CN118024445B - Modification optimization method and system for blending type interpenetrating network thermoplastic elastomer - Google Patents
Modification optimization method and system for blending type interpenetrating network thermoplastic elastomer Download PDFInfo
- Publication number
- CN118024445B CN118024445B CN202410436305.0A CN202410436305A CN118024445B CN 118024445 B CN118024445 B CN 118024445B CN 202410436305 A CN202410436305 A CN 202410436305A CN 118024445 B CN118024445 B CN 118024445B
- Authority
- CN
- China
- Prior art keywords
- modification
- variation
- result
- optimal solution
- solution space
- 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.)
- Active
Links
- 238000012986 modification Methods 0.000 title claims abstract description 190
- 230000004048 modification Effects 0.000 title claims abstract description 190
- 238000005457 optimization Methods 0.000 title claims abstract description 65
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000002156 mixing Methods 0.000 title claims abstract description 37
- 229920002725 thermoplastic elastomer Polymers 0.000 title claims abstract description 37
- 238000002474 experimental method Methods 0.000 claims description 90
- 238000011156 evaluation Methods 0.000 claims description 45
- 230000004913 activation Effects 0.000 claims description 32
- 230000035772 mutation Effects 0.000 claims description 24
- 230000000051 modifying effect Effects 0.000 claims description 16
- 238000004458 analytical method Methods 0.000 claims description 14
- 230000006978 adaptation Effects 0.000 claims description 9
- 238000004519 manufacturing process Methods 0.000 claims description 9
- 230000001360 synchronised effect Effects 0.000 claims description 7
- 230000003044 adaptive effect Effects 0.000 claims description 2
- 230000000694 effects Effects 0.000 abstract description 20
- 238000012545 processing Methods 0.000 abstract description 15
- 229920001971 elastomer Polymers 0.000 abstract description 10
- 230000009467 reduction Effects 0.000 abstract description 8
- 238000005516 engineering process Methods 0.000 abstract description 7
- 239000011159 matrix material Substances 0.000 abstract description 6
- 229920005989 resin Polymers 0.000 abstract description 6
- 239000011347 resin Substances 0.000 abstract description 6
- 238000013461 design Methods 0.000 abstract description 5
- 238000002715 modification method Methods 0.000 abstract description 4
- 230000006870 function Effects 0.000 description 59
- 230000008569 process Effects 0.000 description 25
- 230000006872 improvement Effects 0.000 description 10
- 239000000463 material Substances 0.000 description 10
- 238000004422 calculation algorithm Methods 0.000 description 9
- 230000008901 benefit Effects 0.000 description 5
- 238000009826 distribution Methods 0.000 description 4
- 239000002861 polymer material Substances 0.000 description 4
- 239000000806 elastomer Substances 0.000 description 3
- 239000000203 mixture Substances 0.000 description 3
- 238000007619 statistical method Methods 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 239000000654 additive Substances 0.000 description 2
- 238000013459 approach Methods 0.000 description 2
- 230000003631 expected effect Effects 0.000 description 2
- 239000000945 filler Substances 0.000 description 2
- 238000012216 screening Methods 0.000 description 2
- JBRZTFJDHDCESZ-UHFFFAOYSA-N AsGa Chemical compound [As]#[Ga] JBRZTFJDHDCESZ-UHFFFAOYSA-N 0.000 description 1
- 229910001218 Gallium arsenide Inorganic materials 0.000 description 1
- 239000011157 advanced composite material Substances 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000000919 ceramic Substances 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000003795 chemical substances by application Substances 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000010219 correlation analysis Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013479 data entry Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 239000012760 heat stabilizer Substances 0.000 description 1
- 239000011256 inorganic filler Substances 0.000 description 1
- 229910003475 inorganic filler Inorganic materials 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 239000007769 metal material Substances 0.000 description 1
- 239000003607 modifier Substances 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 238000010206 sensitivity analysis Methods 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 238000002922 simulated annealing Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 239000012745 toughening agent Substances 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
- 238000004073 vulcanization Methods 0.000 description 1
- 239000004636 vulcanized rubber Substances 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a modification optimization method and a modification optimization system for a blending type interpenetrating network thermoplastic elastomer, which relate to the field of data processing, and comprise the following steps: and establishing a modification requirement, wherein the modification requirement comprises a modification optimization direction, a modification value and an expected cost, carrying out modification fitting on the blending type interpenetrating network thermoplastic elastomer according to the modification requirement, and completing modification optimization according to a modification fitting result. The technical problems of performance reduction at an interface and cost effectiveness reduction caused by the fact that the compatibility between matrix resin and a rubber phase of the blending type interpenetrating network thermoplastic elastomer in the prior art is generally poor are solved, modification optimization of the blending type interpenetrating network thermoplastic elastomer is achieved, and therefore a blending structure with more reasonable processing technology and design is improved, and the technical effect of a modification method with higher cost effectiveness is achieved.
Description
Technical Field
The application relates to the field of data processing, in particular to a modification optimization method and system of a blending type interpenetrating network thermoplastic elastomer.
Background
Along with the development of scientific technology, particularly on the basis of traditional materials, new materials are developed according to the research results of modern technology, but the new materials can be divided into four major categories of high polymer materials, advanced composite materials, inorganic nonmetallic materials (such as ceramics, gallium arsenide semiconductors and the like) and metal materials according to the materials, the high polymer materials can be divided into natural high polymer materials and synthetic high polymer materials, wherein elastomer blending modification is a commercializable and frequently applied method, and the method can lead the compounded and vulcanized rubber to finally obtain the comprehensive performance which cannot be achieved by a single elastomer, thereby meeting the commercial demands. Furthermore, by adjusting the proportions of the components, the properties of the blend can be continuously varied over a range. Thus, the potential applications of the blends are significantly greater than for the single novel elastomer, whereas the prior art blended interpenetrating network thermoplastic elastomers have technical problems of reduced performance at the interface and reduced cost effectiveness due to the generally poor compatibility between the matrix resin and the rubber phase.
