CN111382876B - Ground fire transfer orbit design initial value acquisition method and system based on evolutionary algorithm - Google Patents
Ground fire transfer orbit design initial value acquisition method and system based on evolutionary algorithm Download PDFInfo
- Publication number
- CN111382876B CN111382876B CN202010130385.9A CN202010130385A CN111382876B CN 111382876 B CN111382876 B CN 111382876B CN 202010130385 A CN202010130385 A CN 202010130385A CN 111382876 B CN111382876 B CN 111382876B
- Authority
- CN
- China
- Prior art keywords
- initial value
- track
- fire transfer
- ground fire
- orbit
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/10—Machine learning using kernel methods, e.g. support vector machines [SVM]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Medical Informatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method and a system for quickly acquiring an initial value of a ground fire transfer track design based on an evolutionary algorithm. Before the track is precisely transferred, the track initial value in the dominant direction is rapidly calculated based on an evolutionary algorithm by directly utilizing a previously established track database, so that more efficient subsequent searching can be supported. The method mainly comprises 3 steps: 1) According to the GTOP database, a model initial value sensitive matrix characteristic value is used as a sample label, and a learning sample library is established; 2) Constructing a track optimization support vector machine (TO_SVM) TO learn a ground fire transfer track learning sample library; 3) And running the TO_SVM according TO the ground fire transfer orbit TO be designed, and giving an orbit design initial value (a launch date, an arrival date and a launch orbit residual speed) of the dominant direction c for subsequent accurate design.
Description
Technical Field
The invention relates to a method and a system for quickly acquiring an initial value of a ground fire transfer track design based on an evolutionary algorithm. The method can rapidly calculate the initial value of the orbit in the dominant direction before precisely transferring the orbit by adopting an indirect method and an evolutionary algorithm so as to carry out more efficient search subsequently.
Background
Because the relative position and speed of the earth and the Mars are continuously changed and the carrying capacity of the rocket is limited, the parameters (the launching time, the arrival time, the launching energy, the arrival braking energy and the like) of the ground fire transfer orbit of the Mars detector need to be optimally designed. And (3) generally traversing all possible combinations of emission time and arrival time, adopting a conical curve splicing method, selecting a proper track from all possible tracks as a design initial value, and then adopting an accurate dynamics model for optimization to obtain a final accurate ground fire transfer track solution. The method for searching the design initial value by traversing and splicing the conical curves has large calculated amount and cannot be applied to the spacecraft.
Disclosure of Invention
The invention aims at: the method for quickly acquiring the initial design values of the ground fire transfer track based on the evolutionary algorithm is provided, and a plurality of initial design values (transmitting date and arrival date) of the dominant direction can be quickly given for subsequent accurate design and use under the condition of no traversal. The method can solve the defect of unidirectional evolution of the traditional method, thereby greatly improving the calculation efficiency of the precise transfer track of the ground fire.
In order to solve the technical problems, the invention is realized by the following technical scheme:
a method for quickly acquiring an initial value of a ground fire transfer orbit design based on an evolutionary algorithm comprises the following steps:
step one: according to the GTOP database, adopting a model initial value sensitive matrix characteristic value as a sample label, and establishing a ground fire transfer orbit learning sample library;
step two: constructing a track optimization support vector machine TO_SVM, and learning a ground fire transfer track learning sample library;
step three: aiming at the ground fire transfer orbit TO be designed, an orbit optimization support vector machine TO_SVM in the second operation step is operated, and an orbit design initial value in the dominant direction is given for subsequent use, wherein the orbit design initial value comprises a transmitting date and an arrival date.
Further, the GTOP database refers to an international track optimization competition.
Further, according to the transmission date D d Date of arrival A d Emission energy C d 3 Reaching braking energy C a 3 Constructing sample points X as sample features i Namely the following formula:
where l is the number of samples.
Further, use of sample pointsX i And model initial value sensitive matrix eigenvalue A i The ground fire transfer track learning sample library is constructed as follows:
further, the built learning sample library is utilizedSolving for the optimal solution (ω) of the quadratic convex optimization process * ,b * ) Obtaining a track optimization classification decision function:
f(x)＝sgn((ω * ) T φ(x)+b * )
the orbit optimization classification decision function is an orbit optimization support vector machine TO_SVM obtained after training;
wherein sgn is a sign function and f (x) has a value range of [0,1].
