CN107609323A - A kind of method that proving ground road conditions period is calculated based on genetic algorithm - Google Patents
A kind of method that proving ground road conditions period is calculated based on genetic algorithm Download PDFInfo
- Publication number
- CN107609323A CN107609323A CN201711068583.1A CN201711068583A CN107609323A CN 107609323 A CN107609323 A CN 107609323A CN 201711068583 A CN201711068583 A CN 201711068583A CN 107609323 A CN107609323 A CN 107609323A
- Authority
- CN
- China
- Prior art keywords
- population
- calculating
- test field
- genetic algorithm
- parent 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.)
- Granted
Links
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Measuring Or Testing Involving Enzymes Or Micro-Organisms (AREA)
Abstract
The invention provides a kind of method that proving ground road conditions period is calculated based on genetic algorithm, it is characterised in that comprises the following steps：Step 1, initial population is obtained by the way of random assignment, using the initial population as parent population, sets initial evolution number t=0；Step 2, equivalent is carried out to parent population test site damage Y is calculatedsi；Step 3, object function is calculated；Step 4, the relative adaptability degrees of each period in parent population are calculated；Step 5, carry out genetic operator to operate to obtain progeny population, concurrently set evolution number t=t+1；Step 6, if progeny population is unsatisfactory for the condition of convergence or evolution number is less than maximum evolution number, makes progeny population substitute parent population, return to step 2, if progeny population meets the condition of convergence and evolution number is more than or equal to maximum evolution number, exported progeny population as purpose population；Step 7, the period that the value minimum of object function is selected from purpose population exports as a result.
Description
Technical Field
The invention relates to a method for calculating road condition cycle number of a test field, in particular to a method for calculating road condition cycle number of an automobile test field based on a genetic algorithm.
Background
Modern automotive designs must be market-oriented, and products with "over-or under-designed" life are often uneconomical and less competitive in the market, so the user's requirements for use should be considered in either the automotive design, development or testing stages. The automobile reliability test is an important means for examining and evaluating the durability of the automobile, but most of the traditional test bases are biased to strength tests, but not life tests, so that the automobile component is not broken under the worst working condition, and the automobile reliability test can meet the general engineering requirements. Obviously, these tests rely on experience, habits, and do not reasonably consider the situation of the user, mainly by inference, rather than based on scientific principles.
The fatigue endurance test specification of the vehicle structure test field must be associated with the use condition of a user, and the unreasonable test specification which is disjointed with the use condition of the user cannot guide the development of test work, and can mislead the development of a vehicle type and even the development failure. The test field test specification is generally formulated by performing equivalent transformation of the test specification, that is, an effective test specification of a certain test field is transformed into a test specification of another test field according to principles such as a linear damage accumulation theory and fatigue damage equivalent equivalence.
At present, two modes are generally adopted for calculating the number of working condition cycles of a test field: firstly, rely on experimental engineer's experience to judge, but it is great to rely on experience to confirm the error, and subjective influence is big, and the result is inaccurate, also lacks the direction during adjustment, can cause very big influence to the judged result of cycle number. The second method is the least squares method. When the number of channels included in a sub-specification is greater than its sub-specification Fan Geshu, the above mathematical model is an over-determined equation set and has no exact solution. The closest solution is generally found by fitting a discrete solution with a least squares optimization algorithm, but when there is an abnormal value, a local optimal solution is easily found, but a global optimal solution cannot be obtained.
Disclosure of Invention
The present invention is made to solve the above problems, and an object of the present invention is to provide a method for calculating a number of road condition cycles in an automobile test field based on a genetic algorithm.
The invention provides a method for calculating road condition cycle number of an automobile test field based on a genetic algorithm, which is characterized by comprising the following steps of:
step 1, generating k cycle numbers as an initial chromosome by adopting a random assignment mode, wherein the set of the k cycle numbers is an initial population p (t) = [ beta ] 1 ,β 2 ,…,β k ]Setting an initial evolution frequency t =0 by using the initial population p (t) as a parent population;
Step 2, carrying out equivalent calculation on the parent population according to the following formula (1) to obtain the damage Y of the test field si ，
In the formula (1), X ij Indicating damage of the lane i on the road type j, Y i Represents the target lesion, beta, of channel i i Represents the number of cycles of the road i;
step 3, according to the formula (2), calculating to obtain an objective function F (beta) i )，
F(β i )＝|Y si -Y i | (2)
In the formula (2), Y si For test field Damage, Y i Represents a target lesion of channel i;
step 4, calculating the relative fitness of each cycle number in the parent population according to the following formula (3),
p＝F(β i )/∑F(β i ) (3)
in the formula (3), F (. Beta.) i ) As a target function, Σ F (β) i ) Is the sum of fitness of all cycle numbers in the parent population;
step 5, performing corresponding genetic operator operation on the parent population to obtain an offspring population, and setting the evolution times t = t +1;
step 6, if the child population does not meet the convergence condition or the evolutionary times are less than the maximum evolutionary times, the child population is made to replace the parent population, the step 2 is returned,
if the filial generation population meets the convergence condition and the evolution times are more than or equal to the maximum evolution times, outputting the filial generation population as a target population;
and 7, selecting the cycle number with the minimum value of the target function from the target population as a result and outputting the result.
