Multi-dimensional variable cycle engine bypass ratio dynamic prediction method

Abstract

This invention discloses a multi-dimensional dynamic prediction method for bypass ratio of a variable cycle engine. Its core lies in establishing a multi-dimensional prediction model and predicting data based on this model: establishing a baseline dynamic sub-model of bypass ratio with respect to area ratio and its rate of change; constructing a physical correction sub-model; and coupling the inlet pressure ratio to the baseline bypass ratio output by the baseline dynamic sub-model to correct the predicted bypass ratio corresponding to the actual inlet pressure ratio ε. This ultimately forms a multi-dimensional prediction model that simultaneously considers area ratio, area ratio rate of change, and inlet pressure ratio. This invention innovatively introduces three dimensions: area ratio AR, area ratio rate of change dAR / dt, and inlet pressure ratio ε, and constructs a prediction model. The obtained model can accurately characterize the inherent hysteresis loop and nonlinear phenomena in the dynamic adjustment of RVABI, overcoming the limitations of steady-state models.

Description

Technical Field

[0001] This invention relates to the field of data prediction model construction technology, and in particular to a dynamic prediction method for the bypass ratio of a multi-dimensional variable cycle engine. Background Technology

[0002] Currently, research and engineering design of RVABI (Remote Aperture-Vehicle-Brain) systems are mostly based on steady-state aerodynamic analysis, and its bypass ratio (B) is usually considered a single-valued function of the area ratio (AR, i.e., the ratio of the outer bypass outlet area A1 to the inner bypass outlet area A2). However, in the actual mode transition process of an engine, the RVABI is in continuous motion, which is a dynamic process. In this dynamic process, due to the inertial effect of the flow field, airflow separation, reattachment, and other complex transient physical phenomena, the relationship between the bypass ratio (B) and the area ratio (AR) will exhibit significant nonlinear and hysteretic characteristics. That is, under the same area ratio, the bypass ratio will vary significantly depending on the direction of adjustment (opening up or closing down) and the adjustment speed. In addition, the total pressure ratio (ε) at the inner and outer bypass inlets caused by the change of the engine operating point is also a key variable affecting the bypass ratio.

[0003] Current technologies lack high-precision predictive models that can simultaneously and accurately describe the dynamic impact of the three core factors—area ratio (AR), rate of change of area ratio (dAR / dt), and inlet pressure ratio (ε)—on the bypass ratio. This results in the inability to accurately obtain the RVABI dynamic characteristics during engine control system design, thereby limiting the optimization accuracy of the control law in the mode transition process and affecting the engine's transient performance and stability. Therefore, constructing a bypass ratio prediction method that can reflect the dynamic hysteresis and nonlinear characteristics of RVABI and couple the effects of multiple factors has significant engineering value for improving the intelligent control level of variable cycle engines. Summary of the Invention

[0004] To address the problem of insufficient prediction accuracy of dynamic duct ratio for post-duct variable area ejectors in existing technologies, this invention proposes a multi-dimensional dynamic prediction method. This method establishes for the first time a unified prediction model that simultaneously couples the area ratio, the rate of change of the area ratio, and the inlet pressure ratio, which can accurately characterize the nonlinear hysteresis characteristics of RVABI during dynamic adjustment.

[0005] In a first aspect, the present invention provides a method for dynamic prediction of bypass ratio of a multi-dimensional variable cycle engine, comprising the following steps:

[0006] S10: Construct a multi-dimensional prediction model, wherein the multi-dimensional factors include area ratio AR, area ratio change rate dAR / dt, and inlet pressure ratio ε; S11: Construct a benchmark dynamic sub-model, which uses the area ratio AR and its area ratio change rate dAR / dt as generalized parameters to characterize the dynamic relationship between the benchmark bypass ratio and the area ratio AR and its area ratio change rate dAR / dt under a preset benchmark inlet pressure ratio ε0. S12: Construct a physical correction sub-model, which is coupled with the inlet pressure ratio ε and is used to correct the benchmark bypass ratio B_ε0 output by the benchmark dynamic sub-model to the predicted bypass ratio corresponding to the true inlet pressure ratio ε. S20: Input the real-time area ratio AR, the real-time area ratio change rate dAR / dt, and the real-time inlet pressure ratio ε into the multi-dimensional prediction model to obtain the dynamic duct ratio prediction value B_ε.