Disclosure of Invention
The application solves the technical problems of performance reduction at interfaces and cost effectiveness reduction caused by the fact that the compatibility between matrix resin and rubber phase of the blending type interpenetrating network thermoplastic elastomer is generally poor in the prior art by providing the modifying and optimizing method and the modifying and optimizing system for the blending type interpenetrating network thermoplastic elastomer, and realizes the technical effects of improving the processing technology and designing a more reasonable blending structure and obtaining a modifying method with higher cost effectiveness.
The application provides a modification optimization method of a blending type interpenetrating network thermoplastic elastomer, which is applied to a modification optimization system of the blending type interpenetrating network thermoplastic elastomer, and comprises the following steps: establishing a modification requirement, wherein the modification requirement comprises a modification optimization direction, a modification value and an expected cost, and performing modification fitting of the blending type interpenetrating network thermoplastic elastomer by using the modification requirement, and specifically comprises the following steps:
S1: acquiring a historical experiment database, wherein the historical experiment database is constructed by collecting modified experiment data and historical production data, each data in the historical experiment database corresponds to authentication data, and the authentication data characterizes the trust degree of the data;
s2: the historical experiment database is cleaned according to the authentication data, modification parameters are established through the cleaned historical experiment database, and an optimal solution space is configured;
s3: establishing an fitness function according to the modification requirement, wherein the fitness function is an equilibrium evaluation function for modifying a result and modifying cost, and optimizing and searching is carried out in the optimal solution space through the fitness function;
S4: establishing an activation function, and judging a variation direction after one round of search is finished by using the activation function, wherein the variation direction comprises variation in a best solution space and variation in a historical experiment database;
s5: performing modification parameter variation according to the discrimination result, updating the optimal solution space according to the variation result, and continuously iterating to finish modification fitting;
and finishing modification optimization according to the modification fitting result.
In a possible implementation, the following process is performed: performing value degree analysis based on a historical experiment database on the optimal solution space, and generating a first probability constraint according to an analysis result;
configuring a calibration probability constraint through big data, and taking the calibration probability constraint as a second probability constraint;
Setting a value threshold of an optimal solution space according to the first probability constraint and the second probability constraint;
And completing the judgment of the variation direction according to the activation function and the value threshold.
In a possible implementation manner, the determining of the variation direction is completed according to the activation function and the value threshold, and the following processing is performed: generating a random number through the activation function, and judging whether the random number is within the value threshold range;
If the variation instruction is within the value threshold range, generating a variation instruction in the optimal solution space;
Performing fine tuning variation on parameters in the optimal solution space by using the variation instruction in the optimal solution space;
If the random variation instruction is not in the value threshold range, generating a random variation instruction of the historical experiment database;
And carrying out parameter random variation of the historical experiment database by using the random variation instruction of the historical experiment database.
In a possible implementation manner, the modification parameter variation is performed according to the discrimination result, and the following processing is performed: if the judging result is that the parameter selects the fine tuning variation, establishing random parameter probability through the important duty ratio of the modified parameter in the optimal solution space;
randomly selecting variation parameters through the random parameter probability, executing the preset step variation of the selected parameters, and adding the selected parameters to a tabu list;
And evaluating the variation result of the selected parameter based on the fitness function, generating forbidden disturbance according to the evaluation result, comparing the evaluation result with the last fitness evaluation result in the optimal solution space, updating the optimal solution space according to the comparison result, and finishing continuous iteration through the forbidden list.
In a possible implementation, the following process is performed: performing interval parameter homodromous variation constraint of modification parameters based on the historical experiment database;
And when the modification parameter is mutated, if the mutation period is in the mutation early stage, synchronously triggering the modification parameter through the homodromous mutation constraint, and completing the modification parameter mutation according to a synchronous triggering result.
In a possible implementation, the following process is performed: executing iteration records, and constructing a first iteration result of the optimal solution space and a second iteration result of the historical experiment database;
performing variation adaptation evaluation of an optimal solution space and a historical experiment database based on the first iteration result and the second iteration result;
and carrying out self-adaptive updating on the value threshold according to the variation adaptation evaluation result.
In a possible implementation, the following process is performed: establishing a modified solution set according to a modified fitting result, and performing steady-state evaluation on the modified solution set;
And (3) carrying out comprehensive sequence sorting according to the steady-state evaluation result and the modification result, and completing modification optimization based on the selected result of the comprehensive sequence sorting.
The application also provides a modification and optimization system of the blending type interpenetrating network thermoplastic elastomer, which comprises the following steps: the modification fitting module is used for establishing modification requirements, the modification requirements comprise modification optimization directions, modification values and expected cost, and the modification fitting module is used for carrying out modification fitting on the blending type interpenetrating network thermoplastic elastomer according to the modification requirements, and specifically comprises the following steps:
S1: acquiring a historical experiment database, wherein the historical experiment database is constructed by collecting modified experiment data and historical production data, each data in the historical experiment database corresponds to authentication data, and the authentication data characterizes the trust degree of the data;
s2: the historical experiment database is cleaned according to the authentication data, modification parameters are established through the cleaned historical experiment database, and an optimal solution space is configured;
s3: establishing an fitness function according to the modification requirement, wherein the fitness function is an equilibrium evaluation function for modifying a result and modifying cost, and optimizing and searching is carried out in the optimal solution space through the fitness function;
S4: establishing an activation function, and judging a variation direction after one round of search is finished by using the activation function, wherein the variation direction comprises variation in a best solution space and variation in a historical experiment database;
s5: performing modification parameter variation according to the discrimination result, updating the optimal solution space according to the variation result, and continuously iterating to finish modification fitting;
and the modification optimization module is used for completing modification optimization according to the modification fitting result.
