CN117708985A - Machine learning-based high-speed double-spin unsteady aerodynamic prediction method - Google Patents
Machine learning-based high-speed double-spin unsteady aerodynamic prediction method Download PDFInfo
- Publication number
- CN117708985A CN117708985A CN202311745542.7A CN202311745542A CN117708985A CN 117708985 A CN117708985 A CN 117708985A CN 202311745542 A CN202311745542 A CN 202311745542A CN 117708985 A CN117708985 A CN 117708985A
- Authority
- CN
- China
- Prior art keywords
- spin
- double
- aerodynamic force
- moment
- speed
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 27
- 238000010801 machine learning Methods 0.000 title claims abstract description 22
- 238000012360 testing method Methods 0.000 claims abstract description 16
- 238000012549 training Methods 0.000 claims abstract description 15
- 238000010168 coupling process Methods 0.000 claims abstract description 14
- 230000008878 coupling Effects 0.000 claims abstract description 13
- 238000005859 coupling reaction Methods 0.000 claims abstract description 13
- 238000004088 simulation Methods 0.000 claims abstract description 13
- 238000013528 artificial neural network Methods 0.000 claims abstract description 9
- 238000004364 calculation method Methods 0.000 claims description 18
- 230000009977 dual effect Effects 0.000 claims description 7
- 239000002243 precursor Substances 0.000 claims description 6
- 238000012937 correction Methods 0.000 claims description 3
- 239000012530 fluid Substances 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 2
- 239000006185 dispersion Substances 0.000 claims description 2
- 238000013507 mapping Methods 0.000 abstract description 7
- 238000005516 engineering process Methods 0.000 abstract description 3
- 238000013473 artificial intelligence Methods 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 5
- 238000013461 design Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000012938 design process Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/15—Vehicle, aircraft or watercraft design
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/28—Design optimisation, verification or simulation using fluid dynamics, e.g. using Navier-Stokes equations or computational fluid dynamics [CFD]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2113/00—Details relating to the application field
- G06F2113/08—Fluids
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- Computer Hardware Design (AREA)
- Mathematical Optimization (AREA)
- Mathematical Physics (AREA)
- Pure & Applied Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Mathematical Analysis (AREA)
- Software Systems (AREA)
- Fluid Mechanics (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Algebra (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Automation & Control Theory (AREA)
- Aviation & Aerospace Engineering (AREA)
- Computational Mathematics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The utility model provides a high-speed double-spin-bullet unsteady aerodynamic force prediction method based on machine learning, which belongs to the technical field of flight mechanics and artificial intelligence. Generating a high-speed dual-spin elasto aerodynamic modeling sample dataset; dividing a data set into a training set and a testing set, and determining model input and output; establishing a prediction model from the high-speed double-spin state quantity to the aerodynamic force and moment to be applied by using a neural network; the model is used for fast prediction of aerodynamic force and moment of the high-speed double-spin bomb. The invention is based on a machine learning technology, adopts a nonlinear autoregressive neural network, can well learn the mapping relation from the double-spin-bullet flight state parameters to the aerodynamic force and the moment born by the double-spin-bullet flight state parameters, realizes the rapid and accurate prediction of the high-speed double-spin-bullet unsteady aerodynamic force and moment, and reduces the time consumption of large-scale coupling simulation. The method considers the influence of the flight state at the current moment and the historical moment on the aerodynamic force and the moment at the current moment in modeling, considers the influence of the aerodynamic force and the moment at the historical moment, and effectively improves the prediction precision.
Description
Technical Field
The invention belongs to the technical field of flight mechanics and artificial intelligence, and particularly relates to a machine learning-based high-speed double-spin unsteady aerodynamic force prediction method.
Background
In the high-speed double-spin design process, performance parameter calculation is very important to grasp the pneumatic performance and the flight performance of the double-spin. The CFD/RBD coupling method can accurately calculate and simulate the aerodynamic characteristics, the projectile body posture and the movement track of the aircraft, but the CFD analysis is required to be repeatedly invoked in the CFD/RBD coupling calculation, so that the method is extremely low in efficiency and high in calculation condition and resource requirements. An efficient and accurate unsteady aerodynamic model is established, and the CFD module in the coupling simulation is replaced by the model, so that the flight performance of the aircraft is rapidly and accurately predicted. In practice, a certain mapping relation exists between the flying state quantity (such as speed, attack angle and the like) and the aerodynamic force and moment received by the flying state quantity, and the mapping relation can be learned and established by using a machine learning method, so that the rapid prediction of the double-spin unsteady aerodynamic force is realized. Therefore, developing a high-speed bi-spin unsteady aerodynamic force prediction method is necessary and has wide demands.
