional methods relying on static plans or historical average values.

[0092] In the embodiment, the application dynamically corrects the initial value of the cruise remaining time by establishing a correlation model between the flight efficiency coefficient and the time prediction deviation, significantly improving the response capability to real-time running condition changes.

[0093] Further, the specific process of establishing a correlation model between the flight efficiency coefficient and the time prediction deviation is: according to the historical flight data of the direct flight, the predicted cruise remaining time and the corresponding actual cruise remaining time generated when the remaining time is predicted in the cruise stage in each historical flight process are obtained, the cruise remaining time prediction deviation of each prediction is calculated, and the corresponding flight efficiency coefficient of each prediction is extracted synchronously.

[0094] Based on the multiple sets of data composed of the cruise remaining time prediction deviation and the corresponding flight efficiency coefficient, a training data set is constructed.

[0095] Using the training data set, a regression model with the flight efficiency coefficient as input and the cruise remaining time prediction deviation as output is constructed by a machine learning algorithm, and the model is denoted as the correlation model between the flight efficiency coefficient and the time prediction deviation.

[0096] The arrival time prediction module is configured to predict the time length required for the descent approach stage and the landing taxi stage after the cruise stage according to the historical flight data of the flight, combine the corrected cruise remaining time length, calculate the predicted time length from the current time to the arrival of the aircraft at the designated parking position, and infer the arrival time of the aircraft at the designated parking position.

[0097] Further, the specific working process of the arrival time prediction module is as follows: according to the historical flight data of the direct flight, the time consumption of each historical flight in the descent approach stage and the landing taxi stage is counted, and the time length required for the current flight in the descent approach stage and the landing taxi stage after completing the cruise stage is predicted based on the historical time consumption data.

[0098] The corrected cruise remaining time length and the predicted time length required for the descent approach stage and the landing taxi stage are added to obtain the predicted total time length from the current time to the arrival of the aircraft at the designated parking position.

[0099] The predicted arrival time of the aircraft at the designated parking position is calculated based on the current time point and the predicted total time length.

[0100] Further, the specific process of predicting the time length required for the descent approach stage and the landing taxi stage is as follows: the historical time consumption data is statistically analyzed, the optimal fitting probability distribution model corresponding to the historical time consumption of the descent approach stage and the landing taxi stage is determined respectively by using the distribution goodness-of-fit test method, and the corresponding parameters of each distribution model are calculated.

[0101] Based on the determined probability distribution model and its parameters, the time consumption confidence interval of the descent approach stage and the landing taxi stage at a given confidence level is calculated respectively.

[0102] The median of the confidence interval is taken as the predicted value of the time length required for the descent approach stage and the landing taxi stage respectively.

[0103] The above formulas are dimensionless numerical calculations, and the formulas are obtained by collecting a large amount of data to simulate the most recent real situation, and the preset parameters in the formulas are set by the person skilled in the art according to the actual situation.

[0104] The above embodiments can be realized wholly or partially by software, hardware, firmware or any other combination. When realized by software, the above embodiments can be realized in the form of a computer program product wholly or partially.

[0105] Those skilled in the art can understand that the modules and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized in electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0106] In addition, each functional module in each embodiment of the present application can be integrated in one processing module, or each module can exist physically alone, or two or more modules can be integrated in one module.

[0107] The above is merely specific embodiments of the present application, but the protection scope of the present application is not limited thereto, any skilled in the art within the technical scope disclosed in the present application can easily think of changes or replacements, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0108] Finally, the above is only the preferred embodiment of the present application, and is not used to limit the present application, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the protection scope of the present application.

