An aircraft vertical trajectory prediction and optimization method based on a combination model

Abstract

The application discloses a kind of based on combination model's aircraft vertical trajectory prediction and optimization method, the state parameter initialization of current position of aircraft;According to the judgment logic of vertical flight phase, obtain the vertical flight phase where aircraft is currently located, and the vertical trajectory of the whole flight process is calculated in sections according to the processing logic of each vertical flight phase;Based on the flight characteristics of each vertical flight phase and the proportion in the whole flight process, the calculation accuracy and the calculation efficiency are evaluated and analyzed, and the method of selectively using performance database model and first principle model is selected for different flight phases, to form a kind of best combination model, realize the prediction and optimization of the trajectory parameter of the whole flight vertical profile of aircraft from take-off airport or current position to destination airport or alternate airport.The application can effectively improve the calculation accuracy of vertical trajectory prediction and optimization, and provide aircraft state information with higher reliability for each subsequent waypoint prediction of flight plan.

Description

Technical Field

[0001] This invention relates to a method for predicting and optimizing the vertical trajectory of an aircraft in a flight management system. More specifically, this invention relates to a method for predicting and optimizing the vertical trajectory of an aircraft based on a combined model. Background Technology

[0002] With globalization and the booming development of the global economy, the air transport industry in various countries has also developed rapidly. As route networks become increasingly dense and the number of flights increases year by year, airspace traffic is constantly growing, and airspace congestion is becoming increasingly prominent. To address the contradiction between the current scarcity of airspace resources and the needs of civil aviation development, 4D trajectory management is an effective solution for alleviating the congestion in complex airspace. Aircraft vertical trajectory prediction and optimization methods are the core technologies for solving this problem.

[0003] Most aircraft are now equipped with flight management systems, which effectively assist flight crews in performing various flight tasks such as flight planning, integrated navigation, trajectory prediction, flight guidance, and performance calculation. Trajectory prediction can be divided into horizontal trajectory prediction and vertical trajectory prediction, with vertical trajectory prediction and optimization being a crucial sub-function of the flight management system. Based on the aircraft's future flight intentions, and according to the various aircraft states at the start of the horizontal flight plan and vertical trajectory prediction, the system performs rapid simulated flight periodically, using a certain vertical cross-sectional area step size and taking the state at the end of the previous integration as the state at the start of the next integration. The vertical trajectory prediction function can calculate a complete 4D flight trajectory along a specified flight plan. The vertical trajectory is continuous from the departure airport or the aircraft's current position to the destination airport, defined as a series of spatial points followed from the departure airport or the aircraft's current position to the destination airport, where each point is defined by geographical location, altitude, and time. The vertical trajectory prediction function predicts the flight phase, remaining flight distance, estimated arrival time, speed, altitude, remaining fuel and total weight of each subsequent waypoint in the flight plan. It also includes calculated vertical event points, i.e. pseudo waypoints, such as T / C and T / D points.

[0004] Vertical trajectory optimization, also known as performance optimization, optimizes an aircraft's altitude, velocity, and thrust profiles while considering aircraft performance limitations, pilot input constraints, and civil airspace flight rules to meet specific mission objectives. This optimization minimizes total flight time, fuel consumption, and total flight cost. The optimal flight profile represents a trade-off between time and fuel costs, typically expressed using a cost index. The cost index input generally ranges from 0 to 999, with a cost index of 0 corresponding to minimal fuel consumption. Increasing the cost index leads to increased range speed and changes in the aircraft's vertical trajectory. At an appropriate cost index, the increased fuel cost is offset by the reduced time cost. The aviation industry is committed to reducing fuel costs and emissions from fuel consumption, as fuel consumption is directly proportional to pollution levels. Therefore, vertical trajectory optimization is an important solution for reducing environmental emissions and can effectively help mitigate the greenhouse effect.

[0005] The vertical trajectory prediction and optimization functions calculate the optimal flight trajectory based not only on the optimization indices but also on the mathematical model used to predict aircraft performance. Therefore, establishing a mathematical model of the aircraft that closely approximates actual flight is crucial and necessary. Methods for mathematical modeling aircraft typically include performance database-based models and first-principles models. Performance databases are generally stored in the aircraft's non-volatile memory, with data presented in discrete form, typically composed of data tables, polynomials, or other convenient methods for representing data. They store aerodynamic, engine, and performance data. Performance database models access the database based on specific data query conditions and output the desired results through interpolation. First-principles models typically use raw model parameters such as lift, drag, and engine thrust, based on the fundamental equations of flight dynamics for each flight phase, to calculate the required flight performance data in real time. First-principles models start directly from established fundamental physical laws, without relying on historical experience models or parameter fitting methods. Their data calculation results are continuously changing, thus having a clear advantage in higher computational accuracy compared to interpolation calculations of performance database models. However, the calculation process of first-principles models may involve iterative or recursive logic, which introduces more computational steps, thus having the disadvantage of a relatively longer computation cycle.

[0006] For practical engineering applications, a comprehensive evaluation and analysis of computational accuracy and efficiency is required. This method fully considers the trade-off between computational accuracy and efficiency, fully analyzes the attributes of each required performance parameter, and selectively uses first-principles model and performance database model methods for different flight stages based on the degree of influence of each performance parameter in trajectory prediction and performance calculation. The first-principles model method is selected as the processing method for the main flight stages in the vertical trajectory prediction and optimization process, forming an optimal combined model that can effectively improve computational accuracy without seriously affecting computational efficiency, thereby improving the overall level of vertical trajectory prediction and optimization. Summary of the Invention

[0007] The purpose of this invention is to provide a method for predicting and optimizing aircraft vertical trajectory based on a combined model. This method fully considers the trade-off between computational accuracy and efficiency, selectively using first-principles models and performance database models for different flight phases. Based on the optimal combined model, it not only improves the computational accuracy of vertical trajectory prediction and optimization without significantly affecting the computational efficiency of the process, but also periodically predicts future flight conditions during flight, providing highly reliable alerts. This can effectively reduce flight costs, increase airline profits, and reduce the workload of flight crews. Based on reference data, it can improve the economy, comfort, and safety of flight.