Disclosure of Invention
The invention solves the technical problems that: the invention provides a machine learning-based high-speed double-spin unsteady aerodynamic force prediction method, and aims to solve the problem of time consumption of CFD/RBD coupling calculation in high-speed double-spin design and performance evaluation.
In order to achieve the above purpose, the invention adopts the technical scheme that:
a machine learning-based high-speed double-spin unsteady aerodynamic force prediction method comprises the following steps:
step 1): generating a high-speed double-spin unsteady aerodynamic modeling data set;
step 2): the input and output parameters of the model are determined, and the aerodynamic force of the double-spin bullet has obvious unsteady characteristics, so the input parameters comprise the flight state quantity of the current moment and the previous m moments, and the aerodynamic force/moment born by the double-spin bullet at the previous n moments is output as the aerodynamic force/moment born by the double-spin bullet at the current moment;
step 3): dividing the data set in the step 1) into a training set and a testing set, wherein the training set is used for training a model, and the testing set is used for checking the prediction precision of the model;
step 4): establishing a prediction model from the input parameters to the output parameters in the step 2) by using an NARX neural network, and completing the construction of a aerodynamic model for rapid prediction of high-speed double-spin-bullet unsteady aerodynamic force;
step 5): and (3) using the aerodynamic model in the step (4) for fast prediction of the unsteady aerodynamic force and moment of the high-speed double-spin projectile.
In the step 1), the specific steps are as follows:
step 1-1): selecting a certain double-rotating bullet configuration to generate a pneumatic calculation grid of the double-rotating bullet;
step 1-2): determining initial state parameters of the double-spin projectile, including initial speed, angular velocity, attack angle, sideslip angle and attitude angle;
step 1-3): and (3) adopting the pneumatic calculation grid in the step (1-1), and carrying out Computational Fluid Dynamics (CFD)/Rigid Body Dynamics (RBD) coupling simulation on the initial state in the step (1-2) to obtain a double-spin-bullet unsteady aerodynamic modeling data set.
In the step 1-1), the double-rotary bullet is in a 155mm fixed-wing duck-type layout double-rotary bullet configuration; in the step 1-1), the pneumatic calculation grid is an unstructured mixed grid, and the whole set of grid comprises 60000 surface grids and 6613683 individual grids; the boundary layer grid uses a structural grid, the height of the first layer is 4e-7m, 50 layers are added, and the growth rate is 1.17; the other areas are unstructured grids.
In the step 1-2), the initial state parameters of the bispin are as follows: mach number 1.08, precursor roll angle speed 0, back body roll angle speed 1570.7rad/s, pitch angle speed and yaw angle 0, pitch angle 31.98, yaw angle 0, roll angle of the precursor and back body 0, attack angle 0, sideslip angle 0.
In the step 1-3), the space dispersion in the CFD calculation adopts se:Sup>A Roe format, and the turbulence model is an S-A model; solving a seven-degree-of-freedom RBD equation by adopting a four-order variable step Adams pre-estimation correction method; CFD/RBD coupling simulation is carried out within 0-7 seconds of high-speed double-spin flight, and the simulation time step is 0.1ms, so that 70000 samples are obtained in total.
In the step 2), the input parameters include: the aerodynamic force/moment of the double-spin bullet at the current moment and the first 4 moments are output as the aerodynamic force/moment of the double-spin bullet at the current moment.
Wherein the flight state quantity includes: the three-axis speed, the three-axis angular speed, the attitude angle, the attack angle and the sideslip angle of the front body and the rear body respectively; aerodynamic forces/moments experienced by the dual spin include axial forces, lateral forces, normal forces, roll moments, yaw moments, and pitch moments experienced by the front and rear bodies, respectively.
In the step 3), the samples in the step 1) are divided into training sets and testing sets, wherein the training sets account for 75% and the testing sets account for 25%.