The method for calculating the road condition cycle number of the automobile test field based on the genetic algorithm, provided by the invention, can also have the following characteristics: wherein, the genetic operator operation in step 5 is:
When the relative fitness is less than or equal to 0.05, carrying out selection operation on the parent population to obtain a child population,
when the relative fitness is more than or equal to 0.05 and less than or equal to 0.1, performing cross operation in the parent population to obtain a child population,
and when the relative fitness is more than or equal to 0.1, carrying out mutation operation on the parent population to obtain an offspring population.
The method for calculating the road condition cycle number of the automobile test field based on the genetic algorithm, provided by the invention, can also have the following characteristics: wherein the interleaving operation is: the parent population was averaged over two sets of cycles.
The method for calculating the road condition cycle number of the automobile test field based on the genetic algorithm, provided by the invention, can also have the following characteristics: wherein the value space of the cycle number is [ M ] min ,M max ]，M min Is 20 to 550, M max Is 100 to 2500.
The method for calculating the road condition cycle number of the automobile test field based on the genetic algorithm, provided by the invention, can also have the following characteristics: wherein the convergence condition is F (beta) i )→0。
The method for calculating the road condition cycle number of the automobile test field based on the genetic algorithm, provided by the invention, can also have the following characteristics: wherein the maximum number of evolutions is 100.
Action and Effect of the invention
According to the method for calculating the road condition cycle number of the automobile test field based on the genetic algorithm, the genetic algorithm is adopted, so that the calculation precision and accuracy of the road condition cycle number of the automobile test field are high, the evolution times can be set, the optimal result meeting the damage requirement of each channel can be further optimized and selected, and meanwhile, the parallel calculation can be realized, and the calculation efficiency is greatly improved.
Drawings
Fig. 1 is a flowchart of a method for calculating road condition cycle numbers of an automobile test field based on a genetic algorithm according to an embodiment of the present invention.
Detailed Description
In order to make the technical means and functions of the present invention easy to understand, the present invention is specifically described below with reference to the embodiments and the accompanying drawings.
< example >
Fig. 1 is a flowchart of a method for calculating road condition cycle numbers of an automobile test field based on a genetic algorithm according to an embodiment of the present invention.
As shown in fig. 1, the method for calculating the number of road condition cycles in the automobile test field based on the genetic algorithm comprises the following steps:
step 1, generating k cycle numbers as an initial chromosome by adopting a random assignment mode, wherein the set of the k cycle numbers is an initial population p (t) = [ beta ] 1 ,β 2 ,…,β k ]The initial population p (t) is set as a parent population, and the initial number of evolutions t =0. In this embodiment, the space of the number of cycles is [ M ] min ,M max ]，M min Is 20 to 550 of max Is 100 to 2500.
Step 2, carrying out equivalent calculation on the parent population according to the following formula (1) to obtain the damage Y of the test field si ，
In the formula (1), X ij Indicating the impairment of channel i on road type j, Y i Represents the target damage, beta, of channel i i Indicating the number of cycles for road i.
And 3, calculating to obtain an objective function F (beta) according to the formula (2) i )，
F(β i )＝|Y si -Y i | (2)
In the formula (2), Y si For test field Damage, Y i Representing the target lesion of channel i.
Step 4, calculating the relative fitness of each cycle number in the parent population according to the following formula (3),
p＝F(β i )/∑F(β i ) (3)
in the formula (3), F (. Beta.) is i ) As a target function, Σ F (β) i ) Is the sum of fitness of all cycles in the parent population.
And 5, performing corresponding genetic operator operation on the parent population to obtain an offspring population, and setting the evolution times t = t +1.
The genetic operator operates as:
when the relative fitness is less than or equal to 0.05, carrying out selection operation on the parent population to obtain a child population,
when the relative fitness is more than or equal to 0.05 and less than or equal to 0.1, performing cross operation in the parent population to obtain a child population,
And when the relative fitness is more than or equal to 0.1, carrying out mutation operation in the parent population (the cross operation is to carry out two-two combination on each cycle number in the parent population to obtain the average number of the cycles), and obtaining the offspring population.