[0008] The objective of this invention is achieved through the following technical solution:

[0009] A method for predicting and optimizing aircraft vertical trajectory based on a combined model includes the following steps:

[0010] S101. Initialize the status parameters of the aircraft's current position;

[0011] S102. Based on the vertical flight phase judgment logic, obtain the current vertical flight phase of the aircraft. If it is in the pre-flight or takeoff phase, start execution from step S103; if it is in the climb phase, start execution from step S104; if it is in the cruise phase, start execution from step S105; if it is in the descent phase, start execution from step S109; if it is in the advanced phase, start execution from step S110.

[0012] S103. Vertical trajectory prediction during takeoff is processed using a performance database model;

[0013] S104. The vertical trajectory prediction for the climb segment is processed using a first-principles model.

[0014] S105. The vertical trajectory prediction for the cruise segment is processed using a first-principles model.

[0015] S106. If a change in the flight plan is found during the processing of the cruise segment, proceed to step 107; otherwise, proceed to step 109.

[0016] S107. Determine if the bottom point of the decline exists. If it does not exist, proceed to step 110. If it exists, proceed to step 108.

[0017] S108. Construct the descent path;

[0018] S109. The vertical trajectory prediction for the descent segment is processed using a first-principles model.

[0019] S110. For vertical trajectory prediction during the approach phase, a performance database model is used for processing.

[0020] S111. If the go-around phase is activated, the vertical trajectory prediction for the go-around phase is processed using a performance database model; otherwise, proceed to step S113.

[0021] S112. Vertical trajectory prediction for alternate landing routes is processed using a performance database model;

[0022] S113. Based on the vertical 4D trajectory generated by the vertical trajectory prediction at each stage and the optimization index, the vertical flight profile of the entire process is optimized.

[0023] Preferably, in S102, the judgment logic for the vertical flight phase is based on the aircraft's air-to-ground status, guidance mode, distance from the current location to the destination airport, and the relationship between the current altitude and the cruising altitude.

[0024] Preferably, in S103, the takeoff phase is regarded as an integral segment, and an integral calculation is performed according to the altitude integral step size. The unconsumed portions of time, fuel, and distance are added to the initial values ​​of the state parameters. Waypoints falling within the integral segment are interpolated from the performance database to determine the vertical 4D trajectory profile of each waypoint in the takeoff phase.

[0025] Preferably, in S104, a climb phase consists of several segments, including: an initial climb phase that meets airport speed limits; a climb acceleration phase that accelerates from the airport speed limit to the planned indicated airspeed corresponding to the climb speed mode, or a climb acceleration phase generated due to waypoint speed constraints; a climb at a constant planned indicated airspeed until reaching the crossing altitude; a constant Mach number climb phase at an altitude above the crossing altitude until reaching cruise altitude; and may also include a cruise level flight phase generated due to waypoint altitude constraints or airport speed limits.

[0026] The vertical trajectory prediction for the climb segment processes each segment falling into the climb phase sequentially. When predicting the vertical trajectory for any segment in the climb phase, the segment is processed as a vertical integral segment. If a segment involves multiple integral segments, it reverts to the segment termination position when a segment termination is encountered within an integral segment, thus coupling the horizontal and vertical profiles. Based on a complex first-principles model, the changes in time, distance, fuel consumption, and altitude within an integral segment are calculated. The altitude integration step size or time integration step size is adopted according to the change in the aircraft's climb rate, and the integration process continues until the predicted climb peak is reached. This vertical trajectory prediction considers the speed constraints, altitude constraints, time constraints, and airport speed limits at waypoints, as well as the acceleration process during the climb. When encountering waypoint altitude constraints as part of the vertical flight plan, these constraints take precedence over the predicted climb vertical profile.

[0027] Preferably, in S105, a cruise phase consists of several segments, including: a cruise acceleration segment from climb speed to cruise speed; a cruise segment with constant indicated airspeed or constant Mach number; a stepped climb segment and a stepped descent segment; and a cruise deceleration segment from cruise speed to descent speed.

[0028] The vertical trajectory prediction during the cruise phase processes each segment falling within the cruise phase sequentially. When predicting the vertical trajectory for any segment during the cruise phase, the segment is processed as a vertical integral segment. When a segment terminates within an integral segment, the process reverts to the segment termination position, thus coupling the horizontal and vertical profiles together. Based on a complex first-principles model, the cruise fuel flow within an integral segment is calculated using any one of three integration steps: weight, distance, or time. The changes in time, distance, and fuel consumption within an integration step are then calculated. The integration process continues until the prediction reaches the descent peak.

[0029] Preferably, in S108, the descent path is divided into an idle path segment and a geometric path segment. The idle path segment is constructed based on the idle thrust or near-idle thrust and the speed plan used to optimize the operation, starting from the descent bottom point and working backwards in the opposite direction to the flight to the cruise altitude, and the position point for entering the descent phase from the cruise altitude is determined in advance. The geometric path segment is constructed based on altitude constraints, speed constraints, airport altitude restrictions, and flight path angle constraints.

[0030] Preferably, in S109, the descent phase consists of several segments, including: a constant Mach number descent segment flying at the planned Mach number; a constant indicated airspeed descent segment flying at the planned indicated airspeed; a descent acceleration segment decelerating from the planned indicated airspeed corresponding to the descent speed mode to the airport speed limit speed, or due to waypoint speed constraints; a descent deceleration segment; and a constant indicated airspeed segment continuing to descend at the airport speed limit speed to the start of the approach phase; and may also include a cruise level flight segment generated due to waypoint altitude constraints or airport speed limit altitude.

