CN110222300A - A method of calculating cubic System Material parent phase and the alternate orientation relationship of son - Google Patents

A method of calculating cubic System Material parent phase and the alternate orientation relationship of son Download PDF

Info

Publication number
CN110222300A
CN110222300A CN201910424335.9A CN201910424335A CN110222300A CN 110222300 A CN110222300 A CN 110222300A CN 201910424335 A CN201910424335 A CN 201910424335A CN 110222300 A CN110222300 A CN 110222300A
Authority
CN
China
Prior art keywords
matrix
orientation relationship
son
orientation
population
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201910424335.9A
Other languages
Chinese (zh)
Inventor
谢永红
宋子豪
汪浩宇
张德政
阿孜古丽
栗辉
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.)
University of Science and Technology Beijing USTB
Original Assignee
University of Science and Technology Beijing USTB
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 University of Science and Technology Beijing USTB filed Critical University of Science and Technology Beijing USTB
Priority to CN201910424335.9A priority Critical patent/CN110222300A/en
Publication of CN110222300A publication Critical patent/CN110222300A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Theoretical Computer Science (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Pure & Applied Mathematics (AREA)
  • Algebra (AREA)
  • Databases & Information Systems (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Computing Systems (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The present invention provides the method for a kind of calculating cubic System Material parent phase and the alternate orientation relationship of son, comprising: obtains the matrix A 1 of the corresponding storage minimum θ value of parent population using default orientation relationship formula；According to current iteration number and default maximum number of iterations, nonlinear change is carried out to mutagenic factor F and crossover probability CR, and according to the adaptively selected Mutation Strategy of preset strategy adaptation scheme；Then variation is carried out to parent population and crossover operation obtains intermediate population；And the matrix A 2 that minimum θ value is stored corresponding to intermediate population is obtained using default orientation relationship formula；It is selected in matrix A 1 and matrix A 2, and updates parent；The corresponding minimum θ value of population obtained according to last time iteration obtains parent phase and the alternate orientation relationship of son.Method of the invention accurately calculates parent phase and the alternate orientation relationship of son by the differential evolution algorithm of new Mutation Strategy and parameter adaptive, and is effectively shortened by using multi-core parallel concurrent calculation and calculate the time.