Step 6, if the child population does not meet the convergence condition or the evolutionary times are less than the maximum evolutionary times, the child population is made to replace the parent population, the step 2 is returned,
and if the filial generation population meets the convergence condition and the evolution times are more than or equal to the maximum evolution times, outputting the filial generation population as a target population. In the present embodiment, the convergence condition is F (β) i ) → 0, maximum evolution number is 100.
And 7, selecting the cycle number with the minimum value of the target function from the target population as a result and outputting the result.
Effects and effects of the embodiments
According to the method for calculating the road condition cycle number of the automobile test site based on the genetic algorithm, the genetic algorithm is adopted, so that the calculation precision of the road condition cycle number of the automobile test site is high, the accuracy is good, the evolution times can be set, the optimal result meeting the damage requirement of each channel can be further optimized and selected, meanwhile, the parallel calculation can be realized, and the calculation efficiency is greatly improved.
The above embodiments are preferred examples of the present invention, and are not intended to limit the scope of the present invention.
Claims (6)
1. A method for calculating road condition cycle number of an automobile test field based on a genetic algorithm is characterized by comprising the following steps:
step 1, generating k cycle numbers as an initial chromosome in a random assignment mode, wherein the k cycle numbers are collectedSynthesizing to initial population p (t) = [ beta ] 1 ,β 2 ,…,β k ]Setting an initial evolution frequency t =0 by using the initial population p (t) as a parent population;
step 2, carrying out equivalent calculation on the parent population according to the following formula (1) to obtain the damage Y of the test field si ，
In the formula (1), X ij Indicating the impairment of channel i on road type j, Y i Represents the target damage, beta, of channel i i Represents the number of cycles of the road i;
and 3, calculating to obtain an objective function F (beta) according to the formula (2) i )，
F(β i )＝|Y si -Y i | (2)
In the formula (2), Y si For test field Damage, Y i Representing a target lesion of channel i;
step 4, calculating the relative fitness of each cycle number in the parent population according to the following formula (3),
p＝F(β i )/∑F(β i ) (3)
in the formula (3), F (. Beta.) is i ) As a target function, Σ F (β) i ) The sum of the fitness of all the cycle numbers in the parent population;
step 5, performing corresponding genetic operator operation on the parent population to obtain an offspring population, and setting the evolution frequency t = t +1;
Step 6, if the child population does not satisfy the convergence condition or the number of evolutionary times is less than the maximum number of evolutionary times, the child population is made to replace the parent population, the step 2 is returned to,
if the child population meets the convergence condition and the evolution times are more than or equal to the maximum evolution times, outputting the child population as a target population;
and 7, selecting the cycle number with the minimum value of the target function from the target population as a result and outputting the result.
2. The method for calculating the number of road condition cycles in the automobile test field based on the genetic algorithm as claimed in claim 1, wherein:
wherein said genetic operator in said step 5 operates to:
when the relative fitness is less than or equal to 0.05, carrying out selection operation on the parent population to obtain the child population,
when the relative fitness is more than or equal to 0.05 and less than or equal to 0.1, performing cross operation in the parent population to obtain the child population,
and when the relative fitness is more than or equal to 0.1, performing mutation operation in the parent population to obtain the offspring population.
3. The method for calculating the number of road condition cycles in the automobile test field based on the genetic algorithm as claimed in claim 2, wherein:
Wherein the interleaving operation is: the number of cycles per parent population was averaged in two groups.
4. The method for calculating the number of road condition cycles in the automobile test field based on the genetic algorithm as claimed in claim 1, wherein:
wherein the value space of the cycle number is [ M ] min ,M max ]Said M is min Is 20 to 550, the M max Is 100 to 2500.