In the step 4), the NARX neural network is utilized, the flight state quantity at the current moment and the first 4 moments are used as the input of the network, the aerodynamic force and the moment born by the double-spin bullet at the first 4 moments are used as the output of the NARX network, and the aerodynamic force model for rapidly predicting the high-speed double-spin bullet unsteady aerodynamic force is built and trained according to the step 2).
In the step 5), the specific steps are as follows:
step 5-1): selecting input parameters from the test set data in step 3);
step 5-2): and 5-1) taking the input parameters in the step 5-1) as the input of the aerodynamic model trained in the step 4), and predicting to obtain aerodynamic force and moment born by the double-spin projectile at the current moment.
Compared with the prior art, the invention has the advantages that:
1. the scheme is based on a machine learning technology, a nonlinear autoregressive neural network with belt-based input is adopted, and a mapping model of the double-spin-bullet flight state parameters and aerodynamic forces/moments born by the double-spin-bullet flight state parameters can be established, so that the mapping relation from the double-spin-bullet flight state parameters to the aerodynamic forces and moments born by the double-spin-bullet flight state parameters can be well learned, the rapid and accurate prediction of the high-speed double-spin-bullet unsteady aerodynamic forces and moments is realized, the coupling calculation efficiency is improved, and the time consumption problem of CFD/RBD coupling simulation calculation in the high-speed double-spin-bullet design and performance evaluation is reduced;
2. when the double-spin unsteady aerodynamic force prediction method provided by the scheme is used for establishing an aerodynamic force prediction model, the influence of the flight state at the current moment and the historical moment on the aerodynamic force and the moment at the current moment is considered, the influence of the aerodynamic force and the moment at the historical moment is also considered, and the prediction precision is effectively improved;
3. in the scheme, the mapping relation between the double-spin state quantity and aerodynamic force and moment borne by the double-spin state quantity is established, so that the established prediction model can replace the CFD module to carry out coupling simulation with the RBD module.
Drawings
FIG. 1 is a block diagram of a modeling and prediction flow of a machine learning-based high-speed bi-spin unsteady aerodynamic force prediction method;
FIG. 2 is a schematic diagram of the input and output of a aerodynamic force prediction model according to the present invention;
FIG. 3 is a schematic illustration of the geometry of a dual spin projectile employed in the present invention;
FIG. 4 is a schematic diagram of a dual spin pneumatic computing grid used in the present invention, (a) being a face grid, (b) being a slip grid overall schematic diagram;
FIG. 5 shows partial prediction results of high-speed bispin aerodynamic force and moment on a test set in the invention, (a) shows comparison of real and prediction results of back body normal force, (b) shows comparison of real and prediction results of back body side force, (c) shows comparison of real and prediction results of back body pitching moment, and (d) shows comparison of real and prediction results of back body yaw moment.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1-5, embodiments of the present invention are described in detail.
The high-speed double-spin unsteady aerodynamic force prediction method based on machine learning, as shown in FIG. 1, comprises the following steps:
step 1): generating a high-speed double-spin unsteady aerodynamic modeling data set; the method comprises the following specific steps:
step 1-1): selecting a certain double-rotating bullet configuration to generate a pneumatic calculation grid of the double-rotating bullet;
the double-rotary projectile selected in the embodiment has a 155mm fixed-wing duck-type layout double-rotary projectile, and the geometric outline schematic diagram is shown in figure 3; the pneumatic calculation grid is an unstructured mixed grid, and as shown in fig. 4, the whole set of grid comprises 60000 surface grids and 6613683 individual grids; the boundary layer grid uses a structural grid, the height of the first layer is 4e-7m, 50 layers are added, and the growth rate is 1.17; the other areas are unstructured grids.
Step 1-2): determining initial state parameters of the double-spin projectile, including initial speed, angular velocity, attack angle, sideslip angle and attitude angle;
the initial state parameters of the double spin are as follows: mach number 1.08, precursor roll angle speed 0, back body roll angle speed 1570.7rad/s, pitch angle speed and yaw angle 0, pitch angle 31.98, yaw angle 0, roll angle of the precursor and back body 0, attack angle 0, sideslip angle 0.
Step 1-3): and (3) adopting the pneumatic calculation grid in the step (1-1), and carrying out Computational Fluid Dynamics (CFD)/Rigid Body Dynamics (RBD) coupling simulation on the initial state in the step (1-2) to obtain a double-spin-bullet unsteady aerodynamic modeling data set.