[0007] Preferably, step S20 includes: S21: Real-time acquisition of the real-time area ratio AR, real-time area ratio change rate dAR / dt, and real-time inlet pressure ratio ε of the variable cycle engine; S22: Input the real-time area ratio AR and the real-time area ratio change rate dAR / dt into the baseline dynamic sub-model to calculate the baseline duct ratio prediction value B_ε0; input the baseline duct ratio prediction value B_ε0, the real-time area ratio AR, and the real-time inlet pressure ratio ε into the physical correction sub-model to obtain the dynamic duct ratio prediction value B_ε.

[0008] Preferably, let the area ratio AR = x1, the rate of change of the area ratio d(AR) / dt = x2, and the benchmark dynamic sub-model be expressed as a polynomial function of the area ratio AR and the rate of change of the area ratio d(AR) / dt: (1) Wherein, B_ε0 is the predicted bypass ratio under the reference inlet pressure ratio ε0, and c0 to c6 are the model coefficients of the variable cycle engine determined by fitting dynamic simulation data under the reference inlet pressure ratio ε0. Preferably, the reference inlet pressure ratio ε0 is 0.99.

[0009] Preferably, when the variable cycle engine is a rotary type, the model coefficients c0 to c6 are: [0.0069, 1.607, 0.0028, -2.0733, 0.0149, -0.0004, -0.0001]; When the variable cycle engine is a translational type, the model coefficients c0 to c6 are: [0.0056, 0.9779, 0.0088, -0.7632, -0.1073, 0.0015, -0.0117].

[0010] Preferably, the physical correction sub-model is established based on the variation of the bypass flow rate and the ejection physical mechanism, and its expression is: (2) in, B_ε is the predicted dynamic bypass ratio under the actual inlet pressure ratio ε; B_ε0 is the reference bypass ratio calculated from the reference dynamics sub-model; F(ε) is the bypass flow ratio function, which is used to characterize the scaling ratio of the bypass flow relative to the reference inlet pressure ratio ε0 under the true inlet pressure ratio ε. φ(AR) is the unit entrainment capacity function, used to characterize the entrainment efficiency of the bypass flow on the inner flow; Preferably, the bypass flow ratio function F(ε) is fitted by the following formula:

[0011] Preferably, the unit ejection capability function φ(AR) is fitted by the following formula:

[0012] Secondly, the present invention also provides a prediction system for implementing the above method, comprising: The data acquisition module is used to acquire or receive the real-time area ratio AR, real-time area ratio change rate dAR / dt, and real-time inlet pressure ratio ε of the variable cycle engine in real time. The data processing module pre-stores the coefficients of the baseline dynamic sub-model, the bypass ratio function F(ε), and the unit ejection capacity function φ(AR). It is used to execute the calculation process of the baseline dynamic sub-model and the physical correction sub-model, and outputs the predicted value of dynamic bypass ratio B_ε.

[0013] Thirdly, a multi-dimensional variable cycle engine bypass ratio dynamic prediction device is proposed, comprising: a memory and a processor. The memory stores executable code, which, when executed by the processor, causes the processor to perform the prediction method described in the first aspect.

[0014] Fourthly, embodiments of the invention provide a non-transitory machine-readable storage medium storing executable code, which, when executed by a processor of an electronic device, enables the processor to at least implement the prediction method of the first aspect.

