An intelligent prediction method for an aero-engine flight envelope

Abstract

The present application belongs to the field of on-board state prediction of aero-engines, and particularly relates to an intelligent prediction method for flight envelope of an aero-engine. The specific technical solution is as follows: based on a data plus physical architecture driven digital engineering model of an aero-engine, the shrinkage envelope after performance degradation of the engine is predicted, and at the same time, through digital flight test, the envelope database is expanded, and the boundary limit value is calculated, so that the health state of the engine can be evaluated after the flight of the airplane is finished, and a guide is provided for the decision of the next flight, thereby saving the operation and maintenance cost of the airplane while ensuring the safety of the flight.

Description

Technical Field

[0001] The present invention belongs to the technical field of aircraft engine onboard state prediction, and in particular relates to an intelligent prediction method for an aircraft engine flight envelope. Background Art

[0002] The flight envelope refers to a closed irregular geometric figure that uses flight speed, flight altitude, load factor, etc. as boundary lines to represent the flight range and flight restrictions of an aircraft. Its shape is roughly as follows Figure 1 As shown, the horizontal axis is the aircraft's flight speed or Mach number, and the vertical axis is the flight altitude. The leftmost side of the graph represents the minimum speed limit, the upper side represents the maximum altitude limit, and the right boundary line represents the maximum speed limit. Specifically, the minimum speed limit refers to the minimum speed at which stable flight can be maintained at a certain altitude, that is, the speed at which the lift of the aircraft can balance its gravity when the aircraft takes the maximum lift coefficient. The maximum altitude limit refers to the maximum altitude at which the aircraft can maintain constant speed in a straight line at its characteristic weight and given engine operating state, that is, the flight altitude corresponding to zero maximum residual thrust. The maximum speed limit refers to the maximum level flight speed that can be achieved at a certain altitude within the performance limitations of the engine itself, that is, the maximum level flight speed that can be achieved under the available thrust.

[0003] In addition to ensuring flight safety, the flight envelope also reflects the performance of the aircraft to a certain extent. Specifically, the larger the flight envelope of the aircraft, the better the performance of the aircraft. For the same type of engine, when it is first used after leaving the factory, due to the existence of processing tolerances and assembly tolerances, the flight envelopes of different products of the same model deviate slightly from the design state; and in the subsequent use, due to the flight environment, fatigue of the control mechanism and mutual wear between the mating parts during operation, the structure and state of the engine change, causing the overall performance of the engine to decline, thereby causing the flight envelope to shrink ( Figure 1 The dashed line in the middle represents the flight envelope after performance degradation. This reduces the sensitivity of the flight envelope protection system, thereby compromising flight safety. Therefore, periodic flight envelope updates and maintenance are crucial for supporting flight decision-making and ensuring flight safety.

[0004] There are many factors that affect the flight envelope, which can be roughly divided into three categories: environmental parameters, aircraft structure and control parameters, and engine performance parameters. Among them, environmental parameters mainly include atmospheric temperature and atmospheric pressure affected by altitude, as well as meteorological parameters such as thunderstorms, ice and snow, atmospheric turbulence, wind shear, and visibility. These parameters not only determine the flow state around the aircraft airfoil, but also the performance of the engine. Aircraft structure parameters mainly include aircraft structural strength, aircraft configuration, and wing surface state. Control parameters refer to the angle of attack, pitch angle, bank angle, and elevator angle controlled by the pilot according to the flight mission, which determine the current flight direction of the aircraft. Engine performance parameters mainly refer to parameters such as thrust and fuel consumption rate.

[0005] During use, aircraft are subject to environmental corrosion and structural changes (such as wear on the wings causing changes in the airfoil flow), which can shrink the flight envelope. During flight, the influence of control parameters on the flight envelope is real-time. For example, the flight envelope range will change during different flight missions such as climbing, descending, and diving. Because the influence of flight angle on lift and drag is a fixed mathematical and physical relationship, the influence of control parameters on the flight envelope does not shrink due to external factors such as the environment. In addition, since aircraft weight determines the minimum required lift, this parameter also affects the flight envelope. However, since its influence on the flight envelope is a single linear relationship, it can be considered a constant in flight envelope prediction technology solutions to simplify the problem.