Description

A method of calculating cubic System Material parent phase and the alternate orientation relationship of son
Technical field
The present invention relates to cubic System Material technical field, a kind of calculating cubic System Material parent phase and sub- phase meta position are particularly related to To the method for relationship.
Background technique
In phase transformation, it is particularly important to the property of its determining phase-change product to determine that the orientation of original parent phase tissue is returned, really Determining orientation relationship then is one of the basis of original parent phase recurrence.Application No. is 201710068225.4 patents to disclose a kind of stand The determination method of square based material parent phase and the alternate orientation relationship of son, determined based on differential evolution algorithm parent phase and sub- phase meta position to Relationship；But this method Mutation Strategy used in variation and overlaping stages and key control parameter can not adaptive changes；
And on, reach good performance with sub- phase meta position solving the problems, such as parent phase in order to make differential evolution algorithm, it needs All available Mutation Strategies are attempted in variation and overlaping stages, finely tune corresponding key control parameter.Many documents are It points out, the performance of raw differential evolution algorithm is highly dependent on the selection of trial vector generation strategy and the setting of parameter.Although For specific problem, suitable strategy and corresponding control parameter can be found out, but it may need to spend a large amount of meter Evaluation time.
Meanwhile in the different evolutionary phases, different strategies and parameter setting have unused global and local search energy Power, therefore how automatically learning strategy and parameter setting are necessary.
Summary of the invention
The technical problem to be solved in the present invention is to provide a kind of calculating cubic System Material parent phases and sub alternate orientation relationship Method, the characteristics of according to differential evolution algorithm, by introduce parameter adaptive strategy and adaptively selected Mutation Strategy scheme come Optimization algorithm calculated performance, and the problem for spending the time long is calculated for original method, using calculating matrix deformation method and multicore Parallel calculating method, which greatly shortens, calculates the time.
In order to solve the above technical problems, the embodiment of the present invention provides a kind of calculating cubic System Material parent phase and sub- phase meta position To the method for relationship, the described method comprises the following steps:
Step 1: obtaining the corresponding storage minimum θ value of parent population according to orientation point data and default orientation relationship formula Matrix A 1；Wherein, θ value indicates the declinate of the theoretical orientation and actual orientation of all orientation points；
Step 2: being carried out according to current iteration number and default maximum number of iterations to mutagenic factor F and crossover probability CR Nonlinear change, and Mutation Strategy is selected in tactful pond according to preset strategy adaptation scheme；
Step 3: being become according to selected Mutation Strategy and the F after nonlinear change and CR to parent population Different and crossover operation obtains intermediate population；And the centre is obtained according to the orientation point data and default orientation relationship formula The matrix A 2 of minimum θ value is stored corresponding to population；
Step 4: selecting the minimum θ value in matrix A 1 and matrix A 2, selected most in parent population and intermediate population The corresponding individual of small θ value, obtains progeny population；
Step 5: repeating step 2 to step 4 after the number of iterations reaches default maximum number of iterations, according to last The corresponding minimum θ value of the population that an iteration obtains obtains parent phase and the alternate orientation relationship of son.
Further, the default orientation relationship formula are as follows:
Wherein, θ is the declinate of the theoretical orientation and actual orientation of all orientation points, and N is orientation point number, and trace () is Track taking operation, arccos () are inverse cosine function, and V is orientation relationship to be sought, and M is parent phase orientation information, and MM is orientation point Eulerian angles, SjAnd SkThe respectively equivalent matrix of parent phase and the corresponding crystallography symmetrical factor of son, j and k respectively indicate parent phase With the crystallographic symmetry factor of sub- phase.
Further, described according to current iteration number and default maximum number of iterations, mutagenic factor F is carried out non-linear Variation specially carries out nonlinear change to mutagenic factor F using following equation:
Wherein, F mutagenic factor, t are current iteration number, and T is default maximum number of iterations, FmaxAnd FminRespectively make a variation The maximum value and minimum value of factor F, and Fmax=0.9, Fmin=0.4.
Further, the preset strategy adaptation scheme, comprising:
Candidate policy pond is generated, includes the first Mutation Strategy and the second Mutation Strategy in the candidate policy pond；
Initialize the selected probability p of the first Mutation Strategy1With the selected probability p of the second Mutation Strategy2=1-p1；
According to the scale Np of current population, a size is randomly generated as Np and element is equal on range [0,1] The vector of even distribution；If the value of j-th of element is greater than or equal to p in the vector1, then j-th in current population Body just applies the first Mutation Strategy, otherwise, just applies the second Mutation Strategy.
Further, the preset strategy adaptation scheme, further includes:
In the trial vector generated to the first Mutation Strategy of application, follow-on vector can be successfully entered by selection course Number ns1With the vector number nf of failure being dropped1It is counted respectively；The examination that the second Mutation Strategy of application is generated simultaneously It tests in vector, follow-on vector number ns can be successfully entered by selection course2With the vector number nf of failure being dropped2 It is counted respectively；