[0031] The vertical trajectory prediction for the descent segment processes each segment falling into the descent phase sequentially. When predicting the vertical trajectory for any segment of the descent segment, the segment is processed as a vertical integral segment. When a segment termination is encountered within an integral segment, the process reverts to the segment termination position, thus coupling the horizontal and vertical profiles together. Based on a complex first-principles model, the changes in time, distance, fuel consumption, and altitude within an integral segment are calculated, using an altitude integration step size. The integration process continues until the predicted start point of the approach phase is reached. This vertical trajectory prediction considers the speed constraints, altitude constraints, time constraints, and airport speed limits at waypoints, as well as the deceleration process during descent. When encountering waypoint altitude constraints, these constraints take precedence over the predicted descent vertical profile.

[0032] Preferably, in S110, the approach phase is divided into two parts according to altitude intervals, that is, the approach phase is regarded as multiple integration segments, and the vertical trajectory prediction value of the waypoints falling within the integration segments is determined by interpolation.

[0033] Preferably, in S112, the vertical trajectory prediction of the alternate landing route considers both go-around and direct landing types. Based on the alternate landing flight plan and the alternate landing cruise altitude, the vertical flight phase that should be included in the alternate landing flight plan is determined. The entire alternate landing flight plan uses a performance database model to calculate the changes in time, distance, fuel, and altitude of the alternate landing route, as well as the predicted information of the alternate airport.

[0034] The beneficial effects of this invention are as follows:

[0035] a) This invention provides a high-precision vertical trajectory prediction and optimization method for aircraft based on a combined model. It fully considers atmospheric wind and temperature models, as well as various constraints and limitations. The predicted vertical trajectory shows a high degree of fit with the aircraft's actual flight trajectory. Based on the optimal combined model, it not only improves the computational accuracy of vertical trajectory prediction and optimization but also does not significantly affect the computational efficiency of the process. Furthermore, it periodically predicts future flight states during flight, providing highly reliable aircraft state information for each subsequent waypoint in the flight plan. This effectively assists the flight crew in making decisions and improves flight safety. Accurate prediction information also helps in the advance arrangement of air traffic control, effectively improving airspace utilization efficiency.

[0036] (b) This invention provides a high-precision vertical trajectory prediction and optimization method for aircraft based on a combined model. Based on a certain optimization index, the velocity profile, altitude profile and thrust profile of the entire flight process can be continuously optimized, which can effectively reduce flight costs, increase airline profits, reduce the workload of flight crews, and improve the economy, comfort and safety of flight based on reference data. Attached Figure Description

[0037] Figure 1 This is the top-level logic flowchart of the present invention used to control the entire vertical trajectory prediction process.

[0038] Figure 2 This is a low-level logic flowchart for the next level of control segment processing in this invention.

[0039] Figure 3 This is a typical 4D vertical flight profile diagram and an example of the flight phase provided by the present invention.

[0040] Figure 4 This is a schematic diagram of the vertical profile of the climb considering waypoint height and speed constraints, as provided in this invention.

[0041] Figure 5 This is a schematic diagram of a cruise vertical profile including planned steps and optimal steps, provided by the present invention.

[0042] Figure 6 This is a schematic diagram of the descent path constructed according to the present invention.

[0043] Figure 7 This is a schematic diagram illustrating the division of a region with uniform properties, as provided in this invention, into separate integration intervals. Detailed Implementation

[0044] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments.

[0045] See Figure 1As shown in the figure, the aircraft vertical trajectory prediction and optimization method based on a combined model, as illustrated in this embodiment, includes the following steps:

[0046] S101. Initialize the state parameters of the aircraft's current position. To start the vertical trajectory prediction and optimization function, the values ​​of the state parameters to be integrated must be initialized, including the necessary prerequisite data and data that can improve accuracy, to reflect the aircraft's state at the start of each vertical trajectory prediction.

[0047] Initialization parameters include horizontal flight plan data, human-machine interface (HMI) input data, sensor data, navigation data, and other calculated data. Horizontal flight plan data includes the main flight plan and alternate flight plan segment data; HMI input data mainly includes cruise altitude, zero-fuel weight, cost index, and the selected speed mode and performance coefficient for the climb, cruise, and descent phases; sensor data mainly includes actual measured wind and temperature data; navigation data includes current position, pressure-corrected altitude, and standard atmospheric altitude; other calculated data includes initial position and altitude, total distance to the destination airport considering the circular transition between segments, system time, total fuel weight, and total aircraft weight, etc. The difference in calculated data between active and inactive waiting periods for the current segment needs to be considered.

[0048] Starting from the aircraft's current position, the prediction process accumulates changes to these variables until the integration process ends. One integration step in vertical trajectory prediction is defined as an integration segment, or an incrementing unit. Each incrementing unit typically consists of four basic parameters, the values ​​of which are updated at the beginning and end of each prediction integration segment. A prediction integration segment is defined as the incrementing unit or interval that can be predicted. Initial conditions are established at the beginning of an integration segment, and vertical trajectory prediction establishes termination or end conditions for that segment. The vertical trajectory consists of a series of such integration segments connected end-to-end. The termination condition of one integration segment becomes the initial condition for the next. The four basic parameters are:

[0049] • T = Duration, usually measured in seconds (s)

[0050] W = Total weight of the aircraft, usually expressed in pounds (lbs).

[0051] H = Height, usually expressed in feet (ft).

[0052] • X = Distance to the destination airport, usually expressed in nautical miles (nm).