One or more technical schemes provided by the application have at least the following technical effects or advantages:
The application provides a modification optimization method and a modification optimization system for a blending type interpenetrating network thermoplastic elastomer, relates to the technical field of data processing, solves the technical problems of performance reduction at an interface and cost effectiveness reduction caused by poor compatibility between matrix resin and a rubber phase of the blending type interpenetrating network thermoplastic elastomer in the prior art, realizes modification optimization of the blending type interpenetrating network thermoplastic elastomer, improves a processing technology, designs a more reasonable blending structure, and achieves the technical effect of obtaining a modification method with higher cost effectiveness.
Drawings
In order to more clearly illustrate the technical solution of the embodiments of the present application, the following description will briefly refer to the accompanying drawings of the embodiments of the present application, in which flowcharts are used to illustrate operations performed by systems according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in order precisely. Rather, the various steps may be processed in reverse order or simultaneously, as desired. Also, other operations may be added to or removed from these processes.
FIG. 1 is a schematic flow chart of a modification and optimization method of a blend-type interpenetrating network thermoplastic elastomer provided by the embodiment of the application;
fig. 2 is a schematic structural diagram of a modification and optimization system of a blend-type interpenetrating network thermoplastic elastomer according to an embodiment of the application.
Detailed Description
The foregoing description is only an overview of the present application, and is intended to be implemented in accordance with the teachings of the present application in order that the same may be more clearly understood and to make the same and other objects, features and advantages of the present application more readily apparent.
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be described in further detail with reference to the accompanying drawings, and the described embodiments should not be construed as limiting the present application, and all other embodiments obtained by those skilled in the art without making any inventive effort are within the scope of the present application.
In the following description, reference is made to "some embodiments" which describe a subset of all possible embodiments, but it is to be understood that "some embodiments" may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict, the term "first\second" being referred to merely as distinguishing between similar objects and not representing a particular ordering for the objects. The terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or server that comprises a list of steps or elements is not necessarily limited to those steps or elements that are expressly listed or inherent to such process, method, article, or apparatus, but may include other steps or modules that may not be expressly listed or inherent to such process, method, article, or apparatus, and unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. The terminology used herein is for the purpose of describing embodiments of the application only.
The embodiment of the application provides a modification and optimization method of a blending type interpenetrating network thermoplastic elastomer, which is applied to a modification and optimization system of the blending type interpenetrating network thermoplastic elastomer, as shown in fig. 1, and comprises the following steps:
step A100, establishing modification requirements, wherein the modification requirements comprise modification optimization directions, modification values and expected cost, and performing modification fitting of the blending type interpenetrating network thermoplastic elastomer by using the modification requirements, and specifically comprises the following steps:
Firstly, based on the specific performance improvement point required by the product, the direction of modification optimization is determined, and the modification optimization direction contained in the established modification requirement can comprise the improvement of the tensile strength and the elongation at break of the TPV so as to enhance the mechanical property of the TPV; enhancing the heat resistance of the TPV to expand its application in high temperature environments; the processability of the TPV is improved, and the production efficiency is improved; illustratively, the corresponding modification values may be a modification value of at least 20% improvement in tensile strength, at least 15% improvement in elongation at break, at least 10 ℃ improvement in heat resistance (such as vicat softening point), 5 ℃ expansion in processing temperature range, etc., while the expected cost may be set to be not more than 20% of the total cost of the modifier, filler and other additives, and further, the blended interpenetrating network thermoplastic elastomer is modified and fitted by the following method, and illustratively, it is considered that a specific rubber toughening agent is added to improve tensile strength and elongation at break, while a heat-resistant filler such as a heat stabilizer or an inorganic filler is added to enhance heat resistance. In addition, the processing technique is optimized, such as mixing time, vulcanization temperature, pressure, etc., to improve the processing properties.
S1: acquiring a historical experiment database, wherein the historical experiment database is constructed by collecting modified experiment data and historical production data, each data in the historical experiment database corresponds to authentication data, and the authentication data characterizes the trust degree of the data;
The historical experiment database contains all relevant modified experiment data and historical production data, the modified experiment data and the historical production data are used for covering different modification schemes, technological parameters, material formulas and the like, so that comprehensive information support is provided, in order to avoid the conclusion that errors are caused by data errors, each data in the historical experiment database is correspondingly provided with authentication data, the accuracy of the data can be ensured by comparing data from different sources, the reasonable range of the authentication data and the like, the authentication data is an important basis for evaluating the trust degree of the data, and data with high trust degree can be preferentially selected for analysis and fitting when the data in the historical experiment database are utilized.
S2: the historical experiment database is cleaned according to the authentication data, modification parameters are established through the cleaned historical experiment database, and an optimal solution space is configured;
Firstly, setting a trust threshold based on the trust degree of historical authentication data, screening data with higher trust degree of the authentication data, only reserving data exceeding the trust degree threshold, deleting repeated data entries in a historical experiment database, thereby ensuring that each data point in the database is unique, filling or interpolating the deleted data in the historical experiment database according to data under other similar conditions, identifying abnormal values in the historical experiment database by using a statistical method or domain knowledge, deleting, replacing or marking the abnormal values as suspicious values, and the like, and finally carrying out standardized or normalized processing on the data according to the requirements to eliminate dimension influence, improve the accuracy of data analysis, and thus finishing the data cleaning of the historical experiment database.
Further, the washed historical experiment database is subjected to statistical analysis to identify modification parameters (such as types and amounts of additives, processing parameters and the like) closely related to the material performance, and then a reasonable parameter range or space is determined based on the established modification parameters so as to search and explore in the subsequent modification optimization process, and the optimal solution space is configured by factors such as process feasibility, cost limitation, expected performance improvement and the like.