According to statistical result, every iteration 50 times, to Probability p1And p2It is once updated with following formula:
p2=1-p1
Updating Probability p1And p2Afterwards, ns is reset1、ns2、nf1And nf2Corresponding counter.
Further, first Mutation Strategy and second Mutation Strategy respectively indicate are as follows:
First Mutation Strategy: Vi,G=Xr1,G+F*(Xr2,G-Xr3,G)
Second Mutation Strategy: Vi,G=Xi,G+F*(Xbest,G-Xi,G)+F*(Xr1,G-Xr2,G)
Wherein, F is zoom factor, and X is population, and V is the offspring data after variation.
Optionally, Probability p1And p2Initial value be 0.5.
Further, described according to current iteration number and default maximum number of iterations, crossover probability CR is carried out non-thread Property variation, specially using following equation to crossover probability CR progress nonlinear change:
Wherein, CR is crossover probability, and t is current iteration number, and T is default maximum number of iterations, CRmax=0.9 and CRmin =0.3 is the maximum value and minimum value of crossover probability CR respectively.
Further, the method also includes: when being calculated using the default orientation relationship formula, SjAnd SkIt is defeated What is entered is the splicing of multiple matrixes respectively；Wherein,
SjIt is multiplied to obtain the first matrix with M after splicing, first with function mat2cell () by the first matrix-split at more Then multiple second matrixes are merged into a third matrix using cell2mat () function, by third matrix by a second matrix As SjSubstitute into progress next step calculating in the default orientation relationship formula；
SkIt is multiplied to obtain the 4th matrix with MM after splicing, first with function mat2cell () by the 4th matrix-split at more Then multiple 5th matrixes are merged into the 6th matrix using cell2mat () function, by the 6th matrix by a 5th matrix As SkSubstitute into progress next step calculating in the default orientation relationship formula.
Further, the method also includes: when being calculated using the default orientation relationship formula, using multicore The mode of parallel computation is calculated；
And in iteration, increase restrictive condition, when the value being calculated is greater than customized value, after no longer carrying out The calculating of continuous orientation point, directly progress next iteration.
The advantageous effects of the above technical solutions of the present invention are as follows:
The present invention optimized by introducing parameter adaptive strategy and adaptively selected Mutation Strategy scheme raw differential into The calculated performance for changing algorithm is calculated parent phase and sub alternate orientation relationship using the differential evolution algorithm after optimization, improved Computational accuracy；And in calculating process, by the way that multiple small-sized matrixes are merged into large-scale matrix, then in large-scale matrix into The mode that the medium-sized matrix of row calculates, has been greatly shortened the calculating time.Simultaneously in iteration, increase restrictive condition, when calculating When the value arrived is greater than customized value, with regard to no longer carrying out the calculating of subsequent orientation point, parameter iteration again is directly changed.This Outside, loop parallelization is further shortened and calculates the time by the method that the present invention is also calculated by using multi-core parallel concurrent；In reality In operation, client assigns the task to multiple cores, runs simultaneously in a cycle, returns to knot after waiting all end of runs Fruit, then result is integrated, improve arithmetic speed when multiple circular flows.To quickly and be accurately calculated cube Based material parent phase and the alternate orientation relationship of son provide basis for the recurrence of original parent phase.
Detailed description of the invention
Fig. 1 is the method for calculating cubic System Material parent phase and the alternate orientation relationship of son that first embodiment of the invention provides Flow diagram；
Fig. 2 is the method for calculating cubic System Material parent phase and the alternate orientation relationship of son that second embodiment of the invention provides Flow diagram；
Fig. 3 is that parent phase returns calculating raw data sample figure；
Fig. 4 is the convergent tendency figure of Q-SADE algorithm.
Specific embodiment
To keep the technical problem to be solved in the present invention, technical solution and advantage clearer, below in conjunction with attached drawing and tool Body embodiment is described in detail.
The present invention for existing cubic System Material parent phase and the alternate orientation relationship of son determination method, precision it is not high enough and Time longer problem is calculated, a kind of improved method for calculating cubic System Material parent phase and the alternate orientation relationship of son is provided, it should The specific visible the following example of the realization process of method.
First embodiment
As shown in Figure 1, the present embodiment provides a kind of method of calculating cubic System Material parent phase and the alternate orientation relationship of son, it should Method the following steps are included:
S101 obtains the corresponding storage minimum θ value of parent population according to orientation point data and default orientation relationship formula Matrix A 1；
It should be noted that θ value indicates the declinate of the theoretical orientation and actual orientation of all orientation points in above-mentioned steps； The raw data format of point data is orientated as shown in figure 3, data are included X value, Y value and are incremented by with 0.15 step-length；Phase column generation Table phase is different mutually with different digital representations；Three Eulerian angles are used to describe the orientation information of each point；BC value represents each orientation The contrast of point；BS value and MAD value are parameters needed for Material Field, respectively represent the sensitivity of diffraction belt edge and average angle is inclined Difference.Orientation information is mainly used in the present embodiment, i.e., the data of three Eulerian angles arrange to be calculated.
