CN113627025A - Optimized design method for sheet forming process - Google Patents

Optimized design method for sheet forming process Download PDF

Info

Publication number
CN113627025A
CN113627025A CN202110934614.7A CN202110934614A CN113627025A CN 113627025 A CN113627025 A CN 113627025A CN 202110934614 A CN202110934614 A CN 202110934614A CN 113627025 A CN113627025 A CN 113627025A
Authority
CN
China
Prior art keywords
particle
response surface
particles
ratio
forming process
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.)
Granted
Application number
CN202110934614.7A
Other languages
Chinese (zh)
Other versions
CN113627025B (en
Inventor
赵子衡
肖颖
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.)
Xiangtan University
Original Assignee
Xiangtan University
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 Xiangtan University filed Critical Xiangtan University
Priority to CN202110934614.7A priority Critical patent/CN113627025B/en
Publication of CN113627025A publication Critical patent/CN113627025A/en
Application granted granted Critical
Publication of CN113627025B publication Critical patent/CN113627025B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00Computer-aided design [CAD]
    • G06F30/20Design optimisation, verification or simulation
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00Computer-aided design [CAD]
    • G06F30/10Geometric CAD
    • G06F30/17Mechanical parametric or variational design
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computing arrangements based on biological models
    • G06N3/004Artificial life, i.e. computing arrangements simulating life
    • G06N3/006Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F2111/00Details relating to CAD techniques
    • G06F2111/06Multi-objective optimisation, e.g. Pareto optimisation using simulated annealing [SA], ant colony algorithms or genetic algorithms [GA]
    • YGENERAL 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
    • Y02TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02PCLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30Computing systems specially adapted for manufacturing

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Geometry (AREA)
  • General Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Computer Hardware Design (AREA)
  • Biophysics (AREA)
  • Computing Systems (AREA)
  • Artificial Intelligence (AREA)
  • Biomedical Technology (AREA)
  • Health & Medical Sciences (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Pure & Applied Mathematics (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

Compared with the prior art, the optimized design method for the sheet forming process comprises the following steps: s1, determining a decision space and an objective function; s2, experimental design: sampling in the decision space by a Latin hypercube test design method to obtain uniformly distributed sample points, and carrying out simulation analysis on each sample point; s3, constructing a response surface agent model; s4, testing the precision of the response surface proxy model; s5, optimizing and solving: performing optimization solution on the response surface agent model based on a multi-objective gradient enhanced particle swarm algorithm to obtain a balance curve of the objective function, and selecting a satisfactory solution on the balance curve according to specific engineering requirements; and S6, simulation verification. Compared with the prior art, the method is based on the multi-target gradient enhanced particle swarm algorithm, the plate forming optimization efficiency can be effectively improved, the optimal forming process parameter combination can be rapidly determined, the part processing quality is guaranteed, and the production efficiency is improved.