[0006] There are two main issues with envelope prediction methods based on engine performance parameters. First, for traditional thrust prediction methods, mathematical and physical models rely on the equations for the common workings of engine components and component characteristic data. However, component characteristic data gradually degrades during actual engine operation. Tracking characteristic degradation in real time and establishing full-lifecycle performance tracking predictions is difficult to achieve in a short period of time using mathematical and physical models. Furthermore, mathematical and physical models are based on standard design conditions, while aircraft engines vary in performance due to manufacturing and processing variations. Therefore, these methods struggle to reflect individual differences in engine performance. Data-driven methods, on the other hand, suffer from poor model interpretability, making them difficult to implement onboard for complex thermomechanical systems like aircraft engines. Furthermore, because data-driven methods lack physical mechanisms, they have stricter and more demanding data requirements, limiting model accuracy and making predictions unsafe. Offline database-based flight envelope prediction methods are based on a large amount of fault data within the flight envelope and a small amount outside its boundaries. Due to the limited data near or beyond the boundaries and the high cost of artificially generating fault data, these methods also have limitations and low accuracy.

[0007] Therefore, if an intelligent prediction method for aircraft engine flight envelope can be provided, it will have excellent application prospects. Summary of the Invention

[0008] In order to solve the above technical problems, the present invention proposes an intelligent prediction method for aircraft engine flight envelope.

[0009] To achieve the above-mentioned object, the technical solution adopted by the present invention is: an intelligent prediction method for an aircraft engine flight envelope, comprising the following steps:

[0010] S1. Obtain flight parameters during the first k flights, where k is greater than or equal to 1;

[0011] S2, performing data processing on the flight parameters in step S1;

[0012] S3. Predicting the thrust at all time data points in the flight based on the performance prediction digital model to form a thrust data set;

[0013] S4, training performance degradation model, flight speed model;

[0014] S5. Conduct digital flight tests based on the performance degradation model and flight speed model, and expand the flight data within the envelope and the data near the envelope boundary;

[0015] S6. Calculate the thrust limit boundary based on the envelope database;

[0016] S7. Calculate the area of ​​the envelope after shrinkage and compare it with the factory standard parameters to estimate the remaining life value.

[0017] Preferably, in step S1, the flight parameters include atmospheric environment parameters, flight speed / Mach number, state parameters and airborne performance parameters;

[0018] The atmospheric environment parameters include atmospheric temperature, atmospheric pressure and atmospheric density;

[0019] The state parameters include the oil rod position, rotation speed, guide vane angle and tail nozzle area;

[0020] The airborne performance parameters include high pressure compressor outlet temperature and turbine outlet temperature.

[0021] Preferably, in step S3, atmospheric environment parameters, state parameters and airborne performance parameters are used as inputs of the performance prediction digital model, and thrust is used as output of the performance prediction digital model.

[0022] Preferably, in step S4, the atmospheric environment parameters and state parameters are used as model inputs, and the predicted thrust is used as model output to train the engine performance degradation model;

[0023] And / or, the atmospheric environment parameters and state parameters are used as model inputs, and the flight speed / Mach number is used as model output to train the flight speed model.

[0024] Preferably, in step S5, the flight altitude in step S2 is fixed, and the data set is input into the performance degradation model and flight speed model in step S4 to obtain the thrust and Mach number data set at the altitude; the altitude interval [0, h max ] thrust and Mach number data sets for all altitudes within .

[0025] Preferably, in step S6, the upper, left and right boundaries of the envelope area are determined:

[0026] The left boundary is obtained by calculating the minimum thrust, which is consistent with the factory setting;

[0027] The right boundary is obtained by calculating the maximum thrust. At the same altitude, when the Mach number continues to increase and the thrust cannot continue to increase or the increase is not obvious, the thrust at this time is the maximum thrust of the right boundary.

[0028] The upper boundary is obtained by the upper boundary's limit thrust. At the same Mach number, as the altitude increases, when the thrust cannot continue to decrease or decreases to the minimum thrust, the thrust at this time is the upper boundary's limit thrust.

[0029] Preferably, the step S7 includes the following steps:

[0030] S71. Calculate the decay factor coefficient under thrust:

[0031]

[0032] Where, ε F is the decay factor coefficient under thrust;

[0033] is the shrinkage envelope data domain area under thrust;

[0034] S F It is the envelope data area under the thrust at the time of leaving the factory;

[0035] S72. Calculate the remaining service life under thrust:

[0036] RUL F =L*ε F

[0037] Where RUL F is the remaining life under thrust.

