CN102063550A - Intelligent design system of cold extrusion piece with machine intelligence involved design decision - Google Patents
Intelligent design system of cold extrusion piece with machine intelligence involved design decision Download PDFInfo
- Publication number
- CN102063550A CN102063550A CN2011100024036A CN201110002403A CN102063550A CN 102063550 A CN102063550 A CN 102063550A CN 2011100024036 A CN2011100024036 A CN 2011100024036A CN 201110002403 A CN201110002403 A CN 201110002403A CN 102063550 A CN102063550 A CN 102063550A
- Authority
- CN
- China
- Prior art keywords
- design
- cold
- knowledge
- extruded part
- decision
- 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
Links
- 238000013461 design Methods 0.000 title claims abstract description 168
- 238000000641 cold extrusion Methods 0.000 title claims abstract description 17
- 238000000034 method Methods 0.000 claims abstract description 62
- 238000011156 evaluation Methods 0.000 claims abstract description 25
- 238000004458 analytical method Methods 0.000 claims abstract description 22
- 239000002184 metal Substances 0.000 claims abstract description 5
- 230000007547 defect Effects 0.000 claims abstract description 4
- 238000009826 distribution Methods 0.000 claims abstract description 4
- 238000000465 moulding Methods 0.000 claims abstract description 4
- 238000005457 optimization Methods 0.000 claims description 36
- 238000004422 calculation algorithm Methods 0.000 claims description 20
- 230000006870 function Effects 0.000 claims description 17
- 238000012706 support-vector machine Methods 0.000 claims description 17
- 230000002068 genetic effect Effects 0.000 claims description 14
- 230000008569 process Effects 0.000 claims description 14
- 238000004364 calculation method Methods 0.000 claims description 13
- 238000013507 mapping Methods 0.000 claims description 11
- 238000004088 simulation Methods 0.000 claims description 11
- 238000007493 shaping process Methods 0.000 claims description 10
- 238000012360 testing method Methods 0.000 claims description 10
- 230000007246 mechanism Effects 0.000 claims description 9
- 238000012549 training Methods 0.000 claims description 7
- 230000002950 deficient Effects 0.000 claims description 4
- 238000012938 design process Methods 0.000 claims description 4
- 230000007774 longterm Effects 0.000 claims description 4
- 210000000349 chromosome Anatomy 0.000 claims description 3
- 238000010438 heat treatment Methods 0.000 claims description 3
- 230000008676 import Effects 0.000 claims description 3
- 239000000463 material Substances 0.000 claims description 3
- 230000000704 physical effect Effects 0.000 claims description 3
- 108090000623 proteins and genes Proteins 0.000 claims description 3
- 239000000126 substance Substances 0.000 claims description 3
- 238000006467 substitution reaction Methods 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 238000000151 deposition Methods 0.000 claims description 2
- 235000013569 fruit product Nutrition 0.000 claims description 2
- 238000005516 engineering process Methods 0.000 description 5
- 230000000694 effects Effects 0.000 description 4
- 238000009825 accumulation Methods 0.000 description 3
- 230000006399 behavior Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000007812 deficiency Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 230000001537 neural effect Effects 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 210000004556 brain Anatomy 0.000 description 1
- 230000005465 channeling Effects 0.000 description 1
- 239000000470 constituent Substances 0.000 description 1
- 238000012217 deletion Methods 0.000 description 1
- 230000037430 deletion Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 238000003780 insertion Methods 0.000 description 1
- 230000037431 insertion Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012067 mathematical method Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 201000009240 nasopharyngitis Diseases 0.000 description 1
- 238000012913 prioritisation Methods 0.000 description 1
Images
Landscapes
- General Factory Administration (AREA)
Abstract
The invention discloses an intelligent design system of a cold extrusion piece with machine intelligence involved design decision, comprising a design type expert system module, an engineering analysis module and a product performance evaluation module. The interior of the design type expert system module includes a great deal of expert-level knowledge and experience in the cold extrusion molding field, solves the problems in the cold extrusion molding field by utilizing the human expert knowledge and methods solving the problems, reasons and judges by utilizing the knowledge and the experience, and simulates the decision procedure of the human experts. The engineering analysis module is used for establishing a functional relationship between a design variable and an optimized target of an initial design scheme of the cold extrusion piece, optimizing the initial design scheme and drawing the optimized graphic information. The product performance evaluation module is used for evaluating the qualities of the formed parts by judging the concentrated strain, the forming load, the metal flow, the stress strain distribution, the forming defect and the mould filling condition, and defining the relative rule for each evaluated object to be evaluated. In the invention, the intelligent design system simulates the human intelligence to design the cold extrusion piece, reduces the technical requirements of users, and increases the working efficiency.