S102 carries out mutagenic factor F and crossover probability CR non-according to current iteration number and default maximum number of iterations Linear change, and Mutation Strategy is selected in tactful pond according to preset strategy adaptation scheme；
It should be noted that the control parameter of traditional differential evolution algorithm DE is by being manually set, in searching process In remain unchanged, therefore be unable to satisfy in each stage that algorithm performance is to several particular/special requirements is adopted, using the difference of parameter adaptive Divide evolution algorithm ADE that result precision can be improved.There are three key control parameters in optimization algorithm: population scale Np, intersecting generally Rate CR, mutagenic factor F.Wherein, control parameter Np is the value in original DE algorithm；
And for parameter F, at algorithm search initial stage, F value is larger, is conducive to expand search space, keeps the more of population Sample；Under the algorithm later period, convergent situation, F value is smaller, is conducive to scan for around optimal region, such energy Enough improve rate of convergence and search precision；
It for parameter CR, also needs to keep population diversity at evolution initial stage, reinforces local search in later stage of evolution and adopt Dynamic increment method is taken, makes parameter CR with evolutionary generation linear increment；
And in order to make raw differential evolution algorithm reach good performance, it needs to attempt in variation and overlaping stages all Available Mutation Strategy constitutes SaDE algorithm.The method of this adaptively selected Mutation Strategy is according to several available variations Policy selection is suitably applied in current population.
S103 makes a variation to parent population according to selected Mutation Strategy and the F after nonlinear change and CR And crossover operation, obtain intermediate population；And it is obtained corresponding to intermediate population according to orientation point data and default orientation relationship formula Storage minimum θ value matrix A 2；
S104 selects the minimum θ value in matrix A 1 and matrix A 2, selects minimum in parent population and intermediate population The corresponding individual of θ value, obtains progeny population；
S105, after repeating S102 to S104 until the number of iterations reaches default maximum number of iterations, according to changing for the last time The corresponding minimum θ value of population that generation obtains obtains parent phase and the alternate orientation relationship of son.
The characteristics of scheme of the present embodiment is according to differential evolution algorithm, by introducing parameter adaptive strategy and adaptive choosing It selects Mutation Strategy scheme and carrys out optimization algorithm calculated performance, parent phase is calculated by the differential evolution algorithm after optimization and son is alternate Orientation relationship can effectively improve computational accuracy.
Second embodiment
As shown in Fig. 2, the present embodiment provides a kind of method of calculating cubic System Material parent phase and the alternate orientation relationship of son, it should Method the following steps are included:
S201 obtains the orientation point data of cubic System Material；
It should be noted that orientation point data raw data format as shown in figure 3, data include X value, Y value and with 0.15 step-length is incremented by；Phase column represent phase, different mutually with different digital representations；Three Eulerian angles are used to describe each point Orientation information；BC value represents the contrast of each orientation point；BS value and MAD value are parameters needed for Material Field, are respectively represented The sensitivity of diffraction belt edge and average angular displacement.Orientation information, i.e., the number of three Eulerian angles are mainly used in the present embodiment It is calculated according to column.
S202, initialization population are calculated according to orientation information and default orientation relationship formula, generate storage minimum θ The matrix A 1 of value；
It it should be noted that the present embodiment is to acquire according to following orientation relationship formula in specific V, get θ can Minimum, V are the orientation relationship required by us:
Wherein, trace () is track taking, and arccos () is anticosine, Sj, SkFor one in known 24 3*3 matrixes, M For parent phase orientation, N is orientation point number.In calculating orientation relationship, the main optimization object of difference algorithm represents position to pass The matrix V of system.
θ value indicates the declinate of the theoretical orientation and actual orientation of all orientation points；Initialization population data, i.e. initialization square The corresponding matrix of battle array V and M, scale 6*30, first three is classified as V, and rear three are classified as M, Population Size 30, according to above-mentioned formula meter The corresponding minimum θ value of each individual in population is calculated, is 30 by population scale, matrix A 1 can be obtained as the matrix of 1*30.
S203 forms Mutation Strategy pond and adaptively selects Mutation Strategy in Mutation Strategy pond；
It should be noted that need to be attempted in variation and overlaping stages all to make raw differential algorithm reach superperformance Available Mutation Strategy, constitute SaDE algorithm.The method of this adaptively selected Mutation Strategy is according to several available changes Different policy selection is suitably applied in current population.Mainly there are following steps:
1, candidate policy pond is constituted.
In trial vector generating process in algorithm, each individual is selected in candidate policy a kind of next with certain probability Complete mutation process.It selects two kinds of Mutation Strategies as candidate, is " rand/1/bin " and " best/2/bin " respectively, they divide It is not represented as:
Wherein F is zoom factor, and X is population, and V is the offspring data after variation.Wherein tactful " rand/1/bin " is usual It can show good diversity, and the convergence that tactful " best/2/bin " has been typically exhibited.
2, the probability of Mutation Strategy is initialized.