[0038] Correspondingly: Application of intelligent prediction method for aircraft engine flight envelope in aircraft engine remaining life prediction.

[0039] Accordingly: an electronic device comprising:

[0040] one or more processors;

[0041] a storage device for storing one or more programs;

[0042] When the one or more programs are executed by the one or more processors, the one or more processors implement an intelligent prediction method for an aircraft engine flight envelope.

[0043] Correspondingly: A computer-readable medium storing a computer program, wherein the computer program, when executed by a processor, implements an intelligent prediction method for an aircraft engine flight envelope.

[0044] Compared with the prior art, the present invention has the following beneficial effects:

[0045] This invention, based on a digital engineering model of aeroengines driven by a data-plus-physics architecture, predicts the shrinkage envelope of engine performance after degradation. By simultaneously expanding the envelope database and calculating boundary limits through digital flight testing, it can assess the engine's health after the flight, providing guidance for decision-making on the next flight. This reduces aircraft operation and maintenance costs while ensuring flight safety. It can also assess the degree of performance degradation and remaining life, supporting decisions about engine replacement and decommissioning, thereby improving the safety and cost-effectiveness of aircraft operation and maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 Schematic diagram of the geometric shape of the flight envelope of the aircraft engine of the present invention;

[0047] Figure 2 This is a schematic diagram of the intelligent prediction process of the flight envelope of the present invention;

[0048] Figure 3 It is a schematic diagram of the internal structure of the performance degradation model and the flight speed model of the present invention. DETAILED DESCRIPTION

[0049] The following will be combined with the accompanying drawings to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Unless otherwise specified, the technical means used in the embodiments are conventional means well known to those skilled in the art.

[0050] The core of the present invention's solution is the impact of engine performance degradation on the flight envelope, that is, the prediction scheme for the flight envelope shrinkage range. Since the minimum speed, maximum speed and maximum flight altitude in the flight envelope are calculated from thrust, aircraft weight, lift and atmospheric parameters, the key to calculating the shrinkage range lies in the prediction of thrust degradation. Due to structural design and weight limitations, the engine thrust cannot be directly measured during airborne flight, so real-time prediction of thrust degradation is impossible. The engine thrust is mainly calculated and predicted using mathematical and physical models or data-driven methods. In addition, a method for matching and predicting the current envelope state is established by establishing an offline fault flight envelope database method.

[0051] like Figure 2 As shown, the present invention discloses an intelligent prediction method for an aircraft engine flight envelope, comprising the following steps:

[0052] S1. Obtain the flight parameters of all time data points during the first k flights, where k is greater than or equal to 1. The flight parameters are parameters collected by sensors installed on the engine, including atmospheric environment parameters, flight speed / Mach number, state parameters, and some airborne performance parameters. The atmospheric environment parameters include atmospheric temperature, atmospheric pressure, and atmospheric density, etc. These parameters are mainly determined by the flight altitude. When the flight altitude is constant, the atmospheric temperature, atmospheric pressure, and atmospheric density parameters are determined. The state parameters include throttle lever position, rotation speed, guide vane angle, and tail nozzle area, etc. The airborne performance parameters include high-pressure compressor outlet temperature, turbine outlet temperature, etc.

[0053] It should be noted that when k is greater than or equal to 2, the engine performance status within these k flights is considered consistent, meaning that the engine performance has not degraded or has undergone negligible degradation during the first k flights. For example, suppose an aircraft has flown 20 times since leaving the factory. Its current performance status, i.e., the performance at the 20th flight, has changed relative to the factory state, while the performance status between the 15th and 20th flights remains consistent. In this case, k can be 5.

[0054] S2. Process the flight parameters collected in step S1, mainly removing null value data and abnormal data.

[0055] S3. Predict the thrust values ​​for all time data points during the first k flights using the performance prediction digital model to form a thrust dataset. The thrust values ​​are used as the final output parameter of the performance prediction digital model. The performance prediction digital model is trained using neural network technology using factory-tested engine parameters. This model incorporates the physical laws governing the performance parameters of the engine under initial conditions and can be used to predict performance parameters such as engine thrust that are not captured by onboard sensors.