Further, the third step is TO the ground fire transfer orbit TO be designed, the orbit optimization support vector machine TO_SVM in the second step is operated, and the orbit design initial value of the dominant direction is given, specifically:
(1) Generating a sample point set aiming at a ground fire transfer track needing to be designed
(2) By using a trained support vector machine TO_SVM, the method comprises the following steps ofPerforming tag classification, sample point set classified as 1 +.>The initial value is designed for the track in the dominant direction.
The ground fire transfer track to be designed transmits a date zone [ C d1 ,C d2 ]Arrival date section [ C ] a1 ,C a2 ]。
Further, the sample point setIn particular, wherein
X j ＝[D d (j),D a (j)]。
Furthermore, the invention also provides a system for acquiring the initial design value of the ground fire transfer rail, which comprises the following steps:
and a sample library establishing module: according to the GTOP database, adopting a model initial value sensitive matrix characteristic value as a sample label, and establishing a ground fire transfer orbit learning sample library;
the support vector machine building module: constructing a track optimization support vector machine TO_SVM, and learning a ground fire transfer track learning sample library;
the initial value determination module for track design: aiming at the ground fire transfer orbit needing TO be designed, the orbit optimization support vector machine TO_SVM gives an orbit design initial value of the dominant direction for subsequent use, wherein the orbit design initial value comprises a transmitting date and an arrival date.
Compared with the prior art, the method has the advantages that:
(1) According to the invention, before the track is accurately transferred, the track initial value in the dominant direction is rapidly calculated based on the evolution algorithm by directly utilizing the track database established in advance, so that more efficient subsequent searching can be supported.
(2) The method can solve the defect of unidirectional evolution of the traditional method, thereby greatly improving the calculation efficiency of the precise transfer track of the ground fire.
Drawings
FIG. 1 is a flow chart of the method of the present invention.
Detailed Description
The method for quickly acquiring the initial value of the ground fire transfer track design based on the evolution algorithm can quickly calculate the initial value of the track in the dominant direction before precisely transferring the track by adopting the indirect method and the evolution algorithm so as to perform more efficient subsequent searching.
As shown in fig. 1, the steps are as follows:
step one: according to the GTOP database, adopting a model initial value sensitive matrix characteristic value as a sample label, and establishing a ground fire transfer orbit learning sample library;
according to the transmission date D, each track data disclosed by the international track optimization competition (GTOP) d Date of arrival A d Emission energy C d 3 Reaching braking energy C a 3 The 4 parameters are taken as sample characteristics to construct a sample point X i Namely the following formula:
where l is the number of samples.
Then, calculating a model initial value sensitive matrix characteristic value A of each sample point i 。
Using sample points X i And model initial value sensitive matrix eigenvalue A i The ground fire transfer track learning sample library is constructed as follows:
step two:constructing a track optimization support vector machine (TO_SVM) TO learn a ground fire transfer track learning sample library;
the support vector machine (Support Vector Machine, SVM) method of one of the evolutionary algorithms is a machine learning method aiming at the problem of small sample classification, and is proposed according to the principle of minimizing structural risk in the statistical learning theory, and is widely applied to the current hot spot fields such as compressed sensing, sparse optimization, pattern recognition, feature extraction, image processing, medical diagnosis and the like due to the fact that the method has the capability of obtaining a global optimal solution and good generalization. The SVM essence is a convex quadratic optimization problem optimizing method.
The track optimization problem is essentially a nonlinear optimization problem. The secondary convex optimization process is as follows:
s.t.y i (ω T φ(x i )+b)≥1-ξ i ，i＝1，2，...，l，
ξ＝(ξ 1 ，ξ 2 ，...，ξ l ) T ≥0 l ，
in the formula, xi= (xi) 1 ，ξ 2 ，...，ξ l ) For non-negative relaxation variables, each component corresponds to the degree of misclassification of the sample point, C is a penalty parameter, φ (x i ) Is a sample point transfer function. This requires constructing classification hyperplanes in a high-dimensional feature space for such an optimization problem to arrive at an orbit-optimized classification decision function. Utilizing the learning sample library established in the step oneTo find the optimal solution (ω) of the above equation * ，b * ) Obtaining a track optimization classification decision function:
f(x)＝sgn((ω * ) T φ(x)+b * )
the above formula is the orbit optimization support vector machine (TO_SVM) obtained after training, wherein sgn is a sign function, and the value range of f (x) is [0,1]. The method can be used for optimizing the initial value of the track without the sample label in the next step.
Step three:aiming at the ground fire transfer orbit (emitting date interval, reaching date interval) to be designed, the SVM in the second step is operated, and an orbit design initial value (emitting date and reaching date) in the dominant direction is selected for subsequent accurate design and use.