5. The method for calculating the number of road condition cycles in the automobile test field based on the genetic algorithm as claimed in claim 1, wherein:
wherein the convergence condition is F (beta) i )→0。
6. The method for calculating the number of road condition cycles in the automobile test field based on the genetic algorithm as claimed in claim 1, wherein:
wherein the maximum number of evolutions is 100.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711068583.1A CN107609323B (en) | 2017-11-03 | 2017-11-03 | Method for calculating road condition cycle number of automobile test field based on genetic algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711068583.1A CN107609323B (en) | 2017-11-03 | 2017-11-03 | Method for calculating road condition cycle number of automobile test field based on genetic algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107609323A true CN107609323A (en) | 2018-01-19 |
| CN107609323B CN107609323B (en) | 2021-08-10 |
Family
ID=61084711
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711068583.1A Active CN107609323B (en) | 2017-11-03 | 2017-11-03 | Method for calculating road condition cycle number of automobile test field based on genetic algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107609323B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114676598A (en) * | 2022-05-25 | 2022-06-28 | 岚图汽车科技有限公司 | Acceleration method and device for durability test of whole vehicle road of vehicle body system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7672910B1 (en) * | 2005-12-12 | 2010-03-02 | The Mathworks, Inc. | System and method for performing non-linear constrained optimization with a genetic algorithm |
| CN101718651A (en) * | 2009-11-18 | 2010-06-02 | 上海理工大学 | Evaluation method of residual strength and service life under strengthening and damaging combined action |
| CN102156066A (en) * | 2011-03-28 | 2011-08-17 | 上海理工大学 | Method for predicating fatigue life of mobile S-N (Stress-Life) curve on basis of strengthening and damage |
| CN103870647A (en) * | 2014-03-14 | 2014-06-18 | 西安工业大学 | Operation workshop scheduling modeling method based on genetic algorithm |
-
2017
- 2017-11-03 CN CN201711068583.1A patent/CN107609323B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7672910B1 (en) * | 2005-12-12 | 2010-03-02 | The Mathworks, Inc. | System and method for performing non-linear constrained optimization with a genetic algorithm |
| CN101718651A (en) * | 2009-11-18 | 2010-06-02 | 上海理工大学 | Evaluation method of residual strength and service life under strengthening and damaging combined action |
| CN102156066A (en) * | 2011-03-28 | 2011-08-17 | 上海理工大学 | Method for predicating fatigue life of mobile S-N (Stress-Life) curve on basis of strengthening and damage |
| CN103870647A (en) * | 2014-03-14 | 2014-06-18 | 西安工业大学 | Operation workshop scheduling modeling method based on genetic algorithm |
Non-Patent Citations (3)
| Title |
|---|
| 吴道俊: "车辆疲劳耐久性分析、试验与优化关键技术研究", 《中国博士学位论文全文数据库 工程科技Ⅱ辑（月刊）》 * |
| 宁凡光等: "与用户数据相关的试验场规范研究", 《农业装备与车辆工程》 * |
| 张新宇: "车辆结构试验场疲劳耐久试验规范及台架试验研究", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑（月刊）》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114676598A (en) * | 2022-05-25 | 2022-06-28 | 岚图汽车科技有限公司 | Acceleration method and device for durability test of whole vehicle road of vehicle body system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107609323B (en) | 2021-08-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Hayes et al. | Increased accuracy of artificial selection by using the realized relationship matrix | |
| WO2021241115A1 (en) | Deterioration estimation device, model generation device, deterioration estimation method, model generation method, and program | |
| WO2009120083A1 (en) | Automated filling of conditional probability data in a decision support apparatus that uses a bayesian belief network | |
| CN102252898A (en) | Method for testing accelerated life of electronic product based on life-stress model | |
| CN111125964B (en) | Sewage treatment process proxy model construction method based on Kriging interpolation method | |
| Yano et al. | Generating feasible test paths from an executable model using a multi-objective approach | |
| JP2014203228A (en) | Project management support system | |
| CN107609323A (en) | A kind of method that proving ground road conditions period is calculated based on genetic algorithm | |
| CN114578087B (en) | Wind speed uncertainty measurement method based on non-dominant sorting and stochastic simulation algorithm | |
| US20030237058A1 (en) | Method and system for the automatic design of experiments | |
| Yen et al. | Testing forecast accuracy of expectiles and quantiles with the extremal consistent loss functions | |
| CN102255297A (en) | Short-circuit current limiter additionally-arranging method and device | |
| Nagai et al. | Optimization of ridge parameters in multivariate generalized ridge regression by plug-in methods | |
| CN110286646B (en) | Numerical control machine tool assembly importance evaluation method | |
| CN106548301B (en) | Power consumer clustering method and device | |
| Bijma et al. | Genetic gain of pure line selection and combined crossbred purebred selection with constrained inbreeding | |
| CN109857581B (en) | Method for optimizing and selecting software reliability growth model | |
| Chiang et al. | Kernel estimation of rate function for recurrent event data | |
| CN109284320B (en) | Automatic regression diagnosis method on big data platform | |
| Kloubert | Fast point estimate method for correlated multimodally distributed input variables | |
| CN111605587B (en) | Train, automatic train operation system and operation parameter optimization method and system thereof | |
| Shahraki et al. | A combined algorithm for solving reliability-based robust design optimization problems | |
| CN115049223A (en) | Power grid equipment parameter risk rating method based on multi-dimensional data mining | |
| Yao et al. | Testing the effectiveness of principal components in adjusting for relatedness in genetic association studies | |
| CN112653129B (en) | Transient power angle stability margin estimation method, device and system |
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 |