Description
Technical field
The present invention relates to a kind of intelligent design system of cold-extruded part.
Background technology
Design is the analysis of a complexity, comprehensive and decision making, and its constituent can be divided three classes: intelligence creation, analytical calculation and design are expressed.Wherein, intelligence is created the work that the reasoning comprise in the design process, comprehensive, analogy etc. need realize by thinking activities, is the stage of global design design, also is the part of core the most in the design; Analytical calculation is by mathematical method, specifically is the behavior of product to be analyzed or performance is carried out emulation and simulation by all kinds of analysis software, to determine the feasibility of scheme; It is that the result who will design explains out with the form of figure, literal or numeral that design is expressed.Traditional CAD system is two class design efforts after Aided Design person finishes mostly: aspect analytical calculation, be representative with the finite element analysis technology mainly, comprise engineering calculation system, analogue system, design system and the simulation system of all kinds of special uses; Design expression aspect comprises a two-dimensional graphics technology of designing and developing in early days and the product three-dimensional modeling technology that has been widely adopted today, and the value-added functionality of being brought by the product three-dimensional expression, as interference checking, virtual assembling etc.Even traditional CAD system two class work after the Aided Design personnel have finished forcefully, but its intelligent level is still waiting to improve, most CAD software all is to exist as independently analyzing design platform, still not having the design platform of having integrated canonical analysis calculating and design expression class CAD software occurs, the co-ordination of existing number of C AD software only is confined to realize by software interface separately mostly, the application need designer of interface has very high secondary development ability, and this has improved the designer greatly and has utilized interface to realize the threshold of software platform collaborative work.And create as the intelligence of design core, traditional CAD system does not provide strong assisting and supporting to it.Under the current design pattern, intelligence is created intellectual service work such as the reasoning relate to, comprehensive, association, analogy, study and is still finished by the human expert.And the leading design of human expert has its limitation, be in particular in: the designer independently carries out the accumulation that design effort needs the knowledge and experience of higher level, and the accumulation of these knowledge and experiences usually needs very years to grasp, and this pursues efficiently with commercial production that principle contradicts; And, in the actual process of dealing with problems of old designer, often can't clearly explain and specifically use which knowledge, implicit unclear knowledge can't be converted into accurate phy symbol, make old designer knowledge accurately, intactly can't be passed to new designer, thereby cause the knowledge and experience of long-term accumulation to run off.
Design for cold-extruded part also is so, and the technological deficiency that existing cold-extruded part design exists is: with the leading design of human expert,, inefficiency very high to technical requirements of users.
Summary of the invention
For overcome existing cold-extruded part with the leading design of human expert, to very high, the ineffective deficiency of technical requirements of users, the invention provides a kind of simulating human intelligent design cold-extruded part, user's technical requirement reduction, the machine intelligence of high efficiency participated in the cold-extruded part intelligent design system of design decision.
The technical solution adopted for the present invention to solve the technical problems is:
A kind of machine intelligence participates in the cold-extruded part intelligent design system of design decision, and described intelligent design system comprises:
The design mode expert system module, the knowledge and the experience of a large amount of cold-extrusion shaping domain expert levels contained in inside, the method of utilizing human expert's knowledge and dealing with problems is handled the cold-extrusion shaping field question, utilize described knowledge and experience, carry out reasoning and judgement, simulating human expert's decision process, described design mode expert system module comprises:
Knowledge base, set is used to design the domain knowledge of cold-extruded part and expert's practical experience, comprises product example storehouse, design rule storehouse, design parameter storehouse and numerical method storehouse; The product example storehouse is the work set of the medium-term and long-term example of accumulate of designer, produces new object module when being used for based on case-based reasoning, simultaneously as original learning model so that call when needing in designing; The design rule storehouse is the main place that the knowledge and experience relevant with cold-extruded part stored, and knowledge and experience is converted into the manageable sign format of computing machine so that describe and editor; Chemical constitution, the physical property of various cold extrusion die materials, hot-working and the heat treatment process parameter and the relevant mechanical properties technical information of recommendation are mainly being deposited in the design parameter storehouse; Numerical method stock is being put all kinds of numerical algorithm knowledge that Intelligentized design method can be used;
Inference machine is used for after the user imports the performance requirement of cold-extruded part, by inference mechanism is tactful by inference knowledge base is searched for;
Dynamic data base is used for the information in when work dynamic memory design process, comprises initial designs requirement, intermediate result and the final plan of cold-extruded part, controls it by inference mechanism and refreshes;
Explanation facility is used for problem that user or slip-stick artist are proposed, provides an answer; The engineering analysis module is used to set up the design variable of cold-extruded part preliminary design scheme and the funtcional relationship between the optimization aim, then preliminary design scheme is optimized, and is drawn out graphical information after the optimization, comprising:
The simulation analysis unit is drawn up the representational test figures of many groups with finite element software as tool mould, with as follow-up training sample when setting up Nonlinear Mapping;
The calculation optimization unit, be used for these many group test figures as training data, the method of employing support vector machine is set up the mapping between design parameter and the optimization aim, utilize this mapping to dope many group test figures again, utilize multi-group data original and that dope to simulate funtcional relationship between design parameter and the optimization aim, this function is used for calculation optimization, adopt the excellent design parameter of genetic algorithm, with the parameterized preliminary design scheme of design parameter substitution after optimizing, the cold-extruded part design proposal after being optimized;
The Parametric Drawing unit, be used for depositing drawing information in the product example storehouse;
The properties of product evaluation module, whether be used to estimate the scheme that obtains after the optimization adheres to specification, by judging that shaping load, metal flow, ess-strain distribution, forming defects and mould filling situation estimate the quality of cold-extrusion shaping spare, all can define corresponding criterion for evaluation to each evaluation object.