Specifically, CFD/RBD coupling simulation is carried out on the high-speed double-spin flying within 0-7 seconds by using the initial state parameters in the step 1-2), and the simulation time step is 0.1ms, so that 70000 samples are obtained in total; the space discrete in the CFD calculation adopts se:Sup>A Roe format, and the turbulence model is an S-A model; the seven-degree-of-freedom RBD equation solving adopts a four-order variable step Adams pre-estimation correction method.
Step 2): determining input and output parameters of the model, wherein the input parameters comprise the flight state quantity of the current moment and the previous m moments because of the obvious unsteady characteristic of the aerodynamic force of the double-spin bullet, and the aerodynamic force/moment born by the double-spin bullet at the previous n moments is output as the aerodynamic force/moment born by the double-spin bullet at the current moment, and the input parameters are shown in the figure 2;
the input parameters include: the aerodynamic force/moment of the double-spin bullet at the current moment and the first 4 moments are output as the aerodynamic force/moment of the double-spin bullet at the current moment.
Wherein the flight state quantity includes: the three-axis speed, the three-axis angular speed, the attitude angle, the attack angle and the sideslip angle of the front body and the rear body respectively; aerodynamic forces/moments experienced by the dual spin include axial forces, lateral forces, normal forces, roll moments, yaw moments, and pitch moments experienced by the front and rear bodies, respectively.
Step 3): dividing the data set in the step 1-3) into a training set and a testing set, wherein the training set is used for training a model, and the testing set is used for checking the prediction precision of the model; wherein the training set accounts for 75% and the testing set accounts for 25%.
Step 4): and (3) establishing a prediction model from input parameters to output parameters in the step (2) by using an NARX neural network, taking the aerodynamic force and moment of the double-spin bullet at the current moment and the first 4 moments as the input of the network, taking the aerodynamic force/moment of the double-spin bullet at the current moment as the output of the NARX network, establishing a high-speed double-spin bullet unsteady aerodynamic force rapid prediction aerodynamic force model, training, and completing the establishment of the unsteady aerodynamic force model for the high-speed double-spin bullet aerodynamic force/moment rapid prediction.
Step 5): the aerodynamic model in the step 4) is used for fast prediction of high-speed double-spin unsteady aerodynamic force and moment; the method comprises the following specific steps:
step 5-1): selecting input parameters from the test set data in step 3);
step 5-2): and 5-1) taking the input parameters in the step 5-1) as the input of the aerodynamic model trained in the step 4), and predicting to obtain aerodynamic force and moment born by the double-spin projectile at the current moment. Referring to fig. 5, partial prediction results of high-speed dual spin aerodynamic forces and moments on the test set (taking dual spin post-body as an example) are given: (a) comparing the true and predicted results of the back body normal force, (b) comparing the true and predicted results of the back body side force, (c) comparing the true and predicted results of the back body pitching moment, and (d) comparing the true and predicted results of the back body yawing moment.
According to the technical scheme, based on a machine learning technology, a nonlinear autoregressive neural network based on out-of-band input is adopted, so that the mapping relation from the double-spin-bullet flight state parameters to aerodynamic forces and moments born by the double-spin-bullet flight state parameters can be well learned, and the rapid and accurate prediction of high-speed double-spin-bullet unsteady aerodynamic forces and moments is realized. When the double-spin-bullet aerodynamic prediction model is built, the influence of the flight state at the current moment and the historical moment on aerodynamic force and moment at the current moment is considered, the influence of aerodynamic force and moment at the historical moment is also considered, and the prediction precision is effectively improved.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof. The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned.
Furthermore, it should be understood that although the present disclosure describes embodiments, not every embodiment is provided with a separate embodiment, and that this description is provided for clarity only, and that the disclosure is not limited to the embodiments described in detail below, and that the embodiments described in the examples may be combined as appropriate to form other embodiments that will be apparent to those skilled in the art.