[0015] In this embodiment of the invention, a method, system, and application for dynamic prediction of bypass ratio of a variable cycle engine based on multiple dimensions are presented, which have the following beneficial effects: 1. High accuracy of dynamic prediction: The innovative introduction of the area ratio change rate dAR / dt as a key dynamic variable enables the model to accurately characterize the inherent hysteresis loop and nonlinear phenomena in the dynamic adjustment of RVABI, breaking through the limitations of the steady-state model. 2. Coupling of multi-dimensional physical factors: For the first time, the three core influencing factors, area ratio AR, area ratio change rate dAR / dt, and inlet pressure ratio ε, are unified in a single prediction framework, establishing the most comprehensive RVABI bypass ratio prediction model to date, which greatly improves the model's adaptability and prediction accuracy under different flight conditions. 3. Clear physical mechanism and strong extrapolation: The pressure ratio correction part is not a pure mathematical black box fitting, but is derived from a clear physical mechanism based on the change of bypass flow and the enhancement of ejection capacity, so that the model still has good predictive ability and engineering credibility outside the training data range. 4. Good versatility and scalability: The proposed "generalized parametric dynamics sub-model" framework is applicable to RVABI with different motion forms (translation and rotation), and can be quickly applied by changing the model coefficients; 5. High engineering practical value: After verification under multiple sets of variable motion cycle and transformer ratio conditions, the overall prediction accuracy of the method of this invention exceeds 96%. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the rotating rear duct variable area ejector (RVABI) used in an embodiment of the present invention; Figure 2 This is a schematic diagram illustrating the hysteresis loop phenomenon exhibited by the bypass ratio B as a function of the area ratio AR during the dynamic adjustment of RVABI. Figure 3 The figure shows a comparison between the prediction results and the high-precision simulation results of the dynamic sub-model (Equation 1) of this invention for different motion cycles (T=0.5s, T=5s). Figure 4 This is a family of schematic diagrams showing the variation of duct ratio B with area ratio AR under different inlet pressure ratios ε; Figure 5 A schematic diagram illustrating the multiple relationship between the outflow rate at different pressure ratios and the outflow rate at the initial pressure ratio; Figure 6 The figure shows a comparison and verification between the fitted curve of the bypass flow ratio function F(ε) and the simulation data. Figure 7 This is a flowchart illustrating the complete modeling and online prediction process of the multi-dimensional variable cycle engine bypass ratio dynamic prediction method of the present invention. Figure 8 The figure shows the prediction accuracy verification results of the complete model (Equation 1 + Equation 2) of this invention under multiple complex working conditions (variable motion period and variable inlet pressure ratio). Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] The terminology used in the embodiments of this invention is for the purpose of describing particular embodiments only and is not intended to limit the invention. The singular forms “a,” “the,” and “the” as used in the embodiments of this invention and the appended claims are also intended to include the plural forms, and “multiple” generally includes at least two unless the context clearly indicates otherwise.

[0019] Depending on the context, the words "if" or "if" as used here can be interpreted as "when," "when," "in response to determination," or "in response to detection." Similarly, depending on the context, the phrases "if determination" or "if detection (of the stated condition or event)" can be interpreted as "when determination," "in response to determination," "when detection (of the stated condition or event)," or "in response to detection (of the stated condition or event)." Variable cycle engines optimize performance and efficiency in different flight phases by changing the bypass ratio. The rear variable area bypass injector (RVABI) is the key aerodynamic mechanism for achieving continuous and precise adjustment of the bypass ratio.

[0020] Existing technologies lack a high-precision predictive model that can simultaneously and accurately describe the dynamic impact of the three core factors—area ratio (AR), rate of change of area ratio (dAR / dt), and inlet pressure ratio (ε)—on the bypass ratio. This results in the inability to accurately obtain the RVABI dynamic characteristics during engine control system design, thereby limiting the optimization accuracy of the control law during mode transition and affecting the engine's transient performance and stability. Therefore, this invention constructs a predictive model based on the three dimensions of area ratio (AR), rate of change of area ratio (dAR / dt), and inlet pressure ratio (ε), which can achieve dynamic prediction of the bypass ratio of variable cycle engines. This has significant engineering value for improving the intelligent control level of variable cycle engines.

[0021] The implementation principles of the methods, systems, devices, and media of this invention are similar, and will not be repeated here.