S3: establishing an fitness function according to the modification requirement, wherein the fitness function is an equilibrium evaluation function for modifying a result and modifying cost, and optimizing and searching is carried out in the optimal solution space through the fitness function; in the modification and optimization process, establishing the fitness function is a crucial step, the fitness function can be used for evaluating the effect of the modification scheme and carrying out balanced evaluation according to modification requirements and expected cost, and meanwhile, the fitness function combines modification results (such as improvement of material performance) with modification cost so as to quantify the comprehensive benefits of different modification schemes.
Firstly, defining a specific form of a fitness function according to modification requirements, wherein the fitness function is applied to reflect the trade-off relation between modification results and modification cost, and the fitness function can be defined as follows:；
Wherein, To adapt,/>For modifying effect,/>For the expected modification effect,/>For modification cost,/>Is the expected cost.
The modification effect refers to the improvement degree of the material performance after actual modification, and the expected modification effect is the performance improvement target set in the modification requirement. The modification cost comprises the total cost required by modifying agent, filling material, process adjustment and the like, the expected cost is the maximum cost which we are willing to pay for modification, the optimization search is further carried out in an optimization solution space through a fitness function, the optimization solution space is a reasonable range configured according to modification parameters, the optimization solution space defines the boundary and conditions of the search, in the search process, the fitness value of each scheme can be calculated according to different modification schemes (namely different parameter combinations), and further various algorithms and methods such as a genetic algorithm, a particle swarm optimization algorithm, a simulated annealing algorithm and the like can be adopted for optimizing search, the optimization solution space is searched for by iteration on the basis, the modification scheme which maximizes the fitness function value is searched for, meanwhile, in each iteration, the algorithm can evaluate and select the modification scheme according to the fitness value, and the optimization solution is gradually approximated, so that the optimization search is completed, and the modification scheme with the best economic benefit and practical value can be obtained.
S4: establishing an activation function, and judging a variation direction after one round of search is finished by using the activation function, wherein the variation direction comprises variation in a best solution space and variation in a historical experiment database; in a possible implementation manner, the step a100 further includes a step a110 of performing a value degree analysis based on a historical experiment database on the optimal solution space, and generating a first probability constraint according to an analysis result; step A120 is executed, and a calibration probability constraint is configured through big data and is used as a second probability constraint;
The direction of variation in the search process can be determined by introducing an activation function, which is commonly used in machine learning and decision processes in neural networks that can control the search, so that the activation function is applied to the direction discrimination of variation in modification optimization, the activation function is first defined, the activation function can be a nonlinear function, and the output of the activation function generally represents the probability or weight of a certain operation or decision.
Specifically, in this embodiment, the input of the activation function may be a certain evaluation index or score of the variation in the optimal solution space and the variation in the historical experiment database, and the output is a probability value between 0 and 1, which indicates the possibility of selecting the variation direction, and illustratively, when the output of the activation function approaches 1, the variation in the optimal solution space is selected, and if the output of the activation function approaches 0, the variation in the historical experiment database is selected, further, the optimal solution space is subjected to a value degree analysis based on the historical experiment database, the performance of each parameter combination in the optimal solution space in the historical experiment database is evaluated by analyzing the data in the historical experiment database, the different values given to the parameters in the optimal solution space are determined based on the good performance of the different parameter combinations in the historical time period, the first probability constraint is determined according to the value degree analysis result, the first probability constraint is used for reflecting the probability that each parameter combination is selected in the optimal solution space, and then the first probability constraint is used as the second probability constraint to output, which means that the optimal solution space can be calibrated by using a statistical method, a machine or a learning algorithm, the optimal solution constraint is used for calibrating and the optimal solution space can be well calibrated, and the optimal solution space can be well established by combining with the optimal solution constraint and the optimal solution constraint, and the optimal solution is calibrated, and the optimal solution space is well-defining the optimal solution space has a certain performance with a certain performance.
Step A130 is executed, and a value threshold of an optimal solution space is set according to the first probability constraint and the second probability constraint;
Firstly, determining a value threshold of an optimal solution space according to a first probability constraint and a second probability constraint, wherein the value threshold can be a range dynamically adjusted according to the probability constraint, namely, converting the probability constraint into a probability distribution, and determining the probability of a parameter combination exceeding the set value threshold according to the probability distribution so as to take the probability as an optimal solution, and in one possible implementation manner, the step A130 further comprises the step A131 of executing iterative recording to construct a first iterative result of the optimal solution space and a second iterative result of a historical experiment database;
the process of performing iterative recording is: in each iteration, different parameter combinations exist, and the fitness value corresponding to the different parameter combinations is calculated. Meanwhile, the values of different parameter combinations in the optimal solution space and the performance performances of the parameter combinations in the historical experiment database are required to be recorded, and further, when a first iteration result of the optimal solution space is constructed, the effect of each parameter combination can be evaluated according to the result of the first round of search and the fitness function. And screening out the optimal solution meeting the conditions by combining the first probability constraint and the second probability constraint, updating the value threshold of the optimal solution space, and gradually converging the optimal solution space to the vicinity of the optimal solution along with continuous iteration.
Meanwhile, a second iteration result of the historical experiment database is required to be constructed, namely, a new round of search result is required to be added into the historical experiment database, corresponding data are synchronously updated, and the historical experiment database can continuously accumulate new data and experience on the basis, so that richer and more accurate information support is provided for subsequent iteration and optimization.
Step A132 is executed, and the variation adaptation evaluation of the optimal solution space and the historical experiment database is carried out based on the first iteration result and the second iteration result; step A133 is executed to perform adaptive updating of the value threshold according to the variation adaptation evaluation result.