[0056] Specifically, the performance prediction digital model is an existing technology, which is obtained by training an aero-engine digital engineering model based on data plus physics drive. The technical solution of the present invention adopts the performance prediction digital model in the applicant's prior patent technology with application number 202111315645.0. The model is based on deep learning technology, the physical operation mechanism of aero-engines and operation data, and has the advantages of high precision, small data requirements and strong physical interpretability. During the training process, the performance prediction digital model extracts the physical laws of engine operation and, based on the test data during the ground test, uses the engine environmental state parameters, state parameters and airborne performance parameters such as high-pressure compressor outlet temperature and turbine outlet temperature as input parameters of the performance prediction digital model. The airborne unmeasurable parameter, namely thrust, is used as the final output parameter of the model. Based on the condition that the physical laws do not decay, the migration of the engine's physical characteristics to the air engine can be realized.

[0057] By substituting the corresponding airborne parameters into the trained digital performance prediction model as input, the engine's thrust can be calculated during in-flight flight. It should be noted that airborne parameters refer to all parameters collected by onboard sensors during flight, specifically atmospheric environmental parameters, state parameters, and airborne performance parameters during k sorties. This model enables digital flight fault testing and predicts fault data outside the flight envelope. This reduces aircraft maintenance costs while ensuring the safety of daily flight operations and enriches the envelope database.

[0058] It should be noted that the performance prediction digital model in step S3 is trained based on ground data. The model input parameters are measured and collected both on the ground and on the airborne process, while the model output parameter is thrust, which is only measured and collected during the ground test process.

[0059] S4. Use the thrust data set, atmospheric environment parameters, and state parameters predicted for the first k sorties as data sources to train the performance degradation model; use the flight speed / Mach number, atmospheric environment parameters, and state parameters as data sources to train the flight speed model.

[0060] The atmospheric environment parameters and state parameters of the first k flights are used as input parameters of the performance degradation model, and the predicted thrust is used as the model output parameter. The performance degradation model is trained to obtain a performance degradation model.

[0061] The atmospheric environment parameters and state parameters of the first k flights are used as input parameters of the flight speed model, and the flight speed / Mach number is used as the model output parameter. The flight speed model is trained to obtain the flight speed model.

[0062] The performance degradation model is also trained using a digital engineering model of aero-engines based on data and physics. That is, the model structure of the performance degradation model is the same as that of the performance prediction digital model, such as Figure 3 As shown in FIG. 6, the difference is that the input parameters and the final output parameters are different. The input parameters of the performance degradation model discard the airborne performance parameters and only retain the atmospheric environmental parameters and the state parameters as the input parameters, and the thrust is taken as the final output parameter. The airborne performance parameters are removed in the performance degradation model, and the purpose is to facilitate the envelope library expansion operation in the subsequent step S5. At the same time, since the thrust has the most direct influence on the envelope, the final output parameter in the performance degradation model retains the thrust.

[0063] The flight speed model is trained based on the data plus physical driving of the aero-engine digital engineering model, that is, the model structure of the flight speed model is the same as that of the performance prediction digital model, as shown in FIG. 7. Figure 3 As shown in FIG. 7, the difference is that the input parameters and the final output parameters are different. The input parameters of the flight speed model also discard the airborne performance parameters and only retain the atmospheric environmental parameters and the state parameters as the input parameters, and the flight speed / Mach number is taken as the final output parameter.

[0064] S5, performing digital flight test according to the engine performance degradation model and the flight speed model, expanding the flight data in the envelope range and the data near the envelope boundary, thereby expanding the health envelope database and the fault envelope database.

[0065] Specifically, the digital test of the thrust and the Mach number of the engine is performed at different flight altitudes, that is, for the i th flight, i = 1, 2, …, k, the flight altitudes of all time data points in the data set in step S2 are replaced by a fixed altitude h, while the state parameters in the data set remain unchanged, and the atmospheric temperature, atmospheric pressure and atmospheric density and other parameter values are replaced by the values corresponding to the fixed altitude h, and then the state parameters in the data set and the atmospheric environmental parameters corresponding to the fixed altitude h are input into the performance degradation model and the flight speed model in step S4 respectively, to obtain the thrust and Mach number data set at the altitude h. By analogy, the thrust and Mach number data sets at all altitudes in the i th flight altitude interval [0, h max ] are obtained. Until the thrust and Mach number data sets at all altitudes in the k flight altitude interval [0, h max ] are completed. h max is the maximum flight altitude, which is the maximum value of the flight altitude in the flight envelope when leaving the factory. In this way, the data set composed of the flight altitude, the thrust and the Mach number is obtained, and the flight envelope data space is expanded.