The specific process is as follows:
1) For the ground fire transfer track to be designed (emission date section [ C) d1 ，C d2 ]Reach toDate zone [ C a1 ，C a2 ]) Generating a set of sample pointsWherein the method comprises the steps of
X j ＝[D d (j)，D a (j)]
2) Using the SVM model trained in the step 2 toPerforming tag classification, classifying sample point set +.1 (i.e., f (x) =1)>I.e. to design an initial value (launch date, arrival date) for the track in the dominant direction.
The invention can rapidly give out a plurality of track design initial values (transmitting date and arrival date) of dominant direction for subsequent accurate design and use under the condition of no traversal. The method can solve the defect of unidirectional evolution of the traditional method, thereby greatly improving the calculation efficiency of the precise transfer track of the ground fire.
Claims (5)
1. The method for quickly acquiring the initial value of the ground fire transfer orbit design based on the evolutionary algorithm is characterized by comprising the following steps:
step one: according to the GTOP database, adopting a model initial value sensitive matrix characteristic value as a sample label, and establishing a ground fire transfer orbit learning sample library;
according to the transmitting date D d Date of arrival D a Emission energy C d 3 Reaching braking energy C a 3 Constructing sample points X as sample features i Namely the following formula:
wherein l is the number of samples;
using sample points X i Sum dieInitial value sensitive matrix eigenvalue A i The ground fire transfer track learning sample library is constructed as follows:
step two: constructing a track optimization support vector machine TO_SVM, and learning a ground fire transfer track learning sample library;
using an established library of learning samplesSolving for the optimal solution (ω) of the quadratic convex optimization process * ,b * ) Obtaining a track optimization classification decision function:
f(x)＝sgn((w * ) T φ(x)+b * )
the orbit optimization classification decision function is an orbit optimization support vector machine TO_SVM obtained after training;
wherein sgn is a sign function, and the value range of f (x) is [0,1];
step three: aiming at the ground fire transfer orbit TO be designed, a orbit optimization support vector machine TO_SVM in the second operation step is operated, and an orbit design initial value in the dominant direction is given for subsequent use, wherein the orbit design initial value comprises an emission date and an arrival date;
the method comprises the following steps:
(1) Generating a sample point set aiming at a ground fire transfer track needing to be designed
Wherein X is j ＝[D d (j),D a (j)]；
(2) By using a trained support vector machine TO_SVM, the method comprises the following steps ofPerforming tag classification, sample point set classified as 1 +.>The initial value is designed for the track in the dominant direction.
2. The rapid acquisition method for the initial value of the ground fire transfer orbit design based on the evolutionary algorithm according to claim 1, which is characterized by comprising the following steps: the GTOP database refers to an international track optimization competition.
3. The rapid acquisition method for the initial value of the ground fire transfer orbit design based on the evolutionary algorithm according to claim 1, which is characterized by comprising the following steps: the ground fire transfer track to be designed transmits a date zone [ C d1 ,C d2 ]Arrival date section [ C ] a1 ,C a2 ]。
4. A ground fire transfer track design initial value acquisition system implemented according to the ground fire transfer track design initial value rapid acquisition method of claim 1, characterized by comprising:
and a sample library establishing module: according to the GTOP database, adopting a model initial value sensitive matrix characteristic value as a sample label, and establishing a ground fire transfer orbit learning sample library;
the support vector machine building module: constructing a track optimization support vector machine TO_SVM, and learning a ground fire transfer track learning sample library;
the initial value determination module for track design: aiming at the ground fire transfer orbit needing TO be designed, the orbit optimization support vector machine TO_SVM gives an orbit design initial value of the dominant direction for subsequent use, wherein the orbit design initial value comprises a transmitting date and an arrival date.
