CN116738583B - Solid rocket engine charging configuration constraint design method - Google Patents

Solid rocket engine charging configuration constraint design method Download PDF

Info

Publication number
CN116738583B
CN116738583B CN202311029449.6A CN202311029449A CN116738583B CN 116738583 B CN116738583 B CN 116738583B CN 202311029449 A CN202311029449 A CN 202311029449A CN 116738583 B CN116738583 B CN 116738583B
Authority
CN
China
Prior art keywords
elite
constraint
sample point
design
file
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
Application number
CN202311029449.6A
Other languages
Chinese (zh)
Other versions
CN116738583A (en
Inventor
高经纬
马帅超
张为华
杨家伟
王东辉
张德权
武泽平
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
National University of Defense Technology
Original Assignee
National University of Defense Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by National University of Defense Technology filed Critical National University of Defense Technology
Priority to CN202311029449.6A priority Critical patent/CN116738583B/en
Publication of CN116738583A publication Critical patent/CN116738583A/en
Application granted granted Critical
Publication of CN116738583B publication Critical patent/CN116738583B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

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
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F02—COMBUSTION ENGINES; HOT-GAS OR COMBUSTION-PRODUCT ENGINE PLANTS
    • F02K—JET-PROPULSION PLANTS
    • F02K9/00—Rocket-engine plants, i.e. plants carrying both fuel and oxidant therefor; Control thereof
    • F02K9/08—Rocket-engine plants, i.e. plants carrying both fuel and oxidant therefor; Control thereof using solid propellants
    • F02K9/10—Shape or structure of solid propellant charges
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F02—COMBUSTION ENGINES; HOT-GAS OR COMBUSTION-PRODUCT ENGINE PLANTS
    • F02K—JET-PROPULSION PLANTS
    • F02K9/00—Rocket-engine plants, i.e. plants carrying both fuel and oxidant therefor; Control thereof
    • F02K9/08—Rocket-engine plants, i.e. plants carrying both fuel and oxidant therefor; Control thereof using solid propellants
    • F02K9/24—Charging rocket engines with solid propellants; Methods or apparatus specially adapted for working solid propellant charges
    • 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
    • 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
    • Y02T10/00—Road transport of goods or passengers
    • Y02T10/10—Internal combustion engine [ICE] based vehicles
    • Y02T10/40—Engine management systems

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Geometry (AREA)
  • Theoretical Computer Science (AREA)
  • General Engineering & Computer Science (AREA)
  • Evolutionary Computation (AREA)
  • General Physics & Mathematics (AREA)
  • Mechanical Engineering (AREA)
  • Combustion & Propulsion (AREA)
  • Chemical & Material Sciences (AREA)
  • Computer Hardware Design (AREA)
  • Computational Mathematics (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mathematical Optimization (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Mathematical Analysis (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Automation & Control Theory (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses a method for restricting and designing a loading configuration of a solid rocket engine, which comprises the following steps: performing experimental design on design variables to obtain initial sample points, and storing a plurality of optimal individuals into elite files; constructing a proxy model; determining a sampling criterion according to the feasibility of each sample point in the current elite file, searching a new sample point meeting the sampling criterion, and iteratively updating the agent model and the elite file. The invention is applied to the technical field of configuration design, and based on a proxy model technology and a non-precise search-based efficient three-stage constraint sampling technology, by introducing feasibility and optimality of elite archive balance targets, the global feasible optimal point of a model can be positioned while the optimization efficiency is ensured, so that the design efficiency of the charge configuration of the solid rocket engine is obviously improved, and the problems of low optimization efficiency and difficult processing of constraint conditions of the charge configuration design of the solid rocket engine are effectively solved.