[0053] For each set of parameters mentioned above, a joint integral can be applied, where the change within the predicted integration segment needs to be defined as follows:

[0054] ·dT = time change of the integral segment

[0055] ·dW = weight change of the integral segment

[0056] ·dH = Change in height of the integral segment

[0057] ·dX = Change in distance of the integral segment

[0058] Given an integration segment with an initial condition referenced "i" and a termination condition referenced "f", the following relation exists for each integration segment:

[0059] T f =T i +dT W f =W i -dW H f =H i ±dH X f =X i -dX

[0060] S102. Based on the vertical flight phase judgment logic, obtain the current vertical flight phase of the aircraft. If it is in the pre-flight or takeoff phase, start execution from step S103; if it is in the climb phase, start execution from step S104; if it is in the cruise phase, start execution from step S105; if it is in the descent phase, start execution from step S109; if it is in the advanced phase, start execution from step S110.

[0061] Based on the data required for vertical trajectory prediction, and the sequence and transition relationship of vertical flight stages throughout the entire flight process, the vertical 4D flight profile of the entire flight process can be calculated segment by segment according to the processing logic of each vertical flight stage. Based on the flight characteristics of each vertical flight stage and its proportion in the entire flight process, a comprehensive evaluation and analysis of calculation accuracy and efficiency is carried out. For different flight stages, the performance database model and the first-principles model are selectively used to form an optimal combination model.

[0062] The judgment logic for the vertical flight phase is mainly based on the aircraft's air-to-ground status, the guidance mode selected on the flight panel, the distance between the current position and the destination airport, and the relationship between the current altitude and the cruising altitude.

[0063] The vertical flight phase refers to a certain period of flight, which mainly includes eight flight phases: pre-flight, takeoff, climb, cruise, descent, approach, go-around, and post-flight. The transition between each flight phase is handled according to the planned flight distance and the aircraft's flight intention. For example, if the planned total flight distance is short and the cruise altitude is high, without considering the minimum cruise flight time, there may be a direct transition from the climb phase to the descent phase. The pre-flight phase refers to the phase before flight, during which the aircraft is on the ground, completes system power-up and engine start-up, connects various avionics devices, pushes the aircraft off the apron, and taxis to the runway threshold; the takeoff phase involves retracting flaps and landing gear, accelerating to a stable speed; the climb phase extends from the end of the takeoff phase to cruise altitude, i.e., the T / C point; the cruise phase extends from the end of the climb phase to the beginning of the descent phase, i.e., the T / D point, including cruise step climb and step descent; the descent phase involves the aircraft leaving the designated cruise altitude and gradually decreasing altitude, extending to the start point of the approach phase; the approach phase extends from the initial flap deployment point to the landing point; the go-around phase begins at the go-around point or the position where the flight crew initiates the go-around and terminates at the altitude constraints defined in the go-around procedure; the post-flight phase generally refers to the taxiing process after completing the flight and landing at the destination airport. Figure 3 The image shows a typical 4D vertical flight profile of an aircraft, which provides indications of the flight phases and the speed and altitude constraints at waypoints and airport speed limits that may be encountered in the vertical profile.

[0064] S103. Vertical trajectory prediction during takeoff is processed using a performance database model.

[0065] Because the takeoff phase involves low speed, short duration, and relatively small fuel consumption compared to the overall flight, this method treats the takeoff phase as a single unit, without further subdivision. The vertical trajectory prediction during takeoff is based on a simple performance database model, calculating the changes in time, fuel consumption, distance, and altitude. The reason for not using a complex first-principles model is that, after comprehensive analysis and multiple experimental comparisons, a trade-off between computational accuracy and efficiency is considered. Furthermore, the vertical trajectory prediction and optimization functions are periodically updated, and a certain degree of accuracy loss in the initial prediction phase is acceptable. Moreover, this accuracy loss disappears after the takeoff phase and does not accumulate in subsequent flight phases. This process involves interpolating from the performance database to predict the acceleration altitude input from the start of the takeoff prediction to 1500ft above the default runway (or airport) elevation, which is the altitude at the start of the climb phase. Therefore, during the takeoff phase, an integral calculation is performed according to the altitude integral step size, and the unconsumed portions of time, fuel, and distance are added to the initial values ​​of the state parameters. Waypoints falling within the integral segment are interpolated from the performance database to determine the vertical 4D trajectory profile of each waypoint during the takeoff phase.

[0066] S104. The vertical trajectory prediction for the climb segment is processed using a first-principles model.

[0067] A complete climb typically consists of several segments, and a single segment may span two or more flight phases. The initial climb phase is to meet airport speed restrictions, which are altitude-dependent speed limits below a certain altitude; for example, the speed cannot exceed 250 knots below 10,000 ft. The next phase is an acceleration phase from the airport speed limit to the planned indicated airspeed corresponding to the climb speed pattern, or a climb acceleration phase generated due to waypoint speed constraints. The next phase is a climb at a constant planned indicated airspeed until reaching the crossover altitude, where the crossover altitude is the altitude at which the indicated airspeed's vacuum speed equals the Mach number's vacuum speed. Above the crossover altitude, the speed must be changed from indicated airspeed to Mach number, resulting in a constant Mach number climb phase until reaching cruising altitude. The final phase may also include a cruising phase generated due to waypoint altitude constraints or airport speed restrictions.

[0068] Vertical trajectory prediction for the climb segment is typically based on a climb vertical profile at a specified climb altitude and velocity, using methods such as... Figure 2The illustrated logic process sequentially processes each segment falling into the climb phase. When predicting the vertical trajectory for any segment during the climb phase, the segment undergoes vertical integration. A single segment may involve multiple integration segments. When a segment terminates within an integration segment, the process reverts to the segment termination position; therefore, the horizontal and vertical profiles are coupled. Based on a complex first-principles model, the changes in time, distance, fuel consumption, and altitude within an integration segment are calculated. Depending on the aircraft's rate of climb, either an altitude integration step or a time integration step is used. The integration process continues until the predicted climb peak (T / C point) is reached, i.e., the cruise altitude. This process considers waypoint speed constraints, altitude constraints, time constraints, and airport speed limits, while also taking into account acceleration during the climb. Descending behavior during the climb should be prevented. When encountering waypoint altitude constraints as part of the vertical flight plan, these constraints take precedence over the optimal climb vertical profile.