Assuming that the probability of each individual applications strategy " rand/1/bin " is p in current population1, another strategy The probability of " best/2/bin " is p2=1-p1.Probability is set as 0.5, i.e. p1=p2=0.5.Therefore in initial population, The probability of both strategies of each individual applications is identical.For the scale Np of population, it is big to be randomly generated a Np Small vector, element therein are uniformly distributed on range [0,1].If the value of j-th of element is greater than or equal to p in vector1, J-th of individual in so current population is with regard to application strategy " rand/1/bin ", conversely, tactful " best/2/bin " will be answered With.
3, Adaptive evolution is most suitable Mutation Strategy.
After assessing all newly-generated trial vectors, application strategy " rand/1/bin " generate test to In amount, it can be successfully entered follow-on vector number by selection course and be denoted as ns1, failure the vector number being dropped be denoted as nf1；Correspondingly, in the trial vector that application strategy " best/2/bin " generates, follow-on vector number scale can be successfully entered For ns2, failure the vector number being dropped be denoted as nf2.So, Probability p1Following formula will be employed to be updated:
p2=1-p1
Above-mentioned expression formula respectively indicates, and Utilization strategies " rand/1/bin " generate the success rate and Utilization strategies of trial vector The success rate that " best/2/bin " generates trial vector accounts for the percentage of summation.In the present invention, iteration every 50 times primary to update p1、p2.Once p1、p2After being updated, counter ns is just reset1、ns2、nf1And nf2To avoid the product in the pervious study stage Tired some possible deleterious effects, reduce these problems to the greatest extent and have an impact to final result.This adaptive process energy It is enough to be gradually evolved into most suitable Mutation Strategy in the different study stages.
S204 adaptively carries out nonlinear change adjustment to mutagenic factor F；
It should be noted that traditional difference algorithm parameter is remained unchanged, Wu Faman by being manually set in searching process Result can be improved using the difference algorithm ADE of parameter adaptive to several particular/special requirements is adopted in algorithm performance in foot each stage Precision.There are three key control parameter population scale Np, control crossover probability parameter CR, mutagenic factor parameter F in optimization algorithm. Controlling Np is the value in original DE algorithm；And for parameter F, at algorithm search initial stage, F value is larger, is conducive to expand search Space keeps the diversity of population；Under the algorithm later period, convergent situation, F value is smaller, is conducive to the week in optimal region It encloses and scans for, can be improved rate of convergence and search precision in this way.So adjustment becomes when calculating phase relation using ADE algorithm The different factor F's specific formula is as follows:
Wherein, t is current evolutionary generation, and T is maximum evolutionary generation, FmaxAnd FminThe respectively maximum value of mutagenic factor F And minimum value.
S205 carries out population at individual variation according to selected Mutation Strategy and parameter F adjusted；
S206 adaptively carries out nonlinear change adjustment to crossover probability CR；
It should be noted that for parameter CR, also need to keep population diversity at evolution initial stage, in later stage of evolution plus Strong local search takes dynamic increment method, makes parameter CR with evolutionary generation linear increment, specific formula is as follows:
Wherein, t is current evolutionary generation, and T is maximum evolutionary generation, CRmax=0.9 and CRmin=0.3 is to intersect generally respectively The maximum value and minimum value of rate factor CR.
S207 carries out population crossover operation according to parameter CR adjusted；
S208 carries out matrix calculating according to default orientation relationship formula and filial generation, generates the matrix A 2 of storage minimum θ value；
S209 carries out selection operation in matrix A 1, A2 and updates parent；
S210 judges whether to reach the number of iterations, when not up to the number of iterations, returns to S203 and continues iteration, when When reaching the number of iterations, then enter S211；
S211 obtains parent phase and the alternate orientation relationship of son to be asked according to final minimum θ value.
In addition, being related to the operation of multiple minor matrixs when applying difference algorithm in actual phase calculation, needing It calculates more than one hundred million times, speed can be very slow, so multiple small-sized matrixes are merged into large-scale matrix in the present invention, then in large-scale square The calculating that medium-sized matrix is carried out in battle array, is greatly shortened the time.Simultaneously in iteration, increase restrictive condition, when what is be calculated When value is greater than customized value, with regard to no longer carrying out the calculating of subsequent orientation point, parameter iteration again is directly changed.
Specifically, in the present embodiment when being calculated using orientation relationship formula, due to Sj, SkMatrix group be combined into 24*24 Kind, in conjunction with the number of iterations and phase point number, calculation times are up to more than one hundred million times.In practical implementations, Sj, SkInput is complete The matrix of 24 3*3 in portion splices, SjAfter being multiplied after splicing with parent phase orientation M (Metzler matrix size is 3*3), the square of 72*3 is obtained Battle array, the matrix of 24 3*3 is split into using function mat2cell (), then utilizes cell2mat () function by 24 squares The one medium-sized matrix that battle array is merged into 3*72 carries out next step calculating.SkIt is multiplied after splicing with MM (matrix size 3*3), same benefit It is split into the matrix of 24 3*3 with function mat2cell (), is then merged 24 matrixes using cell2mat () function It is calculated at a medium-sized matrix of 3*72.When small-sized matrix being spliced to form medium-sized matrix in this way effectively shortening calculating Between.