Description

Solid rocket engine charging configuration constraint design method
Technical Field
The invention relates to the technical field of configuration design, in particular to a solid rocket engine charging configuration constraint design method.
Background
The solid rocket engine is one of widely used power systems of spacecrafts such as missiles and rockets. The quality of the engine design directly affects the performance of the spacecraft. Charge geometry design is an important component of engine design and is one of the most difficult technologies. The main task is to determine the optimal geometric dimension of the shaped charge under the given constraint limit condition, so that the combustion surface moves back in the working process of the engine to meet the stability requirement of the engine.
The existing common charge constraint design method mainly comprises a manual method and an optimization method. The manual method is to manually adjust and search the geometric parameters of the powder charge according to the experience of engineers so as to meet the design constraint requirements. The optimization method is to construct a charge design optimization model, and an optimization algorithm is adopted to conduct iterative search optimization under constraint limit.
Currently, the manual method is most widely used in the industrial production sector, and better results are generally obtained because the factory has a large number of engineers with accumulated case data and abundant experience. However, the manual method based on factory cases and engineer experience can only adopt manual adjustment and search for the charging geometric parameters and constraint limitation, so that the method not only needs to rely on a great deal of engineering experience as a basis, but also has the defects of long period, low efficiency and the like due to manual iteration.
The optimization design method does not need too much engineering experience, and can avoid complicated manual iteration. The implementation steps of the optimization design method are as follows:
1) Establishing an optimization model
Firstly, defining design variables, optimization targets and constraint variables of an optimization problem, and establishing a parameterized model of a charging configuration. The geometric parameters of the charge are generally used as design variables, the design variables can uniquely determine the available geometric configuration of the charge and set constraint variables according to the overall design requirement index, and the optimization target is the combustion surface stability of the explosive column combustion process;
2) Selection optimization method
In the selection of the optimization method, an evolutionary algorithm is generally combined with a local search method to optimize the model, and the method is difficult to simulate the combustion face migration law by using a high-precision model because a large number of iterations are involved and constraint limits cannot be reliably processed. The optimization method based on the agent model technology can predict and output the overall situation through a small number of sample points in the design space, greatly reduces the simulation times of the model in the optimization process, improves the search efficiency, and can ensure high-precision optimization by adopting an effective constraint processing method, so that the introduction of a high-precision combustion surface migration model becomes possible, and improves the optimization efficiency while ensuring the optimization precision.
Compared with a manual method, the existing optimization design method saves time and workload, but only needs a large amount of data for iterative optimization based on the evolution algorithm, combustion surface simulation can be performed only by adopting a low-precision analysis method, and effective constraint processing is difficult. Even if the optimization method based on the agent model and the high-precision combustion surface migration simulation model are adopted for design, hundreds of high-precision simulations are still needed, and the calculation cost is still relatively high. In addition, the geometric parameters of the powder charge have constraint limitation in the design process, and the general evolutionary algorithm and the agent model optimization method are difficult to effectively solve the problem of constraint optimization. The feasibility rule method is easy to realize and does not need a complicated parameter adjusting process. The method is widely applied to evolutionary algorithms such as differential evolutionary algorithm and particle swarm algorithm. However, the constraint processing is too strict, so that some high-quality infeasible solutions at constraint boundaries are abandoned, and the exploration capacity of the algorithm for the feasible region boundaries is reduced.
Disclosure of Invention
Aiming at the problems that in the prior art, the optimizing efficiency of the solid rocket engine charging configuration design is low and constraint conditions are difficult to process, the invention provides a solid rocket engine charging configuration constraint design method, which is based on a proxy model technology, and is based on a non-accurate searching efficient three-stage constraint sampling technology, and the global feasible optimal point of a model can be positioned while the optimizing efficiency is ensured by introducing the feasibility and the optimality of elite archive balance targets, so that the solid rocket engine charging configuration design efficiency is remarkably improved.
In order to achieve the purpose, the invention provides a solid rocket engine charging configuration constraint design method, which comprises the following steps:
step 1, establishing a constraint optimization model of a solid rocket engine charging configuration;
step 2, on the basis of the constraint optimization model, carrying out experimental design on design variables of the loading configuration of the solid rocket engine to obtain a group of initial sample points, and storing a plurality of optimal individuals in the current sample points into elite files;
step 3, constructing a proxy model of the solid rocket engine charge configuration constraint design based on the current partial or all sample points;
step 4, determining a sampling criterion according to the feasibility of each sample point in the elite file, and searching a new sample point meeting the sampling criterion;
step 5, judging whether the new sample point is better than the worst sample point in the elite file based on the agent model:
if yes, deleting the worst sample point in the elite file, and updating the new sample point to the elite file, and then carrying out step 6;
otherwise, go to step 6;
step 6, judging whether the optimal sample points in the elite file are continuous or notMThe next time not updated:
if yes, taking the optimal sample point in the elite file as a design result of a loading configuration of the solid rocket engine, and outputting the result;
