CN-122024973-A - Intelligent design system and method for low-carbon high-performance paving material proportion

CN122024973ACN 122024973 ACN122024973 ACN 122024973ACN-122024973-A

Abstract

The invention discloses a low-carbon high-performance paving material proportioning intelligent design system and a method, in particular relates to the technical field of computer aided design and intelligent optimization algorithms, and aims to solve the problems that the existing intelligent optimization method is easy to fall into local optimization and insufficient in global optimizing capability due to fragmentation of a design space when multi-source and strong coupling constraint is processed; the method is implemented by analyzing the violation state of an initial population to constraint, identifying constraint coupling groups which lead to space fragmentation based on violation anti-co-occurrence, dividing and identifying sub-populations corresponding to different fragment areas according to the constraint coupling groups, calculating the closeness of the coding entropy of each sub-population and the average constraint boundary to evaluate the evolution potential of each sub-population, distributing differentiated optimizing resources according to the coded entropy and the average constraint boundary, performing independent optimizing calculation, simultaneously monitoring the convergence situation of the evolution track of each sub-population, and executing the directional migration of a proportioning scheme crossing the sub-population when the similar areas tend to be determined among the sub-populations and premature convergence exists based on the situation.

Inventors

Rao Lingli
ZHAO YU
ZHANG YA
LI FENG
GUO QIAN
ZHANG YUJIE
LONG JIANXU
ZHOU FANG
HAO ZENGTAO
TANG HONG
LI BIN
He Feixue

Assignees

贵州交通职业大学

Dates

Publication Date: 20260512
Application Date: 20260410

Claims (10)

1. The intelligent design method for the low-carbon high-performance paving material proportion is characterized by comprising the following steps of: S1, acquiring a multisource constraint condition of paving materials, and constructing an initial candidate proportion population; S2, analyzing violation states of initial candidate matching population to multi-source constraint conditions, and identifying and dividing constraint coupling groups which lead to design space fragmentation based on co-occurrence of the violation states; s3, dividing candidate matching population according to constraint coupling groups, and generating and identifying sub-populations corresponding to different fragmentation feasible domains; s4, calculating the closeness of the coding entropy and the average constraint boundary of the candidate matching scheme in each sub-population, and evaluating the evolution potential value of each sub-population based on the closeness of the coding entropy and the average constraint boundary; s5, distributing optimizing calculation resources for each sub-population according to the evolution potential value, performing independent optimizing calculation, and simultaneously monitoring and analyzing the evolution track convergence situation of each sub-population in the independent optimizing calculation process; s6, executing the proportioning scheme directional migration operation of the cross-sub population based on the evolution track convergence situation, and updating the candidate proportioning population.
2. The intelligent design method for the low-carbon high-performance paving material proportion according to claim 1, wherein S1 comprises the following steps: acquiring and analyzing multisource constraint conditions, and determining the allowable value range of each component of the paving material; Randomly generating a plurality of candidate proportioning schemes in an allowable value range; judging the feasibility of each candidate matching scheme according to the multi-source constraint condition; and selecting candidate matching schemes with different feasibility judging results to form an initial candidate matching population.
3. The intelligent design method for the low-carbon high-performance paving material proportion according to claim 1, wherein the step S2 comprises the following steps: Recording the violation state of each multisource constraint condition by each candidate matching scheme in the initial candidate matching population; counting the co-occurrence frequency of simultaneous violation of any two multi-source constraint conditions by the same batch of candidate matching schemes; the multi-source constraint conditions of which the co-occurrence frequency exceeds a preset frequency threshold are divided into a constraint coupling group.
4. The intelligent design method for the low-carbon high-performance paving material proportion according to claim 1, wherein the step S3 comprises the following steps: Aiming at each constraint coupling group, analyzing the overall satisfaction condition of each candidate matching scheme in the candidate matching population to the multi-source constraint conditions in the constraint coupling group; the candidate proportioning schemes with consistent overall meeting conditions are classified into the same set; Each set corresponds to a sub-population and an identification is assigned to the corresponding sub-population to associate its corresponding set of constrained couplings.
5. The intelligent design method for the low-carbon high-performance paving material proportion according to claim 1, wherein the step S4 comprises the following steps: Counting the coding feature distribution of the candidate proportioning scheme in each sub population and calculating coding entropy; calculating the distance from each candidate matching scheme to the boundary of the multi-source constraint condition, and taking an average value to obtain the average constraint boundary closeness; And combining the coding entropy with the average constraint boundary closeness according to a preset rule, and calculating to obtain an evolution potential value of the sub-population.
6. The intelligent design method for the low-carbon high-performance paving material proportion is characterized by comprising the steps of counting coding characteristic distribution of candidate proportion schemes in the intelligent design method and calculating coding entropy, wherein the coding characteristic distribution comprises the steps of coding content percentages of components in each candidate proportion scheme into discretized coding bits, counting occurrence frequencies of different discrete values of the sub-population on all the coding bits, and calculating the coding entropy of the sub-population based on the occurrence frequencies according to an information entropy formula.
7. The intelligent design method for the low-carbon high-performance paving material proportion is characterized by combining the coding entropy and the average constraint boundary closeness according to a preset rule, calculating to obtain an evolution potential value of the sub-population, and comprises the steps of respectively carrying out normalization processing on the coding entropy and the average constraint boundary closeness, carrying out weighted summation on the normalized coding entropy and the normalized average constraint boundary closeness according to a preset weight, and obtaining a result, namely the evolution potential value of the sub-population.
8. The intelligent design method for the low-carbon high-performance paving material proportion according to claim 1, wherein the step S5 comprises the following steps: according to the evolution potential value proportion of each sub-population, independent iterative computation times are distributed for each sub-population; Each sub population independently updates and screens the candidate matching scheme in the allocated iterative calculation times; in the independent optimizing calculation process of each sub population, recording the optimal fitness and coding diversity of each generation of candidate matching scheme; And analyzing and generating evolution track convergence situations of each sub-population according to the recorded changes of the optimal fitness and the coding diversity.
9. The intelligent design method for the low-carbon high-performance paving material proportion according to claim 1, wherein the step S6 comprises the following steps: Judging whether two sub-populations tend to be similar in a target space and at least one sub-population tends to be premature and converged or not based on the evolution track convergence situation of each sub-population; when the judgment is yes, selecting a candidate matching scheme with highest fitness from the sub-population tending to premature convergence as a migration individual; And adding the migrated individuals into another sub-population, and replacing the candidate matching scheme with the lowest adaptability in the other sub-population to update the candidate matching population.
10. An intelligent design system for low-carbon high-performance paving material proportion, which is used for realizing the intelligent design method for the low-carbon high-performance paving material proportion according to any one of claims 1-9, and is characterized by comprising the following steps: the constraint acquisition module is used for acquiring multi-source constraint conditions of paving materials and constructing an initial candidate proportioning population; The constraint coupling module is used for analyzing the violation state of the initial candidate matching population to the multi-source constraint condition, and identifying and dividing constraint coupling groups which lead to the fragmentation of the design space based on the co-occurrence of the violation state; The subgroup dividing module is used for dividing the candidate matching population according to the constraint coupling group, and generating and identifying sub-populations corresponding to different fragmentation feasible domains; the potential evaluation module is used for calculating the closeness of the coding entropy and the average constraint boundary of the candidate matching scheme in each sub-population, and evaluating the evolution potential value of each sub-population based on the closeness of the coding entropy and the average constraint boundary; the independent optimizing module is used for distributing optimizing calculation resources for each sub-population according to the evolution potential value, performing independent optimizing calculation, and simultaneously monitoring and analyzing the evolution track convergence situation of each sub-population in the independent optimizing calculation process; And the population updating module is used for executing the proportioning scheme directional migration operation of the cross-sub population based on the evolution track convergence situation and updating the candidate proportioning population.