Meanwhile the method that the present embodiment also uses multi-core parallel concurrent to calculate, loop parallelization is shortened into the time.In practical fortune In row, client assigns the task to multiple cores, runs simultaneously in a cycle, returns the result after waiting all end of runs, Result is integrated again, improves arithmetic speed when multiple circular flows.
Original differential evolution algorithm calculated result and the improved Q-SADE calculated result such as following table of the embodiment of the present invention Shown in one:
One phase relation calculation result table of table
And the convergent tendency of the improved Q-SADE algorithm of the embodiment of the present invention is then as shown in Figure 4, it is seen that the embodiment of the present invention Improved Q-SADE algorithm about starts to restrain after iteration 90 times；Convergence precision and in terms of compared to biography The differential evolution algorithm of system will be got well, while being calculated the time and being greatly shortened.
The scheme of the present embodiment optimizes original by introducing parameter adaptive strategy and adaptively selected Mutation Strategy scheme The calculated performance of beginning differential evolution algorithm calculates parent phase and sub alternate position using the differential evolution algorithm after optimization to pass System, improves computational accuracy；And in calculating process, by the way that multiple small-sized matrixes are merged into large-scale matrix, then big The mode that medium-sized matrix calculating is carried out in type matrix, has been greatly shortened the calculating time.Simultaneously in iteration, increase limitation item Part, with regard to no longer carrying out the calculating of subsequent orientation point, directly changes parameter weight when the value being calculated is greater than customized value New iteration.In addition, the method that the present invention is also calculated by using multi-core parallel concurrent, by loop parallelization come when further shortening calculating Between；In actual operation, client assigns the task to multiple cores, runs simultaneously in a cycle, waits all end of runs After return the result, then result is integrated, improves arithmetic speed when multiple circular flows.
In addition, it should be noted that, it should be understood by those skilled in the art that, the embodiment of the embodiment of the present invention can provide For method, apparatus or computer program product.Therefore, it is real that complete hardware embodiment, complete software can be used in the embodiment of the present invention Apply the form of example or embodiment combining software and hardware aspects.Moreover, the embodiment of the present invention can be used it is one or more its In include computer usable program code computer-usable storage medium (including but not limited to magnetic disk storage, CD-ROM, Optical memory etc.) on the form of computer program product implemented.
The embodiment of the present invention be referring to according to the method for the embodiment of the present invention, terminal device (system) and computer program The flowchart and/or the block diagram of product describes.It should be understood that flowchart and/or the block diagram can be realized by computer program instructions In each flow and/or block and flowchart and/or the block diagram in process and/or box combination.It can provide these Computer program instructions to general purpose computer, Embedded Processor or other programmable data processing terminal devices processor with A machine is generated, so that generating by the instruction that computer or the processor of other programmable data processing terminal devices execute For realizing the function of being specified in one or more flows of the flowchart and/or one or more blocks of the block diagram Device.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing terminal devices In computer-readable memory operate in a specific manner, so that instruction stored in the computer readable memory generates packet The manufacture of command device is included, which realizes in one side of one or more flows of the flowchart and/or block diagram The function of being specified in frame or multiple boxes.These computer program instructions can also be loaded at computer or other programmable datas It manages on terminal device, so that executing series of operation steps on computer or other programmable terminal equipments to generate computer The processing of realization, so that the instruction executed on computer or other programmable terminal equipments is provided for realizing in flow chart one The step of function of being specified in a process or multiple processes and/or one or more blocks of the block diagram.
Although preferred embodiments of the present invention have been described, it is created once a person skilled in the art knows basic Property concept, then additional changes and modifications can be made to these embodiments.So it includes excellent that the following claims are intended to be interpreted as It selects embodiment and falls into all change and modification of range of embodiment of the invention.
It should also be noted that, herein, relational terms such as first and second and the like are used merely to one Entity or operation are distinguished with another entity or operation, without necessarily requiring or implying between these entities or operation There are any actual relationship or orders.
Moreover, the terms "include", "comprise" or any other variant thereof is intended to cover non-exclusive inclusion, to make Obtaining process, method, article or terminal device including a series of elements not only includes those elements, but also including not bright The other element really listed, or further include element inherent to the process, method, article, or terminal device.? Do not have in the case where more limiting, the element limited by sentence "including a ...", it is not excluded that including the element There is also other identical elements in process, method, article or terminal device.
The above is a preferred embodiment of the present invention, it is noted that for those skilled in the art For, without departing from the principles of the present invention, several improvements and modifications can also be made, these improvements and modifications It should be regarded as protection scope of the present invention.