otherwise, returning to the step 3.
Compared with the prior art, the invention has the following beneficial technical effects:
1. the invention can effectively process the constraint conditions existing in the process of designing the charge configuration of the solid rocket engine, has high design universality, can effectively avoid design optimization from leaving a feasible region, and has excellent design result performance;
2. compared with the general optimization method, the method has higher efficiency and faster design speed, greatly reduces the time consumption degree of charge design, and can effectively meet the rapid design requirement of engine charge.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to the structures shown in these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a method of constrained design of a solid rocket engine charge configuration in an embodiment of the present invention;
fig. 2 is a schematic illustration of a front and rear spar charge configuration in an example embodiment of the invention, wherein: (a) is a front wing section view of a front and rear wing column charge configuration, (b) is an overall axial section view of the front and rear wing column charge configuration, and (c) is a rear wing section view of the front and rear wing column charge configuration;
FIG. 3 is a plot of mass convergence of a grain in an embodiment of the invention;
FIG. 4 is a chart showing convergence of standard deviation with iteration number in an embodiment of the present invention;
FIG. 5 is a schematic diagram of engine combustion design results in an example of an embodiment of the invention.
The achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all embodiments of the invention. 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.
In addition, the technical solutions of the embodiments of the present invention may be combined with each other, but it is necessary to be based on the fact that those skilled in the art can implement the technical solutions, and when the technical solutions are contradictory or cannot be implemented, the combination of the technical solutions should be considered as not existing, and not falling within the scope of protection claimed by the present invention.
The embodiment discloses a solid rocket engine charge configuration constraint design method, which comprises the steps of firstly adopting an optimized Latin hypercube design method to carry out experimental design on design variables of solid rocket engine charge, obtaining a group of initial sample points, and constructing an initial proxy model. In the subsequent iteration process, the three-stage constraint sampling method based on the inaccurate search idea is used for constraint sampling and continuously updating the agent model, so that the accuracy of the agent model and the constraint model is continuously improved, the sampling and positioning of the truly feasible optimal point is promoted, and the efficient design of the charge configuration is realized. By providing a three-stage constraint sampling method based on an inaccurate search idea, the feasibility and the optimality of balanced optimization are ensured on the premise that the optimization result meets constraint condition setting. The design performance is improved while the simulation times are reduced, so that the design combustion face is more stable. Provides a rapid and accurate design method for the charge design of the solid rocket engine.
Referring to fig. 1, the method for designing the charge configuration constraint of the solid rocket engine in this embodiment specifically includes the following steps:
step 1, establishing a constraint optimization model of a solid rocket engine charging configuration;
step 2, on the basis of a constraint optimization model, carrying out experimental design on design variables of a loading configuration of the solid rocket engine to obtain a group of initial sample points, and storing a plurality of optimal individuals in the current sample points into elite files;
step 3, constructing a proxy model of the solid rocket engine charge configuration constraint design based on the current partial or all sample points;
step 4, establishing a three-stage constraint sampling method based on an inaccurate search idea, determining a sampling criterion according to the feasibility of each sample point in the current elite file, and searching a new sample point meeting the sampling criterion;
step 5, judging whether the new sample point is better than the worst sample point in the current elite archive based on the agent model:
if yes, deleting the worst sample point in the current elite file, and updating the new sample point to the elite file, and then carrying out step 6;
otherwise, go to step 6;
step 6, judging whether the optimal sample points in the elite file are continuousM(e.g.M=10) times without updating:
if yes, taking the optimal sample point in the current elite file as a design result of the loading configuration of the solid rocket engine, and outputting the result;
otherwise, returning to the step 3.
In the implementation process of the step 1, a constraint optimization model of the loading configuration of the solid rocket engine is specifically:
(1)
wherein ,representation->The design variable is->Sample dot, & lt>Representing the objective function of the design of the charge configuration of a solid rocket engine, < >>The>Design variable function of the individual inequality constraint, +.>The>Design of individual equality constraintsVariable function->、/>Respectively representing the inequality constraint and the number of inequality constraints, +.>、/>Respectively represent design variable +.>Lower limit, upper limit of (c).
In this embodiment, the standard deviation of the charge combustion area of the solid rocket engine is used as the objective function of the constraint optimization model. The design variable is a certain size parameter of the solid rocket engine charge, taking front and rear wing column type charges as an example, and the design variable can be one, two, more or all of the size parameters of rear wing length, rear wing width, rear wing inclination angle, rear wing number, front wing length, front wing width, front wing inclination angle, charge outer diameter, grain quality and the like.
The inequality constraint is that a certain or excessive size parameter of the solid rocket engine charge meets a certain inequality constraint, for example, an inequality constraint can be set for the grain quality of the solid rocket engine charge so as to require the grain quality of the solid rocket engine charge to be larger than a certain set value. Or an equality constraint can be set on the grain quality of the solid rocket engine charge so as to require that the grain quality of the solid rocket engine charge obtained by design is equal to a certain set value.
In the specific implementation process of the step 2, the optimization pull Ding Chao cube experiment design method adopting recursive evolution is adopted to generateN(N=2n) And a number of sample points.