Further, in the described properties of product evaluation module, by the part behind the finite element software aftertreatment simulation forming, derive stress, deformation result to third party's post processor, this post processor is write according to the interpretational criteria of definition by the developer, by the complete overall performance of estimating product credibly of this program, defective as fruit product through this evaluation, then turn back to the optimizing phase again, iteration optimization finally obtains optimization solution again.
Further again, the inference machine of described inference machine is shaped on four classes: based on case-based reasoning, rule-based reasoning, based on support vector machine reasoning and genetic algorithm.
Further, described based on the support vector machine reasoning, be used for setting up finite element prediction of result model according to several groups of typical case's The results, by seeking structuring risk minimum, can realize minimizing of empiric risk and fiducial range; Finally transform into a quadratic form optimizing problem, what obtain will be global optimum's point, and practical problems is transformed into the feature space of higher-dimension by nonlinear transformation, the linear decision function of structure is realized the non-linear decision function in the former space in higher dimensional space.
Described genetic algorithm, be used to optimize the cold-extruded part design parameter, its coding rather than value with design parameter makes the calculation optimization process can use for reference chromosome and the Concept of Gene in the biology as operand, the heredity and the evolution mechanism of biology in natural imitation circle, as search information, use the search information of a plurality of search points with target function value simultaneously.
Beneficial effect of the present invention mainly shows: simulating human intelligent design cold-extruded part, the technical requirement reduction to the user, high efficiency, can guarantee the optimization of design proposal.
Description of drawings
Fig. 1 is the architecture of cold-extruded part Intelligentized design method.
Fig. 2 is based on the cold-extruded part Intelligentized design method workflow of integrated solution strategies.
Embodiment
Below in conjunction with accompanying drawing the present invention is further described.
See figures.1.and.2, a kind of machine intelligence participates in the cold-extruded part intelligent design system of design decision, and described intelligent design system comprises:
Module I---design mode expert system module. it comprises knowledge base, inference machine, dynamic data base and explanation facility.The knowledge and the experience of a large amount of cold-extrusion shaping domain expert levels contained in its inside, can utilize human expert's knowledge and the method for dealing with problems to handle this field question.This module is to utilize these knowledge and experiences, carries out reasoning and judgement, and simulating human expert's decision process is so that solve the challenge that those need the human expert to handle;
Knowledge base in the above-mentioned design mode expert system module has been gathered and has been used to design the domain knowledge of cold-extruded part and expert's practical experience in the Intelligentized design method.Comprise common cold extrusion product example storehouse, design rule storehouse, design parameter storehouse and numerical method storehouse.The product example storehouse is the work set of the medium-term and long-term cold-extruded part example of accumulate of designer, produces new object module when these examples can be used for based on case-based reasoning, simultaneously as original learning model so that call when needing in designing; The design rule storehouse is the main place that the knowledge and experience relevant with cold-extruded part stored, and knowledge and experience must be converted into the manageable sign format of computing machine so that describe and editor; Chemical constitution, the physical property of various cold extrusion die materials, the hot-working and the technical information such as heat treatment process parameter and relevant mechanical properties of recommendation are mainly being deposited in the design parameter storehouse; Numerical method stock is being put all kinds of numerical algorithm knowledge that Intelligentized design method can be used.User, expert and knowledge engineer can be by the knowledge in the system of knowledge base management system managerial knowledge storehouse, and the function of knowledge base management system is: the foundation of knowledge base with cancel, insertion and deletion, the modification of knowledge, the retrieval of knowledge and the confidentiality measure of search and knowledge etc. of knowledge.This method relates to the support vector machine knowledge relevant with genetic algorithm.Support vector machine is used for setting up finite element prediction of result model according to several groups of typical case's The results, compare with neural net method, the support vector machine characteristics are: specially at limited sample situation, its target is exactly to obtain the optimum solution under the existing information and be not only the optimal value of sample number when being tending towards infinity; Support vector machine can realize minimizing of empiric risk and fiducial range by seeking structuring risk minimum; Algorithm of support vector machine transforms into a quadratic form optimizing problem the most at last, and in theory, what obtain will be global optimum's point, solve unavoidable local extremum problem in neural net method; And support vector machine is transformed into practical problems the feature space of higher-dimension by nonlinear transformation, the linear decision function