[0022] After introducing the basic principles of the present invention, various non-limiting embodiments of the present invention are described in detail below. It should be noted that the embodiments provided by the present invention are only shown to facilitate understanding of the spirit and principles of the present invention, and the embodiments of the present invention are not limited in any way. On the contrary, the embodiments of the present invention can be applied to any applicable system.

[0023] This embodiment uses a certain type of rotating rear duct variable area ejector as an example, with a maximum rotation angle of 24.25°. It should be understood that the description herein is for illustrative purposes only and does not constitute a limitation.

[0024] I. Problem Statement and System Modeling This invention addresses the problem of accurately predicting the dynamic characteristics of the bypass ratio during mode transitions in variable cycle engines, where the rear-ducted variable area ejector (RVABI) is in motion. The geometric movement of the RVABI causes a change in the area ratio AR, which in turn alters the bypass ratio B. The area ratio AR is defined as A1 / A2, and the inlet pressure ratio ε is defined as P1 / P2. The core challenge lies in establishing a high-precision model of B = f(AR, dAR / dt, ε).

[0025] First, in step S11, a nonlinear aerodynamic simulation model of the RVABI and its flow channel is established. The Fluent simulation software is used for simulation calculations, and the motion process of the area regulator is simulated using a dynamic mesh method. The simulation results yield the bypass ratio of the RVABI under corresponding area ratios and rates of change of area ratio, which is used to generate training and validation data. The control variables are the actuation commands of the RVABI (such as rotation angle), and the state variables are the bypass ratio B, internal and external bypass flow rates, pressure, etc.

[0026] To enable those skilled in the art to fully understand and reproduce the present invention, the specific implementation of the simulation data generation process is now described in detail to ensure the sufficiency of the disclosure.

[0027] To construct a high-precision dynamic prediction model for bypass ratio, it is first necessary to obtain a simulation dataset with physical fidelity that covers the complete working envelope of RVABI. This dataset is generated through high-precision computational fluid dynamics (CFD) numerical simulation, with the specific steps as follows: S111. Construction and Verification of High-Fidelity Aerodynamic Simulation Model 1. Three-dimensional geometric modeling: Based on the actual geometric dimensions of RVABI and its upstream and downstream channels, a precise fluid computational domain model is established using three-dimensional Solidworks software. The computational domain must completely include key areas such as the afterburner outlet, outer bypass duct, inner bypass duct, and mixing zone to ensure the physical integrity of the flow path.

[0028] 2. Mesh Generation and Independence Verification: Import the 3D geometric model into preprocessing software (such as ANSYS ICEM CFD) for structured / unstructured mesh generation. Locally refine the mesh in areas with large flow gradients, such as the RVABI throat and near adjustable components. Mesh independence verification is required: Gradually increase the global mesh count (e.g., from 200,000, 230,000 to 290,000). Calculate the bypass ratio for a steady-state condition under the same boundary conditions. When the bypass ratio changes by less than 0.5% after doubling the mesh count, the mesh resolution is considered to meet the accuracy requirements, and the mesh used at this point is the final computational mesh.

[0029] 3. Physical Model and Boundary Condition Settings: Import the mesh model into Fluent software. Set the solver to a pressure-based transient solver. Both the inner and outer bypass gases are incompressible fluids, and the numerical calculation model is the Realizable k-ɛ model, which has high accuracy in simulating the mixing flow field of the ejector. The Simple algorithm is used to couple the pressure, and all parameters in the flow equations are discretized using Second Order Upwind. The boundary conditions are set as follows: Internal parameters: Total pressure 140200Pa, total temperature 1094.8K.

[0030] Outer bypass inlet: given total pressure 138800Pa, total temperature 419.4K.

[0031] Outlet: Given static back pressure conditions of 134784.5 Pa.

[0032] Wall surface: Adopt non-slip, heat-insulated wall surface conditions.

[0033] 4. Benchmark Steady-State Validation: Run steady-state calculations under multiple operating conditions with a fixed area ratio, and compare the calculated key parameters such as bypass ratio and flow rate with existing experimental data or high-precision literature data. If the error is within the acceptable engineering range (e.g., ±5%), it proves that the simulation model has sufficient physical fidelity and can be used for subsequent dynamic simulations.