In this embodiment, the concept of elite file is also introduced, so as to store a plurality of last sample points after each iteration, that is, store a part of sample points with the minimum objective function value, where the objective function value of the sample points is obtained by substituting the design variable parameters in the sample points into simulation software for simulation. And at the beginning of iteration, directly setting part or all of sample points as elite individuals in the elite file, wherein the number of the sample points of the elite file is kept unchanged all the time in the iteration process.
In the specific implementation process of the step 3, a radial basis interpolation method is adopted to construct a proxy model of the solid rocket engine charge configuration constraint design, which is:
(2)
wherein ,representing the sample point to be predicted as +.>Agent model of->Representing the number of sample points currently used to build the proxy model,/->Representing the basis function coefficients>Representing a Gaussian basis function +.>The Euclidean distance from an unknown sample point to a known sample point is expressed in the following specific form:
(3)
wherein ,representing shape parameters; shape parameter value +.>After the determination, in order to calculate the corresponding basis function coefficients +.>Interpolation conditions or least squares fitting are used to fit +.>The training samples are brought into the basic form (2) of the approximation model by introducing interpolation conditions +.>The system of linear equations for the basis function coefficients can be found as:
(4)
by solving the above linear equation set, the basis function coefficients can be obtainedωThe method comprises the following steps:
(5)
wherein ,representing the coefficient vector of the basis function>Representing the coefficient matrix calculated by taking all sample inputs into the radial basis method, each element in the matrix being calculated using equation (3), the sum of the coefficients is +.>Representing the objective function vector for all samples.
In the implementation process of step 4, a three-stage constraint sampling method based on an inaccurate search idea is set according to the feasibility of each sample point in the current elite archive, and specifically:
when all the sample points in the elite file are not feasible, the sampling criterion is to find the sample points with smaller violated constraint values;
when the sample point part in the elite file is not feasible, the sampling criterion is to find a sample point with smaller violation constraint value and better objective function value;
when all the sample points in the elite file are feasible, the sampling criterion is to find the sample points with better objective function values.
Further specifically, a constraint conflict function is introduced in the process of three-stage constraint sampling, and the constraint conflict function is an important index capable of effectively representing the distance between a current sample point and a feasible region and is a common method of a constraint optimization algorithm. Since the equality constraint belongs to a strong constraint, difficulties are brought to the feasible region of the sample point search positioning problem. The equation is therefore generally constrained to the following process:
(6)
wherein ,representing a small number of parameters, the original equality constraint can be converted into an inequality form by applying equation (6), in which case a sample point +.>Violation of the constraint optimization model +.>Constraint conflict value of individual constraint->The method comprises the following steps:
(7)
wherein ,the>Are not individually provided withDesign variable function of equality constraint ++>The>A design variable function constrained by the equations;
at this time, the sample pointConstraint conflict value of violation of all constraints +.>The method comprises the following steps:
(8)
notably, due to the difference in scale characteristics of different variables, constraints, partial constraints deviate slightly from the feasible domain within the design domain, resulting in large conflict values, resulting in constraint conflict values for these small constraints for this sample pointPlays a decisive role. Therefore, constraint conflicts for each constraint need to be normalized within the population to indicate the relative size of the solution constraint conflicts within the current population. The specific normalization is divided into the following two steps:
(1) Calculating the maximum conflict value of all sample points in elite file aiming at single constraintThe method comprises the following steps:
(9)
wherein ,representing an elite file;
(2) Based on maximum punchingBurst valueNormalizing to obtain sample point->Normalized constraint conflict value of all constraints in violation of constraint optimization model +.>The method comprises the following steps:
(10)
when (when)When it is, then indicate sample point +.>Within the feasible region, the sample point can be judged>To be viable, otherwise determine the sample point +.>It is not feasible. And->The larger the description sample point +.>The larger the relative constraint conflict in the population.
In this embodiment, the three-phase constraint sampling is based on the spreading of the samples under different conditions of the elite file, and the sampling criteria (sampling for achieving purposes) are different.
When all the sample points in the elite archive are not viable, i.e., at least one strong constraint exists, the filled target (the target to find the next new sample point) is near the boundary of the strong constraint, and the sample with the smaller constraint violation than the sample with the smallest constraint violation in the elite archive is selected. The sampling criterion at this time is to find sample points with smaller violated constraint values, namely:
(11)
wherein ,representing the new sample point searched, +.>A normalized constraint conflict value representing a new sample point,representing the smallest normalized constraint conflict value in the elite archive.
When a sample point portion in the elite file is not feasible, the new sample point searched will be diverted to a sample with a relatively lower constraint violation and a better target value than the best sample in the elite file, so that the sample around the constraint boundary that is not feasible with high target performance can be sampled. The sampling criterion at this time is to find a sample point with smaller violated constraint value and better objective function value, specifically:
(12)
wherein ,representing the optimal target value in elite file, +.>Representing the set of all infeasible sample points in elite archive, +.>A normalized constraint violation minimum representing a infeasible sample point in the elite archive.