Because the optimal solution space has larger variation probability, the optimal solution space has better effect, if the optimal effect and the expected effect are not matched, the historical experiment database has better performance effect, so that the threshold value can be adjusted in a self-adaptive manner, if the optimal effect and the expected effect are matched, the threshold value can be increased continuously, the probability of the threshold value is higher, the variation adaptation evaluation of the optimal solution space and the historical experiment database is completed, the value threshold value can be updated in a self-adaptive manner according to the variation evaluation result, and the updating time length and the updating time period can be set, so that more reliable and efficient technical support is provided for actual production.
And executing step A140, and completing the judgment of the variation direction according to the activation function and the value threshold.
In a possible implementation manner, step a140 further includes step a141 of generating a random number by the activation function, and determining whether the random number is within the range of the value threshold; executing step A142, if the variation instruction is within the value threshold range, generating a variation instruction in the optimal solution space; executing step A143, and performing fine tuning variation in the parameter selection in the optimal solution space by using the variation instruction in the optimal solution space; executing step A144, if the random variation instruction is not in the value threshold range, generating a random variation instruction of the historical experiment database; and executing step A145, and carrying out parameter random mutation of the historical experiment database by using the random mutation instruction of the historical experiment database.
The step of completing the judgment of the variation direction of the modification parameters by using the activation function and the value threshold value means that firstly, the generation of random numbers is carried out based on the activation function, and meanwhile, whether the generated random numbers are in the value threshold value range is judged, namely, the random numbers are compared with the value threshold value set according to the first probability constraint and the second probability constraint, if the random numbers are in the threshold value range (namely, the random numbers are smaller than or equal to the threshold value), the variation in the optimal solution space is selected, and if the random numbers are in the value threshold value range, the search process is considered to tend to search for a better optimal solution in the current optimal solution space. At this time, a variation instruction in the optimal solution space is generated, the variation instruction guides an algorithm to perform fine tuning variation in the optimal solution space, any parameter or parameter combination in the optimal solution space is selected according to the variation instruction in the optimal solution space to perform fine tuning variation, and meanwhile, the amplitude and the mode of the fine tuning variation can be designed according to a specific optimization algorithm and modification requirements and are used for keeping searching for a better solution in an adjacent area of the current solution.
If the random number exceeds the value threshold range, more global explorations are needed to be introduced in the searching process, random variation instructions of the historical experiment database are generated, one or more parameters in the historical experiment database are selected randomly according to the random variation instructions of the historical experiment database to perform variation, and the variation usually involves larger parameter adjustment so as to expand the searching range and avoid falling into a local optimal solution, and then the flexibility and the efficiency of the modification and optimization process are improved.
S5: performing modification parameter variation according to the discrimination result, updating the optimal solution space according to the variation result, and continuously iterating to finish modification fitting; in one possible implementation manner, the step a100 further includes a step a150 of establishing a random parameter probability by an important duty ratio of the modification parameter in the optimal solution space if the discrimination result is that the parameter selects the fine tuning variation;
When the judging result is that the parameters are selected to be finely adjusted and mutated, the random parameter probability is established according to the important duty ratio of the modified parameters in the optimal solution space, the influence degree of each parameter on the whole modification effect is firstly evaluated, the mutation probability is distributed according to the influence degrees, firstly, the importance degree of each parameter in the optimal solution space is evaluated through methods such as analysis of historical experimental data, sensitivity analysis and correlation analysis among the parameters, so that the important duty ratio of the modified parameters in the optimal solution space is completed, and the important duty ratio of each parameter is further directly used as the probability of the modified parameters in random selection, so that the establishment of the random parameter probability is completed.
Executing step A160, namely randomly selecting variation parameters according to the random parameter probability, executing the preset step variation of the selected parameters, and adding the selected parameters to a tabu list; and executing the step A170, evaluating the variation result of the selected parameter based on the fitness function, generating forbidden disturbance according to the evaluation result, comparing the evaluation result with the last fitness evaluation result in the optimal solution space, updating the optimal solution space according to the comparison result, and finishing continuous iteration through the forbidden list.
Firstly, the parameters which need to be mutated can be set as probability distribution through factors such as value analysis, first probability constraint and second probability constraint of a historical experiment database based on probability output of an activation function, parameters which need to be mutated are randomly selected based on the probability distribution, mutation operation is carried out on the selected parameters by adding or subtracting a preset step length on the basis of the current value of the parameters, the mutated parameters are added into a tabu list to ensure that the parameters cannot be repeatedly selected in the following iteration, the performance of the mutated parameter combination is further evaluated by using a fitness function, the fitness function is used for reflecting the balance of a modification effect and a modification cost, so that different parameter combinations are subjected to fair comparison, and forbidden disturbance is generated according to the evaluation result of the fitness function, if the quality of knowledge is remarkably improved due to mutation of any one parameter, the parameters can be removed (forbidden) from the tabu list, and the purpose of skipping out local optimal solution is achieved by allowing the parameters to be selected again in the future iteration.
Comparing the adaptability evaluation result of the current variation parameter with the adaptability of the last position (namely the lowest adaptability) parameter in the optimal solution space, if the adaptability of the new parameter is better than the last position parameter in the optimal solution space, replacing, if the adaptability of the new parameter is better, adding the new parameter into the optimal solution space, correspondingly updating the value threshold and other constraint conditions of the optimal solution space, thereby completing the updating of the optimal solution space, repeating the steps, but considering the limitation of a tabu list in the process of selecting the parameter in each iteration, ensuring the searching process to have the capability of local fine searching and keeping the global exploratory property. By iterating continuously, the optimal solution space will gradually converge to the vicinity of the optimal solution.
In one possible implementation manner, the step a100 further includes a step a180 of performing interval parameter homodromous constraint of modification parameters based on the historical experiment database; and executing the step A190, when the modification parameter variation is carried out, if the variation period is in the earlier variation period, synchronously triggering the modification parameter through the homodromous variation constraint, and completing the modification parameter variation according to the synchronous triggering result.