[0066] It should be noted that, since the range of the use process envelope shrinks, the flight altitude and Mach number data in the calculated data set outside the factory flight envelope range can be removed, thereby reducing the subsequent calculation.

[0067] S6. Calculate the thrust limit boundary based on the envelope database.

[0068] According to the envelope database generated in step S5, the envelope data space is limited to Figure 1 The data space is within the factory flight envelope. Due to engine performance degradation, there are healthy envelope data and out-of-envelope fault data within this data space.

[0069] By determining the parameter limits of the upper, left, and right boundaries, the envelope range can be determined and the flight envelope diagram under thrust can be obtained. The specific implementation method is as follows:

[0070] The minimum thrust required is determined by the aircraft weight configuration, so the minimum thrust requirement after performance degradation is the same as that at the factory. Figure 1 The left boundary of the envelope, after performance degradation, requires a higher Mach number under the same altitude and conditions to achieve minimum thrust, so the left boundary of the envelope moves to the right.

[0071] The maximum thrust that can be achieved is less than the factory envelope maximum thrust under the same conditions due to performance degradation. The right boundary can be determined by calculating the maximum thrust. The core idea is that at the same altitude, when the Mach number continues to increase, when the thrust cannot continue to increase or the increase is not significant, the thrust in this situation is the right boundary limit thrust.

[0072] Specifically, assuming that at a fixed height h1, the minimum thrust is F min1 The thrust value factory data corresponds to the Mach number Ma1 or flight speed v1, that is, the left boundary of the non-factory flight envelope under h1. Filter the thrust and Mach number data at h1 from the data set obtained in step S5, and find F min1 Nearby (eg F min1 ±2) corresponds to the Ma1' value after the decay. Similarly, the thrust is sorted from small to large. When the thrust does not increase for the first time, the corresponding Mach value is the right boundary. For example, when F is first satisfied, n -F n-1 <0.5, F n The corresponding Mach number is the right boundary.

[0073] For the upper limit, when the thrust cannot be reduced further or reaches the minimum thrust requirement as the altitude increases at the same Mach number, this thrust is the upper limit thrust. This gives the thrust limit range, or the engine shrinkage envelope data range after performance degradation.

[0074] For the upper limit, when the thrust cannot be reduced further or reaches the minimum thrust requirement as the altitude increases at the same Mach number, this thrust is the upper limit thrust. This gives the thrust limit range, or the engine shrinkage envelope data range after performance degradation.

[0075] Furthermore, the upper boundary height limit can be solved by the original height limit. First, let the minimum Mach number of the factory envelope height limit boundary be Ma′ min Then, based on the updated left boundary line (i.e., the data set obtained in step S5), the Mach number is calculated as Ma′ min The height H' max , this height can be approximately considered as the updated height limit.

[0076] Furthermore, for the parameter restriction of the right boundary, the state parameters of the factory envelope boundary can be input into the performance degradation model in step S4 to obtain the predicted thrust F′ under this condition. max , note that F′ here max This is the right boundary performance limit value after degradation, which is greater than the thrust limit value F' corresponding to the boundary at the factory. max The size of F′ is reduced. Then, based on the same height, find F′ max The corresponding minimum Mach number is the value of the right boundary line.

[0077] It should be noted that the envelope boundary in the solution of the present invention is a discrete data point rather than a continuous smooth curve, and the discrete boundary curve can be found by methods such as support vector machines.

[0078] S7. Calculate the area of ​​the envelope after shrinkage and compare it with the factory standard parameters to estimate the remaining life value and provide decision support for subsequent flight missions. This includes the following steps:

[0079] S71. Calculate the area of ​​the thrust shrinkage envelope data domain using discrete integration methods or other area calculation methods based on the thrust shrinkage envelope data range. By calculating the area between it and the factory envelope The engine's degradation factor can be calculated by the ratio of:

[0080]

[0081] Where, ε F is the decay factor coefficient under thrust.

[0082] is the shrinking envelope data domain area under thrust.

[0083] S F It is the envelope data area under the thrust at factory.

[0084] S72. Based on the rated total engine life L, the remaining life under each thrust can be calculated:

[0085] RUL F =L*ε F

[0086] Where RUL F is the remaining life under thrust.