When all sample points in the elite file are feasible, the new sample point searched for will be selected with the sample having the better target value than the best sample in the elite file. Once a new sample point is found, the search is stopped, avoiding an exhaustive search for inaccurate models, which may help the algorithm jump out of local optima. The sampling criterion at this time is to find a sample point with a better objective function value, specifically:
(13)
after determining the sampling criteria based on the feasibility of each sample point in the current elite archive, new sample points can be sampled, and the next new sample point is determined by solving the following equation (14)
(14)
wherein ,representing go +.>The most recent model-like built after sub-sampling, < >>Representing a sample set for approximate modeling, +.>Representing the distribution according to the current proxy model and sample points. The sampling criterion determined in the mode can be used for processing to obtain the next new sample point for high-precision simulation.
In the implementation process of step 5, the process of judging whether the new sample point is better than the worst sample point in the current elite archive based on the proxy model is specifically as follows:
defining a sample point with the maximum high-precision target simulation value in the current elite file as a worst sample point in the current elite file;
substituting the new sample point into the proxy model to obtain a low-precision target simulation value corresponding to the new sample point;
comparing the low-precision target simulation value of the new sample point with the high-precision target simulation value of the worst sample point in the current elite file, and if the low-precision target simulation value of the new sample point is smaller, judging that the new sample point is better than the worst sample point in the current elite file;
the high-precision target simulation value of the sample point is obtained through simulation.
Since the proxy model is a simple mathematical model built based on radial basis functions, equation (14) can be processed using conventional heuristic algorithms. The searching of the sample points in the formula (14) is realized through an optimization algorithm (particle swarm algorithm), the optimization algorithm finds the sample points meeting the sampling criteria according to the set sampling criteria, but the sample points meeting the sampling criteria do not necessarily meet the constraint domain limit and the objective function optimal limit, so that the situation that the optimal solution is not updated may occur in the iterative process (i.e. in the step 5), but even if the new sample points are not updated into elite files, the optimization algorithm searches for a new sample point again for calculation.
The method of designing the solid rocket motor charge configuration constraints in this embodiment is further described below with reference to specific examples.
Taking the design of an engine wing column type grain configuration as an example, the charging geometry is shown in fig. 2.
Giving engineering examples, wherein the continuous variable in the design variables is the geometric parameters of the charge, including the lengths of front and rear wing columns and />Width ∈of front wing>And width of rear wing->Depth-> and />Front wing dip +.>And rear wing dip->And the corresponding number of front and rear wings-> and />The method comprises the steps of carrying out a first treatment on the surface of the The discrete variable is the number of wings; constraint variable is grain design quality +.>The variable ranges are shown in table 1.
The design objective is to minimize the standard deviation of the combustion area by optimization in order to obtain smooth thrust performance. Aiming at the engineering calculation example, a multi-constraint agent model optimization method is applied to search, and the specific steps are as follows:
selecting in design space by adopting optimized Latin hypercube experimental design methodSample spots->Constructing an initial proxy model of targets and constraints for designing the number of variables;
searching the next sampling point by adopting a three-stage constraint sampling method and constructing an elite file;
high-precision simulation is carried out on the new sampling point to obtain a combustion surface curve;
updating the agent model sample points and elite files;
and (3) judging termination, outputting the current design parameters if the termination condition is met, otherwise, performing the next sampling according to the new proxy model.
TABLE 1 design variables for front and rear wing column charges and ranges thereof
Design cases of front and rear wing columns of an engine charging:
the initial sample point number is set to 20, and the convergence process of the target and constraint conditions is shown in fig. 3 and 4. In fig. 3 and 4, the open square points represent invalid samples, the solid square points represent valid samples, and the dots are optimal samples found in the optimization process. As shown in fig. 3 and 4, the method of the present invention finds the optimal result of the grain design in less than 120 iterations. And the constraint condition is converged to the boundary of the feasible region, which shows that the method has the capability of processing the constraint problem.
The best sample data results after the design is completed are shown in table 2. The standard deviation of the combustion area in the design process reaches the expected optimal design result. As shown in fig. 5, the combustion zone obtained using the design method of the present invention achieves standard-compliant stability. Compared with a general evolutionary algorithm, the method is higher in efficiency, the obtained result is more reliable, and the method has potential of engineering application.
TABLE 2 cases front and rear wingpost charge design results
As can be seen from the above examples, the present invention can obtain a better design result by adopting 116 calculations, and the number of calculations required by the method based on the intelligent optimization algorithm is far greater than that required by the present invention. The statistics of the number of iterations required for both are shown in table 3. By adopting constraint condition processing technology and efficient non-accurate sampling method, the method can obtain the charging geometric parameters with excellent performance by far less simulation times than that of the general intelligent optimization method, thereby greatly improving the design efficiency and fully verifying the effectiveness of the invention.
TABLE 3 simulation times required for different charge design methods
The foregoing description is only of the preferred embodiments of the present invention and is not intended to limit the scope of the invention, and all equivalent structural changes made by the description of the present invention and the accompanying drawings or direct/indirect application in other related technical fields are included in the scope of the invention.