[0069] Figure 4 This is a schematic diagram of a typical vertical profile of a climb that takes into account the speed and altitude constraints at waypoints during the climb phase.

[0070] S105. The vertical trajectory prediction for the cruise segment is processed using a first-principles model.

[0071] A complete cruise typically consists of several segments: a cruise acceleration segment from climb speed to cruise speed; a cruise segment at constant indicated airspeed or Mach number; a stepped climb segment and a stepped descent segment; and a cruise deceleration segment from cruise speed to descent speed. Vertical trajectory prediction for each cruise segment is usually based on the optimal speed vertical profile at a specified cruise altitude and provides one or more pre-planned cruise steps and a calculated optimal cruise step. Stepped climbs and descents can be pre-planned by the flight crew to perform stepped climbs or descents at designated waypoints, or the optimal step position and altitude can be calculated by flight management functions to change the optimal step for cruise altitude, where the optimal step position is determined based on the aircraft's weight. The termination checks and handling of a cruise segment should include a check to determine whether the calculated optimal step climb point or the planned step climb point has been reached.

[0072] Considering the stepped ascent and descent, as well as the acceleration and deceleration processes that may occur at the start and end of cruise, use, for example Figure 2The logical process shown sequentially processes each segment falling into the cruise phase. When predicting the vertical trajectory for any segment in the cruise phase, the segment is processed as a vertical integral segment. When a segment terminates within an integral segment, the process reverts to the segment termination position, coupling the horizontal and vertical profiles together. Based on a complex first-principles model, the cruise fuel flow within an integral segment is calculated, using any one of three integration steps: weight, distance, or time. The changes in time, distance, and fuel consumption within an integral step are then calculated. The integration process continues until the prediction reaches the descent peak (T / D point), i.e., the prediction for the descent phase begins. Stepped ascent and descent can be considered as altitude constraints during the cruise process, and a similar vertical trajectory prediction method as the ascent and descent phases can be used. Therefore, the process of stepped ascent and descent is incorporated into the trajectory prediction and optimization of the cruise phase.

[0073] Figure 5 This is a schematic diagram of a typical cruise vertical profile that includes the planned cruise steps and the calculated optimal cruise steps.

[0074] S106. If a change in the flight plan is detected during the processing of the cruise segment, proceed to step 107; otherwise, proceed to step 109.

[0075] S107. Determine if the bottom point (B / D point) exists. If it does not exist, proceed to step 110. If it exists, proceed to step 108.

[0076] S108. Constructing the Descent Path. The descent path is a pre-planned descent profile designed to ensure a perfect landing at the destination airport during the descent and approach phases. The calculated descent profile is a sequence of paired distances and altitudes to the destination airport, referenced to the ground. The vertical trajectory prediction and optimization function should refer to this descent path as the planned vertical profile for the descent phase, while the vertical guidance function uses this reference path as the target path for vertical navigation during the descent phase. The descent path generally consists of an idle path segment and a geometric path segment. It should be constructed based on idle thrust or near-idle thrust and a speed plan for optimized operations, starting from the B / D point and working backwards in the opposite direction of flight to the cruise altitude. The entry point from cruise altitude into the descent phase, i.e., the T / D point, can be determined in advance. The idle path segment, also known as the performance path segment, is unconstrained and requires no ATC intervention. The aircraft can perform continuous descent operations (CDO) within this segment, achieving fuel / time efficient descent maneuvers. The geometric path segment is primarily constructed based on altitude constraints, speed constraints, airport altitude restrictions, and flight path angle constraints. Its main characteristic is that altitude and distance satisfy a certain angular relationship, but the idle path segment does not have this significant characteristic. If there are no altitude constraints or airport speed restrictions during the descent phase, the entire descent path is constructed based on the idle path segment. The constructed descent path should be flyable by the aircraft; if this is not possible, appropriate instructions should be provided to the flight crew. Figure 6 The diagram shown is a typical completed descent path.

[0077] S109. The vertical trajectory prediction for the descent segment is processed using a first-principles model.

[0078] Vertical trajectory prediction along the descent path needs to consider anomalies such as aircraft deviation from the descent path. This is primarily because the aircraft may descend prematurely or delayed, meaning the initial descent position may not be as expected. This includes two scenarios: the aircraft being above or below the descent path. The simulation aims to demonstrate how the vertical guidance function can reacquire and track the descent path. The descent phase is similar to the climb phase, but in a reverse sense, and generally consists of several sub-phases: a constant Mach number descent phase following the planned descent Mach number; a constant airspeed descent phase following the planned indicated airspeed; a descent acceleration phase from the planned indicated airspeed corresponding to the descent speed pattern to the airport speed limit speed, or due to waypoint speed constraints; a descent deceleration phase; and a constant airspeed phase continuing the descent at the airport speed limit speed to the start of the approach phase; it may also include a cruise phase due to waypoint altitude constraints or airport speed limit altitudes.