Claims (10)

1. a kind of method for calculating cubic System Material parent phase and the alternate orientation relationship of son, which is characterized in that the method includes with Lower step:
Step 1: obtaining the square of the corresponding storage minimum θ value of parent population according to orientation point data and default orientation relationship formula Battle array A1；Wherein, θ value indicates the declinate of the theoretical orientation and actual orientation of all orientation points；
Step 2: being carried out to mutagenic factor F and crossover probability CR non-thread according to current iteration number and default maximum number of iterations Property variation, and Mutation Strategy is selected in tactful pond according to preset strategy adaptation scheme；
Step 3: according to selected Mutation Strategy and the F after nonlinear change and CR to parent population carry out variation and Crossover operation obtains intermediate population；And the intermediate population is obtained according to the orientation point data and default orientation relationship formula The matrix A 2 of corresponding storage minimum θ value；
Step 4: selecting the minimum θ value in matrix A 1 and matrix A 2, minimum θ is selected in parent population and intermediate population It is worth corresponding individual, obtains progeny population；
Step 5: repeating step 2 to step 4 after the number of iterations reaches default maximum number of iterations, according to last time The corresponding minimum θ value of the population that iteration obtains obtains parent phase and the alternate orientation relationship of son.
2. the method for calculating cubic System Material parent phase and the alternate orientation relationship of son as described in claim 1, which is characterized in that institute State default orientation relationship formula are as follows:
Wherein, θ is the declinate of the theoretical orientation and actual orientation of all orientation points, and N is orientation point number, and trace () is track taking Operation, arccos () are inverse cosine function, and V is orientation relationship to be sought, and M is parent phase orientation information, and MM is the Europe of orientation point Draw angle, SjAnd SkThe respectively equivalent matrix of parent phase and the corresponding crystallography symmetrical factor of son, j and k respectively indicate parent phase and son The crystallographic symmetry factor of phase.
3. the method for calculating cubic System Material parent phase and the alternate orientation relationship of son as described in claim 1, which is characterized in that institute It states according to current iteration number and default maximum number of iterations, nonlinear change is carried out to mutagenic factor F, specially using following Formula carries out nonlinear change to mutagenic factor F:
Wherein, F mutagenic factor, t are current iteration number, and T is default maximum number of iterations, FmaxAnd FminRespectively mutagenic factor The maximum value and minimum value of F, and Fmax=0.9, Fmin=0.4.
4. the method for calculating cubic System Material parent phase and the alternate orientation relationship of son as described in claim 1, which is characterized in that institute State preset strategy adaptation scheme, comprising:
Candidate policy pond is generated, includes the first Mutation Strategy and the second Mutation Strategy in the candidate policy pond；
Initialize the selected probability p of the first Mutation Strategy1With the selected probability p of the second Mutation Strategy2=1-p1；
According to the scale Np of current population, a size is randomly generated as Np and element uniformly divides on range [0,1] The vector of cloth；If the value of j-th of element is greater than or equal to p in the vector1, then j-th of individual in current population is just Using the first Mutation Strategy, otherwise, the second Mutation Strategy is just applied.
5. the method for calculating cubic System Material parent phase and the alternate orientation relationship of son as claimed in claim 4, which is characterized in that institute State preset strategy adaptation scheme, further includes:
In the trial vector generated to the first Mutation Strategy of application, follow-on vector number can be successfully entered by selection course ns1With the vector number nf of failure being dropped1It is counted respectively；Simultaneously to application the second Mutation Strategy generate test to In amount, follow-on vector number ns can be successfully entered by selection course2With the vector number nf of failure being dropped2Respectively It is counted；
According to statistical result, every iteration 50 times, to Probability p1And p2It is once updated with following formula:
p2=1-p1
Updating Probability p1And p2Afterwards, ns is reset1、ns2、nf1And nf2Corresponding counter.
6. the method for calculating cubic System Material parent phase and the alternate orientation relationship of son as claimed in claim 5, which is characterized in that institute It states the first Mutation Strategy and second Mutation Strategy respectively indicates are as follows:
First Mutation Strategy:
Second Mutation Strategy:
Wherein, F is zoom factor, and X is population, and V is the offspring data after variation.
7. the method for calculating cubic System Material parent phase and the alternate orientation relationship of son as claimed in claim 6, which is characterized in that general Rate p1And p2Initial value be 0.5.
8. the method for calculating cubic System Material parent phase and the alternate orientation relationship of son as described in claim 1, which is characterized in that institute It states according to current iteration number and default maximum number of iterations, nonlinear change is carried out to crossover probability CR, under specially utilizing Column formula carries out nonlinear change to crossover probability CR:
Wherein, CR is crossover probability, and t is current iteration number, and T is default maximum number of iterations, CRmax=0.9 and CRmin= 0.3 is the maximum value and minimum value of crossover probability CR respectively.
9. the method for calculating cubic System Material parent phase and the alternate orientation relationship of son as described in claim 1, which is characterized in that institute State method further include: when being calculated using the default orientation relationship formula, SjAnd SkInput is multiple matrixes respectively Splicing；Wherein,
SjIt is multiplied to obtain the first matrix with M after splicing, first with function mat2cell () by the first matrix-split at multiple Then multiple second matrixes are merged into a third matrix using cell2mat () function by two matrixes, using third matrix as SjSubstitute into progress next step calculating in the default orientation relationship formula；
SkIt is multiplied to obtain the 4th matrix with MM after splicing, first with function mat2cell () by the 4th matrix-split at multiple Then multiple 5th matrixes are merged into the 6th matrix using cell2mat () function by five matrixes, using the 6th matrix as SkSubstitute into progress next step calculating in the default orientation relationship formula.
10. the method for calculating cubic System Material parent phase and the alternate orientation relationship of son as claimed in claim 9, which is characterized in that The method also includes: when being calculated using the default orientation relationship formula, using multi-core parallel concurrent calculating by the way of into Row calculates；
And in iteration, increase restrictive condition, it is subsequent with regard to no longer carrying out when the value being calculated is greater than customized value The calculating of orientation point, directly progress next iteration.
CN201910424335.9A 2019-05-21 2019-05-21 A method of calculating cubic System Material parent phase and the alternate orientation relationship of son Pending CN110222300A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910424335.9A CN110222300A (en) 2019-05-21 2019-05-21 A method of calculating cubic System Material parent phase and the alternate orientation relationship of son