[0034] S112. Dynamic Process Simulation Based on Dynamic Mesh Technology To simulate the motion process of the RVABI area adjuster, dynamic mesh technology is used. The specific implementation method is as follows: 1. Motion Region Definition and Mesh Update Method: Define the adjustable geometry of the RVABI (such as the area adjuster) and its surrounding local mesh region as the rigid body motion region. Set an interface between this region and the external stationary mesh region. The mesh update method combines "mesh smoothing" and "local mesh re-division". Mesh smoothing method: During component movement, without adding or deleting meshes, the system adjusts the position of mesh nodes to correct node displacements in deformed meshes, keeping mesh quality within a reasonable range. When the component movement is small and the mesh only undergoes elastic deformation without severe distortion, the system automatically and smoothly adjusts node coordinates based on mesh quality, displacement constraints, and other conditions. This method is suitable for scenarios where components undergo small-amplitude translational or rotational movements with limited range of motion.

[0035] Local mesh re-division method: When the component movement is large, causing the local mesh quality (such as distortion) to fall below a set threshold (such as 0.3), the system automatically re-divides the mesh in that local area to maintain computational stability. During mesh reconstruction, the mesh size is kept consistent with the initial mesh size, i.e., the mesh size on the mixer surface is less than 0.4 mm, and the maximum mesh size during mesh reconstruction is 0.7 mm.

[0036] 2. Definition of Motion Law (UDF): User-defined functions (UDFs) are used to program and assign specific motion laws to the moving parts. In this embodiment, to generate data covering a wide range of dynamic processes, various excitation signals are defined, such as: sinusoidal motion of the area regulator at different frequencies and amplitudes w(t) = A. sin(2πf (t), step motion, ramp motion, etc. In UDF, the relationship between the angular velocity (dAR / dt) of the moving parts and time needs to be precisely defined.

[0037] 3. Transient Calculation Settings: Set a sufficiently small time step (e.g., 1e-5 seconds) to ensure that the displacement of the moving part is less than the size of its nearest smallest mesh cell within each time step, satisfying the Courant number condition for dynamic mesh calculation. Each dynamic simulation case starts from the steady-state initial flow field. During the motion, the solver synchronously updates the mesh position and solves the transient Navier-Stokes equations at each time step, completely recording the flow field evolution throughout the entire motion cycle.

[0038] S113. Simulation Data Acquisition and Dataset Construction During transient calculations, Fluent's monitoring and reporting capabilities are used to collect and output data for each time step in real time, including: Input variables: the area ratio AR(t) at the current time step, the instantaneous area ratio change rate dAR / dt(t) calculated by the UDF function, and the inlet real-time pressure ratio ε(t) calculated by the solver.

[0039] Output variables (target values): Instantaneous flow rates of the inner channel and outer channel obtained through surface integral calculation, and based on this, the real-time bypass ratio B(t) = outer channel flow rate / inner channel flow rate is calculated.

[0040] Exporting the [AR, dAR / dt, ε, B] data pairs collected synchronously in time series constitutes a high-fidelity dynamic simulation sample for model training and validation. By changing the inlet total pressure, total temperature boundary conditions, and the motion law of moving parts, repeating steps S2-S3 above, samples covering different pressure ratios ε and different dynamic processes (AR, dAR / dt) can be generated, ultimately forming a large-scale, high-dimensional dataset.

[0041] This detailed implementation ensures the repeatability of the simulation data generation process, providing a reliable data foundation for the subsequent establishment of prediction models.

[0042] II. Establishment and Coefficient Calibration of the Baseline Dynamics Sub-model (Offline) In step S12, the baseline operating condition is first determined. In this embodiment, the inlet pressure ratio ε0 = 0.99 is taken. The baseline model adopts a generalized nonlinear parameter modeling method, as shown in Equation 1. The coefficients (c1, c2, etc.) are solved by the least squares method. Under this baseline pressure ratio, a sinusoidal motion of a typical period (e.g., T = 0.2s) is applied to RVABI, and high-precision dynamic numerical simulation is performed. Real-time area ratio AR, area ratio change rate dAR / dt, and duct ratio B data are collected throughout the process.