[0079] Vertical trajectory prediction for the descent segment uses, for example Figure 2 The illustrated logic process sequentially handles each segment falling into the descent phase. When predicting the vertical trajectory for any segment of the descent phase, the segment undergoes vertical integration. Upon encountering a segment termination within an integration segment, the process reverts to the segment termination position, coupling the horizontal and vertical profiles together. Based on a complex first-principles model, the changes in time, distance, fuel consumption, and altitude within an integration segment are calculated, typically using an altitude integration step size. The integration process continues until the predicted start point of the approach phase is reached. This process considers waypoint speed constraints, altitude constraints, time constraints, and airport speed limits, as well as the deceleration process during descent. Climbing behavior during descent should be prevented; when encountering waypoint altitude constraints, these constraints take precedence over the optimal descent profile.

[0080] Figure 6 It is not only a schematic diagram of a completed descent path, but also essentially a vertical flight profile of the aircraft as it flies along the descent path.

[0081] S110. For vertical trajectory prediction during the approach phase, a performance database model is used for processing.

[0082] The approach phase is characterized by low speed, short duration, and relatively low fuel consumption compared to the overall flight. However, considering the significant deceleration during approach, simple processing can directly impact the vertical trajectory prediction information for the destination airport. Therefore, this method divides the approach phase into multiple integral segments based on altitude intervals. Waypoints falling within these segments are interpolated to determine their predicted vertical trajectory values. This method uses a simple performance database model to calculate the changes in time, distance, fuel consumption, and altitude within an integral segment. The reason for not using a complex first-principles model is based on comprehensive analysis and multiple experimental comparisons, considering a trade-off between computational accuracy and efficiency. Furthermore, the approach phase is not treated as a single integral segment like the takeoff phase; it is already segmented. Given the relatively short distance and flight duration of the approach phase, and the periodic updates to the vertical trajectory prediction and optimization functions, the accuracy loss caused by using the performance database model remains within acceptable limits as the flight progresses. The integration process continues until the predicted arrival at the destination airport, and the predicted value at the end of the integration is the predicted information for the destination airport.

[0083] S111. If the go-around phase is activated, the vertical trajectory prediction for the go-around phase is processed using a performance database model; otherwise, proceed to step S113.

[0084] Since the go-around phase requires activation by the flight crew, no predictions are provided before activation. Typically, the prediction for the go-around phase begins at the go-around point or the flight crew's initial position when initiating the go-around and terminates at the altitude constraints defined in the go-around procedure. The processing of vertical trajectory prediction during the go-around phase is similar to that of the takeoff phase, using a performance database model to process the go-around phase (i.e., one integral step), which will not be elaborated further.

[0085] S112. Vertical trajectory prediction for alternate landing routes is processed using a performance database model.

[0086] The alternate landing route handling considers two types of alternate landings: go-around and direct flight. Based on the alternate landing flight plan and the alternate landing cruise altitude, the vertical flight phase that should be included in the alternate landing flight plan is determined. The entire alternate landing flight plan uses a performance database model to calculate the changes in time, distance, fuel, and altitude of the alternate landing route, as well as the predicted information of the alternate airport. The reason for using a performance database model to calculate the entire alternate landing route, rather than a first-principles method, is mainly because the alternate airport is generally selected to be closer to the destination airport, that is, the alternate landing flight plan distance is relatively short, and the flight time of the alternate landing route is relatively short. Considering the trade-off between calculation accuracy and calculation efficiency, after comprehensive analysis and experimental comparison, the performance database model is sufficient to support the required vertical trajectory prediction accuracy.

[0087] S113. Based on the vertical 4D trajectory generated by the vertical trajectory prediction at each stage and the optimization index, the vertical flight profile of the entire process is optimized.

[0088] Based on the optimization indices for the vertical trajectory, and considering aircraft performance limitations, waypoint constraints, and civil airspace flight rules, this method calculates recommended aircraft state parameters using a first-principles model or performance database model to optimize the vertical flight profile to meet mission objectives. The optimized state parameters are then used as the initial state parameters for the aircraft in subsequent periodic updates of the vertical trajectory prediction process, thus optimizing the vertical flight profile throughout the entire process.

[0089] Figure 2 This is a low-level logic flowchart for the next level of control segment processing in this invention. It is a general method for predicting the vertical trajectory during the three flight phases of climb, cruise, and descent. It illustrates that this method divides the vertical trajectory of the three flight phases into regions of uniform nature and processes them at single integral intervals. When the aircraft is in any of the three flight phases, short-term vertical trajectory prediction based on the current vertical mode should be considered. This prediction is used to correlate with the vertical guidance function and to provide a reference for vertical guidance.

[0090] S201. Obtain data for the next flight segment. The vertical trajectory prediction logic for the three flight phases (climb, cruise, and descent) is processed according to the flight plan segments. This means that a flight phase may contain multiple segments, and a single segment may span two or more flight phases. Therefore, the segment processing is coupled horizontally and vertically, requiring the horizontal and vertical profiles to be correlated. The primary task of controlling the logic flow of segment processing is to retrieve the data for the next flight segment from the buffer storing the flight plan segments.

[0091] S202. Set Termination Conditions. This procedure selects the termination conditions to apply to the segment acquired in S201. "Termination" is a general term encompassing the following two situations, defined as:

[0092] a) The end of the same attribute region (i.e., the integral type changes) or

[0093] b) Points where prediction data needs to be saved (e.g., waypoints, T / C points, T / D points, speed change points, optimal cruise step points, and other vertical event points).

[0094] For setting termination conditions, both horizontal and vertical termination points should be defined before integration begins to couple horizontally and vertically. After interpolating an integration segment or termination point, one or both of the horizontal and vertical terminations will be sorted according to the next valid value. A horizontal termination is defined as the distance to the destination airport; a vertical termination is generally defined by the vertical segment type and associated with the integration segment in the form of an altitude value. A vertical termination may span multiple horizontal terminations until the accumulated integral value reaches the vertical termination criterion; conversely, a horizontal termination may also span multiple vertical terminations until the accumulated integral value reaches the horizontal termination criterion. For example, when a waypoint with altitude constraints arrives at the climb or descent segment integration process, the horizontal and vertical terminations are sorted simultaneously.