of structure is realized the non-linear decision function in the former space in higher dimensional space, it has solved problem of dimension cleverly, and its algorithm complex and sample dimension are irrelevant.Genetic algorithm is used to optimize the cold-extruded part design parameter, it makes the calculation optimization process can use for reference notions such as chromosome in the biology and gene with the coding of design parameter rather than value as operand, therefore mechanism such as biological heredity and evolution have its uniqueness in can natural imitation circle; Genetic algorithm as search information, does not need some other supplementary such as objective function derivative value just can determine the direction of search with target function value, and this makes genetic algorithm that higher adaptability be arranged; Genetic algorithm can use the search information of a plurality of search points can avoid search procedure to sink into locally optimal solution and stagnate simultaneously, and its hereditary capacity makes it possess the ability of search globally optimal solution.
Inference machine in the above-mentioned design mode expert system module is the control center of expert system, be responsible for the theory of artificial intelligence and method are applied to cold-stamped CAD field, thereby further liberate intellectual service work such as the reasoning of deviser in design, comprehensive, association, analogical learning.Import the performance requirement of product the user after, according to certain inference strategy knowledge base is searched for by inference mechanism, and the preliminary design scheme of output cold-extruded part, simultaneously as guiding the foundation of adjusting design and optimal design.In follow-up engineering analysis module, also need during prediction finite element model simulation result and optimal design parameter inference machine to utilize relevant knowledge experience participation work in the knowledge base.This method design inference machine is shaped on four classes: based on case-based reasoning, rule-based reasoning, based on the reasoning and the genetic algorithm of support vector machine.Knowledge-based inference problem to be solved is how in problem solving process, selects and utilization knowledge, finishes problem solving, and the knowledge in the knowledge base all will just can find application by inference machine.
The memory location of dynamic data when dynamic data base is this method work.Information in the design process comprises that initial designs requirement, intermediate result and final plan all are written to database, controls it by inference mechanism and refreshes.
The effect of explanation facility in expert system module is the problem that user or slip-stick artist are proposed, provide one clearly, completely, understandable answer, promptly rational explanation is made in its behavior and result.On the one hand, many users may not understand the knowledge of knowledge base, even do not trust, if module intelligence is given the conclusion of finding the solution that goes wrong, and necessary, the definite explanation of shortage, its conclusion is regarded as the prophesy of "black box" easily and is difficult to be accepted by the user so, and the procurement process that provides in the course of the work conclusion provides detailed explanation, can increase the acceptance of designer to this Intelligentized design method greatly; On the other hand, the knowledge of knowledge base the inside has certain limitation, the knowledge engineer can not be converted into sign format to each the bar knowledge in human expert's brain to entirely accurate and be stored in the knowledge base, this will cause Intelligentized design method to be runed counter in result and the actual conditions doing to release intellectual service work back, such situation appears, just need to determine wrong root, be beneficial to the knowledge engineer and find and revise defective and mistake in the knowledge base at any time, so, explanation to reasoning process is the main means in knowledge engineer's maintenance knowledge storehouse, also is that development and debugging bring great convenience simultaneously.Another big effect of explaining is to naive user and has just begun to utilize this Intelligentized design method to carry out the new hand's channeling conduct and the education of cold-extruded part design.Therefore, explanation facility has compensated the limitation of knowledge base.
Module ii---engineering analysis module, the function of engineering analysis module are to set up the design variable of preliminary design scheme and the funtcional relationship between the optimization aim, then preliminary design scheme are optimized, and are drawn out graphical information after the optimization.This module is made up of simulation analysis, calculation optimization and Parametric Drawing three parts.The effect of simulation analysis is to simulate several groups of representational test figures with finite element software before support vector machine is set up mapping between design parameter and the optimization aim, with these several groups of test figures as training data, the method of employing support vector machine is set up the mapping between design parameter and the optimization aim, utilize this mapping to dope many group test figures again, utilize multi-group data original and that dope to simulate funtcional relationship between design parameter and the optimization aim.This function is used for calculation optimization, uses genetic algorithm optimization here, the design parameter after can being optimized.With the parameterized preliminary design scheme of design parameter substitution after optimizing, the design proposal after can being optimized, Parametric Drawing and deposit drawing information in the product example storehouse simultaneously.