Firstly, analyzing data in a historical experiment database, identifying parameter combinations which tend to change in the same direction (increase or decrease) at the same time in a modification process as interval parameter homodromous variation constraint, then, when modification parameter variation is carried out, identifying the early variation period in the variation period, wherein the early variation period generally refers to the beginning stage of a search process, an algorithm is exploring different areas of a solution space and is not converged near an optimal solution, when the variation period is in the early variation period, the variation of related modification parameters is synchronously triggered according to the interval parameter homodromous variation constraint, namely, if one parameter is mutated in a certain direction, other related parameters are mutated in the same direction according to the homodromous variation constraint, and according to a synchronous triggering result, the actual variation operation such as parameter adjustment value, parameter relation among modification parameters or new parameter introduction is carried out on the modification parameters.
Further, according to the random parameter probability, interval parameter homodromous mutation constraint or other mutation strategies, mutation operation is carried out on the selected parameters, if the mutation result is superior to the existing solutions in the optimal solution space, the new solutions are added into the optimal solution space, the value range, the threshold value or other constraint conditions of the optimal solution space are correspondingly updated, then the steps of parameter mutation, mutation result evaluation and optimal solution space updating are repeatedly executed, the iterative optimization process is continued, each iteration guides the searching direction based on the previous searching result and fitness evaluation, the global optimal solution is gradually approximated, finally, through continuous iteration and optimization, when the performance of the solutions in the optimal solution space is not remarkably improved or reaches the preset iteration times, the modification fitting can be considered to be completed, and at the moment, the solutions in the optimal solution space represent the optimal combination of the modification parameters, and the modification fitting is completed.
And (C) executing the step A200, and completing modification optimization according to the modification fitting result. In one possible implementation manner, the step a200 further includes a step a210 of establishing a modified solution set according to the modified fitting result, and performing steady-state evaluation on the modified solution set; and executing the step A220, carrying out comprehensive sequence sorting according to the steady state evaluation result and the modification result, and completing modification optimization based on the selected result of the comprehensive sequence sorting.
Based on the modified fitting results, i.e. the performance of the modified parameters under different conditions, a modified solution set is created comprising a plurality of potential solutions, which is obtained by simulation or experiment under different parameter combinations and conditions, and further, the stability and reliability of each solution in the modified solution set is evaluated by analyzing the robustness, sensitivity, and performance variations under different scenarios of the solution, i.e. whether the performance under different environments and operating conditions is consistent or reliable steady state.
And combining the steady state evaluation result and the modification result, and comprehensively sequencing the solutions in the modification solution set according to the factors such as the quality of the modification effect, the stability of steady state performance, the difficulty of realization and the like. The purpose of identifying solutions with good modification effects and higher steady-state performance is achieved, and the optimal modification solution is further selected as a final optimization scheme based on the selected results of comprehensive sequence ordering. The optimal solution is to achieve the optimal balance between the modification effect and the steady-state performance, finally adjust the modification parameters according to the optimal solution, implement modification measures and complete the modification optimization process.
The embodiment of the application solves the technical problems of performance reduction at an interface and cost effectiveness reduction caused by the poor compatibility between matrix resin and rubber phase of the thermoplastic elastomer of the blending type interpenetrating network in the prior art, realizes modification and optimization of the thermoplastic elastomer of the blending type interpenetrating network, improves a processing technology, designs a more reasonable blending structure, and achieves the technical effect of obtaining a modification method with higher cost effectiveness.
In the above, the modification and optimization method of the blended type interpenetrating network thermoplastic elastomer according to the embodiment of the present invention is described in detail with reference to fig. 1. Next, a modified optimization system of a blended type interpenetrating network thermoplastic elastomer according to an embodiment of the present invention will be described with reference to fig. 2.
The modification and optimization system of the blending type interpenetrating network thermoplastic elastomer is used for solving the technical problems that the performance at an interface is reduced and the cost efficiency is reduced due to the fact that the compatibility between matrix resin and a rubber phase of the blending type interpenetrating network thermoplastic elastomer is generally poor in the prior art, modifying and optimizing the blending type interpenetrating network thermoplastic elastomer are achieved, and therefore a blending structure with a more reasonable processing technology and design is improved, and the technical effect of a modification method with higher cost efficiency is achieved. The modification and optimization system of the blending type interpenetrating network thermoplastic elastomer comprises the following components: a modification fitting module 10 and a modification optimizing module 20.
A modification fitting module 10, wherein the modification fitting module 10 is configured to establish a modification requirement, the modification requirement includes a modification optimization direction, a modification value and an expected cost, and the modification fitting of the blend type interpenetrating network thermoplastic elastomer is performed according to the modification requirement, and specifically includes:
S1: acquiring a historical experiment database, wherein the historical experiment database is constructed by collecting modified experiment data and historical production data, each data in the historical experiment database corresponds to authentication data, and the authentication data characterizes the trust degree of the data;
s2: the historical experiment database is cleaned according to the authentication data, modification parameters are established through the cleaned historical experiment database, and an optimal solution space is configured;
s3: establishing an fitness function according to the modification requirement, wherein the fitness function is an equilibrium evaluation function for modifying a result and modifying cost, and optimizing and searching is carried out in the optimal solution space through the fitness function;
S4: establishing an activation function, and judging a variation direction after one round of search is finished by using the activation function, wherein the variation direction comprises variation in a best solution space and variation in a historical experiment database;
s5: performing modification parameter variation according to the discrimination result, updating the optimal solution space according to the variation result, and continuously iterating to finish modification fitting;
And the modification optimization module 20 is used for completing modification optimization according to the modification fitting result by the modification optimization module 20.