Claims (8)

1. The method for restricting and designing the charging configuration of the solid rocket engine is characterized by comprising the following steps:
step 1, establishing a constraint optimization model of a solid rocket engine charging configuration;
step 2, on the basis of the constraint optimization model, carrying out experimental design on design variables of the loading configuration of the solid rocket engine to obtain a group of initial sample points, and storing a plurality of optimal individuals in the current sample points into elite files;
step 3, constructing a proxy model of the solid rocket engine charge configuration constraint design based on the current partial or all sample points;
step 4, determining a sampling criterion according to the feasibility of each sample point in the elite file, and searching a new sample point meeting the sampling criterion, wherein the method specifically comprises the following steps:
when all the sample points in the elite file are not feasible, the sampling criterion is to find the sample points with smaller violated constraint values;
when the sample point part in the elite file is not feasible, the sampling criterion is to search for a sample point with smaller violation constraint value and better objective function value;
when all the sample points in the elite file are feasible, the sampling criterion is to find the sample point with better objective function value;
the process for judging whether one sample point is feasible is as follows:
defining sample pointsXViolating the first constraint optimization modellConstraint conflict value for each constraintG l (X) The method comprises the following steps:
，
wherein ,g l (X) Design of charge configuration of solid rocket enginelA design variable function of the inequality constraint,h l (X) Design of charge configuration of solid rocket enginelA design variable function constrained by an equation,representing a small amount of parameters;
calculating the maximum conflict value of all current sample points aiming at single constraintThe method comprises the following steps:
，
wherein ,Xelite Representing an elite file;
based on the maximum collision valueNormalization processing is carried out to obtain sample pointsXNormalized constraint conflict values violating all constraints in the constraint optimization modelG nor (X) The method comprises the following steps:
，
when (when)When the sample point is determinedXTo be viable, otherwise determine sample pointsXIs not feasible;
step 5, judging whether the new sample point is better than the worst sample point in the elite file based on the agent model:
if yes, deleting the worst sample point in the elite file, and updating the new sample point to the elite file, and then carrying out step 6;
otherwise, go to step 6;
step 6, judging whether the optimal sample points in the elite file are continuous or notMThe next time not updated:
if yes, taking the optimal sample point in the elite file as a design result of a loading configuration of the solid rocket engine, and outputting the result;
otherwise, returning to the step 3.
2. The method for constrained design of a solid rocket engine charge configuration according to claim 1, wherein in step 1, the constrained optimization model is:
，
wherein ,representation ofnThe design variable isx 1 、x 2 、···、x n The point of the sample is at the point of the sample,f(X) Representing an objective function of the solid rocket engine charge configuration design,g j (X) Design of charge configuration of solid rocket enginejA design variable function of the inequality constraint,h k (X) Design of charge configuration of solid rocket enginekA design variable function constrained by an equation,p、mthe inequality constraint and the number of inequality constraints are expressed respectively,L i 、U i representing design variables respectivelyx i Lower limit, upper limit of (c).
3. The solid rocket motor charge configuration constraint design method according to claim 1, wherein in step 3, a radial basis interpolation method is adopted to construct a proxy model of the solid rocket motor charge configuration constraint design.
4. A solid rocket engine charge configuration constraint design method according to claim 1, 2 or 3, wherein when all sample points in the elite file are not feasible, the sampling criterion is to find sample points with smaller violation constraint values, in particular:
searching a sample point with a constraint conflict value smaller than the minimum constraint conflict value in the elite file, namely:
，
wherein ,X 0 representing the new sample point that was searched for,G nor (X 0 ) A normalized constraint conflict value representing a new sample point,representing the smallest normalized constraint conflict value in the elite archive.
5. A solid rocket engine charge configuration constraint design method according to claim 1, 2 or 3, wherein when the sample point part in the elite archive is not feasible, the sampling criterion is to find a sample point with smaller violation constraint value and better objective function value, specifically:
，
wherein ,X 0 representing the new sample point that was searched for,G nor (X 0 ) A normalized constraint conflict value representing a new sample point,f min (X elite ) Represents the optimal target value in the elite file,representing a set of all infeasible sample points in an elite archive,a normalized constraint violation minimum representing a infeasible sample point in the elite archive.
6. A solid rocket engine charge configuration constraint design method according to claim 1, 2 or 3, wherein when all the sample points in the elite file are feasible, the sampling criterion is to find the sample point with the better objective function value, specifically:
，
wherein ,X 0 representing the new sample point that was searched for,G nor (X 0 ) A normalized constraint conflict value representing a new sample point,f min (X elite ) Representing the optimal target value in the elite file.
7. A solid rocket engine charge configuration constraint design method according to claim 1, 2 or 3, wherein in step 5, the process of judging whether the new sample point is better than the worst sample point in the elite file based on the proxy model is specifically as follows:
defining a sample point with the maximum high-precision target simulation value in the current elite file as a worst sample point in the current elite file;
substituting the new sample point into the proxy model to obtain a low-precision target simulation value corresponding to the new sample point;
comparing the low-precision target simulation value of the new sample point with the high-precision target simulation value of the worst sample point in the elite file, and judging that the new sample point is better than the worst sample point in the elite file if the low-precision target simulation value of the new sample point is smaller;
the high-precision target simulation value of the sample point is obtained through simulation.
8. A method of constrained design of a solid rocket engine charge configuration according to claim 1, 2 or 3, wherein in step 2, the design variables of the solid rocket engine charge configuration are experimentally designed by using an optimized latin hypercube design method, so as to obtain a set of initial sample points.
CN202311029449.6A 2023-08-16 2023-08-16 Solid rocket engine charging configuration constraint design method Active CN116738583B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202311029449.6A CN116738583B (en) 2023-08-16 2023-08-16 Solid rocket engine charging configuration constraint design method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202311029449.6A CN116738583B (en) 2023-08-16 2023-08-16 Solid rocket engine charging configuration constraint design method