[0043] Substituting the collected data into the generalized parametric polynomial model shown in formula (1), and fitting it using the least squares method, we can obtain a set of model coefficients c0~c6 for this specific rotational RVABI. In this example, the fitting result is: [c0, c1, c2, c3, c4, c5, c6] = [0.0069, 1.607, 0.0028, -2.0733, 0.0149,-0.0004, -0.0001] At this point, the baseline dynamic sub-model has been established. Figure 3 The results of using this model to predict other motion cycle (T=0.5s, 5s) conditions are shown. Compared with the simulation values, the accuracy is as high as 98%, which verifies the model's powerful ability to describe dynamic characteristics.

[0044] III. Modeling and Determination of Correction Function for Pressure Ratio Influence (Offline) Step S13 aims to extend the baseline model to any inlet pressure ratio ε. Under the same RVABI configuration and motion laws, dynamic simulations are performed for multiple different inlet pressure ratios ε (e.g., 1.03, 1.07, 1.11). Figure 4 As shown.

[0045] Analysis of simulation data revealed that: (1) The circumferential flow rate m_w increases with increasing pressure ratio ε, but its variation can be described by a function F(ε) that is weakly correlated with the area ratio AR, and the following fit is obtained:

[0046] This function reflects the change in the duct flow capacity with pressure ratio. A comparison of the fitting results with simulation data is shown below. Figure 5 As shown.

[0047] (2) The inward flow rate m_n is affected by the entrainment effect of the outward flow, and the unit entrainment capacity is defined as:

[0048] Analysis shows that φ(AR) mainly depends on the area ratio AR, and the fitted result is: φ(AR) = 0.0392 + 0.0125 / AR Based on the above physical relationship, the formula for correcting the predicted value B_ε0 under the reference pressure ratio ε0 to the actual pressure ratio ε can be derived, namely, equation (2).

[0049] 4. Online Dynamic Prediction Process Step S20 is the online application stage, and its process is as follows: Figure 7 As shown, the specific steps are as follows: (1) Real-time data input: The engine controller acquires or calculates the following parameters in real time: RVABI current actuation position (converted to area ratio AR).

[0050] RVABI current actuation speed (converted to area ratio change rate dAR / dt).

[0051] Total pressures at the inlet of the inner and outer culverts: P1, P2 (pressure ratio ε is calculated).

[0052] (2) Calculate the reference bypass ratio: Input the real-time area ratio AR and dAR / dt obtained in real time into the calibrated reference dynamic sub-model (Equation 1) to calculate the "reference bypass ratio" B_0.99 when the inlet pressure ratio is 0.99 under the current motion state.

[0053] (3) Pressure ratio correction: Substitute the B_0.99 obtained in the previous step, along with the real-time inlet pressure ratio ε and the real-time area ratio AR, into the pressure ratio correction model (Equation 2):

[0054] The predicted dynamic bypass ratio B_ε, which ultimately reflects the actual flight conditions (current AR, dAR / dt, ε), is calculated.

[0055] V. System Verification The prediction system is implemented as a software algorithm module, which is embedded in the engine's Full Authority Digital Electronic Controller (FADEC) during verification. The data acquisition module interfaces with the sensor and controller bus; the model storage and calculation module stores the calibrated coefficients and functions; the calculation core operates according to the above process; and the result output module transmits the bypass ratio B_ε to the control law.

[0056] To verify the overall performance of the method of this invention, tests were conducted under various complex operating conditions without prior training, including different RVABI motion cycles and different flight states (corresponding to different inlet pressure ratios ε). The verification results are as follows: Figure 8 As shown, the predicted values ​​of the method of the present invention closely match the high-fidelity simulation values ​​throughout the entire dynamic process, with a comprehensive prediction accuracy exceeding 96%, which is significantly better than the traditional steady-state method and fully meets the accuracy requirements of engineering control.