Claims

1. A real-time optimization management system based on flight track predicted arrival stand time, characterized in that, The method comprises the following steps: a remaining duration initial estimation module: according to the distance of the non-stop flight in the cruise phase and the actual flight duration, the duration required to complete the remaining distance is preliminarily estimated, and the initial value of the cruise remaining duration is obtained; a flight efficiency analysis module: based on the flight data of the cruise phase, flight efficiency factors are extracted from three dimensions of track directness, speed control and energy management, and each flight efficiency factor is weighted and fused to obtain the flight efficiency coefficient of the flight distance; a remaining duration correction module: the flight efficiency coefficient is input into the correlation model between the flight efficiency coefficient and the time prediction deviation, the cruise remaining duration compensation is obtained, and the cruise remaining duration is corrected combined with the initial value of the cruise remaining duration to obtain the corrected cruise remaining duration; an arrival time prediction module: according to the historical flight data of the flight, the time required for the descent approach phase and the landing taxi phase after the cruise phase is predicted, the estimated time from the current time to the arrival of the aircraft at the parking position is calculated combined with the corrected cruise remaining duration, and the arrival time of the aircraft at the parking position is inferred; the specific process of extracting the flight efficiency factor from the track directness dimension in the flight efficiency analysis module is as follows: the actual track of the non-stop flight in the cruise phase is obtained; the complete planned track of the non-stop flight is extracted from the database, and the track segment corresponding to the flight distance is intercepted as the planned track of the flight distance; the actual track and the planned track are compared to determine whether there is an intersection point; if there is no intersection point, the ratio of the length of the planned track to the length of the actual track is calculated, and the ratio is taken as the flight efficiency factor based on the track directness; if there is an intersection point, the following steps are performed: the actual track is divided into several sub-segments according to the number of intersection points; for each sub-segment, compare its length with the length of the corresponding segment in the planned track: if the actual sub-segment length is greater than the corresponding planned length, mark the sub-segment as a detour segment; if the actual sub-segment length is equal to the corresponding planned length, mark it as an equidistant segment; if the actual sub-segment length is less than the corresponding planned length, mark it as a straight flight segment; count the cumulative length and number of all straight flight segments; calculate the first ratio of the cumulative length of the straight flight segment to the total length of the actual track and the second ratio of the number of straight flight segments to the total number of actual track sub-segments; the first ratio and the second ratio are weighted and averaged, and the result is taken as the flight efficiency factor based on the track directness; the specific process of extracting the flight efficiency factor from the speed control dimension in the flight efficiency analysis module is as follows: in the cruise phase, a plurality of continuous time windows are divided according to a preset rule; the speed monitoring data of the non-stop flight in each time window is obtained, and the corresponding actual average ground speed in each time window is calculated; the planned flight speed data of the non-stop flight in the cruise phase is extracted from the database, and the corresponding planned average ground speed of each time window is determined accordingly; for each time window, the ratio of the actual average ground speed to the planned average ground speed is calculated; the ratios corresponding to all time windows are accumulated, and the arithmetic mean value is calculated, which is taken as the flight efficiency factor based on the speed control; The specific process of extracting flight efficiency factors from the energy management dimension in the flight efficiency analysis module is as follows: in the flown range of the cruise stage, a plurality of monitoring time points are selected according to the equal time interval rule; the altitude and true airspeed corresponding to each monitoring time point are extracted from the flight data of the direct flight in the flown range of the cruise stage, the energy altitude of each monitoring time point is calculated based on the energy altitude formula, and the energy altitude trend curve of the flown range of the cruise stage is drawn according to the time sequence; the cruise stage is divided into a gradient rising process, a stable cruising process and a slow descending process in turn; according to the energy altitude trend curve, the curve segments corresponding to each flight process are intercepted respectively; the linear fitting of each curve segment is performed to obtain the fitting straight line slope as the energy altitude change slope of the process; the energy altitude change slopes corresponding to each flight process are compared with the pre-stored energy altitude change expected slopes of the corresponding processes in the database, and the slope coincidence degrees of each process are calculated respectively; The arithmetic mean of the slope coincidence degrees of all flight processes is calculated, and the mean value is taken as the flight efficiency factor based on energy management. 2.The flight track based estimated time of arrival gate occupancy time real-time optimization management system of claim 1, wherein: The specific working process of the remaining time initial estimation module is as follows: Extracting the pre-stored planned total distance and planned total time of the non-stop direct flight in the cruise phase from the database, calculating the planned time per unit distance in the cruise phase ; acquire a flight distance of the direct flight from a cruise phase start position to a current position as a distance of a cruise phase flown distance , and acquire an actual flight duration corresponding to the flown distance ; According to the distance of the flown distance The planned time of the unit distance , the planned flight time corresponding to the flown distance is calculated Wherein ; According to the flight path information of the flown range of the cruise stage, the position to be reached by the direct flight at the end of the cruise stage is predicted as the cruise stage end position; obtaining a flight distance of the non-stop flight from the current position to the end position of the cruise phase as a distance of the remaining range of the cruise phase ; The initial estimate of the remaining duration of the cruise phase is calculated based on the following formula : ； wherein, represents the deviation amount of the actual flight time per unit distance in the flown distance from the planned flight time.