Description

Optimized design method for sheet forming process
Technical Field
The application relates to the technical field of sheet forming optimization, in particular to an optimization design method for a sheet forming process.
Background
The sheet metal stamping forming is to utilize the characteristic of plastic deformation of a metal sheet, apply certain pressure on the sheet metal on a stamping device by means of a die, and enable the sheet metal to generate plastic deformation under the action of the pressure, so as to obtain formed parts in different shapes. The process parameters influencing the sheet forming are very many, including the material parameters of the sheet, the stamping speed, the blank holder force, the clearance between the dies, the chamfer angle of the dies, the friction coefficient and the like. If these parameters are not properly selected, the machining accuracy and surface quality of the part can be affected, and failure in the form of wrinkling, cracking, and bouncing can even occur.
The sheet forming is a complex highly nonlinear process, the optimization process is complex in calculation and time-consuming in solving, optimization solving is carried out on the basis of an intelligent algorithm in the existing sheet forming optimization design, process parameters are repeatedly adjusted through experience to carry out experiments until a satisfactory forming result is obtained, the method is time-consuming and labor-consuming, the determined process parameters are not the optimal parameter combination, and the intelligent optimization algorithm is very time-consuming in solving, so that the optimization efficiency is influenced.
Therefore, how to provide a method for optimally designing a sheet forming process, which is based on a multi-objective gradient enhanced particle swarm algorithm, can effectively improve the sheet forming optimization efficiency, quickly determine the optimal forming process parameter combination, ensure the part processing quality and improve the production efficiency, has become a technical problem to be solved by technical personnel in the field.
Disclosure of Invention
In order to solve the technical problems, the application provides a plate forming process optimization design method which is based on a multi-objective gradient enhanced particle swarm algorithm, can effectively improve plate forming optimization efficiency, quickly determines an optimal forming process parameter combination, ensures part processing quality and improves production efficiency.
The technical scheme provided by the application is as follows:
the application provides a sheet forming process optimization design method, which comprises the following steps: s1, determining a decision space and an objective function: performing analog simulation on the part forming process, analyzing the forming characteristics of the part, determining an optimization target and an objective function, and determining design parameters and an approximate value range of the part forming through single-factor experimental analysis; s2, experimental design: sampling in the decision space by a Latin hypercube test design method to obtain uniformly distributed sample points, and performing simulation analysis on each sample point to obtain a simulation value of the target function; s3, constructing a response surface proxy model: constructing a response surface agent model based on the data of the sample points, and constructing a mapping relation between the design parameters and the target function through the response surface agent module; s4, response surface proxy model precision test: performing precision analysis on the response surface agent model, and if the precision meets the engineering requirement, jumping to the next step; if the precision does not meet the engineering requirement, jumping back to the step S2, increasing the number of the sample points by a Latin hypercube test design method, and continuing to key the response surface proxy model until the precision of the response surface proxy model meets the engineering requirement; s5, optimizing and solving: performing optimization solution on the response surface agent model based on a multi-objective gradient enhanced particle swarm algorithm to obtain a balance curve of the objective function, and selecting a satisfactory solution on the balance curve according to specific engineering requirements; s6, simulation verification: and carrying out simulation verification on the satisfactory solution, and verifying the reliability of the optimization result.
Further, in a preferred embodiment of the present invention, the step of performing an optimization solution on the response surface proxy model based on the multi-objective gradient enhanced particle swarm optimization includes:
s501, initializing population and initializing parameters: setting the steepest descent term coefficient c1Global learning factor c2And a repulsive direction coefficient c3Maximum number of iterations kmaxThe ratio S of the judgment of the congested area, the upper limit u of the search spacenAnd ulConvergence precision xi of the algorithm, and upper and lower limits v of the particle velocitymaxAnd vmin(ii) a Initializing a population and an external set, and randomly generating an initial position and an initial speed of each particle in the population;
s502, calculating a fitness value, evaluating particles, calculating the fitness value of each particle, comparing all solutions according to a Pareto domination relation, and storing non-inferior solutions in the external set;
s503, calculating a first ratio of the number of the non-inferior solutions in each hypercube in the external set to the total number of the solutions in the external set;
s504, calculating a second ratio of the number of the non-inferior solutions in the hypercube to the total number of the non-inferior solutions in the external set, comparing and analyzing the second ratio with the crowded area judgment ratio S, and updating the speed and the position of each particle according to a comparison result.
Further, in a preferred embodiment of the present invention, the step of performing an optimization solution on the response surface proxy model based on the multi-objective gradient enhanced particle swarm optimization further includes:
s505, calculating the fitness value of each particle after updating;
s506, comparing the fitness value of each particle in the population with the non-inferior solutions in the external set according to a Pareto dominant solution, storing the non-inferior solutions, and deleting the dominant solution;
s507, when the number of the non-inferior solutions of the external set reaches a set value, maintaining the external set according to a self-adaptive grid method of a multi-target particle swarm algorithm, storing superior non-inferior solutions, and removing poor solutions;
and S508, judging whether the algorithm meets the condition of iteration stop, if so, stopping iteration and outputting a result, otherwise, skipping to the step S503 to continue circulation.
Further, in a preferred aspect of the present invention, in the step S504, the comparing and analyzing of the second ratio and the congested area determination ratio S includes:
if the second ratio is smaller than or equal to the crowded area judgment ratio, the rejection condition is not met, all the particles are indicated to do guiding motion, the global optimal solution is selected according to a self-adaptive grid method of the multi-target particle swarm algorithm, and the speed and the position of the particles are updated through a guiding motion updating rule.
Further, in a preferred embodiment of the present invention, the guiding motion updating rule is specifically: and calling a guiding movement speed updating formula to update the speed of each particle, and calling a position updating formula to update the position of each particle.
Further, in a preferred embodiment of the present invention, the formula for updating the guiding movement speed is specifically:
vi(k+1)=wvi(k)+r1c1gspeed(k)+r2c2(gBESTi(k)-xi(k))
in the formula: w is the inertial weight; 1,2,3.. m; m is the total number of particles in the particle swarm; k, the iteration number, represents the kth iteration; r is1And r2∈[0,1]Two random numbers in between; v. ofi(k) -the guiding motion velocity of the ith particle, the kth iteration; greed (k) -kth iteration, the descent direction searched for by the ith particle at the current position; gBESTi(k) The best positions searched so far by all particles, called global optimal solution; x is the number ofi(k) -the position of the ith particle.
Further, in a preferred embodiment of the present invention, the location update formula is specifically:
xi(k+1)=xi(k)+vi(k+1)
further, in a preferred aspect of the present invention, in the step S504, the comparing and analyzing the second ratio with the congested area judgment ratio S further includes:
if the second ratio is larger than the crowded area judgment ratio, the rejection condition is met, a rejection center is determined, particles close to the rejection center are selected according to the crowded area judgment ratio to do rejection movement, the speed and the position of the particles are updated according to a rejection movement update rule, the rest particles do guide movement, and the speed and the position of the rest particles are updated according to a guide movement update rule.
Further, in a preferred embodiment of the present invention, the repulsive-motion update rule is specifically: and calling a repulsive movement updating formula to update the speed of the particles, and calling a position updating formula to update the positions of the particles.
Further, in a preferred embodiment of the present invention, the repulsive movement velocity update formula is specifically:
vi(k+1)=wvi(k)+r1c1gspeed(k)+r2c3respeedi(k)
in the formula: w is the inertial weight; r is1And r2∈[0,1]Two random numbers in between; v. ofi(k) -the ith particle, the repulsive motion velocity of the kth iteration; greed (k) -kth iteration, the descent direction searched for by the ith particle at the current position; respeed (k) -the direction of repulsion of the ith particle in the population.
Compared with the prior art, the optimized design method for the sheet forming process provided by the invention comprises the following steps: s1, determining a decision space and an objective function: performing analog simulation on the part forming process, analyzing the forming characteristics of the part, determining an optimization target and an objective function, and determining design parameters and an approximate value range of the part forming through single-factor experimental analysis; s2, experimental design: sampling in the decision space by a Latin hypercube test design method to obtain uniformly distributed sample points, and performing simulation analysis on each sample point to obtain a simulation value of the target function; s3, constructing a response surface proxy model: constructing a response surface agent model based on the data of the sample points, and constructing a mapping relation between the design parameters and the target function through the response surface agent module; s4, response surface proxy model precision test: performing precision analysis on the response surface agent model, and if the precision meets the engineering requirement, jumping to the next step; if the precision does not meet the engineering requirement, jumping back to the step S2, increasing the number of the sample points by a Latin hypercube test design method, and continuing to key the response surface proxy model until the precision of the response surface proxy model meets the engineering requirement; s5, optimizing and solving: performing optimization solution on the response surface agent model based on a multi-objective gradient enhanced particle swarm algorithm to obtain a balance curve of the objective function, and selecting a satisfactory solution on the balance curve according to specific engineering requirements; s6, simulation verification: and carrying out simulation verification on the satisfactory solution, and verifying the reliability of the optimization result. Because the convergence speed of the intelligent algorithm is low in the sheet forming process, the efficiency of the intelligent algorithm is influenced in the forming optimization process, a multi-objective optimization algorithm with high convergence, namely a multi-objective gradient enhanced particle swarm algorithm, is designed for forming optimization; in the optimization design method, firstly, the sheet forming process is analyzed, and the target and the influence factor which need to be optimized are determined; then sampling in a design space by an experimental method to obtain a sample; then, simulating the sample to obtain a value of the target function; and then, a mapping relation between the design parameters and the objective function is constructed by using the response surface proxy model, and finally, the response surface proxy model is optimized and solved by using the multi-objective gradient enhanced particle swarm algorithm, so that the efficiency of the sheet forming optimization design can be improved, and the production cost is reduced. Compared with the prior art, the technical scheme provided by the invention is based on the multi-target gradient enhanced particle swarm algorithm, so that the plate forming optimization efficiency can be effectively improved, the optimal forming process parameter combination can be rapidly determined, the part processing quality is ensured, and the production efficiency is improved.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a flow chart of steps of a sheet forming process optimization design method according to an embodiment of the present invention;
FIG. 2 is a flowchart of the steps of performing optimization solution on a response surface proxy model based on a multi-objective gradient enhanced particle swarm algorithm according to an embodiment of the present invention;
FIG. 3 is a frame flowchart of the method for optimally designing the sheet forming process according to the embodiment of the present invention;
FIG. 4 is a framework flowchart for performing optimization solution on a response surface proxy model based on a multi-objective gradient enhanced particle swarm optimization algorithm according to an embodiment of the present invention;
FIG. 5 is a schematic diagram illustrating velocity and position updates of particles in guiding motion according to an embodiment of the present invention;
fig. 6 is a schematic diagram of the velocity and position update when the particles do repulsive movement according to the embodiment of the present invention.
Detailed Description
In order to make those skilled in the art better understand the technical solutions in the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
It will be understood that when an element is referred to as being "fixed" or "disposed" on another element, it can be directly on the other element or be indirectly disposed on the other element; when an element is referred to as being "connected to" another element, it can be directly connected to the other element or be indirectly connected to the other element.
It will be understood that the terms "length," "width," "upper," "lower," "front," "rear," "first," "second," "vertical," "horizontal," "top," "bottom," "inner," "outer," and the like, as used herein, refer to an orientation or positional relationship indicated in the drawings that is solely for the purpose of facilitating the description and simplifying the description, and do not indicate or imply that the device or element being referred to must have a particular orientation, be constructed and operated in a particular orientation, and thus should not be construed as limiting the application.
Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of the present application, "plurality" or "a plurality" means two or more unless specifically limited otherwise.
It should be understood that the structures, ratios, sizes, and the like shown in the drawings are only used for matching the disclosure of the specification, so as to be understood and read by those skilled in the art, and are not used to limit the practical limit conditions of the present application, so that the modifications of the structures, the changes of the ratio relationships, or the adjustment of the sizes, do not have the technical essence, and the modifications, the changes of the ratio relationships, or the adjustment of the sizes, are all within the scope of the technical contents disclosed in the present application without affecting the efficacy and the achievable purpose of the present application.
As shown in fig. 1 to 6, the method for optimally designing a sheet metal forming process provided in the embodiment of the present application includes the following steps: s1, determining a decision space and an objective function: performing analog simulation on the part forming process, analyzing the forming characteristics of the part, determining an optimization target and an objective function, and determining design parameters and an approximate value range of the part forming through single-factor experimental analysis; s2, experimental design: sampling in the decision space by a Latin hypercube test design method to obtain uniformly distributed sample points, and performing simulation analysis on each sample point to obtain a simulation value of the target function; s3, constructing a response surface proxy model: constructing a response surface agent model based on the data of the sample points, and constructing a mapping relation between the design parameters and the target function through the response surface agent module; s4, response surface proxy model precision test: performing precision analysis on the response surface agent model, and if the precision meets the engineering requirement, jumping to the next step; if the precision does not meet the engineering requirement, jumping back to the step S2, increasing the number of the sample points by a Latin hypercube test design method, and continuing to key the response surface proxy model until the precision of the response surface proxy model meets the engineering requirement; s5, optimizing and solving: performing optimization solution on the response surface agent model based on a multi-objective gradient enhanced particle swarm algorithm to obtain a balance curve of the objective function, and selecting a satisfactory solution on the balance curve according to specific engineering requirements; s6, simulation verification: and carrying out simulation verification on the satisfactory solution, and verifying the reliability of the optimization result.
The invention provides an optimized design method for a sheet forming process, which specifically comprises the following steps: s1, determining a decision space and an objective function: performing analog simulation on the part forming process, analyzing the forming characteristics of the part, determining an optimization target and an objective function, and determining design parameters and an approximate value range of the part forming through single-factor experimental analysis; s2, experimental design: sampling in the decision space by a Latin hypercube test design method to obtain uniformly distributed sample points, and performing simulation analysis on each sample point to obtain a simulation value of the target function; s3, constructing a response surface proxy model: constructing a response surface agent model based on the data of the sample points, and constructing a mapping relation between the design parameters and the target function through the response surface agent module; s4, response surface proxy model precision test: performing precision analysis on the response surface agent model, and if the precision meets the engineering requirement, jumping to the next step; if the precision does not meet the engineering requirement, jumping back to the step S2, increasing the number of the sample points by a Latin hypercube test design method, and continuing to key the response surface proxy model until the precision of the response surface proxy model meets the engineering requirement; s5, optimizing and solving: performing optimization solution on the response surface agent model based on a multi-objective gradient enhanced particle swarm algorithm to obtain a balance curve of the objective function, and selecting a satisfactory solution on the balance curve according to specific engineering requirements; s6, simulation verification: and carrying out simulation verification on the satisfactory solution, and verifying the reliability of the optimization result. Because the convergence speed of the intelligent algorithm is low in the sheet forming process, the efficiency of the intelligent algorithm is influenced in the forming optimization process, a multi-objective optimization algorithm with high convergence, namely a multi-objective gradient enhanced particle swarm algorithm, is designed for forming optimization; in the optimization design method, firstly, the sheet forming process is analyzed, and the target and the influence factor which need to be optimized are determined; then sampling in a design space by an experimental method to obtain a sample; then, simulating the sample to obtain a value of the target function; and then, a mapping relation between the design parameters and the objective function is constructed by using the response surface proxy model, and finally, the response surface proxy model is optimized and solved by using the multi-objective gradient enhanced particle swarm algorithm, so that the efficiency of the sheet forming optimization design can be improved, and the production cost is reduced. Compared with the prior art, the technical scheme provided by the invention is based on the multi-target gradient enhanced particle swarm algorithm, so that the plate forming optimization efficiency can be effectively improved, the optimal forming process parameter combination can be rapidly determined, the part processing quality is ensured, and the production efficiency is improved.
Specifically, in the embodiment of the present invention, the step of performing an optimization solution on the response surface proxy model based on the multi-objective gradient enhanced particle swarm optimization includes:
s501, initializing population and initializing parameters: setting the steepest descent term coefficient c1Global learning factor c2And a repulsive direction coefficient c3Maximum number of iterations kmaxThe ratio S of the judgment of the congested area, the upper limit u of the search spacenAnd ulConvergence precision xi of the algorithm, and upper and lower limits v of the particle velocitymaxAnd vmin(ii) a Initializing a population and an external set, and randomly generating an initial position and an initial speed of each particle in the population;
s502, calculating a fitness value, evaluating particles, calculating the fitness value of each particle, comparing all solutions according to a Pareto domination relation, and storing non-inferior solutions in the external set;
s503, calculating a first ratio of the number of the non-inferior solutions in each hypercube in the external set to the total number of the solutions in the external set;
s504, calculating a second ratio of the number of the non-inferior solutions in the hypercube to the total number of the non-inferior solutions in the external set, comparing and analyzing the second ratio with the crowded area judgment ratio S, and updating the speed and the position of each particle according to a comparison result.
S505, calculating the fitness value of each particle after updating;
s506, comparing the fitness value of each particle in the population with the non-inferior solutions in the external set according to a Pareto dominant solution, storing the non-inferior solutions, and deleting the dominant solution;
s507, when the number of the non-inferior solutions of the external set reaches a set value, maintaining the external set according to a self-adaptive grid method of a multi-target particle swarm algorithm, storing superior non-inferior solutions, and removing poor solutions;
and S508, judging whether the algorithm meets the condition of iteration stop, if so, stopping iteration and outputting a result, otherwise, skipping to the step S503 to continue circulation.
The multi-target gradient enhanced particle swarm optimization is a classical multi-target particle swarm optimization enhanced based on a multi-target gradient synchronous descent method, and both the improvement and the programming of the optimization are based on the multi-target particle swarm optimization; the particle swarm algorithm is a bionic algorithm obtained by simulating a foraging process of a bird swarm, and the multi-target gradient value enhanced particle swarm algorithm is equivalent to the fact that a radar is installed on each bird, and the radar can just guide each bird to find the nearest food source: when the bird group is too densely eaten, a natural enemy of birds is placed in a dense area to promote the bird group to spread from the crowded area according to a certain rule, and the radar is a rapid descending direction searched by a multi-target gradient synchronous descending method at the current position of the particles, and the rapid descending direction is combined with a speed updating direction of a particle swarm algorithm to improve the convergence speed of the algorithm; and the 'natural enemy' mechanism is a rejection mechanism for the algorithm construction, when the algorithm generates a crowded area, in order to avoid falling into a local optimal solution, a rejection center is selected in the crowded area, and particles in the crowded area are diffused to the periphery, so that the solution diversity is ensured. The convergence efficiency of the enhanced multi-target particle swarm algorithm is remarkably improved.
Specifically, in an embodiment of the present invention, the population is constructed from particles, each of which represents one possible solution in the optimization problem.
Specifically, in the embodiment of the present invention, in the step S504, the comparing and analyzing of the second ratio and the congested area judgment ratio S includes:
if the second ratio is smaller than or equal to the crowded area judgment ratio, the rejection condition is not met, all the particles are indicated to do guide motion, a global optimal solution is selected according to a self-adaptive grid method of a multi-target particle swarm algorithm, and the speed and the positions of the particles are updated through a guide motion updating rule;
if the second ratio is larger than the crowded area judgment ratio, the rejection condition is met, a rejection center is determined, particles close to the rejection center are selected according to the crowded area judgment ratio to do rejection movement, the speed and the position of the particles are updated according to a rejection movement update rule, the rest particles do guide movement, and the speed and the position of the rest particles are updated according to a guide movement update rule.
Specifically, in the embodiment of the present invention, the step of selecting the particles close to the repulsion center specifically includes the following steps:
selecting a repulsion center, and if the proportion of the total number of the non-inferior solutions in the ith hypercube to the total number of the external solutions exceeds the judgment proportion s of the crowded area, randomly selecting a particle in the hypercube, taking the particle as the repulsion center of the surrounding particles, and taking the point as the center of the nearby particles to do repulsion motion;
calculating the distances from all the particles to the repulsion center, and arranging the particles from small to large according to the distance;
and selecting the particles with smaller distance from all the particles according to the same crowded area judgment ratio s from small to large distance from the particles to the repulsion center to do repulsive movement, and conducting guiding movement to the rest.
Specifically, in an embodiment of the present invention, the guiding motion updating rule is specifically:
and calling a guiding movement speed updating formula to update the speed of each particle, and calling a position updating formula to update the position of each particle.
Specifically, in the embodiment of the present invention, the update rule for the repulsive movement specifically includes: and calling a repulsive movement updating formula to update the speed of the particles, and calling a position updating formula to update the positions of the particles.
Specifically, in the embodiment of the present invention, the guiding movement speed updating formula is specifically:
vi(k+1)=wvi(k)+r1c1gspeed(k)+r2c2(gBESTi(k)-xi(k))
in the formula: w is the inertia weight, which enables the particle to keep the velocity inertia of the last iteration in the iteration process of the algorithm; 1,2,3.. m; m is the total number of particles in the particle swarm; k, the iteration number, represents the kth iteration; r is1And r2∈[0,1]Two random numbers in between; v. ofi(k) -the guiding motion velocity of the ith particle, the kth iteration; greed (k) -kth iteration, the descent direction searched for by the ith particle at the current position; gBESTi(k) The best positions searched so far by all particles, called global optimal solution; x is the number ofi(k) -the position of the ith particle.
In the guiding movement speed updating formula, a first term is an inertia term, a second term is a descending direction searched by a multi-target gradient synchronous descending method, and a third term is a global optimal term; the maintenance strategy of the multi-target gradient enhanced particle swarm algorithm external set and the selection method of the global optimal value are consistent with those of the multi-target particle swarm algorithm, a self-adaptive grid method is adopted, and the algorithm improvement comprises the following steps: and (3) guiding motion, namely, improving the convergence speed of the multi-target particle swarm algorithm based on the convergence performance of the multi-target gradient synchronous descent method, namely, adding the descending direction searched by the multi-target gradient synchronous descent method into a speed updating formula of the multi-target particle swarm algorithm, and deleting the local optimal item in the formula to obtain a speed updating formula of the particles as guiding motion, namely the guiding motion speed updating formula.