[0057] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

Claims

1. A method for dynamic prediction of bypass ratio of a multi-dimension variable cycle engine, characterized in that, include: S10: Construct a multi-dimensional prediction model, wherein the multi-dimensional factors include area ratio AR, area ratio change rate dAR / dt, and inlet pressure ratio ε; S11: Establish a nonlinear aerodynamic simulation model of RVABI and its flow channel, and generate training data and validation data; S12: Construct a benchmark dynamic sub-model, which uses the area ratio AR and its area ratio change rate dAR / dt as generalized parameters to characterize the dynamic relationship between the benchmark bypass ratio and the area ratio AR and its area ratio change rate dAR / dt under a preset benchmark inlet pressure ratio ε0. S13: Construct a physical correction sub-model, which is coupled with the inlet pressure ratio ε and is used to correct the benchmark bypass ratio B_ε0 output by the benchmark dynamic sub-model to the predicted bypass ratio corresponding to the actual inlet pressure ratio ε. S20: Input the real-time area ratio AR, the real-time area ratio change rate dAR / dt, and the real-time inlet pressure ratio ε into the multi-dimensional prediction model to obtain the dynamic duct ratio prediction value B_ε.

2. The multi-dimension based variable cycle engine bypass ratio dynamic prediction method of claim 1, wherein, Step S20 includes: S21: Real-time acquisition of the real-time area ratio AR, real-time area ratio change rate dAR / dt, and real-time inlet pressure ratio ε of the variable cycle engine; S22: Input the real-time area ratio AR and the real-time area ratio change rate dAR / dt into the baseline dynamic sub-model to calculate the baseline duct ratio prediction value B_ε0; input the baseline duct ratio prediction value B_ε0, the real-time area ratio AR, and the real-time inlet pressure ratio ε into the physical correction sub-model to obtain the dynamic duct ratio prediction value B_ε.

3. The multi-dimension based variable cycle engine bypass ratio dynamic prediction method of claim 1, wherein The area ratio AR = x1, the rate of change of the area ratio d(AR) / dt = x2, and the baseline dynamic sub-model are expressed as polynomial functions of the area ratio AR and the rate of change of the area ratio d(AR) / dt: Wherein, B_ε0 is the predicted bypass ratio under the reference inlet pressure ratio ε0, and c0 to c6 are the model coefficients of the variable cycle engine determined by fitting dynamic simulation data under the reference inlet pressure ratio ε0.

4. The method of claim 3, wherein, The reference inlet pressure ratio ε0 is 0.

99.

5. The method for dynamic prediction of bypass ratio of a multi-dimensional variable cycle engine according to claim 3 or 4, characterized in that, When the variable cycle engine is a rotary type, the model coefficients c0 to c6 are: [0.0069, 1.607, 0.0028, -2.0733, 0.0149, -0.0004, -0.0001]; When the variable cycle engine is a translational type, the model coefficients c0 to c6 are: [0.0056, 0.9779, 0.0088, -0.7632, -0.1073, 0.0015, -0.0117].

6. The method of claim 1, wherein, The physical correction sub-model is established based on the variation of the bypass flow rate and the ejection physical mechanism, and its expression is as follows: in, B_ε is the predicted dynamic bypass ratio under the actual inlet pressure ratio ε; B_ε0 is the reference bypass ratio calculated from the reference dynamics sub-model; F(ε) is the bypass flow ratio function, which is used to characterize the scaling ratio of the bypass flow relative to the reference inlet pressure ratio ε0 under the true inlet pressure ratio ε. φ(AR) is the unit entrainment capacity function, used to characterize the entrainment efficiency of the bypass flow on the inner flow.

7. The method of claim 6, wherein, The external flow ratio function F(ε) is fitted by the following formula: ； The unit ejection capability function φ(AR) is fitted by the following formula: φ(AR) = 0.0392 + 0.0125 / AR.

Images