Next, the specific configuration of the modified fitting module 10 will be described in detail. As described above, the modified fitting module 10 may further include: the value analysis unit is used for carrying out value analysis based on a historical experiment database on the optimal solution space, and generating a first probability constraint according to an analysis result; the constraint calibration unit is used for configuring and calibrating probability constraints through big data and taking the probability constraints as second probability constraints; the threshold setting unit is used for setting a value threshold of the optimal solution space according to the first probability constraint and the second probability constraint; the first judging unit is used for completing the judgment of the variation direction according to the activation function and the value threshold.
Next, the specific configuration of the first discrimination unit will be described in detail. As described above, the determining the mutation direction according to the activation function and the value threshold may further include: the second judging unit is used for generating a random number through the activation function and judging whether the random number is in the value threshold range or not; the third judging unit is used for generating a variation instruction in the optimal solution space if the third judging unit is in the value threshold range; the fine tuning mutation unit is used for carrying out fine tuning mutation on the parameters in the optimal solution space according to the mutation instruction in the optimal solution space; the fourth judging unit is used for generating a random variation instruction of the historical experiment database if the random variation instruction is not in the value threshold range; the random variation unit is used for carrying out parameter random variation of the historical experiment database according to the random variation instruction of the historical experiment database.
Next, the specific configuration of the modified fitting module 10 will be described in detail. As described above, the modification parameter variation according to the discrimination result, the modification fitting module 10 may further include: the fifth judging unit is used for establishing random parameter probability through the important duty ratio of the modified parameters in the optimal solution space when the judging result is the fine tuning variation selected by the parameters; the random selection unit is used for carrying out random selection of variation parameters according to the random parameter probability, executing preset step length variation of the selected parameters and adding the selected parameters to a tabu list; the first evaluation unit is used for evaluating the variation result of the selected parameter based on the fitness function, generating forbidden disturbance according to the evaluation result, comparing the evaluation result with the last fitness evaluation result in the optimal solution space, updating the optimal solution space according to the comparison result, and finishing continuous iteration through the forbidden list.
Next, the specific configuration of the modified fitting module 10 will be described in detail. As described above, the modified fitting module 10 may further include: the variation constraint unit is used for performing interval parameter homodromous variation constraint of the modification parameters based on the historical experiment database; and the synchronous triggering unit is used for synchronously triggering the modification parameters through the homodromous mutation constraint when the modification parameters are mutated and the mutation period is in the mutation earlier stage, and finishing the modification parameter mutation according to the synchronous triggering result.
Next, a specific configuration of the threshold setting unit will be described in detail. As described above, the threshold setting unit may further include: the iteration recording unit is used for executing iteration recording and constructing a first iteration result of the optimal solution space and a second iteration result of the historical experiment database; the second evaluation unit is used for performing variation adaptation evaluation of the optimal solution space and the historical experiment database based on the first iteration result and the second iteration result; and the updating unit is used for carrying out self-adaptive updating on the value threshold according to the variation adaptation evaluation result.
Next, the specific configuration of the modification-optimization module 20 will be described in detail. As described above, the modification and optimization module 20 may further include: the third evaluation unit is used for establishing a modified solution set according to the modified fitting result and performing steady state evaluation on the modified solution set; the sequence ordering unit is used for carrying out comprehensive sequence ordering according to the steady state evaluation result and the modification result, and finishing modification optimization based on the selected result of the comprehensive sequence ordering.
The modification and optimization system of the blending type interpenetrating network thermoplastic elastomer provided by the embodiment of the invention can execute the modification and optimization method of the blending type interpenetrating network thermoplastic elastomer provided by any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.
Although the present application makes various references to certain modules in a system according to embodiments of the present application, any number of different modules may be used and run on a user terminal and/or server, including units and modules that are merely partitioned by functional logic, but are not limited to the above-described partitioning, so long as the corresponding functionality is enabled; in addition, the specific names of the functional units are also only for distinguishing from each other, and are not used to limit the protection scope of the present application.
The above embodiments do not limit the scope of the present application. It will be apparent to those skilled in the art that various modifications, combinations, and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the scope of the present application.
Claims (1)
1. The modification and optimization method of the blending type interpenetrating network thermoplastic elastomer is characterized by comprising the following steps of:
Establishing a modification requirement, wherein the modification requirement comprises a modification optimization direction, a modification value and an expected cost, and performing modification fitting of the blending type interpenetrating network thermoplastic elastomer by using the modification requirement, and specifically comprises the following steps:
S1: acquiring a historical experiment database, wherein the historical experiment database is constructed by collecting modified experiment data and historical production data, each data in the historical experiment database corresponds to authentication data, and the authentication data characterizes the trust degree of the data;
s2: the historical experiment database is cleaned according to the authentication data, modification parameters are established through the cleaned historical experiment database, and an optimal solution space is configured;
s3: establishing an fitness function according to the modification requirement, wherein the fitness function is an equilibrium evaluation function for modifying a result and modifying cost, and optimizing and searching is carried out in the optimal solution space through the fitness function;
S4: establishing an activation function, and judging a variation direction after one round of search is finished by using the activation function, wherein the variation direction comprises variation in a best solution space and variation in a historical experiment database;
s5: performing modification parameter variation according to the discrimination result, updating the optimal solution space according to the variation result, and continuously iterating to finish modification fitting;
Finishing modification optimization according to the modification fitting result;
The method for determining the mutation direction comprises the following steps:
Performing value degree analysis based on a historical experiment database on the optimal solution space, and generating a first probability constraint according to an analysis result;
configuring a calibration probability constraint through big data, and taking the calibration probability constraint as a second probability constraint;
Setting a value threshold of an optimal solution space according to the first probability constraint and the second probability constraint;
Completing the judgment of the variation direction according to the activation function and the value threshold;
completing the variation direction discrimination according to the activation function and the value threshold, including:
generating a random number through the activation function, and judging whether the random number is within the value threshold range;
If the variation instruction is within the value threshold range, generating a variation instruction in the optimal solution space;
Performing fine tuning variation on parameters in the optimal solution space by using the variation instruction in the optimal solution space;
If the random variation instruction is not in the value threshold range, generating a random variation instruction of the historical experiment database;
carrying out parameter random variation of the historical experiment database by using the random variation instruction of the historical experiment database;
and carrying out modification parameter variation according to the discrimination result, wherein the modification parameter variation comprises the following steps:
If the judging result is that the parameter selects the fine tuning variation, establishing random parameter probability through the important duty ratio of the modified parameter in the optimal solution space;
randomly selecting variation parameters through the random parameter probability, executing the preset step variation of the selected parameters, and adding the selected parameters to a tabu list;
Evaluating the variation result of the selected parameters based on the fitness function, generating forbidden disturbance according to the evaluation result, comparing the evaluation result with the last fitness evaluation result in the optimal solution space, updating the optimal solution space according to the comparison result, and finishing continuous iteration through the forbidden list;
the modification parameter variation comprises the following steps:
performing interval parameter homodromous variation constraint of modification parameters based on the historical experiment database;
When modification parameter variation is carried out, if the variation period is in the earlier variation period, synchronous triggering of modification parameters is carried out through the homodromous variation constraint, and modification parameter variation is completed according to a synchronous triggering result;
The adaptive updating of the value threshold value comprises the following steps:
executing iteration records, and constructing a first iteration result of the optimal solution space and a second iteration result of the historical experiment database;
performing variation adaptation evaluation of an optimal solution space and a historical experiment database based on the first iteration result and the second iteration result;
Performing value threshold self-adaptive updating according to the variation adaptation evaluation result;
establishing a modified solution set according to a modified fitting result, and performing steady-state evaluation on the modified solution set;
And (3) carrying out comprehensive sequence sorting according to the steady-state evaluation result and the modification result, and completing modification optimization based on the selected result of the comprehensive sequence sorting.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410436305.0A CN118024445B (en) | 2024-04-11 | 2024-04-11 | Modification optimization method and system for blending type interpenetrating network thermoplastic elastomer |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410436305.0A CN118024445B (en) | 2024-04-11 | 2024-04-11 | Modification optimization method and system for blending type interpenetrating network thermoplastic elastomer |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118024445A CN118024445A (en) | 2024-05-14 |
| CN118024445B true CN118024445B (en) | 2024-06-21 |
Family
ID=90995340
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410436305.0A Active CN118024445B (en) | 2024-04-11 | 2024-04-11 | Modification optimization method and system for blending type interpenetrating network thermoplastic elastomer |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118024445B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110533221A (en) * | 2019-07-29 | 2019-12-03 | 西安电子科技大学 | Multipurpose Optimal Method based on production confrontation network |
| CN117610435A (en) * | 2024-01-22 | 2024-02-27 | 天津路联智通交通科技有限公司 | Automatic proportioning method and system for civil and architectural construction mixed materials |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108133272A (en) * | 2018-01-15 | 2018-06-08 | 大连民族大学 | A kind of method of complex network community detection |
-
2024
- 2024-04-11 CN CN202410436305.0A patent/CN118024445B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110533221A (en) * | 2019-07-29 | 2019-12-03 | 西安电子科技大学 | Multipurpose Optimal Method based on production confrontation network |
| CN117610435A (en) * | 2024-01-22 | 2024-02-27 | 天津路联智通交通科技有限公司 | Automatic proportioning method and system for civil and architectural construction mixed materials |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118024445A (en) | 2024-05-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108509335B (en) | Software test data generation method based on genetic algorithm optimization | |
| CN112540849B (en) | Parameter configuration optimization method and system for distributed computing operation | |
| CN107766245A (en) | The online sort method of variable dynamics combined test use-case priority based on OTT strategies | |
| CN114490316B (en) | Loss function-based automatic generation method for unit test case | |
| CN111286599A (en) | Alloy heat treatment process optimization method based on machine learning | |
| JP2009099051A (en) | Parametric multi-objective optimization apparatus, method, and program | |
| CN118024445B (en) | Modification optimization method and system for blending type interpenetrating network thermoplastic elastomer | |
| CN1185848A (en) | Time-series signal predicting apparatus | |
| Yu et al. | Domination-based ordinal regression for expensive multi-objective optimization | |
| CN117037960A (en) | Performance control method and system for sulfur aluminum iron series cementing material | |
| CN116258165A (en) | Multi-target neural architecture searching method integrating convolution and self-attention | |
| CN109034479B (en) | Multi-target scheduling method and device based on differential evolution algorithm | |
| CN115248769A (en) | Test case optimization method, device, equipment and storage medium | |
| Molina et al. | Analysis among winners of different IEEE CEC competitions on real-parameters optimization: Is there always improvement? | |
| Gal et al. | Improvements to the bacterial memetic algorithm used for fuzzy rule base extraction | |
| CN116010291A (en) | Multipath coverage test method based on equalization optimization theory and gray prediction model | |
| CN107908557B (en) | Embedded software credible attribute modeling and verifying method | |
| Soliman et al. | A self-adaptive strategy for controlling parameters in differential evolution | |
| CN116090773A (en) | Flexible job shop scheduling method based on improved wolf algorithm | |
| CN113378276B (en) | Composite foundation intelligent design method based on genetic algorithm and gene expression programming | |
| Berghammer et al. | Set-based multi-objective optimization, indicators, and deteriorative cycles | |
| Liu et al. | Automatic calibration of numerical models using fast optimisation by fitness approximation | |
| Tanabe et al. | Non-elitist evolutionary multi-objective optimizers revisited | |
| Rusu et al. | On Combining Several Optimization Algorithms for Improving the Efficiency of IC Verification Pprocess | |
| CN114882961B (en) | Firing curve prediction method based on raw material weight as model parameter selection condition |
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 |