[0095] S203. Selecting the Integration Type: This process evaluates the type of the uniform property region to be integrated and determines the required integration type. For vertical trajectory prediction and optimization, as mentioned above, the first partition divides the flight into flight phases; takeoff and go-around phases are not further divided like other phases. While the approach phase is divided into integration segments, it is a relatively special partitioning process. The next partition divides the climb, cruise, and descent phases into vertical segments, each characterized by a single speed pattern. Each segment extends from the end of the previous segment to its designated target altitude, or, in the case of a cruise segment, to the planned step climb point, or to the optimal climb point or T / D point. The next partition divides each vertical segment into uniform property regions. In these regions, at least one variable in the aircraft's equations of motion can be considered a constant, and these regions also approximate regions where the flight crew or automatic guidance system will use drastically different control methods. The main significance of this partition is that it allows for relatively simple integration algorithms for each region. This method provides 10 integration segment types for handling uniform property regions; see the table describing the integration segment step size for details.

[0096] The final flight division divides each region of uniformity into separate integration intervals. For example, during the climb phase of a target velocity region, assuming it's divided into altitude intervals of 1000 feet, if the region spans 5000 feet, then the integration will process five consecutive intervals to accumulate the total for that region, such as... Figure 7 This is what is shown as a climbing process without height constraints.

[0097] S204. Integrate within one integration step. Based on the selected integration type and using first-principles calculations, and considering the atmospheric wind-temperature model, calculate the data changes of the incremental units within one integration step and add them to the previously accumulated calculations. Integrating within one integration step, different values ​​of the integration step will lead to different calculation results. A smaller integration step results in higher calculation accuracy, but also a sharp increase in computational load. This method requires a trade-off between computational accuracy and efficiency; therefore, the integration step is carefully selected after numerous experiments. The choice of integration step varies depending on the integration type, as shown in Table 1.

[0098] Table 1

[0099]

[0100] S205. Has the termination condition been met? This step determines whether the termination condition set in S202 has been exceeded after integration of one integration step, i.e., whether the set termination value has been passed in either the horizontal distance or the vertical height. If termination is detected, the data at the termination point is stored and processed in S206; if the set termination value has not been passed in either the horizontal distance or the vertical height, the same integration type can be used to integrate for another integration step, and the termination condition can be determined again.

[0101] S206. Process termination and store data. Typically, when a horizontal or vertical termination is detected, the integral value will exceed the precise termination criterion. Therefore, this process interpolates backward to the precise termination point within the last integration interval. At this point, the required data (e.g., predicted height, weight, speed, time, etc.) is stored in the appropriate location. If appropriate, the next termination of the same type can be set before continuing.

[0102] S207. Whether the terminal is the end of a flight segment is used to determine whether it is a horizontal termination point. The termination data has already been processed and stored in Processing Termination and Storing Data 206. This step is used to determine whether it is a horizontal termination. If it is a horizontal termination, then proceed to Whether it is the last segment of the current flight phase 209 for further logical judgment.

[0103] S208. Whether it ends in a uniform attribute region is used to determine whether it is a vertical termination. If it does not end in a uniform attribute region, you can use the same integration type as before to integrate for another integration step. If it ends in a uniform attribute region, you need to select a new integration type and integrate according to the new integration type.

[0104] S209. Is this the last segment of the current flight phase? If it is, then segment processing within this flight phase is complete, and the transition logic for the vertical flight phase needs to be used to determine the upcoming flight phase based on the vertical trajectory prediction. This requires re-execution of the following... Figure 2 The process shown involves entering a new flight phase to process flight segments. If the flight segment is not the last segment of the current flight phase, meaning there are still unprocessed segments within the current flight phase, then the next unprocessed segment needs to be acquired, and the segments falling within that flight phase are processed sequentially until all segments within that flight phase have been processed.

[0105] It is understood that those skilled in the art can make equivalent substitutions or modifications to the technical solution and inventive concept of the present invention, and all such substitutions or modifications should fall within the protection scope of the appended claims.

Claims

1. A method for predicting and optimizing aircraft vertical trajectory based on a combined model, characterized in that... Includes the following steps: S101. Initialize the status parameters of the aircraft's current position; S102. Based on the vertical flight phase judgment logic, obtain the current vertical flight phase of the aircraft. If it is in the pre-flight or takeoff phase, start execution from step S103; if it is in the climb phase, start execution from step S104; if it is in the cruise phase, start execution from step S105; if it is in the descent phase, start execution from step S109; if it is in the advanced phase, start execution from step S110. S103. Vertical trajectory prediction during takeoff is processed using a performance database model; S104. The vertical trajectory prediction for the climb segment is processed using a first-principles model. S105. The vertical trajectory prediction for the cruise segment is processed using a first-principles model. S106. If a change in the flight plan is found during the processing of the cruise segment, proceed to step 107; otherwise, proceed to step 109. S107. Determine if the bottom point of the decline exists. If it does not exist, proceed to step 110. If it exists, proceed to step 108. S108. Construct the descent path; S109. The vertical trajectory prediction for the descent segment is processed using a first-principles model. S110. Vertical trajectory prediction during the approach phase is processed using a performance database model; S111. If the go-around phase is activated, the vertical trajectory prediction for the go-around phase is processed using a performance database model; otherwise, proceed to step S113. S112. Vertical trajectory prediction for alternate landing routes is processed using a performance database model; S113. Based on the vertical 4D trajectory generated by the vertical trajectory prediction at each stage and the optimization index, the vertical flight profile of the entire process is optimized.