Claims (10)
1. The high-speed double-spin unsteady aerodynamic force prediction method based on machine learning is characterized by comprising the following steps of:
step 1): generating a high-speed double-spin unsteady aerodynamic modeling data set;
step 2): the input and output parameters of the model are determined, and the aerodynamic force of the double-spin bullet has obvious unsteady characteristics, so the input parameters comprise the flight state quantity of the current moment and the previous m moments, and the aerodynamic force/moment born by the double-spin bullet at the previous n moments is output as the aerodynamic force/moment born by the double-spin bullet at the current moment;
step 3): dividing the data set in the step 1) into a training set and a testing set, wherein the training set is used for training a model, and the testing set is used for checking the prediction precision of the model;
step 4): establishing a prediction model from the input parameters to the output parameters in the step 2) by using an NARX neural network, and completing the construction of a aerodynamic model for rapid prediction of high-speed double-spin-bullet unsteady aerodynamic force;
step 5): and (3) using the aerodynamic model in the step (4) for fast prediction of the unsteady aerodynamic force and moment of the high-speed double-spin projectile.
2. The machine learning-based high-speed double-spin unsteady aerodynamic force prediction method according to claim 1, characterized by comprising the following steps: in the step 1), the specific steps are as follows:
step 1-1): selecting a certain double-rotating bullet configuration to generate a pneumatic calculation grid of the double-rotating bullet;
step 1-2): determining initial state parameters of the double-spin projectile, including initial speed, angular velocity, attack angle, sideslip angle and attitude angle;
step 1-3): and (3) adopting the pneumatic calculation grid in the step (1-1), and carrying out Computational Fluid Dynamics (CFD)/Rigid Body Dynamics (RBD) coupling simulation on the initial state in the step (1-2) to obtain a double-spin-bullet unsteady aerodynamic modeling data set.
3. The machine learning-based high-speed double-spin unsteady aerodynamic force prediction method according to claim 2, characterized by comprising the following steps: in the step 1-1), the double-rotary bullet is in a 155mm fixed-wing duck-type layout double-rotary bullet configuration; the pneumatic calculation grid is an unstructured mixed grid, and the whole set of grid comprises 60000 surface grids and 6613683 individual grids; the boundary layer grid uses a structural grid, the height of the first layer is 4e-7m, 50 layers are added, and the growth rate is 1.17; the other areas are unstructured grids.
4. The machine learning-based high-speed double-spin unsteady aerodynamic force prediction method according to claim 2, characterized by comprising the following steps: in the step 1-2), the initial state parameters of the bispin are as follows: mach number 1.08, precursor roll angle speed 0, back body roll angle speed 1570.7rad/s, pitch angle speed and yaw angle 0, pitch angle 31.98, yaw angle 0, roll angle of the precursor and back body 0, attack angle 0, sideslip angle 0.
5. The machine learning-based high-speed double-spin unsteady aerodynamic force prediction method according to claim 2, characterized by comprising the following steps: in the step 1-3), the space dispersion in the CFD calculation adopts se:Sup>A Roe format, and the turbulence model is an S-A model; solving a seven-degree-of-freedom RBD equation by adopting a four-order variable step Adams pre-estimation correction method; CFD/RBD coupling simulation is carried out within 0-7 seconds of high-speed double-spin flight, and the simulation time step is 0.1ms, so that 70000 samples are obtained in total.
6. The machine learning-based high-speed double-spin unsteady aerodynamic force prediction method according to claim 1, characterized by comprising the following steps: in the step 2), the input parameters include: the aerodynamic force/moment of the double-spin bullet at the current moment and the first 4 moments are output as the aerodynamic force/moment of the double-spin bullet at the current moment.
7. The machine learning-based high-speed bi-spin unsteady aerodynamic force prediction method of claim 6, characterized by: the flight state quantity includes: the three-axis speed, the three-axis angular speed, the attitude angle, the attack angle and the sideslip angle of the front body and the rear body respectively; aerodynamic forces/moments experienced by the dual spin include axial forces, lateral forces, normal forces, roll moments, yaw moments, and pitch moments experienced by the front and rear bodies, respectively.
8. The machine learning-based high-speed double-spin unsteady aerodynamic force prediction method according to claim 1, characterized by comprising the following steps: in the step 3), the training set accounts for 75% and the testing set accounts for 25%.
9. The machine learning-based high-speed bi-spin unsteady aerodynamic force prediction method of claim 6, characterized by: in the step 4), the NARX neural network is utilized, the flight state quantity at the current moment and the first 4 moments are used as the input of the network, the aerodynamic force and the moment born by the double-spin bullet at the first 4 moments are used as the output of the NARX network, and the aerodynamic force model for rapidly predicting the high-speed double-spin bullet unsteady aerodynamic force is built and trained according to the step 2).