Publications (2)

Publication Number Publication Date
CN116738583A CN116738583A (en) 2023-09-12
CN116738583B true CN116738583B (en) 2023-10-31

Family

ID=87906534

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202311029449.6A Active CN116738583B (en) 2023-08-16 2023-08-16 Solid rocket engine charging configuration constraint design method

Country Status (1)

Country Link
CN (1) CN116738583B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117556551B (en) * 2024-01-11 2024-04-09 中国人民解放军国防科技大学 Lightweight design method, device, equipment and medium for engine combustion chamber shell

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7499890B1 (en) * 2003-10-14 2009-03-03 At&T Mobility Ii Llc Cellsite budgetmaster
CN105975651A (en) * 2016-04-12 2016-09-28 哈尔滨工业大学 Genetic particle swarm multi-disciplinary design optimization algorithm-based guided missile parameter design method
CN111783251A (en) * 2020-07-16 2020-10-16 中国人民解放军国防科技大学 Method for designing overall parameters of solid rocket engine
CN111881614A (en) * 2020-09-28 2020-11-03 中国人民解放军国防科技大学 Solid rocket engine charging characterization method
CN112052521A (en) * 2020-09-18 2020-12-08 中国人民解放军国防科技大学 Solid engine charging configuration design method based on continuous-discrete mixing optimization
CN112149228A (en) * 2020-09-25 2020-12-29 中国人民解放军国防科技大学 Progressive matching design method for performance of solid rocket engine
CN112528441A (en) * 2021-02-18 2021-03-19 中国人民解放军国防科技大学 Throat-plug type variable thrust engine overall parameter design method, device and equipment
CN113255082A (en) * 2021-06-21 2021-08-13 中国人民解放军国防科技大学 Integrated optimization design method for throat plug spray pipe profile of variable-thrust solid engine
CN113297686A (en) * 2021-07-27 2021-08-24 中国人民解放军国防科技大学 Solid rocket engine data fusion design method, device, equipment and medium
CN115238421A (en) * 2022-09-23 2022-10-25 中国人民解放军国防科技大学 Method and device for designing charging configuration of multi-pulse gas generator and computer equipment