Whether module ii I---properties of product evaluation module is used for estimating optimization back model and adheres to specification.This module has comprised the whole appraisement system of cold-extrusion shaping design: the criterion, method and the specific targets that comprise evaluation.In the Intelligentized design method life's work, whole appraisement system is stored in the knowledge base with the form of knowledge.Performance evaluation is to estimate the quality of drip molding by situations such as judgement shaping load, metal flow, ess-strain distribution, forming defects and mould fillings, all can define corresponding criterion for evaluation to each evaluation object.Main method is by the part behind the finite element software aftertreatment simulation forming, derive stress, deformation result to third party's post processor, this post processor is write according to the interpretational criteria of definition by the developer, can the complete overall performance of estimating product credibly by this program.Suppose that product is defective through this evaluation, then need to turn back to again the optimizing phase, iteration optimization finally obtains optimization solution again.
As shown in Figure 1 the cold-extruded part Intelligentized design method is divided into three big modules: design mode expert system module, engineering analysis module and properties of product evaluation module.Both realize the intellectuality of cold-extruded part conceptual design by these three modules, simultaneously the simulation analysis result had been carried out automatic A+E.Wherein, the design mode expert system module is the core of method, and it comprises knowledge base, inference machine, dynamic data base and explanation facility.The function of this module is the design of initial scheme and provides design decision based on knowledge for whole Intelligentized design method.The present invention has preset the method for designing of two kinds of initial scheme, and knowledge-based inference method and human expert be design method (original model is made an amendment or brand-new design) voluntarily.Still do not have the cold-extruded part of much changes for model in some standardization, essential characteristic and the case library, can adopt the knowledge-based inference method; And for other part because to such an extent as to case library model limited amount and cold-extruded part changing features differ greatly and do not have enough knowledge for reasoning, preset human expert's design method voluntarily, to remedy the scarcity of knowledge base knowledge.The initial scheme that designs will be admitted to the engineering analysis module and make optimal design, and the part that designs is delivered to the performance evaluation module again and in conjunction with the fuzzy overall evaluation technology combination property of cold-extruded part design proposal etc. estimated.
These three modules are not only separate but also connect each other, the design mode expert system module provides the product design scheme of relevant performance parameters such as structure and function for engineering analysis, the module of engineering analysis simultaneously can be the expert system module service again, in order to the rationality and the reliability of check expert system institute output scheme.The properties of product module can be made further analysis and evaluation according to the result of engineering analysis to the design proposal of expert system module output, accumulate experience by evaluation procedure simultaneously, thereby further enrich the knowledge base in the expert system module, improve the intelligent level of Intelligentized design method.
The cold-extruded part Intelligentized design method workflow that present embodiment proposes comprises following content as shown in Figure 2:
1. generation preliminary design scheme.Adopt two kinds of diverse ways, if rationalistic method then adopts inference machine to infer preliminary design scheme according to the knowledge and experience of knowledge base; If the human expert is design method voluntarily, then can realize by revising the method that has model or design brand-new model;
2. determine design parameter.If the similar part of existing desire design cold-extruded part can be determined design parameter by the knowledge experience inference of knowledge base so in the knowledge base; If other part then need user's assistant analysis to determine design parameter, and this decision behavior of user can be learnt by expert system, and this experience is stored in the knowledge base, for later use;
3. set up the funtcional relationship between design parameter and the optimization aim.At first adopt the test design method contrived experiment, utilize finite element software emulation to obtain the training data of support vector machine, obtain the mapping between design parameter and the optimization aim behind the support vector machine training data, utilize the predictive ability of mapping to obtain many group FEM data predicted values, set up funtcional relationship then;
4. prioritization scheme.Optimized Algorithm adopts genetic algorithm.
5. optimization evaluation of result.Need make evaluation to the design of the metal plastic forming product of technological process formulation, the design of whole formation system and generalities from the angle of manufacturability.