3. The flight track based estimated time of arrival stand occupancy time real-time optimization management system of claim 1, wherein: The specific process of obtaining the flown range flight efficiency coefficient in the flight efficiency analysis module is as follows: The weights of the flight efficiency factors based on the track directness, speed control and energy management are extracted from the database; The numerical values of the flight efficiency factors are linearly weighted and fused with the corresponding weights to obtain the flight efficiency coefficient of the flown range.

4. The flight track based estimated time of arrival stand assignment time real-time optimization management system of claim 1, wherein: The specific working process of the remaining time correction module is as follows: According to the historical flight data of the direct flight, a correlation model between the flight efficiency coefficient and the time prediction deviation is established; The flight efficiency coefficient of the current flown range is input into the correlation model to obtain the corresponding cruise remaining time compensation amount; The initial value of the cruise remaining time is algebraically added to the compensation amount to obtain the corrected cruise remaining time.

5. The flight track based estimated time of arrival stand assignment real-time optimization management system of claim 4, wherein: The specific process of establishing the correlation model between the flight efficiency coefficient and the time prediction deviation is as follows: According to the historical flight data of the direct flight, the predicted cruise remaining time and the corresponding actual cruise remaining time generated when the remaining time is predicted in the cruise stage in each historical flight process are obtained, the cruise remaining time prediction deviation of each prediction is calculated, and the corresponding flight efficiency coefficient is extracted synchronously; Based on the plurality of groups of data composed of the cruise remaining time prediction deviation and the corresponding flight efficiency coefficient, a training data set is constructed; By using the training data set, a regression model taking the flight efficiency coefficient as the input and taking the cruise remaining time prediction deviation as the output is constructed by a machine learning algorithm, and the model is recorded as the correlation model between the flight efficiency coefficient and the time prediction deviation.

6. The flight track based estimated time of arrival stand assignment time real-time optimization management system of claim 1, wherein: The specific working process of the arrival time prediction module is as follows: According to historical flight data of the direct flight, time consumption of each flight in the descent approach phase and the landing taxiing phase is counted, and based on the historical time consumption data, time length required by the current flight in the descent approach phase and the landing taxiing phase after completing the cruising phase is predicted; The corrected cruising remaining time length and the predicted time length required in the descent approach phase and the landing taxiing phase are added to obtain a predicted total time length from the current time to the arrival of the aircraft at the specified entry point; The predicted total time length and the current time point are combined to calculate the predicted arrival time of the aircraft at the entry point.

7. The flight track based estimated time of arrival stand occupancy time real-time optimization management system of claim 6, wherein: The specific process of predicting the time length required in the descent approach phase and the landing taxiing phase is as follows: The historical time consumption data is statistically analyzed, and a distribution goodness-of-fit test method is used to determine the optimal fitting probability distribution model corresponding to the historical time consumption in the descent approach phase and the landing taxiing phase, respectively, and the corresponding parameters of each distribution model are calculated; Based on the determined probability distribution model and the parameters, the time consumption confidence interval of the descent approach phase and the landing taxiing phase under a given confidence level is calculated, respectively; The median of the confidence interval is taken as the predicted value of the time length required in the descent approach phase and the landing taxiing phase, respectively.

Images