[0087] The envelope prediction scheme proposed in the present invention is an offline flight envelope prediction scheme for the shrinkage caused by engine performance degradation. It can be spread to the whole aircraft and online envelope prediction through the physical relationship between engine performance parameters and whole aircraft control parameters, lift and drag and aircraft weight.

[0088] The embodiments described above are merely descriptions of preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the spirit of the present invention, various deformations, modifications, and substitutions made to the technical solutions of the present invention by ordinary technicians in this field should fall within the scope of protection determined by the claims of the present invention.

Claims

1. An intelligent prediction method for an aircraft engine flight envelope, characterized by: The following steps are involved: S1. Acquire flight parameters during the first k flights, where k is greater than or equal to 1. The flight parameters include atmospheric environment parameters, flight speed / Mach number, state parameters, and airborne performance parameters. The atmospheric environment parameters include atmospheric temperature, atmospheric pressure, and atmospheric density. The state parameters include fuel lever position, rotational speed, guide vane angle, and tail nozzle area. The airborne performance parameters include high-pressure compressor outlet temperature and turbine outlet temperature. S2, performing data processing on the flight parameters in step S1; S3. Predicting the thrust at all time data points in the flight based on the performance prediction digital model to form a thrust data set; S4, training performance degradation model, flight speed model; S5. Conduct digital flight tests based on the performance degradation model and flight speed model, and expand the flight data within the envelope and the data near the envelope boundary; In the i=1, 2, ..., kth flight sortie, the flight altitude in the data set of step S2 is replaced with a fixed altitude h, the state parameters remain unchanged, and the atmospheric environment parameters are replaced with the values ​​corresponding to h. Then, the state parameters and the atmospheric environment parameters corresponding to h are input into the performance degradation model and the flight speed model to obtain the thrust and Mach number data set at the altitude. This is repeated to obtain the kth flight sortie in [0, h max ]Related data sets for all altitudes within the altitude range, expanding the flight envelope data space, h max is the maximum flight altitude, which is the maximum value of the flight altitude in the flight envelope when leaving the factory; S6. Calculate the thrust limit boundary based on the envelope database; The minimum thrust of the left boundary is strictly consistent with the factory state. The thrust and Mach number data at a fixed altitude are screened from the data set obtained in step S5, and the corresponding Mach number within the range of ±2 of the minimum thrust is found. The value is the left boundary value after decay. The thrust is sorted in order from small to large. When F is first met, n -F n-1 <0.5, F n The corresponding Mach number is the right boundary; the upper boundary is the limit boundary determined when the altitude increases to the point where the thrust cannot be further reduced or reaches the minimum thrust at the same Mach number; S7. Calculate the area of ​​the envelope after shrinkage and compare it with the factory standard parameters to estimate the remaining life value; S71. Calculate the decay factor coefficient under thrust: Where, ε F is the decay factor coefficient under thrust; is the shrinkage envelope data domain area under thrust; S F It is the envelope data area under the thrust at the time of leaving the factory; S72. Calculate the remaining service life under thrust: NUMBER F =L*ε F Where RUL F is the remaining life under thrust, and L is the rated total life of the engine.

2. The intelligent prediction method for aircraft engine flight envelope according to claim 1, characterized in that: In step S3, atmospheric environment parameters, state parameters and airborne performance parameters are used as inputs of the performance prediction digital model, and thrust is used as output of the performance prediction digital model.

3. The intelligent prediction method for aircraft engine flight envelope according to claim 1, characterized in that: In step S4, the atmospheric environment parameters and state parameters are used as model inputs, and the predicted thrust is used as model output to train the engine performance degradation model; and / or the atmospheric environment parameters and state parameters are used as model inputs, and the flight speed / Mach number is used as model output to train the flight speed model.

4. Application of the intelligent prediction method for aircraft engine flight envelope according to any one of claims 1 to 3 in aircraft engine remaining life prediction.

5. An electronic device, characterized in that: include: one or more processors; a storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the intelligent prediction method for the flight envelope of an aircraft engine according to any one of claims 1 to 3.

6. A computer-readable medium storing a computer program, characterized in that: When the computer program is executed by a processor, the intelligent prediction method for the flight envelope of an aircraft engine according to any one of claims 1 to 3 is implemented.

Images