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7937945B2 (en) * 2006-10-27 2011-05-10 Kinde Sr Ronald August Combining a series of more efficient engines into a unit, or modular units

Patent Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7499890B1 (en) * 2003-10-14 2009-03-03 At&T Mobility Ii Llc Cellsite budgetmaster
CN105975651A (en) * 2016-04-12 2016-09-28 哈尔滨工业大学 Genetic particle swarm multi-disciplinary design optimization algorithm-based guided missile parameter design method
CN111783251A (en) * 2020-07-16 2020-10-16 中国人民解放军国防科技大学 Method for designing overall parameters of solid rocket engine
CN112052521A (en) * 2020-09-18 2020-12-08 中国人民解放军国防科技大学 Solid engine charging configuration design method based on continuous-discrete mixing optimization
CN112149228A (en) * 2020-09-25 2020-12-29 中国人民解放军国防科技大学 Progressive matching design method for performance of solid rocket engine
CN111881614A (en) * 2020-09-28 2020-11-03 中国人民解放军国防科技大学 Solid rocket engine charging characterization method
CN112528441A (en) * 2021-02-18 2021-03-19 中国人民解放军国防科技大学 Throat-plug type variable thrust engine overall parameter design method, device and equipment
CN113255082A (en) * 2021-06-21 2021-08-13 中国人民解放军国防科技大学 Integrated optimization design method for throat plug spray pipe profile of variable-thrust solid engine
CN113297686A (en) * 2021-07-27 2021-08-24 中国人民解放军国防科技大学 Solid rocket engine data fusion design method, device, equipment and medium
CN115238421A (en) * 2022-09-23 2022-10-25 中国人民解放军国防科技大学 Method and device for designing charging configuration of multi-pulse gas generator and computer equipment

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
基于代理模型的固体火箭发动机装药设计;武泽平;王东辉;胡凡;张为华;;固体火箭技术(03);第321-326页 *

Also Published As

Publication number Publication date
CN116738583A (en) 2023-09-12

Similar Documents

Publication Publication Date Title
CN111783251B (en) Method for designing overall parameters of solid rocket engine
CN116738583B (en) Solid rocket engine charging configuration constraint design method
CN112528441B (en) Throat-plug type variable thrust engine overall parameter design method, device and equipment
CN112052521B (en) Solid engine charging configuration design method based on continuous-discrete mixing optimization
Liu et al. Comparison of infill sampling criteria in kriging-based aerodynamic optimization
CN107391891B (en) Large-aspect-ratio wing optimization design method based on model fusion method
CN112149228B (en) Progressive matching design method for performance of solid rocket engine
CN113297686B (en) Solid rocket engine data fusion design method, device, equipment and medium
Zhao et al. Pigeon-inspired optimization applied to constrained gliding trajectories
Van Nguyen et al. Adaptive multifidelity constraints method for efficient multidisciplinary missile design framework
CN117094090A (en) Solid engine overall performance rapid calculation method for heterogeneous scheme knowledge migration
CN115906286A (en) Rocket design method and device with coupled inner and outer trajectories, electronic equipment and storage medium
CN109783918B (en) Speed reducer optimization design implementation method based on switching of sequential sampling mode
Zhao et al. Optimization of the aerodynamic configuration of a tubular projectile based on blind kriging
CN116776748B (en) Throat bolt type variable thrust engine throat bolt spray pipe configuration design knowledge migration optimization method
Hao et al. Design optimization by integrating limited simulation data and shape engineering knowledge with Bayesian optimization (BO-DK4DO)
Lengyel-Kampmann et al. Generalized optimization of counter-rotating and single-rotating fans
CN116646976A (en) Distributed photovoltaic cluster division method and system based on aggregation hierarchical clustering algorithm
CN113761659B (en) Nested multi-attribute evaluation decision-making method and device for solid-liquid power-like space vehicles
CN114996880A (en) Composite armor structure optimization method based on ANSYS secondary development
Koziel et al. Multi-objective airfoil design using variable-fidelity CFD simulations and response surface surrogates
Amrit et al. Efficient multi-objective aerodynamic optimization by design space dimension reduction and co-kriging
Choi et al. Cost-aware adaptive design of experiment with Nonstationary surrogate model for wind tunnel testing
Epstein et al. Automatic optimization of wing–body–under-the-wing-mounted-nacelle configurations
Bergsma et al. Application of Taylor-series integration to reentry problems with wind

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