5. The ground fire transfer track design initiation value acquisition system of claim 4, wherein:
using sample points X i And model initial value sensitive matrix eigenvalue A i The ground fire transfer track learning sample library is constructed as follows:
wherein according to the transmitting date D d Date of arrival D a Emission energy C d 3 Reaching braking energy C a 3 Constructing sample points X as sample features i Namely the following formula:
where l is the number of samples.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010130385.9A CN111382876B (en) | 2020-02-28 | 2020-02-28 | Ground fire transfer orbit design initial value acquisition method and system based on evolutionary algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010130385.9A CN111382876B (en) | 2020-02-28 | 2020-02-28 | Ground fire transfer orbit design initial value acquisition method and system based on evolutionary algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111382876A CN111382876A (en) | 2020-07-07 |
| CN111382876B true CN111382876B (en) | 2023-09-29 |
Family
ID=71217085
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010130385.9A Active CN111382876B (en) | 2020-02-28 | 2020-02-28 | Ground fire transfer orbit design initial value acquisition method and system based on evolutionary algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111382876B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114137993B (en) * | 2021-11-01 | 2022-07-29 | 北京空间飞行器总体设计部 | Ground fire short transfer orbit launching window searching method with deep space maneuver |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103112600A (en) * | 2013-03-04 | 2013-05-22 | 北京理工大学 | Interplanetary transfer orbit design method |
| CN105631095A (en) * | 2015-12-18 | 2016-06-01 | 中国人民解放军国防科学技术大学 | Search method for multi-constrained earth-moon transfer orbit cluster with equal launch intervals |
| CN107391813A (en) * | 2017-07-03 | 2017-11-24 | 北京航空航天大学 | A kind of energetic optimum moon based on the moon high ladder ground transfer orbit design method |
| CN107844462A (en) * | 2017-10-26 | 2018-03-27 | 北京理工大学 | A kind of interplanetary continuous thrust transfer orbit appraisal procedure |
-
2020
- 2020-02-28 CN CN202010130385.9A patent/CN111382876B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103112600A (en) * | 2013-03-04 | 2013-05-22 | 北京理工大学 | Interplanetary transfer orbit design method |
| CN105631095A (en) * | 2015-12-18 | 2016-06-01 | 中国人民解放军国防科学技术大学 | Search method for multi-constrained earth-moon transfer orbit cluster with equal launch intervals |
| CN107391813A (en) * | 2017-07-03 | 2017-11-24 | 北京航空航天大学 | A kind of energetic optimum moon based on the moon high ladder ground transfer orbit design method |
| CN107844462A (en) * | 2017-10-26 | 2018-03-27 | 北京理工大学 | A kind of interplanetary continuous thrust transfer orbit appraisal procedure |
Non-Patent Citations (1)
| Title |
|---|
| 尚海滨等.基于形状的行星际小推力转移轨道初始设计方法.《宇航学报》.2010,第31卷(第31期),全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111382876A (en) | 2020-07-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109711714B (en) | Manufacturing assembly product quality prediction method based on parallel long-term and short-term memory network | |
| US11836632B2 (en) | Method and system for image classification | |
| US20220188568A1 (en) | Methods and systems for mining minority-class data samples for training a neural network | |
| CN111814871A (en) | Image classification method based on reliable weight optimal transmission | |
| CN108596327B (en) | Seismic velocity spectrum artificial intelligence picking method based on deep learning | |
| CN111897353B (en) | Aircraft maneuvering trajectory prediction method based on GRU | |
| CN111161315B (en) | Multi-target tracking method and system based on graph neural network | |
| CN110728698B (en) | Multi-target tracking system based on composite cyclic neural network system | |
| CN108416270B (en) | Traffic sign identification method based on multi-attribute combined characteristics | |
| CN108021947A (en) | A kind of layering extreme learning machine target identification method of view-based access control model | |
| Roy et al. | A stacked autoencoder neural network based automated feature extraction method for anomaly detection in on-line condition monitoring | |
| Liu et al. | A fault diagnosis intelligent algorithm based on improved BP neural network | |
| CN113255432B (en) | Turbine vibration fault diagnosis method based on deep neural network and manifold alignment | |
| CN109740695A (en) | Image-recognizing method based on adaptive full convolution attention network | |
| CN111382876B (en) | Ground fire transfer orbit design initial value acquisition method and system based on evolutionary algorithm | |
| CN114387473A (en) | Small sample image classification method based on base class sample characteristic synthesis | |
| CN114170461A (en) | Teacher-student framework image classification method containing noise labels based on feature space reorganization | |
| Chen et al. | A semisupervised deep learning framework for tropical cyclone intensity estimation | |
| CN113887342A (en) | Equipment fault diagnosis method based on multi-source signals and deep learning | |
| CN112749675A (en) | Potato disease identification method based on convolutional neural network | |
| CN114882531A (en) | Cross-domain pedestrian re-identification method based on deep learning | |
| CN116310647A (en) | Labor insurance object target detection method and system based on incremental learning | |
| Cebollada et al. | A Deep Learning Tool to Solve Localization in Mobile Autonomous Robotics. | |
| Rethik et al. | Attention Based Mapping for Plants Leaf to Classify Diseases using Vision Transformer | |
| Akrim et al. | A framework for generating large data sets for fatigue damage prognostic problems |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |