CN110704959A - MOEAD (Metal oxide optical insulator deposition) optimization fixture layout method and device based on migration behavior - Google Patents
MOEAD (Metal oxide optical insulator deposition) optimization fixture layout method and device based on migration behavior Download PDFInfo
- Publication number
- CN110704959A CN110704959A CN201910765080.2A CN201910765080A CN110704959A CN 110704959 A CN110704959 A CN 110704959A CN 201910765080 A CN201910765080 A CN 201910765080A CN 110704959 A CN110704959 A CN 110704959A
- Authority
- CN
- China
- Prior art keywords
- population
- environment
- workpiece
- preset
- migration
- 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
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial 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]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a method and a device for optimizing clamp layout by MOEAD based on migration behaviors, and relates to the field of clamp layout. The method comprises the steps of constructing a workpiece position offset model based on a multi-target method, and respectively determining offsets in x, y and z directions as target functions. The method comprises the steps of generating an initial population in a feasible domain space of a workpiece position according to a preset mode, carrying out iterative computation on the initial population according to preset times in an original environment to obtain a non-dominated solution set of the original environment, taking the non-dominated solution set obtained in the original environment as an initial population of a next environment, and continuing to carry out iterative computation corresponding to the next environment according to the preset times to obtain a final non-dominated solution set.
Description
Technical Field
The invention relates to the field of clamp layout, in particular to a method and a device for optimizing clamp layout by MOEAD based on migration behavior.
Background
The machine tool clamp belongs to important equipment in the production and manufacturing process and is used for positioning and clamping a workpiece on a machine tool, so that the workpiece can resist the action of external force in the whole processing process, and the workpiece is always in a correct position relative to the machine tool and a cutter, so that the processing of all characteristics of the workpiece is smoothly completed. The reasonable clamp layout can position the workpiece at the correct position, so that the workpiece is kept in a stable state, and the subsequent procedures of cutting, welding, assembling and the like are guaranteed. Due to the unreasonable layout of the clamp, when a workpiece is influenced by external forces such as cutting force and the like in the subsequent machining process, the position of the workpiece relative to a machine tool and a cutter can deviate, so that the workpiece is machined with a large error, even can be separated from the clamp, and the danger is caused.
The existing workpiece position offset model is to combine the movement and rotation in the x, y and z directions into a target, and then use an algorithm for optimization, which is equivalent to assign the same weight to the offset in the three directions, but the processing requirements of different workpieces are different, so the existing method cannot adapt to workpieces with different processing requirements.
MOEAD (decomposition-based multi-objective evolutionary algorithm) decomposes MOP into N scalar sub-problems. It solves all sub-problems simultaneously by evolving a population of solutions. For each generation population, the population is a set of optimal solutions for each sub-problem selected from all generations. The degree of correlation between two adjacent subproblems is determined by the distance between their aggregate coefficient vectors. For two adjacent sub-problems to optimize it only with the information of the sub-problem adjacent to it, the optimal solution should be very similar.
MOEAD has the following characteristics: (1) MOEAD provides a simple and effective method, and introduces a decomposition method into multi-objective evolutionary computation. For the decomposition method which is often developed in the field of mathematical programming, the method can be really incorporated into EA (evolutionary algorithm), and MOP (multi-objective optimization) problem is solved by using MOEAD framework. (2) Because the MOEAD algorithm optimizes N scalar subproblems simultaneously rather than solving the MOP problem directly as a whole, the difficulty of fitness assignment and diversity control is reduced in the MOEAD framework for the traditional MOEA (multi-objective evolution) algorithm which is not based on decomposition.
The layout problem of the fixture is that a large number of points exist on each contact surface, so that the fixture is easy to fall into local optimum, and therefore, the search capability of the MOEAD needs to be further improved, and the diversity of the population of the MOEAD needs to be enhanced.
It is the original intention and essence of GA (genetic algorithm) to use and simulate the species evolution law and pattern in ecology. In the natural evolution of species, there are some such populations: the populations migrate to and from different types of natural environments and are influenced by different environmental factors, so that the adaptability of the populations to the environments is improved. We abstract this natural ecological phenomenon simply to a model of environmental migration of species, in which we assume that there are species P and K different natural environments Ei, where i is equal to or less than K, and K is a positive integer greater than 1, first let species P enter environment E1, complete the survival tasks of competition, multiplication, elimination, etc., then let species P migrate from environment E1 to the next environment E2, and complete the same tasks, and migrate in different environments in this way in sequence until species P completes the survival task in the last environment EK, at which time we call species P complete a round of environmental migration. After the species P has completed several rounds of environmental migration, the adaptability of the individuals in the species P' after migration to the environment Ei is improved compared with the individuals in the population P before migration in terms of the ecological rule for survival of the candidates.
Disclosure of Invention
In view of the above, the present invention has been developed to provide a method and apparatus for MOEAD optimized gripper layout based on migration behavior that overcomes or at least partially solves the above-mentioned problems.
In order to solve the technical problems, the invention provides the following technical scheme:
in one aspect, the invention provides a method for MOEAD optimization fixture layout based on migration behavior, comprising:
constructing a workpiece position offset model based on a multi-target method, and respectively determining offsets in x, y and z directions as target functions;
generating an initial population in a feasible domain space of a workpiece position according to a preset mode;
performing iterative computation on the initial population according to preset times in the original environment to obtain a non-dominated solution set of the original environment, wherein in the process of one iterative computation, gene recombination is performed on individuals in the population to obtain new individuals, and the individuals obtained after the recombination and the population after the last iteration are screened according to preset rules and the population is updated;
and taking the non-dominated solution set obtained in the original environment as the initial population of the next environment, continuing to perform iterative computation corresponding to the next environment according to the preset times, and obtaining a final non-dominated solution set after a preset termination condition is met.
Optionally, the constructing a workpiece position offset model based on a multi-target method, and determining offsets in the x, y, and z directions as target functions respectively includes:
respectively determining the displacement and the rotation angle of the workpiece positioning point in the x, y and z directions of the workpiece position offset model;
based on the displacement and rotation angle of the workpiece positioning point in the x direction of the workpiece position offset model according to a first objective function
Determining the offset of the workpiece location point in the x-direction, where δ xw 2Indicating the displacement of the anchor point in the x-direction, delta alphaw 2Representing the rotation angle of the positioning point in the x direction;
workpiece positioning point based workpiece position deviationThe displacement and rotation angle of the moving model in the y direction are according to a second objective function
Determining the offset of the workpiece location point in the y-direction, where δ yw 2Indicating the displacement of the anchor point in the y-direction, δ βw 2Representing the rotation angle of the positioning point in the y direction;
based on the displacement and the rotation angle of the workpiece positioning point in the z direction of the workpiece position offset model according to a third objective function
Determining the offset of the workpiece location point in the z-direction, where δ zw 2Indicating the displacement of the anchor point in the z direction, δ γw 2Representing the angle of rotation of the anchor point in the z direction.
Optionally, the generating an initial population in a feasible region space of the workpiece position according to a preset mode includes:
calculating Euclidean distance between any two weight vectors, and searching T weight vectors nearest to each weight vector, wherein T represents the number of weight vectors in a neighborhood, and for each i 11,...,iT}，
Is λiThe most recent T weight vectors, where B (i) represents the set of ordinal numbers of the weight vectors in the i-th weight vector neighborhood, λiRepresents the ith weight vector;
establishing an external population for storing the non-dominated solution found in the searching process, and setting the external population to be empty during initialization;
uniform random acquisition of decision variables x in feasible space1,...,xNGenerating an initial population based on the acquired decision variables and the objective function;
using the chebyshev method, the objective function f (x) is decomposed into N layers of sub-problems:
adjacent relation of jth sub-problem by all sub-problems with respect to λjWeight vector of points, z*Is the minimum vector value of the target function which can be searched at present; where m represents the number of objective functions and h represents the number of objective functions.
Optionally, in the one-time iterative computation process, performing gene recombination on individuals in the population to obtain new individuals, screening the individuals obtained after the recombination and the population after the last iteration according to a preset rule, and updating the population, including:
randomly selecting two serial numbers u, v from the set B (i), and using genetic operator to select xuAnd xvGenerating a new value, screening the value with the minimum offset from the generated value and the values of the neighborhood, and updating the population by using the screened value with the minimum offset.
Optionally, before the taking the non-dominated solution set obtained in the original environment as the initial population of the next environment and continuing performing iterative computation corresponding to the next environment according to the preset number of times, the method further includes:
setting a numerical value p, randomly generating a random number rand, and analyzing whether the random number rand is smaller than p, wherein the rand is located in the range of 0-1, and p is larger than 0 and smaller than 1;
if so, moving the individual in the non-dominant solution set obtained in the original environment to a first moving environment by approaching the individual to other individuals in the non-dominant solution set obtained in the original environment, and taking the first moving environment as a next environment;
if not, randomly selecting a new position around the individuals in the non-dominant solution set obtained in the original environment based on normal distribution for migration to a second migration environment, and taking the second migration environment as the next environment.
Optionally, if the first migration environment is the next environment, the performing, by using the non-dominated solution set obtained in the original environment as the initial population of the next environment, iterative computation corresponding to the next environment according to the preset number of times includes:
each iteration in the first migration environment obtains a new individual in a mode of Xi _ new ═ (1-rand) × Xi + rand Xj; wherein Xi represents the ith individual in the new population, Xj represents the jth individual randomly selected in the neighborhood of the individual Xi, Xi _ new represents the final position of the ith individual Xi after moving to the direction of the neighbor Xj, and i is not equal to j;
after one iteration calculation, screening the individuals obtained after the recombination and the population after the last iteration according to a preset rule and updating the population.
Optionally, if the second migration environment is the next environment, the performing, by using the non-dominated solution set obtained in the original environment as the initial population of the next environment, iterative computation corresponding to the next environment according to the preset number of times includes:
each iteration in the second migration environment obtains a new individual in a mode of Xi _ new ═ Xi + w. (VarMax-VarMin) × (randn)/norm (randn, 2); wherein w represents a nonlinear adaptive coefficient based on the iteration times, w is exp (-10 × rand it/MaxIt), Xi represents the ith individual in the new population, it represents the current iteration times, MaxIt represents the preset times, Xi _ new represents the final position of the ith individual Xi after randomly selecting a position around the position based on normal distribution for migration, VarMax and VarMin represent the upper and lower bounds of the search space of the feasible region space of the current workpiece position respectively, randn represents the normal distribution with the average value of 0 and the variance of 1;
after one iteration calculation, screening the individuals obtained after the recombination and the population after the last iteration according to a preset rule and updating the population.
According to another aspect of the present invention, there is also provided an apparatus for optimizing a layout of a fixture based on a move behavior, including:
the construction module is suitable for constructing a workpiece position offset model based on a multi-target mode and respectively determining offsets in the x direction, the y direction and the z direction as target functions;
the generating module is suitable for generating an initial population in a feasible region space of the workpiece position according to a preset mode;
the calculation module is used for carrying out iterative calculation on the initial population according to preset times in the original environment to obtain a non-dominated solution set of the original environment, wherein in the process of one iterative calculation, gene recombination is carried out on individuals in the population to obtain new individuals, and the individuals obtained after the recombination and the population after the last iterative calculation are screened according to preset rules and the population is updated;
and the migration module is used for taking the non-dominated solution set obtained in the original environment as the initial population of the next environment, continuously performing iterative computation corresponding to the next environment according to the preset times, and obtaining a final non-dominated solution set after a preset termination condition is met.
In the embodiment of the invention, the offsets in the x, y and z directions are taken as the target function, corresponding weights can be respectively distributed to the offsets in the x, y and z directions according to different processing precisions in all directions of the workpiece, and the higher the processing precision of the workpiece is, the larger the corresponding weight is distributed to the offset in the corresponding direction, so that the processing of various parts with different requirements can be met. And migration behaviors are introduced into the MOEAD algorithm, different computing methods are provided in different environments, the diversity of the population is enhanced, the searching capability of the algorithm is improved, a value with small offset is screened out through the preferential elimination of various environments, and the overall fitness of the population is improved. When the clamp layout is optimized, the clamp layout does not fall into local optimization due to the existence of a large number of points on the machined workpiece surface.
The foregoing description is only an overview of the technical solutions of the present invention, and the embodiments of the present invention are described below in order to make the technical means of the present invention more clearly understood and to make the above and other objects, features, and advantages of the present invention more clearly understandable.
The above and other objects, advantages and features of the present invention will become more apparent to those skilled in the art from the following detailed description of specific embodiments thereof, taken in conjunction with the accompanying drawings.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:
FIG. 1 shows a flowchart of a method for MOEAD optimization fixture layout based on migration behavior according to one embodiment of the present invention;
FIG. 2 shows a schematic diagram of a static equilibrium relationship of a workpiece model according to another embodiment of the invention;
FIG. 3 is a flow chart illustrating a method for generating an initial population based on a MOEAD optimized fixture layout for migration behavior according to another embodiment of the present invention;
FIG. 4 shows a schematic diagram of a population migration method according to another embodiment of the present invention;
FIG. 5 shows a schematic view of a milled through slot according to another embodiment of the invention;
FIG. 6 shows a comparison of results before and after MOEAD improvement according to another embodiment of the present invention;
fig. 7 shows a schematic diagram of an apparatus for MOEAD optimized fixture layout based on migration behavior according to yet another embodiment of the present invention.
Detailed Description
Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
In order to solve the technical problem, the embodiment of the invention provides a method for optimizing the layout of a fixture based on MOEAD of a migration behavior. Fig. 1 shows a flow diagram of a method for optimizing a fixture layout based on a move behavior MOEAD according to an embodiment of the invention. Referring to fig. 1, the method for optimizing a layout of a jig based on MOEAD of migration behavior includes steps S101 to S104.
And S101, constructing a workpiece position offset model based on a multi-target method, and respectively determining offsets in x, y and z directions as target functions.
Step S102, generating an initial population in a feasible region space of the workpiece position according to a preset mode.
And step S103, carrying out iterative computation on the initial population according to preset times in the original environment to obtain a non-dominated solution set of the original environment.
In the process of one iteration calculation, gene recombination is carried out on individuals in the population to obtain new individuals, the individuals obtained after the recombination and the population after the last iteration are screened according to a preset rule, and the population is updated.
And step S104, taking the non-dominated solution set obtained in the original environment as the initial population of the next environment, continuing to perform iterative computation corresponding to the next environment according to preset times, and obtaining a final non-dominated solution set after a preset termination condition is met.
In this embodiment, the preset number of times may be set manually, and the preset termination condition is that the preset number of times is reached, that is, the last iteration calculation is terminated after the last iteration calculation is finished. The offsets in the x, y and z directions are taken as three different target functions, and corresponding weights can be respectively distributed to the offsets in the x, y and z directions according to different processing precision of workpieces in all directions, so that the processing of parts with different requirements can be met, and the distributed corresponding weights are larger in the direction with higher processing precision. And a migration behavior is introduced into the MOEAD algorithm and is migrated to different environments, different environments have different iterative calculation methods, the diversity of the population is enhanced, the search capability of the algorithm is improved, and the overall fitness of the population is improved through the preferential elimination of various environments. When the clamp layout is optimized, the clamp layout does not fall into local optimization due to the existence of a large number of points on the machined workpiece surface.
Referring to fig. 2, in an embodiment of the present invention, the step S101 of constructing a workpiece position offset model based on a multi-objective method, and determining offsets in the three directions x, y, and z as objective functions respectively includes determining displacements and rotation angles of a workpiece positioning point in the x, y, and z directions of the workpiece position offset model respectively.
Based on the displacement and rotation angle of the workpiece positioning point in the x direction of the workpiece position offset model according to a first objective functionDetermining the offset of the workpiece location point in the x-direction, where δ xw 2Indicating the displacement of the anchor point in the x-direction, delta alphaw 2Representing the rotation angle of the anchor point in the x-direction.
Based on the displacement and the rotation angle of the workpiece positioning point in the y direction of the workpiece position offset model according to a second objective function
Determining the offset of the workpiece location point in the y-direction, where δ yw 2Indicating the displacement of the anchor point in the y-direction, δ βw 2Indicating the angle of rotation of the anchor point in the y-direction.
Based on the displacement and the rotation angle of the workpiece positioning point in the z direction of the workpiece position offset model according to a third objective function
Determining the offset of the workpiece location point in the z-direction, where δ zw 2Indicating the displacement of the anchor point in the z direction, δ γw 2Representing the angle of rotation of the anchor point in the z direction.
In this embodiment, the objective function f will now be described in conjunction with the description of FIG. 21(x)，f2(x)，f3(x) The steps are as follows:
step 1.1, u positioning elements and v clamping elements are selected for clamping layout, wherein u and v are positive integers larger than 1. The work being subjected to external forces such as working force, gravity, etc. and torqueMachining force rotation, gravity rotation and clamping force. According to the stress condition of the workpiece, the static equilibrium equation of the workpiece can be known
Wherein G isdRepresenting a layout matrix of positioning elements, GjA layout matrix representing the clamping elements is shown,
representing the contact force vector of the positioning element,representing the contact force vector, W, of the clamping elementbIndicating the amount of external force rotation. Gd、Gj、Fj cThe layout matrix of (a) is specifically as follows:
by
The contact force vector of the positioning element can be obtained
Wherein r iscuAs coordinates of the u-th anchor point, nuIs the coordinates of the normal vector of the u positioning point, tuAnd buAnd identifying tangent vectors of the u-th positioning points which are orthogonal to each other, wherein the elements represented by the ellipses in the matrix are the elements with the sequentially increasing element sequence numbers in front of the ellipses.
Step 1.2, the relationship between contact force and local deformation at all positioning elements is
Local deformation at the ith clamping member includes clamping member deformation
Deformation by contact
Is the contact force at the ith contact,is the local stiffness at the ith contact. In fig. 2, ni denotes a normal vector, bi and ti denote two mutually orthogonal tangent vectors,
representing the component of the contact force in the normal vector,
andrepresenting the force component of the contact force on two tangent vectors, the triangle representing the location point.
Step 1.3, the position of the ith contact point between the workpiece and the ith positioning element can be expressed by the following two formulas
Wherein r isw∈R3 ×1、Θw∈R3×1And rfi∈R3×1、Θfi∈R3×1The position and orientation of the workpiece and the positioning element i, respectively, relative to the { GCS }.
Is the position vector of the contact point i in WCS,
is the position vector of the contact point i in { FCS }. T (theta)w)∈R3×3Is the orientation matrix of { WCS } relative to { GCS }, T (Θ)fi)∈R3×3Is the orientation matrix of { FCS } relative to { GCS }. FCS is the positioning (clamping) element coordinate system, WCS is the workpiece coordinate system, and GCS is the global coordinate system.
In the actual clamping process, the workpiece and the ith positioning element are always in contact at the ith contact point, so that
When the directions of the coordinate axes of { GCS } and { WCS } are consistent, the equation is matched
And
is differentiated and simplified to obtain Eci·δqw＝ΔpiWherein
EciIn order to locate the matrix of positions,δqwis the positional deviation of the workpiece due to clamping.
According to the above process, a first objective function, a second objective function, and a third objective function can be obtained.
Furthermore, constraints are set for the objective function:
constraint of static equilibrium
Friction cone constraint NTT(Θc)FcGreater than or equal to 0, friction cone restraining HFcNot less than 0, whichIn, NTRepresents the normal vector matrix, T (Θ)c) An orientation matrix representing the orientation of the positioning elements, FcThe contact force vector of the positioning element.
In the embodiment, the offsets in the x, y and z directions are taken as three different objective functions, corresponding weights can be respectively distributed to the offsets in the x, y and z directions according to different processing precisions in all directions of a workpiece, and the higher the processing precision is, the larger the corresponding weight is distributed to the offset in the corresponding direction, so that the processing of parts with different requirements can be met.
In an embodiment of the present invention, referring to fig. 1 and 3, the step S102 of generating the initial population in a predetermined manner in the feasible region space of the workpiece position includes steps S1021 to S1024.
Step S1021, calculating a euclidean distance between any two weight vectors. Find the nearest T weight vectors of each weight vector, where T represents the number of weight vectors in the neighborhood, for each i 1, 21,...,iT}，
Is λiThe most recent T weight vectors, where B (i) represents the set of ordinal numbers of the weight vectors in the i-th weight vector neighborhood, λiRepresenting the ith weight vector. i represents the number of weight vectors, and N weight vectors correspond to N weight vector numbers.
Step S1022, an external population is established for storing the non-dominated solution found in the search process, and the external population is set to be empty during initialization.
Step S1023, decision variable x is uniformly and randomly acquired in feasible space1,...,xNAn initial population is generated based on the collected decision variables and the objective function. Wherein x is1,...,xNI.e. x1,x2...,xNHere, N denotes the number of the N decision variables x.
Step S1024, decomposing the objective function F (x) into N layers of subproblems by utilizing a Chebyshev method. The chebyshev method is as follows:
adjacent relation of jth sub-problem by all sub-problems with respect to λjWeight vector of points, z*Is the minimum vector value of the target function which can be searched at present; where m represents the number of objective functions, m has a value of 3, and h represents the number of objective functions.
In this embodiment of the present invention,
t in (2) denotes transposition, and is not the same as T in the above T weight vectors. One decision variable and the objective function value of the decision variable represent one individual, and f (x) represents a multi-objective function.
In one embodiment of the present invention, in the process of one iterative computation, gene recombination is performed on individuals in a population to obtain new individuals, and the individuals obtained after the recombination and the population after the last iterative computation are screened according to a preset rule and the population is updated, including,
randomly selecting two serial numbers u, v from the set B (i), and using genetic operator to select xuAnd xvGenerating a new value, screening the value with the minimum offset from the generated value and the values of the neighborhood, and updating the population by using the screened value with the minimum offset.
In this embodiment, x represents a decision variable, and a genetic operator is applied to the decision variable to determine the decision variable from xuAnd xvA new value is generated and compared with the values in the i-th decision variable neighborhood to select the value with the minimum offset. And the minimum value of each decision variable selected after one iteration calculation is used for updating the external population.
In an embodiment of the present invention, before step S104 is executed to use the non-dominated solution set obtained in the original environment as the initial population of the next environment and continue to perform iterative computation corresponding to the next environment according to a preset number of times, the next environment may be selected as follows.
Setting a value p, randomly generating a random number rand, and analyzing whether the random number rand is smaller than p, wherein the rand is within the range of 0-1, and p is larger than 0 and smaller than 1. If the random number rand is smaller than p, moving the individual in the non-dominated solution set obtained in the original environment to other individual in the non-dominated solution set obtained in the original environment to a first moving environment, and taking the first moving environment as the next environment. And if the random number rand is not less than p, randomly selecting a new position around the individuals in the non-dominant solution set obtained in the original environment based on normal distribution for migration to a second migration environment, and taking the second migration environment as the next environment.
Referring to fig. 4, fig. 4 is a schematic diagram of species migration, a species P before migration is migrated through a plurality of environments, namely, an environment E1, an environment E2, an environment Ei and an environment Ek, to obtain a species P' after migration, so that the diversity and adaptability of the species are improved. The number of transitions in fig. 4 is only schematically shown and does not conflict with the present solution.
In the embodiment, a migration behavior is introduced into the MOEAD, the diversity of the population is enhanced, the searching capability of the algorithm is improved, the value with small offset is screened out through the preferential elimination of various environments, and the overall fitness of the population is improved. When the clamp layout is optimized, the clamp layout does not fall into local optimization due to the existence of a large number of points on the machined workpiece surface.
In an embodiment of the present invention, if the first migration environment is the next environment, a specific process of taking the non-dominated solution set obtained in the original environment as the initial population of the next environment and continuing to perform iterative computation corresponding to the next environment according to a preset number of times may be that, in each iteration in the first migration environment, a new individual is obtained in a manner of Xi _ new ═ 1-rand × Xi + rand × Xj. Wherein Xi represents the ith individual in the new population, Xj represents the jth individual randomly selected in the individual Xi neighborhood, Xi _ new represents the final position of the ith individual Xi after moving to the direction of the neighbor Xj, and i is not equal to j. And after one-time iterative computation, screening the individuals obtained after the recombination and the population after the last iteration according to a preset rule and updating the population.
Certainly, when performing the next iterative computation, the updated population obtained after the previous screening may be continuously subjected to the iterative computation in a manner of Xi _ new ═(1-rand) × Xi + rand Xj to obtain a new individual, and corresponding screening is continuously performed according to the preset rule. The preset rule may be a filtering rule in the original environment.
In an embodiment of the present invention, if the second migration environment is the next environment, a specific process of taking the non-dominated solution set obtained in the original environment as the initial population of the next environment and continuing to perform iterative computation corresponding to the next environment according to a preset number of times may be that, in the second migration environment, each iteration obtains a new individual in a manner of Xi _ new ═ Xi + w. (VarMax-VarMin) ×/rand/norm (randn, 2). W represents a nonlinear adaptive coefficient based on the iteration number, w is exp (-10 × rand × it/MaxIt), Xi represents the ith individual in the new population, it represents the current iteration number, MaxIt represents the preset number, Xi _ new represents the final position of the ith individual Xi after randomly selecting a position around the position based on normal distribution for migration, VarMax and VarMin represent the upper and lower bounds of the search space of the feasible region space of the current workpiece position respectively, randn represents the normal distribution with the average value of 0 and the variance of 1. After one iteration calculation, screening the individuals obtained after the recombination and the population after the last iteration according to a preset rule and updating the population.
Certainly, during the next iterative computation, the updated population obtained after the previous screening may be continuously iteratively computed in a manner of Xi _ new ═ Xi + w. (VarMax-VarMin) × (randn/norm (randn, 2)) to obtain a new individual, and corresponding screening is continuously performed according to the preset rule. The preset rule may be a filtering rule in the original environment.
Referring to fig. 5, fig. 5 shows a schematic view of a milled through slot according to another embodiment of the present invention, wherein A, B, C, D, E, F in fig. 5 represents an anchor point, and P1, P2 and P3 represent a clamping point.
Referring to fig. 6, the minimum offsets in the x, y, and z directions are selected from the non-dominated solution sets before and after the improvement of MOEAD, and the results are compared, where the same initial population is common to the algorithms before and after each improvement, and the results after the independent operation of 4 groups are shown in table 1 below, where MOEAD is the original algorithm and imoad is the improved algorithm. The comparison before and after the improvement is shown in fig. 6. In this embodiment, fig. 5 is only a model diagram, and there is no correspondence between the model diagram of fig. 5 and the result shown in fig. 6.
In addition, referring to the comparison of the offset results shown in table 1, the minimum offset of the improved imoad in the three directions of x, y and z is smaller than that of the original MOEAD, the search capability of the improved imoad is improved relative to that of the original MOEAD, and the population adaptation capability obtained after various environments is also better than that of the original MOEAD.
TABLE 1
Referring to fig. 7, the present invention further provides an apparatus for optimizing a layout of a fixture based on a move behavior, and the apparatus 600 for optimizing a layout of a fixture based on a move behavior includes the following modules.
The building module 610 is adapted to build a workpiece position offset model based on a multi-objective mode, and respectively determine offsets in the x, y, and z directions as objective functions.
The generating module 620, coupled to the constructing module 610, generates the initial population in a predetermined manner in a feasible domain space of the workpiece location.
And the calculating module 630 is coupled to the generating module 620, and performs iterative calculation on the initial population according to preset times in the original environment to obtain a non-dominated solution set of the original environment. In the process of one iteration calculation, gene recombination is carried out on individuals in the population to obtain new individuals, and the individuals obtained after recombination and the population after the last iteration are screened according to a preset rule and the population is updated.
And the migration module 640 is coupled with the calculation module 630, and continues to perform iterative calculation corresponding to the next environment according to preset times by using the non-dominated solution set obtained in the original environment as an initial population of the next environment, and obtains a final non-dominated solution set after a preset termination condition is met.
In the embodiment, the offsets in the x, y and z directions are taken as three different objective functions, and corresponding weights can be respectively distributed to the offsets in the x, y and z directions according to different processing precisions in all directions of a workpiece, so that the processing of parts with different requirements can be met, and the distributed weight is larger in the direction with higher processing precision. And migration behaviors are introduced into MOEAD, different computing methods are provided in different environments, the diversity of the population is enhanced, the searching capability of the algorithm is improved, and the overall fitness of the population is improved through the preferential elimination of various environments. When the clamp layout is optimized, the clamp layout does not fall into local optimization due to the existence of a large number of points on the machined workpiece surface.
According to any one or a combination of the above preferred embodiments, the following advantages can be achieved by the embodiments of the present invention:
in the embodiment of the invention, the offsets in the x, y and z directions are taken as the target function, and corresponding weights can be respectively distributed to the offsets in the x, y and z directions according to different processing precisions in all directions of a workpiece, so that the processing of parts with different requirements can be met. And moreover, migration behavior is introduced into MOEAD, the diversity of the population is enhanced, the searching capability of the algorithm is improved, the value with small offset is screened out through the preferential elimination of various environments, and the overall fitness of the population is improved. When the clamp layout is optimized, the clamp layout does not fall into local optimization due to the existence of a large number of points on the machined workpiece surface.
It can be clearly understood by those skilled in the art that the specific working processes of the above-described apparatuses and modules may refer to the corresponding processes in the foregoing method embodiments, and for the sake of brevity, further description is omitted here.
In addition, the functional units in the embodiments of the present invention may be physically independent of each other, two or more functional units may be integrated together, or all the functional units may be integrated in one processing unit. The integrated functional units may be implemented in the form of hardware, or in the form of software or firmware.
Those of ordinary skill in the art will understand that: the integrated functional units, if implemented in software and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes several instructions, so that a computing device (for example, a personal computer, a server, or a network device) executes all or part of the steps of the method according to the embodiments of the present invention when the instructions are executed. And the aforementioned storage medium includes: u disk, removable hard disk, Read Only Memory (ROM), Random Access Memory (RAM), magnetic or optical disk, and other various media capable of storing program code.
Alternatively, all or part of the steps of implementing the foregoing method embodiments may be implemented by hardware (such as a personal computer, a server, or a network device) associated with program instructions, which may be stored in a computer-readable storage medium, and when the program instructions are executed by a processor of the computing device, the computing device executes all or part of the steps of the method according to the embodiments of the present invention.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments can be modified or some or all of the technical features can be equivalently replaced within the spirit and principle of the present invention; such modifications or substitutions do not depart from the scope of the present invention.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments can be modified or some or all of the technical features can be equivalently replaced within the spirit and principle of the present invention; such modifications or substitutions do not depart from the scope of the present invention.
Claims (8)
1. A method for optimizing fixture layout based on MOEAD of migration behavior is characterized by comprising the following steps:
constructing a workpiece position offset model based on a multi-target method, and respectively determining offsets in x, y and z directions as target functions;
generating an initial population in a feasible domain space of a workpiece position according to a preset mode;
performing iterative computation on the initial population according to preset times in the original environment to obtain a non-dominated solution set of the original environment, wherein in the process of one iterative computation, gene recombination is performed on individuals in the population to obtain new individuals, and the individuals obtained after the recombination and the population after the last iteration are screened according to preset rules and the population is updated;
and taking the non-dominated solution set obtained in the original environment as the initial population of the next environment, continuing to perform iterative computation corresponding to the next environment according to the preset times, and obtaining a final non-dominated solution set after a preset termination condition is met.
2. The method of claim 1, wherein the constructing a workpiece position offset model based on a multi-objective method and determining the offset in the x, y and z directions as an objective function respectively comprises:
respectively determining the displacement and the rotation angle of the workpiece positioning point in the x, y and z directions of the workpiece position offset model;
based on the displacement and rotation angle of the workpiece positioning point in the x direction of the workpiece position offset model according to a first objective function
Determining the location point of the workpieceOffset in the x-direction, where δ xw 2Indicating the displacement of the anchor point in the x-direction, delta alphaw 2Representing the rotation angle of the positioning point in the x direction;
based on the displacement and the rotation angle of the workpiece positioning point in the y direction of the workpiece position offset model according to a second objective function
Determining the offset of the workpiece location point in the y-direction, where δ yw 2Indicating the displacement of the anchor point in the y-direction, δ βw 2Representing the rotation angle of the positioning point in the y direction;
based on the displacement and the rotation angle of the workpiece positioning point in the z direction of the workpiece position offset model according to a third objective function
Determining the offset of the workpiece location point in the z-direction, where δ zw 2Indicating the displacement of the anchor point in the z direction, δ γw 2Representing the angle of rotation of the anchor point in the z direction.
3. The method of claim 1, wherein generating the initial population in a predetermined manner in a feasible region space of the workpiece location comprises:
calculating Euclidean distance between any two weight vectors, and searching T weight vectors nearest to each weight vector, wherein T represents the number of weight vectors in a neighborhood, and for each i 11,...,iT}，
Is λiThe most recent T weight vectors, where B (i) represents the set of ordinal numbers of the weight vectors in the i-th weight vector neighborhood, λiRepresents the ith weight vector;
establishing an external population for storing the non-dominated solution found in the searching process, and setting the external population to be empty during initialization;
uniform random acquisition of decision variables x in feasible space1,...,xNGenerating an initial population based on the acquired decision variables and the objective function;
using the chebyshev method, the objective function f (x) is decomposed into N layers of sub-problems:
adjacent relation of jth sub-problem by all sub-problems with respect to λjWeight vector of points, z*Is the minimum vector value of the target function which can be searched at present; where m represents the number of objective functions and h represents the number of objective functions.
4. The method of claim 3, wherein in the iterative calculation process, the gene recombination is performed on the individuals in the population to obtain new individuals, and the individuals obtained after the recombination and the population after the last iterative are screened according to a preset rule and the population is updated, including:
randomly selecting two serial numbers u, v from the set B (i), and using genetic operator to select xuAnd xvGenerating a new value, screening the value with the minimum offset from the generated value and the values of the neighborhood, and updating the population by using the screened value with the minimum offset.
5. The method according to any one of claims 1 to 4, wherein before the step of taking the non-dominated solution set obtained from the original environment as the initial population of the next environment and continuing to perform the iterative computation corresponding to the next environment according to the preset number of times, the method further comprises:
setting a numerical value p, randomly generating a random number rand, and analyzing whether the random number rand is smaller than p, wherein the rand is located in the range of 0-1, and p is larger than 0 and smaller than 1;
if so, moving the individual in the non-dominant solution set obtained in the original environment to a first moving environment by approaching the individual to other individuals in the non-dominant solution set obtained in the original environment, and taking the first moving environment as a next environment;
if not, randomly selecting a new position around the individuals in the non-dominant solution set obtained in the original environment based on normal distribution for migration to a second migration environment, and taking the second migration environment as the next environment.
6. The method according to claim 5, wherein if the first migration environment is a next environment, the performing, by the preset number of times, iterative computation corresponding to the next environment by using the non-dominated solution set obtained in the original environment as an initial population of the next environment further comprises:
each iteration in the first migration environment obtains a new individual in a mode of Xi _ new ═ (1-rand) × Xi + rand Xj; wherein Xi represents the ith individual in the new population, Xj represents the jth individual randomly selected in the neighborhood of the individual Xi, Xi _ new represents the final position of the ith individual Xi after moving to the direction of the neighbor Xj, and i is not equal to j;
after one iteration calculation, screening the individuals obtained after the recombination and the population after the last iteration according to a preset rule and updating the population.
7. The method according to claim 5, wherein if the second migration environment is the next environment, the step of taking the non-dominated solution set obtained from the original environment as the initial population of the next environment to continue the iterative computation corresponding to the next environment according to the preset number of times includes:
each iteration in the second migration environment obtains a new individual in a mode of Xi _ new ═ Xi + w. (VarMax-VarMin) × (randn)/norm (randn, 2); wherein w represents a nonlinear adaptive coefficient based on the iteration times, w is exp (-10 × rand it/MaxIt), Xi represents the ith individual in the new population, it represents the current iteration times, MaxIt represents the preset times, Xi _ new represents the final position of the ith individual Xi after randomly selecting a position around the position based on normal distribution for migration, VarMax and VarMin represent the upper and lower bounds of the search space of the feasible region space of the current workpiece position respectively, randn represents the normal distribution with the average value of 0 and the variance of 1;
after one iteration calculation, screening the individuals obtained after the recombination and the population after the last iteration according to a preset rule and updating the population.
8. An apparatus for MOEAD optimization fixture layout based on migration behavior, comprising:
the construction module is suitable for constructing a workpiece position offset model based on a multi-target mode and respectively determining offsets in the x direction, the y direction and the z direction as target functions;
the generating module is suitable for generating an initial population in a feasible region space of the workpiece position according to a preset mode;
the calculation module is used for carrying out iterative calculation on the initial population according to preset times in the original environment to obtain a non-dominated solution set of the original environment, wherein in the process of one iterative calculation, gene recombination is carried out on individuals in the population to obtain new individuals, and the individuals obtained after the recombination and the population after the last iterative calculation are screened according to preset rules and the population is updated;