Claims (5)
1. a machine intelligence participates in the cold-extruded part intelligent design system of design decision, it is characterized in that:
Described intelligent design system comprises:
The design mode expert system module, the knowledge and the experience of a large amount of cold-extrusion shaping domain expert levels contained in inside, the method of utilizing human expert's knowledge and dealing with problems is handled the cold-extrusion shaping field question, utilize described knowledge and experience, carry out reasoning and judgement, simulating human expert's decision process, described design mode expert system module comprises:
Knowledge base, set is used to design the domain knowledge of cold-extruded part and expert's practical experience, comprises product example storehouse, design rule storehouse, design parameter storehouse and numerical method storehouse; The product example storehouse is the work set of the medium-term and long-term example of accumulate of designer, produces new object module when being used for based on case-based reasoning, simultaneously as original learning model so that call when needing in designing; The design rule storehouse is the main place that the knowledge and experience relevant with cold-extruded part stored, and knowledge and experience is converted into the manageable sign format of computing machine so that describe and editor; Chemical constitution, the physical property of various cold extrusion die materials, hot-working and the heat treatment process parameter and the relevant mechanical properties technical information of recommendation are mainly being deposited in the design parameter storehouse; Numerical method stock is being put all kinds of numerical algorithm knowledge that Intelligentized design method can be used;
Inference machine is used for after the user imports the performance requirement of cold-extruded part, by inference mechanism is tactful by inference knowledge base is searched for;
Dynamic data base is used for the information in when work dynamic memory design process, comprises initial designs requirement, intermediate result and the final plan of cold-extruded part, controls it by inference mechanism and refreshes;
Explanation facility is used for problem that user or slip-stick artist are proposed, provides an answer; The engineering analysis module is used to set up the design variable of cold-extruded part preliminary design scheme and the funtcional relationship between the optimization aim, then preliminary design scheme is optimized, and is drawn out graphical information after the optimization, comprising:
The simulation analysis unit is drawn up the representational test figures of many groups with finite element software as tool mould, with as follow-up training sample when setting up Nonlinear Mapping;
The calculation optimization unit, be used for these many group test figures as training data, the method of employing support vector machine is set up the mapping between design parameter and the optimization aim, utilize this mapping to dope many group test figures again, utilize multi-group data original and that dope to simulate funtcional relationship between design parameter and the optimization aim, this function is used for calculation optimization, adopt the excellent design parameter of genetic algorithm, with the parameterized preliminary design scheme of design parameter substitution after optimizing, the cold-extruded part design proposal after being optimized;
The Parametric Drawing unit, be used for depositing drawing information in the product example storehouse;
The properties of product evaluation module, whether be used for estimating optimization back model adheres to specification, by judging that shaping load, metal flow, ess-strain distribution, forming defects and mould filling situation estimate the quality of drip molding, all can define corresponding criterion for evaluation to each evaluation object.
2. machine intelligence as claimed in claim 1 participates in the cold-extruded part intelligent design system of design decision, it is characterized in that: in the described properties of product evaluation module, by the part behind the finite element software aftertreatment simulation forming, derive stress, deformation result to third party's post processor, this post processor is write according to the interpretational criteria of definition by the developer, by the complete overall performance of estimating product credibly of this program, defective as fruit product through this evaluation, then turn back to the optimizing phase again, again iteration optimization finally obtains optimization solution.
3. machine intelligence as claimed in claim 1 or 2 participates in the cold-extruded part intelligent design system of design decision, and it is characterized in that: the inference machine of described inference machine is shaped on four classes: based on case-based reasoning, rule-based reasoning, based on support vector machine reasoning and genetic algorithm.
4. machine intelligence as claimed in claim 3 participates in the cold-extruded part intelligent design system of design decision, it is characterized in that: described based on the support vector machine reasoning, be used for setting up finite element prediction of result model according to several groups of typical case's The results, by seeking structuring risk minimum, can realize minimizing of empiric risk and fiducial range; Finally transform into a quadratic form optimizing problem, what obtain will be global optimum's point, and practical problems is transformed into the feature space of higher-dimension by nonlinear transformation, the linear decision function of structure is realized the non-linear decision function in the former space in higher dimensional space.