Description

Intelligent design system and method for low-carbon high-performance paving material proportion Technical Field The invention relates to the technical field of computer aided design and intelligent optimization algorithms, in particular to an intelligent design system and method for low-carbon high-performance paving material proportion. Background In the field of road engineering, the proportioning design of paving materials is a key link for ensuring the service performance and engineering economy, and especially in areas with complex geological environment and severe climatic conditions like karst landforms, the paving materials are required to have dual requirements of low carbon, environmental protection and high performance on the maintenance materials of traffic infrastructures. The current pavement material design for such targets needs to meet constraint conditions from various aspects such as environment, mechanics, technology and cost, for example, solid waste recycling utilization rate, various road performance indexes, construction feasible windows and raw material cost limit, and the constraint conditions are mutually interwoven to form a complex design space with high dimensionality and nonlinearity. To cope with the complexity, an intelligent method is adopted, and the matching scheme meeting all constraints and optimizing the comprehensive target is automatically searched in the design space through an algorithm. However, when the strong coupling effect of the multi-source and rigid constraint is faced, the existing intelligent optimization method relying on continuous space assumption can lead the feasible region conforming to all the constraint to be degenerated from a single communication region to a plurality of mutually isolated fragmented regions when the design space is jointly cut by a plurality of constraint conditions, the algorithm population is extremely easy to be limited in a certain local feasible fragment for invalid optimization, and can not be detected and migrated to other isolated regions possibly with better solutions, thereby leading to design failure or obtaining local suboptimal solutions and severely restricting the reliability and global optimizing capability of the intelligent design method under complex engineering scenes. Disclosure of Invention In order to overcome the defects in the prior art, the invention provides a low-carbon high-performance paving material proportioning intelligent design system and method for solving the problems in the background art. In order to achieve the above purpose, the present invention provides the following technical solutions: A low-carbon high-performance paving material proportioning intelligent design method comprises the following steps: S1, acquiring a multisource constraint condition of paving materials, and constructing an initial candidate proportion population; S2, analyzing violation states of initial candidate matching population to multi-source constraint conditions, and identifying and dividing constraint coupling groups which lead to design space fragmentation based on co-occurrence of the violation states; s3, dividing candidate matching population according to constraint coupling groups, and generating and identifying sub-populations corresponding to different fragmentation feasible domains; s4, calculating the closeness of the coding entropy and the average constraint boundary of the candidate matching scheme in each sub-population, and evaluating the evolution potential value of each sub-population based on the closeness of the coding entropy and the average constraint boundary; s5, distributing optimizing calculation resources for each sub-population according to the evolution potential value, performing independent optimizing calculation, and simultaneously monitoring and analyzing the evolution track convergence situation of each sub-population in the independent optimizing calculation process; s6, executing the proportioning scheme directional migration operation of the cross-sub population based on the evolution track convergence situation, and updating the candidate proportioning population. Further, S1 includes: acquiring and analyzing multisource constraint conditions, and determining the allowable value range of each component of the paving material; Randomly generating a plurality of candidate proportioning schemes in an allowable value range; judging the feasibility of each candidate matching scheme according to the multi-source constraint condition; and selecting candidate matching schemes with different feasibility judging results to form an initial candidate matching population. Further, S2 includes: Recording the violation state of each multisource constraint condition by each candidate matching scheme in the initial candidate matching population; counting the co-occurrence frequency of simultaneous violation of any two multi-source constraint conditions by the same batch of candidate matching