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910424335.9A CN110222300A (en) 2019-05-21 2019-05-21 A method of calculating cubic System Material parent phase and the alternate orientation relationship of son

Publications (1)

Publication Number Publication Date
CN110222300A true CN110222300A (en) 2019-09-10

Family

ID=67821437

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910424335.9A Pending CN110222300A (en) 2019-05-21 2019-05-21 A method of calculating cubic System Material parent phase and the alternate orientation relationship of son

Country Status (1)

Country Link
CN (1) CN110222300A (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2000181503A (en) * 1998-12-14 2000-06-30 Tokai Rubber Ind Ltd Adaptive control method for cyclic signal
CN106647262A (en) * 2016-11-28 2017-05-10 中国人民解放军国防科学技术大学 Differential evolution method facing agile satellite multi-object task planning
CN106934453A (en) * 2017-02-08 2017-07-07 北京科技大学 A kind of determination method of cubic System Material parent phase and sub alternate orientation relationship
CN108564592A (en) * 2018-03-05 2018-09-21 华侨大学 Based on a variety of image partition methods for being clustered to differential evolution algorithm of dynamic
CN109284579A (en) * 2018-11-12 2019-01-29 东南大学 Nonlinear system parameter discrimination method based on improved differential evolution algorithm

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2000181503A (en) * 1998-12-14 2000-06-30 Tokai Rubber Ind Ltd Adaptive control method for cyclic signal
CN106647262A (en) * 2016-11-28 2017-05-10 中国人民解放军国防科学技术大学 Differential evolution method facing agile satellite multi-object task planning
CN106934453A (en) * 2017-02-08 2017-07-07 北京科技大学 A kind of determination method of cubic System Material parent phase and sub alternate orientation relationship
CN108564592A (en) * 2018-03-05 2018-09-21 华侨大学 Based on a variety of image partition methods for being clustered to differential evolution algorithm of dynamic
CN109284579A (en) * 2018-11-12 2019-01-29 东南大学 Nonlinear system parameter discrimination method based on improved differential evolution algorithm

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
张延莉: "函数优化问题求解的自适应差分进化算法", 《内蒙古师范大学学报》 *

Similar Documents

Publication Publication Date Title
Yang et al. Multi-objective flower algorithm for optimization
Mouret et al. Quality diversity for multi-task optimization
Ray et al. Infeasibility driven evolutionary algorithm for constrained optimization
Pham Improved Nelder Mead’s simplex method and applications
Safro et al. Multilevel algorithms for linear ordering problems
Altinoz et al. Multiobjective Hooke–Jeeves algorithm with a stochastic Newton–Raphson-like step-size method
Yang et al. ADAPTIVE RANKS CLONE AND k‐NEAREST NEIGHBOR LIST–BASED IMMUNE MULTI‐OBJECTIVE OPTIMIZATION
EP3764250A2 (en) Optimization apparatus, control method for optimization apparatus, and optimization program
Karapetyan et al. A new approach to population sizing for memetic algorithms: a case study for the multidimensional assignment problem
Silva et al. Finding multiple roots of a box-constrained system of nonlinear equations with a biased random-key genetic algorithm
JP2022541370A (en) Data enrichment policy update method, apparatus, device and storage medium
Cao et al. An enhanced whale optimization algorithm with improved dynamic opposite learning and adaptive inertia weight strategy
Nagar et al. Multiobjective optimization for electromagnetics and optics: An introduction and tutorial based on real-world applications
Cook et al. GPU based parallel ising computing for combinatorial optimization problems in VLSI physical design
Liu et al. Joint multi-dimension pruning via numerical gradient update
CN110796268A (en) Method and device for determining hyper-parameters of business processing model
EP3674959A1 (en) Optimization calculation method and information processing apparatus
Chowdhury et al. Improvements to single-objective constrained predator–prey evolutionary optimization algorithm
JP7111966B2 (en) Optimization device and control method for optimization device
Soukaina et al. A hybrid genetic algorithm for solving 0/1 knapsack problem
JP7256378B2 (en) Optimization system and method of controlling the optimization system
CN110222300A (en) A method of calculating cubic System Material parent phase and the alternate orientation relationship of son
CN110782020A (en) Network structure determination method and device and electronic system
Cortés-Antonio et al. Design and implementation of differential evolution algorithm on FPGA for double-precision floating-point representation
Serpa et al. Flexible use of temporal and spatial reasoning for fast and scalable CPU broad‐phase collision detection using KD‐Trees

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
RJ01 Rejection of invention patent application after publication

Application publication date: 20190910

RJ01 Rejection of invention patent application after publication