5. machine intelligence as claimed in claim 3 participates in the cold-extruded part intelligent design system of design decision, it is characterized in that: described genetic algorithm, be used to optimize the cold-extruded part design parameter, its coding rather than value with design parameter makes the calculation optimization process can use for reference chromosome and the Concept of Gene in the biology as operand, the heredity and the evolution mechanism of biology in natural imitation circle, as search information, use the search information of a plurality of search points with target function value simultaneously.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2011100024036A CN102063550A (en) | 2011-01-07 | 2011-01-07 | Intelligent design system of cold extrusion piece with machine intelligence involved design decision |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2011100024036A CN102063550A (en) | 2011-01-07 | 2011-01-07 | Intelligent design system of cold extrusion piece with machine intelligence involved design decision |
Publications (1)
Publication Number | Publication Date |
---|---|
CN102063550A true CN102063550A (en) | 2011-05-18 |
Family
ID=43998824
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN2011100024036A Pending CN102063550A (en) | 2011-01-07 | 2011-01-07 | Intelligent design system of cold extrusion piece with machine intelligence involved design decision |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102063550A (en) |
Cited By (18)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103218665A (en) * | 2011-10-26 | 2013-07-24 | 通用汽车环球科技运作有限责任公司 | Real-time detection and weld quality prediction in vibration welding process |
CN103810102A (en) * | 2014-02-19 | 2014-05-21 | 北京理工大学 | Method and system for predicting software defects |
CN106447046A (en) * | 2016-10-17 | 2017-02-22 | 深圳市彬讯科技有限公司 | House type design scheme evaluation method based on machine learning |
CN106944584A (en) * | 2017-03-21 | 2017-07-14 | 武汉理工大学 | Expert system and method that a kind of four tracks rotary forging press production technology is automatically generated |
CN107067078A (en) * | 2017-05-31 | 2017-08-18 | 江西制造职业技术学院 | A kind of metal corrosion test sample and test specimen automatic measure grading equipment and its ranking method |
CN108197702A (en) * | 2018-02-09 | 2018-06-22 | 艾凯克斯(嘉兴)信息科技有限公司 | A kind of method of the product design based on evaluation network and Recognition with Recurrent Neural Network |
CN108804759A (en) * | 2018-05-03 | 2018-11-13 | 上海大学 | A kind of engineering product design method of case-based reasioning |
CN110134063A (en) * | 2019-04-30 | 2019-08-16 | 济宁科力光电产业有限责任公司 | The automatic control device and method of workpiece shaping can be achieved |
CN110135081A (en) * | 2019-05-20 | 2019-08-16 | 哈尔滨工程大学 | A kind of internal combustion engine tribology Expert System Design method based on artificial intelligence |
CN111651842A (en) * | 2020-06-03 | 2020-09-11 | 中国农业机械化科学研究院 | Intelligent design system and method for hob type chopper |
CN111805958A (en) * | 2020-07-13 | 2020-10-23 | 武汉轻工大学 | Parameter optimization method and system of spiral oil press |
CN112100749A (en) * | 2020-09-29 | 2020-12-18 | 上海卓位信息技术有限公司 | Automobile intelligent development system and construction and application method thereof |
CN113377962A (en) * | 2021-06-18 | 2021-09-10 | 西安石油大学 | Intelligent process simulation method based on image recognition and natural language processing |
CN113591205A (en) * | 2021-07-08 | 2021-11-02 | 武汉理工大学 | Intelligent design method and system for automobile structural part |
CN113780712A (en) * | 2021-04-30 | 2021-12-10 | 中国人民解放军国防科技大学 | Joint combat equipment system application evaluation method and system and computer equipment |
CN115310232A (en) * | 2022-08-17 | 2022-11-08 | 东华大学 | Fan design BOM simulation system and simulation equipment |
CN115841052A (en) * | 2023-02-23 | 2023-03-24 | 江苏甬金金属科技有限公司 | Stainless steel precision extrusion process optimization method and system based on numerical simulation |
CN116415662A (en) * | 2023-06-12 | 2023-07-11 | 四川云申至诚科技有限公司 | Factory expert system based on knowledge discovery |
-
2011
- 2011-01-07 CN CN2011100024036A patent/CN102063550A/en active Pending
Cited By (24)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103218665A (en) * | 2011-10-26 | 2013-07-24 | 通用汽车环球科技运作有限责任公司 | Real-time detection and weld quality prediction in vibration welding process |
CN103218665B (en) * | 2011-10-26 | 2016-03-16 | 通用汽车环球科技运作有限责任公司 | Can the binary classification of item of interest in repetitive process |
CN103810102A (en) * | 2014-02-19 | 2014-05-21 | 北京理工大学 | Method and system for predicting software defects |
CN106447046A (en) * | 2016-10-17 | 2017-02-22 | 深圳市彬讯科技有限公司 | House type design scheme evaluation method based on machine learning |
CN106447046B (en) * | 2016-10-17 | 2019-04-19 | 深圳市彬讯科技有限公司 | A kind of Size Dwelling Design scheme evaluating method based on machine learning |
CN106944584A (en) * | 2017-03-21 | 2017-07-14 | 武汉理工大学 | Expert system and method that a kind of four tracks rotary forging press production technology is automatically generated |