2. The method for predicting and optimizing aircraft vertical trajectory based on a combined model according to claim 1, characterized in that... In S102, the judgment logic for the vertical flight phase is based on the aircraft's air-to-ground status, guidance mode, distance from the current location to the destination airport, and the relationship between the current altitude and the cruising altitude.

3. The method for predicting and optimizing aircraft vertical trajectory based on a combined model according to claim 1, characterized in that... In S103, the takeoff phase is considered as an integral segment. An integral calculation is performed according to the altitude integral step size. The unconsumed portions of time, fuel, and distance are added to the initial values ​​of the state parameters. Waypoints that fall within the integral segment are interpolated from the performance database to determine the vertical 4D trajectory profile of each waypoint in the takeoff phase.

4. The method for predicting and optimizing aircraft vertical trajectory based on a combined model according to claim 1, characterized in that... In S104, a climb phase consists of several segments, including: an initial climb phase that meets airport speed limits; a climb acceleration phase that accelerates from the airport speed limit to the planned indicated airspeed corresponding to the climb speed mode, or a climb acceleration phase generated due to waypoint speed constraints; a climb at a constant planned indicated airspeed until reaching the crossing altitude; a constant Mach number climb phase at an altitude above the crossing altitude until reaching cruise altitude; and may also include a cruise level flight phase generated due to waypoint altitude constraints or airport speed limits. The vertical trajectory prediction for the climb segment processes each segment falling into the climb phase sequentially. When predicting the vertical trajectory for any segment in the climb phase, the segment is processed as a vertical integral segment. If a segment involves multiple integral segments, it reverts to the segment termination position when a segment termination is encountered within an integral segment, thus coupling the horizontal and vertical profiles. Based on a complex first-principles model, the changes in time, distance, fuel consumption, and altitude within an integral segment are calculated. The altitude integration step size or time integration step size is adopted according to the change in the aircraft's climb rate, and the integration process continues until the predicted climb peak is reached. This vertical trajectory prediction considers the speed constraints, altitude constraints, time constraints, and airport speed limits at waypoints, as well as the acceleration process during the climb. When encountering waypoint altitude constraints as part of the vertical flight plan, these constraints take precedence over the predicted climb vertical profile.

5. The method for predicting and optimizing aircraft vertical trajectory based on a combined model according to claim 1, characterized in that... In S105, a cruise phase consists of several segments, including: a cruise acceleration segment from climb speed to cruise speed; a cruise segment with constant indicated airspeed or constant Mach number; a stepped climb segment and a stepped descent segment; and a cruise deceleration segment from cruise speed to descent speed. The vertical trajectory prediction during the cruise phase processes each segment falling within the cruise phase sequentially. When predicting the vertical trajectory for any segment during the cruise phase, the segment is processed as a vertical integral segment. When a segment terminates within an integral segment, the process reverts to the segment termination position, thus coupling the horizontal and vertical profiles together. Based on a complex first-principles model, the cruise fuel flow within an integral segment is calculated using any one of three integration steps: weight, distance, or time. The changes in time, distance, and fuel consumption within an integration step are then calculated. The integration process continues until the prediction reaches the descent peak.

6. The method for predicting and optimizing aircraft vertical trajectory based on a combined model according to claim 1, characterized in that... In S108, the descent path is divided into an idle path segment and a geometric path segment. The idle path segment is constructed based on the idle thrust or near-idle thrust and the speed plan used to optimize operations. It is calculated backward from the descent bottom point along the opposite direction of flight to the cruise altitude, and the position point from the cruise altitude to the descent phase is determined in advance. The geometric path segment is constructed based on altitude constraints, speed constraints, airport altitude restrictions, and flight path angle constraints.

7. The method for predicting and optimizing aircraft vertical trajectory based on a combined model according to claim 1, characterized in that... In S109, the descent phase consists of several segments, including: a constant Mach number descent segment flying at the planned Mach number; a constant indicated airspeed descent segment flying at the planned indicated airspeed; a descent acceleration segment decelerating from the planned indicated airspeed corresponding to the descent speed mode to the airport speed limit speed, or due to waypoint speed constraints; a descent deceleration segment; and a constant indicated airspeed segment continuing to descend at the airport speed limit speed to the start of the approach phase; it may also include a cruise level flight segment generated due to waypoint altitude constraints or airport speed limit altitude. The vertical trajectory prediction for the descent segment processes each segment falling into the descent phase sequentially. When predicting the vertical trajectory for any segment of the descent segment, the segment is processed as a vertical integral segment. When a segment termination is encountered within an integral segment, the process reverts to the segment termination position, thus coupling the horizontal and vertical profiles together. Based on a complex first-principles model, the changes in time, distance, fuel consumption, and altitude within an integral segment are calculated, using an altitude integration step size. The integration process continues until the predicted start point of the approach phase is reached. This vertical trajectory prediction considers the speed constraints, altitude constraints, time constraints, and airport speed limits at waypoints, as well as the deceleration process during descent. When encountering waypoint altitude constraints, these constraints take precedence over the predicted descent vertical profile.

8. The method for predicting and optimizing aircraft vertical trajectory based on a combined model according to claim 1, characterized in that... In S110, the approach phase is divided into two parts according to altitude intervals, that is, the approach phase is regarded as multiple integration segments. The vertical trajectory prediction value of waypoints falling within the integration segments is determined by interpolation.

9. The method for predicting and optimizing aircraft vertical trajectory based on a combined model according to claim 1, characterized in that... In S112, the vertical trajectory prediction of the alternate landing route considers two types of alternate landing: go-around and direct flight. Based on the alternate landing flight plan and the alternate landing cruise altitude, the vertical flight phase that should be included in the alternate landing flight plan is determined. The entire alternate landing flight plan uses a performance database model to calculate the changes in time, distance, fuel, and altitude of the alternate landing route, as well as the predicted information of the alternate airport.

Images