and the migration module is used for taking the non-dominated solution set obtained in the original environment as the initial population of the next environment, continuously performing iterative computation corresponding to the next environment according to the preset times, and obtaining a final non-dominated solution set after a preset termination condition is met.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910765080.2A CN110704959B (en) | 2019-08-19 | 2019-08-19 | MOEAD (Metal oxide optical insulator deposition) optimization fixture layout method and device based on migration behavior |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910765080.2A CN110704959B (en) | 2019-08-19 | 2019-08-19 | MOEAD (Metal oxide optical insulator deposition) optimization fixture layout method and device based on migration behavior |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110704959A true CN110704959A (en) | 2020-01-17 |
| CN110704959B CN110704959B (en) | 2022-04-08 |
Family
ID=69193426
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910765080.2A Active CN110704959B (en) | 2019-08-19 | 2019-08-19 | MOEAD (Metal oxide optical insulator deposition) optimization fixture layout method and device based on migration behavior |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110704959B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111310385A (en) * | 2020-01-22 | 2020-06-19 | 南昌航空大学 | CMODE optimization fixture positioning method based on greedy variation search |
| CN113626954A (en) * | 2021-08-17 | 2021-11-09 | 中国地质大学（武汉） | Multi-target information processing method and system based on decomposition, computer equipment and terminal |
| CN114942917A (en) * | 2022-04-28 | 2022-08-26 | 阿里巴巴（中国）有限公司 | Migration method, device and equipment for internal space layout |
| CN115828641A (en) * | 2023-02-09 | 2023-03-21 | 西南交通大学 | MOEAD algorithm-based horizontal cylinder internal equipment layout method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107122844A (en) * | 2017-03-15 | 2017-09-01 | 深圳大学 | A kind of Multipurpose Optimal Method and system being combined based on index and direction vector |
| CN108038538A (en) * | 2017-12-06 | 2018-05-15 | 西安电子科技大学 | Multi-objective Evolutionary Algorithm based on intensified learning |
| CN109066710A (en) * | 2018-07-13 | 2018-12-21 | 国网安徽省电力有限公司滁州供电公司 | A kind of multi-objective reactive optimization method, apparatus, computer equipment and storage medium |
-
2019
- 2019-08-19 CN CN201910765080.2A patent/CN110704959B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107122844A (en) * | 2017-03-15 | 2017-09-01 | 深圳大学 | A kind of Multipurpose Optimal Method and system being combined based on index and direction vector |
| CN108038538A (en) * | 2017-12-06 | 2018-05-15 | 西安电子科技大学 | Multi-objective Evolutionary Algorithm based on intensified learning |
| CN109066710A (en) * | 2018-07-13 | 2018-12-21 | 国网安徽省电力有限公司滁州供电公司 | A kind of multi-objective reactive optimization method, apparatus, computer equipment and storage medium |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111310385A (en) * | 2020-01-22 | 2020-06-19 | 南昌航空大学 | CMODE optimization fixture positioning method based on greedy variation search |
| CN111310385B (en) * | 2020-01-22 | 2022-08-30 | 南昌航空大学 | CMODE optimization fixture positioning method based on greedy variation search |
| CN113626954A (en) * | 2021-08-17 | 2021-11-09 | 中国地质大学（武汉） | Multi-target information processing method and system based on decomposition, computer equipment and terminal |
| CN113626954B (en) * | 2021-08-17 | 2023-08-18 | 中国地质大学（武汉） | Multi-target information processing method, system, computer equipment and terminal based on decomposition |
| CN114942917A (en) * | 2022-04-28 | 2022-08-26 | 阿里巴巴（中国）有限公司 | Migration method, device and equipment for internal space layout |
| CN115828641A (en) * | 2023-02-09 | 2023-03-21 | 西南交通大学 | MOEAD algorithm-based horizontal cylinder internal equipment layout method |
| CN115828641B (en) * | 2023-02-09 | 2023-05-23 | 西南交通大学 | MOEAD algorithm-based horizontal cylinder internal equipment layout method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110704959B (en) | 2022-04-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110704959B (en) | MOEAD (Metal oxide optical insulator deposition) optimization fixture layout method and device based on migration behavior | |
| Zandieh et al. | An adaptive multi-population genetic algorithm to solve the multi-objective group scheduling problem in hybrid flexible flowshop with sequence-dependent setup times | |
| CN112526927B (en) | Quick optimization compensation method for space positioning error of rotating shaft of five-axis numerical control machine tool | |
| Díaz et al. | A GRASP heuristic for the manufacturing cell formation problem | |
| CN109917754B (en) | Multi-population distribution estimation algorithm-based robot assembly unit multi-target layout optimization method | |
| CN112766659A (en) | Distributed hybrid flow shop scheduling method, medium and system | |
| Chattopadhyay et al. | Principal component analysis and self-organizing map for visual clustering of machine-part cell formation in cellular manufacturing system | |
| CN111596622B (en) | Flexible job shop scheduling method of ECM (electronic model control) rule distribution estimation algorithm | |
| Zacharia et al. | Planning the construction process of a robotic arm using a genetic algorithm | |
| Alatartsev et al. | Constricting insertion heuristic for traveling salesman problem with neighborhoods | |
| Borojević et al. | An integrated intelligent CAD/CAPP platform: Part II-Operation sequencing based on genetic algorithm | |
| Ghosh | Allocating tools to index positions in tool magazines using tabu search | |
| Kontolatis et al. | Optimisation of press-brake bending operations in 3D space | |
| CN112666890B (en) | Curved surface workpiece machining track planning method | |
| CN116468137A (en) | Distributed process planning and workshop scheduling integrated optimization method | |
| CN113963758A (en) | Prediction recommendation method, device and terminal for thermodynamic stable structure of disordered material | |
| Rezaeipanah et al. | Meta-heuristic approach based on genetic and greedy algorithms to solve flexible job-shop scheduling problem | |
| Wang | PCB drill path optimization by improved genetic algorithm | |
| Allaoui et al. | Optimization of unsupervised classification by evolutionary strategies | |
| Viharos et al. | Support Vector Machine (SVM) based general model building algorithm for production control | |
| Dražić | Gaussian variable neighborhood search for the file transfer scheduling problem | |
| US12099346B1 (en) | Large-scale dynamic double-effect scheduling method for flexible job shop based on genetic programming | |
| Shang et al. | A kinematic chain isomorphism identification algorithm using optimized circuit simulation method | |
| Wang et al. | Dynamic Performance Prediction in Batch-Based Assembly System with Bernoulli Machines and Changeovers | |
| van Hassel et al. | Machine learning and genetic algorithms for conformal geometries in design support systems |
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 |