CN107067078A (en) * | 2017-05-31 | 2017-08-18 | 江西制造职业技术学院 | A kind of metal corrosion test sample and test specimen automatic measure grading equipment and its ranking method |
CN108197702A (en) * | 2018-02-09 | 2018-06-22 | 艾凯克斯(嘉兴)信息科技有限公司 | A kind of method of the product design based on evaluation network and Recognition with Recurrent Neural Network |
CN108804759A (en) * | 2018-05-03 | 2018-11-13 | 上海大学 | A kind of engineering product design method of case-based reasioning |
CN110134063A (en) * | 2019-04-30 | 2019-08-16 | 济宁科力光电产业有限责任公司 | The automatic control device and method of workpiece shaping can be achieved |
CN110135081A (en) * | 2019-05-20 | 2019-08-16 | 哈尔滨工程大学 | A kind of internal combustion engine tribology Expert System Design method based on artificial intelligence |
CN111651842A (en) * | 2020-06-03 | 2020-09-11 | 中国农业机械化科学研究院 | Intelligent design system and method for hob type chopper |
CN111805958A (en) * | 2020-07-13 | 2020-10-23 | 武汉轻工大学 | Parameter optimization method and system of spiral oil press |
CN111805958B (en) * | 2020-07-13 | 2022-06-14 | 武汉轻工大学 | Parameter optimization method and system of spiral oil press |
CN112100749A (en) * | 2020-09-29 | 2020-12-18 | 上海卓位信息技术有限公司 | Automobile intelligent development system and construction and application method thereof |
CN113780712A (en) * | 2021-04-30 | 2021-12-10 | 中国人民解放军国防科技大学 | Joint combat equipment system application evaluation method and system and computer equipment |
CN113377962A (en) * | 2021-06-18 | 2021-09-10 | 西安石油大学 | Intelligent process simulation method based on image recognition and natural language processing |
CN113377962B (en) * | 2021-06-18 | 2023-04-07 | 西安石油大学 | Intelligent process simulation method based on image recognition and natural language processing |
CN113591205A (en) * | 2021-07-08 | 2021-11-02 | 武汉理工大学 | Intelligent design method and system for automobile structural part |
CN113591205B (en) * | 2021-07-08 | 2023-10-17 | 武汉理工大学 | Intelligent design method and system for automobile structural part |
CN115310232A (en) * | 2022-08-17 | 2022-11-08 | 东华大学 | Fan design BOM simulation system and simulation equipment |
CN115841052A (en) * | 2023-02-23 | 2023-03-24 | 江苏甬金金属科技有限公司 | Stainless steel precision extrusion process optimization method and system based on numerical simulation |
CN116415662A (en) * | 2023-06-12 | 2023-07-11 | 四川云申至诚科技有限公司 | Factory expert system based on knowledge discovery |
CN116415662B (en) * | 2023-06-12 | 2023-08-11 | 四川云申至诚科技有限公司 | Factory expert system based on knowledge discovery |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102063550A (en) | Intelligent design system of cold extrusion piece with machine intelligence involved design decision | |
Zhang et al. | Building a right digital twin with model engineering | |
Xu et al. | A decision support system for product design in concurrent engineering | |
Hong et al. | Probabilistic electric load forecasting: A tutorial review | |
Golmohammadi | Neural network application for fuzzy multi-criteria decision making problems | |
Suryani et al. | Demand scenario analysis and planned capacity expansion: A system dynamics framework | |
CN111695731B (en) | Load prediction method, system and equipment based on multi-source data and hybrid neural network | |
Hu et al. | New CBR adaptation method combining with problem–solution relational analysis for mechanical design | |
CN105893669A (en) | Global simulation performance predication method based on data digging | |
CN115422814B (en) | Digital twin-driven closed-loop optimization design method for complex electromechanical product | |
Ghoshchi et al. | Machine learning theory in building energy modeling and optimization: a bibliometric analysis | |
CN114548494B (en) | Visual cost data prediction intelligent analysis system | |
CN113449919A (en) | Power consumption prediction method and system based on feature and trend perception | |
Liu et al. | Using fuzzy neural network approach to estimate contractors’ markup | |
Cao et al. | Simulation optimization for a digital twin using a multi-fidelity framework | |
CN201716727U (en) | Geographical simulation system based on remote sensing and GIS | |
Manoj et al. | FWS-DL: forecasting wind speed based on deep learning algorithms | |
CN107491841A (en) | Nonlinear optimization method and storage medium | |
Yan et al. | Application of the quality function deployment method in the mechanical structure design of subsea power devices | |
Preis et al. | Online hydraulic state prediction for water distribution systems | |
Qian et al. | Towards discrete manufacturing workshop-oriented digital twin model: Modeling, verification and evolution | |
Schönmann et al. | Anticipative strategic production technology planning considering cyclic interactions | |
Mele et al. | A new method for the design of knowledge-based engineering systems for manufacturing | |
CN114154768A (en) | Natural gas demand prediction method, device, electronic equipment and medium | |
Kamire et al. | Real Property Value Prediction Capability Using Fuzzy Logic and ANFIS |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20110518 |