And in the whole iterative convergence process, if the algorithm which is not detected by the rejection mechanism falls into a local optimal solution or generates an overcrowded area, the velocity of the particle is updated according to the guiding motion velocity updating formula all the time, and guiding motion is carried out. Under the action of the guiding movement speed updating formula, the particles can converge to the Pareto front surface at a higher speed, and after the updating speed of the particles is finished, the positions of the particles are updated according to the position updating formula.
Specifically, in the embodiment of the present invention, the repulsive movement velocity update formula is specifically:
vi(k+1)=wvi(k)+r1c1gspeed(k)+r2c3respeedi(k)
in the formula: w is the inertia weight, which enables the particle to keep the velocity inertia of the last iteration in the iteration process of the algorithm; r is1And r2∈[0,1]Two random numbers in between; v. ofi(k) -the ith particle, the repulsive motion velocity of the kth iteration; greed (k) -kth iteration, the descent direction searched for by the ith particle at the current position; respeed (k) -the direction of repulsion of the ith particle in the population.
In the repulsive movement velocity updating formula, a first term is an inertia term, a second term is a descending direction searched by a multi-target gradient synchronism descending method, and a third term is a repulsive term; the algorithm improvement further comprises: the exclusion mechanism is used for judging whether the algorithm falls into the local optimal solution or not, and if an overcrowded area appears in the population, the exclusion mechanism is started; and repulsive movement, starting a repulsive mechanism, diffusing the particles in the crowded area to the periphery through the repulsive movement, taking the vertical direction from the current position of the particles to a repulsive center as a repulsive direction, and combining the descending direction of the multi-target gradient synchronous descending method and the inertia item of the multi-target particle swarm algorithm to form a velocity updating formula of the repulsive movement, namely the velocity updating formula of the repulsive movement.
In particular, in an embodiment of the invention, the repulsion direction respondedi(k) Is directed perpendicular to the particle to repel the center rejecti(k) The direction of (a), i.e. the repulsion direction of the ith particle in the population, is the perpendicular direction of the current position of the ith particle pointing to the repulsion center direction.
And if the algorithm detects that an overcrowded area exists in the external set, the repulsion mechanism is started, and the particles in the overcrowded area are iterated according to the repulsion movement speed updating formula, so that the particles in the overcrowded area can rapidly jump out of the overcrowded area, and the diversity of the algorithm is ensured.
Specifically, in the embodiment of the present invention, the location update formula is specifically:
xi(k+1)=xi(k)+vi(k+1)
specifically, in the embodiment of the present invention, in step S4, the method for verifying accuracy of a response surface proxy model specifically includes:
s401, calculating a decision coefficient, and judging the precision of the response surface proxy model by comparing and analyzing the decision coefficient value.
Specifically, in the embodiment of the present invention, the calculation formula of the decision coefficient is specifically:
Figure BDA0003212465980000111
in the formula, n is the number of test points,
Figure BDA0003212465980000112
in order to respond to the observations of the face proxy model,
Figure BDA0003212465980000113
is the mean value of the true response, yiIs the true response value.
Wherein when R is2The closer to 1, the higher the accuracy of the response surface proxy model.
Specifically, in the embodiment of the present invention, in step S4, the method for verifying accuracy of a response surface proxy model specifically further includes:
s402, calculating a root mean square error value, and detecting the model precision by showing the error between the real response value and the model response value through the root mean square error value.
Specifically, in the embodiment of the present invention, the calculation formula of the root mean square error value is specifically:
Figure BDA0003212465980000114
in the formula, k is the number of detection points.
Specifically, in the embodiment of the present invention, in step S4, the method for verifying accuracy of a response surface proxy model specifically further includes:
s403, variance detection: and checking the precision of the response surface agent model by calculating variance values.
Specifically, in the embodiment of the present invention, the variance test requires two models for comparison, and the definition formula specifically is:
Figure BDA0003212465980000121
in the formula: y isiIs an actual value, yi' is the model response value and m is the detection point data.
More specifically, the multi-target gradient-based enhanced particle swarm algorithm can obtain a better convergence effect with fewer iteration times for a single target function or a multi-target function, and all the functions converge to the vicinity of a Pareto optimal solution; secondly, for the diversity of the optimized solution, based on a multi-target gradient enhanced particle swarm algorithm, the rejection mechanism and the rejection motion of the particles are utilized, the convergence to the local optimal solution can be successfully avoided, and the obtained Pareto solution is wide in coverage and has good diversity.
In view of the above, in the sheet forming process optimization design method related to the embodiment of the invention, since the convergence rate of the intelligent algorithm applied in the sheet forming process is low, and the efficiency of the intelligent algorithm is affected in the forming optimization process, the forming optimization is performed by designing a multi-objective optimization algorithm-a multi-objective gradient enhanced particle swarm algorithm with high convergence rate; in the optimization design method, firstly, the sheet forming process is analyzed, and the target and the influence factor which need to be optimized are determined; then sampling in a design space by an experimental method to obtain a sample; then, simulating the sample to obtain a value of the target function; and then, a mapping relation between the design parameters and the objective function is constructed by using the response surface proxy model, and finally, the response surface proxy model is optimized and solved by using the multi-objective gradient enhanced particle swarm algorithm, so that the efficiency of the sheet forming optimization design can be improved, and the production cost is reduced. Compared with the prior art, the technical scheme provided by the invention is based on the multi-target gradient enhanced particle swarm algorithm, so that the plate forming optimization efficiency can be effectively improved, the optimal forming process parameter combination can be rapidly determined, the part processing quality is ensured, and the production efficiency is improved.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims (10)

1. The optimized design method for the sheet forming process is characterized by comprising the following steps of:
s1, determining a decision space and an objective function: performing analog simulation on the part forming process, analyzing the forming characteristics of the part, determining an optimization target and an objective function, and determining design parameters and an approximate value range of the part forming through single-factor experimental analysis;
s2, experimental design: sampling in the decision space by a Latin hypercube test design method to obtain uniformly distributed sample points, and performing simulation analysis on each sample point to obtain a simulation value of the target function;
s3, constructing a response surface proxy model: constructing a response surface agent model based on the data of the sample points, and constructing a mapping relation between the design parameters and the target function through the response surface agent module;
s4, response surface proxy model precision test: performing precision analysis on the response surface agent model, and if the precision meets the engineering requirement, jumping to the next step; if the precision does not meet the engineering requirement, jumping back to the step S2, increasing the number of the sample points by a Latin hypercube test design method, and continuing to key the response surface proxy model until the precision of the response surface proxy model meets the engineering requirement;
s5, optimizing and solving: performing optimization solution on the response surface agent model based on a multi-objective gradient enhanced particle swarm algorithm to obtain a balance curve of the objective function, and selecting a satisfactory solution on the balance curve according to specific engineering requirements;
s6, simulation verification: and carrying out simulation verification on the satisfactory solution, and verifying the reliability of the optimization result.
2. The sheet forming process optimization design method according to claim 1, wherein the step of performing optimization solution on the response surface proxy model based on the multi-objective gradient enhanced particle swarm optimization comprises the following steps:
s501, initializing population and initializing parameters: setting the steepest descent term coefficient c1Global learning factor c2And a repulsive direction coefficient c3Maximum number of iterations kmaxThe ratio S of the judgment of the congested area, the upper limit u of the search spacenAnd ulConvergence precision xi of the algorithm, and upper and lower limits v of the particle velocitymaxAnd vmin(ii) a Initializing a population and an external set, and randomly generating an initial position and an initial speed of each particle in the population;
s502, calculating a fitness value, evaluating particles, calculating the fitness value of each particle, comparing all solutions according to a Pareto domination relation, and storing non-inferior solutions in the external set;
s503, calculating a first ratio of the number of the non-inferior solutions in each hypercube in the external set to the total number of the solutions in the external set;
s504, calculating a second ratio of the number of the non-inferior solutions in the hypercube to the total number of the non-inferior solutions in the external set, comparing and analyzing the second ratio with the crowded area judgment ratio S, and updating the speed and the position of each particle according to a comparison result.
3. The sheet forming process optimization design method according to claim 2, wherein the step of performing optimization solution on the response surface proxy model based on the multi-objective gradient enhanced particle swarm optimization further comprises the steps of:
s505, calculating the fitness value of each particle after updating;
s506, comparing the fitness value of each particle in the population with the non-inferior solutions in the external set according to a Pareto dominant solution, storing the non-inferior solutions, and deleting the dominant solution;
s507, when the number of the non-inferior solutions of the external set reaches a set value, maintaining the external set according to a self-adaptive grid method of a multi-target particle swarm algorithm, storing superior non-inferior solutions, and removing poor solutions;
and S508, judging whether the algorithm meets the condition of iteration stop, if so, stopping iteration and outputting a result, otherwise, skipping to the step S503 to continue circulation.
4. The method as claimed in claim 2, wherein in the step S504, the comparing and analyzing the second ratio with the congested area judgment ratio S includes:
if the second ratio is smaller than or equal to the crowded area judgment ratio, the rejection condition is not met, all the particles are indicated to do guiding motion, the global optimal solution is selected according to a self-adaptive grid method of the multi-target particle swarm algorithm, and the speed and the position of the particles are updated through a guiding motion updating rule.
5. The sheet forming process optimization design method according to claim 4, wherein the guide motion update rule is specifically:
and calling a guiding movement speed updating formula to update the speed of each particle, and calling a position updating formula to update the position of each particle.
6. The sheet forming process optimization design method according to claim 5, wherein the guide movement speed update formula is specifically as follows:
vi(k+1)=wvi(k)+r1c1gspeed(k)+r2c2(gBESTi(k)-xi(k))
in the formula: w is the inertial weight; 1,2,3.. m; m is the total number of particles in the particle swarm; k, the iteration number, represents the kth iteration; r is1And r2∈[0,1]Two random numbers in between; v. ofi(k) -the guiding motion velocity of the ith particle, the kth iteration; greed (k) -kth iteration, the descent direction searched for by the ith particle at the current position; gBESTi(k) The best positions searched so far by all particles, called global optimal solution; x is the number ofi(k) -the position of the ith particle.
7. The sheet forming process optimization design method according to claim 6, wherein the position update formula is specifically:
xi(k+1)=xi(k)+vi(k+1)
8. the method as claimed in claim 5, wherein in the step S504, the comparing and analyzing the second ratio with the congested area judgment ratio S further comprises:
if the second ratio is larger than the crowded area judgment ratio, the rejection condition is met, a rejection center is determined, particles close to the rejection center are selected according to the crowded area judgment ratio to do rejection movement, the speed and the position of the particles are updated according to a rejection movement update rule, the rest particles do guide movement, and the speed and the position of the rest particles are updated according to a guide movement update rule.
9. The sheet forming process optimization design method according to claim 8, wherein the update rule of the repelling motion is specifically as follows:
and calling a repulsive movement updating formula to update the speed of the particles, and calling a position updating formula to update the positions of the particles.
10. The sheet forming process optimization design method according to claim 9, wherein the repulsive movement velocity update formula is specifically:
vi(k+1)=wvi(k)+r1c1gspeed(k)+r2c3respeedi(k)
in the formula: w is the inertial weight; r is1And r2∈[0,1]Two random numbers in between; v. ofi(k) -the ith particle, the repulsive motion velocity of the kth iteration; greed (k) -kth iteration, the descent direction searched for by the ith particle at the current position; respeed (k) -the direction of repulsion of the ith particle in the population.
CN202110934614.7A 2021-08-16 2021-08-16 Optimal design method for sheet forming process Active CN113627025B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110934614.7A CN113627025B (en) 2021-08-16 2021-08-16 Optimal design method for sheet forming process

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110934614.7A CN113627025B (en) 2021-08-16 2021-08-16 Optimal design method for sheet forming process

Publications (2)

Publication Number Publication Date
CN113627025A true CN113627025A (en) 2021-11-09
CN113627025B CN113627025B (en) 2023-05-26

Family

ID=78385503

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110934614.7A Active CN113627025B (en) 2021-08-16 2021-08-16 Optimal design method for sheet forming process

Country Status (1)

Country Link
CN (1) CN113627025B (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104036332A (en) * 2014-06-17 2014-09-10 中国地质大学(武汉) Average gradient value and improved multi-objective particle swarm optimization based robust optimization system
CN104809304A (en) * 2015-05-12 2015-07-29 上海拖拉机内燃机有限公司 Aluminum plate stamping forming process optimization method based on variable-gap blank pressing
CN106909743A (en) * 2017-03-02 2017-06-30 合肥工业大学 McPherson suspension hard spot coordinate optimizing method based on ectonexine nesting multi-objective particle swarm algorithm
US20190359510A1 (en) * 2018-05-23 2019-11-28 Beijing University Of Technology Cooperative optimal control method and system for wastewater treatment process
CN110610225A (en) * 2019-08-28 2019-12-24 吉林大学 Multi-objective particle swarm optimization algorithm based on kriging proxy model plus-point strategy

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104036332A (en) * 2014-06-17 2014-09-10 中国地质大学(武汉) Average gradient value and improved multi-objective particle swarm optimization based robust optimization system
CN104809304A (en) * 2015-05-12 2015-07-29 上海拖拉机内燃机有限公司 Aluminum plate stamping forming process optimization method based on variable-gap blank pressing
CN106909743A (en) * 2017-03-02 2017-06-30 合肥工业大学 McPherson suspension hard spot coordinate optimizing method based on ectonexine nesting multi-objective particle swarm algorithm
US20190359510A1 (en) * 2018-05-23 2019-11-28 Beijing University Of Technology Cooperative optimal control method and system for wastewater treatment process
CN110610225A (en) * 2019-08-28 2019-12-24 吉林大学 Multi-objective particle swarm optimization algorithm based on kriging proxy model plus-point strategy

Non-Patent Citations (9)

* Cited by examiner, † Cited by third party
Title
HONGGUI HAN等: ""Adaptive Gradient Multiobjective Particle Swarm Optimization"", 《IEEE TRANSACTIONS ON CYBERNETICS》 *
孙光永 等: ""拉延成形多目标序列响应面法优化设计方法"", 《力学学报》 *
安治国等: ""基于径向基函数响应面法的板料成形仿真研究"", 《***仿真学报》 *
张克: ""基于梯度信息的粒子群优化算法的改进及应用"", 《中国优秀博硕士学位论文全文数据库(硕士)信息科技辑(月刊) 》 *
王俊伟 等: ""一种带有梯度加速的粒子群算法"", 《控制与决策》 *
祁荣宾 等: ""一种基于梯度信息的多目标优化算法"", 《化工学报》 *
袁恋: ""多目标粒子群优化算法的改进研究"", 《中国优秀博硕士学位论文全文数据库(硕士) 信息科技辑(月刊) 》 *
赵鹏军等: ""基于吸引排斥机制的粒子群优化算法"", 《计算机应用》 *
郑友莲等: ""基于密集距离的多目标粒子群优化算法"", 《湖北大学学报(自然科学版)》 *

Also Published As

Publication number Publication date
CN113627025B (en) 2023-05-26

Similar Documents

Publication Publication Date Title
KR102236046B1 (en) Face detection training method, device and electronic device
CN105160444B (en) Electrical equipment failure rate determining method and system
CN105700549B (en) A kind of unmanned plane Multiple routes planning method based on sequence small survival environment particle sub-group algorithm
CN106709216B (en) Microphone array optimization design method considering acoustic propagation correlation loss
CN105005820B (en) Target assignment optimizing method based on particle swarm algorithm of population explosion
CN113032902B (en) High-speed train pneumatic head shape design method based on machine learning optimization
CN110909773B (en) Client classification method and system based on adaptive particle swarm
CN109104737B (en) Cluster countervailing capacity evaluation method based on time-varying network
CN115310554A (en) Item allocation strategy, system, storage medium and device based on deep clustering
CN108062585A (en) A kind of method that Function Extreme value is calculated based on a flying moth darts into the fire algorithm
CN113627025A (en) Optimized design method for sheet forming process
CN113627075B (en) Projectile pneumatic coefficient identification method based on adaptive particle swarm optimization extreme learning
CN106488482B (en) Wireless sensor network optimizing method based on multi-Agent evolutionary Algorithm
KR20190136770A (en) Method and apparatus for predicting bead geometry of gas metal arc welding
CN116522565B (en) BIM-based power engineering design power distribution network planning method and computer equipment
CN117216692A (en) Training result acceptance method and system
CN111225367A (en) RFID network planning method based on hybrid particle swarm optimization
CN115936773A (en) Internet financial black product identification method and system
CN109657577A (en) A kind of animal detection method based on entropy and motion excursion amount
CN110286383B (en) Radar and infrared sensor deployment method applied to target tracking
CN114330135A (en) Classification model construction method and device, storage medium and electronic equipment
CN114527435A (en) Interference resource allocation method based on interference vector and NSGA-II algorithm
CN111832646A (en) CMCSA (China-computer aided design) based classifier integration weight distribution and self-adaptive adjustment method
Thompson Utilizing the generalized likelihood ratio as a termination criterion
Li et al. Improved Adaptive Wolf Pack Algorithm based on Clustering

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