10. The machine learning-based high-speed double-spin unsteady aerodynamic force prediction method according to claim 1, characterized by comprising the following steps: in the step 5), the specific steps are as follows:
step 5-1): selecting input parameters from the test set data in step 3);
step 5-2): and 5-1) taking the input parameters in the step 5-1) as the input of the aerodynamic model trained in the step 4), and predicting to obtain aerodynamic force and moment born by the double-spin projectile at the current moment.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311745542.7A CN117708985A (en) | 2023-12-18 | 2023-12-18 | Machine learning-based high-speed double-spin unsteady aerodynamic prediction method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311745542.7A CN117708985A (en) | 2023-12-18 | 2023-12-18 | Machine learning-based high-speed double-spin unsteady aerodynamic prediction method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117708985A true CN117708985A (en) | 2024-03-15 |
Family
ID=90155007
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311745542.7A Pending CN117708985A (en) | 2023-12-18 | 2023-12-18 | Machine learning-based high-speed double-spin unsteady aerodynamic prediction method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117708985A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117952037A (en) * | 2024-03-22 | 2024-04-30 | 中国空气动力研究与发展中心超高速空气动力研究所 | High-speed aircraft aerodynamic engineering estimation correction method based on deep learning |
-
2023
- 2023-12-18 CN CN202311745542.7A patent/CN117708985A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117952037A (en) * | 2024-03-22 | 2024-04-30 | 中国空气动力研究与发展中心超高速空气动力研究所 | High-speed aircraft aerodynamic engineering estimation correction method based on deep learning |
| CN117952037B (en) * | 2024-03-22 | 2024-05-28 | 中国空气动力研究与发展中心超高速空气动力研究所 | High-speed aircraft aerodynamic engineering estimation correction method based on deep learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105843073B (en) | A kind of wing structure aeroelastic stability analysis method not knowing depression of order based on aerodynamic force | |
| CN111611652B (en) | Flapping wing aerodynamic characteristic analysis method based on active flexible deformation | |
| CN108090302B (en) | Helicopter flight mechanics simulation method and system | |
| CN117708985A (en) | Machine learning-based high-speed double-spin unsteady aerodynamic prediction method | |
| CN107390531B (en) | The hypersonic aircraft control method of parameter learning finite time convergence control | |
| CN111813146B (en) | Reentry prediction-correction guidance method based on BP neural network prediction voyage | |
| CN114355777B (en) | Dynamic gliding method and system based on distributed pressure sensor and sectional attitude control | |
| CN111176263B (en) | Online identification method for thrust fault of aircraft based on BP neural network | |
| Thirumalainambi et al. | Training data requirement for a neural network to predict aerodynamic coefficients | |
| CN108920811A (en) | A kind of analogy method and system for helicopter flight simulation | |
| CN102566446B (en) | Method for establishing full-envelope mathematical model of unmanned helicopter based on linear model group | |
| CN117634020A (en) | Hypersonic aircraft overall control collaborative design method based on multi-fidelity data fusion | |
| Wang et al. | Glide performance analysis of underwater glider with sweep wings inspired by swift | |
| CN114818477A (en) | Aircraft aerodynamic characteristic prediction method, device and medium | |
| CN114676639A (en) | Aircraft aerodynamic shape optimization method, device and medium based on neural network | |
| Zhu et al. | Design of an RBF surrogate model for low Reynolds number airfoil based on transfer learning | |
| Qin et al. | Construction on aerodynamic surrogate model of stratospheric airship | |
| Li et al. | Deep learning-based surrogate model for flight load analysis | |
| EP3945395A1 (en) | Aircraft wake sensing | |
| CN117556727A (en) | Method and device for predicting aerodynamic force of flapping wing time-varying based on deep learning | |
| CN116205003A (en) | Rotary bullet trajectory parameter prediction method based on reduced order model and multilayer perceptron | |
| CN116933400A (en) | Method for constructing aerodynamic moment model of coupling uncertainty | |
| CN116068894A (en) | Rocket recovery guidance method based on double-layer reinforcement learning | |
| CN109992859A (en) | Helicopter Dynamics analogy method based on Unigine | |
| Gehlert et al. | Vortex force decomposition—forces associated with individual elements of a vorticity field |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |