Concrete component adjustment method and system based on multi-dimensional parameter analysis

Abstract

The application provides a concrete component adjustment method and system based on multi-dimensional parameter analysis, and relates to the technical field of concrete preparation. The method comprises the following steps: constructing a mapping relationship list of concrete components and evaluation parameters; determining a multi-dimensional target evaluation parameter; obtaining associated concrete components; establishing a target evaluation function; obtaining a single evaluation value and a global evaluation value; obtaining an optimization component composition; and adjusting and optimizing the current concrete components by using the optimization component composition. The application can solve the technical problem that the performance of concrete is unstable and it is difficult to meet actual requirements due to the lack of a systematic optimization method in the prior art. By constructing a positive incentive relationship and a negative limit relationship between the concrete components and the evaluation parameters, and establishing individual evaluation functions and global evaluation functions to optimize the concrete components, the technical effects of improving the global optimization, improving the stability of the concrete performance, and further improving the adaptability of the concrete performance to actual requirements are achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of concrete preparation, and particularly relates to a concrete component adjustment method and system based on multi-dimensional parameter analysis. BACKGROUND

[0002] Concrete is a composite material made of a variety of raw materials, and its performance is affected by factors such as the types and proportions of raw materials. In the fields of construction and engineering, concrete is a widely used building material, and its performance directly affects the quality and durability of the building structure. The performance of concrete is affected by different components, and the influence relationship of different components may be different.

[0003] Traditional concrete proportioning design mainly relies on experience, lacks systematic optimization methods, and is prone to unstable concrete performance, which is difficult to meet actual needs. SUMMARY

[0004] The purpose of the present application is to provide a concrete component adjustment method and system based on multi-dimensional parameter analysis, to solve the technical problems in the prior art that lack of systematic optimization methods, which is prone to unstable concrete performance and difficult to meet actual needs.

[0005] In view of the above problems, the present application provides a concrete component adjustment method and system based on multi-dimensional parameter analysis.

[0006] In a first aspect, the present application provides a concrete component adjustment method based on multi-dimensional parameter analysis, which is realized by a concrete component adjustment system based on multi-dimensional parameter analysis, wherein the method comprises: constructing a mapping relationship list of concrete components and evaluation parameters, the mapping relationship list comprising influence relationships of concrete components and evaluation parameters, the influence relationships comprising positive incentive relationships and negative limiting relationships; obtaining a target parameter requirement, performing multi-dimensional decomposition on the target parameter requirement, and determining multi-dimensional target evaluation parameters; comparing the multi-dimensional target evaluation parameters with evaluation parameters of the mapping relationship list to obtain associated concrete components; establishing a target evaluation function based on the positive incentive relationships and the negative limiting relationships, the target evaluation function comprising individual evaluation functions and global evaluation functions; evaluating the associated concrete components by using the individual evaluation functions and the global evaluation functions to obtain individual evaluation values and global evaluation values; determining whether the individual evaluation values and the global evaluation values meet the target parameter requirement, and when they do not meet the target parameter requirement, performing component optimization by using an optimization space to obtain an optimized component group, wherein the optimization space is constructed based on the target evaluation function and a preset optimization rule; and adjusting and optimizing the current concrete components by using the optimized component group.

[0007] In a second aspect, the application further provides a concrete component adjustment system based on multi-dimensional parameter analysis, configured to perform the method for adjusting concrete components based on multi-dimensional parameter analysis as described in the first aspect, wherein the system comprises: a mapping relationship list construction module configured to construct a mapping relationship list of concrete components and evaluation parameters, the mapping relationship list comprising influence relationships between concrete components and evaluation parameters, the influence relationships comprising positive incentive relationships and negative restriction relationships; a multi-dimensional decomposition module configured to obtain a target parameter requirement, perform multi-dimensional decomposition on the target parameter requirement, and determine multi-dimensional target evaluation parameters; an associated concrete component acquisition module configured to compare the multi-dimensional target evaluation parameters with evaluation parameters of the mapping relationship list, and obtain associated concrete components; a target evaluation function establishment module configured to establish a target evaluation function based on the positive incentive relationships and the negative restriction relationships, the target evaluation function comprising individual evaluation functions and global evaluation functions; an evaluation module configured to evaluate the associated concrete components by using the individual evaluation functions and the global evaluation functions, and obtain individual evaluation values and global evaluation values; a component optimization module configured to determine whether the individual evaluation values and the global evaluation values meet the target parameter requirement, perform component optimization in an optimization space when the individual evaluation values and the global evaluation values do not meet the target parameter requirement, and obtain an optimized component composition, wherein the optimization space is constructed based on the target evaluation function and a preset optimization rule; and an adjustment and optimization module configured to adjust and optimize a current concrete component by using the optimized component composition.

[0008] One or more technical solutions provided in the application have at least the following technical effects or advantages:

[0009] A mapping relationship list of concrete components and evaluation parameters is constructed, the mapping relationship list including an influence relationship of the concrete components and the evaluation parameters, the influence relationship including a positive incentive relationship and a negative restriction relationship; a target parameter requirement is obtained, the target parameter requirement is multi-dimensionally decomposed to determine multi-dimensional target evaluation parameters; the multi-dimensional target evaluation parameters are compared with evaluation parameters of the mapping relationship list to obtain associated concrete components; a target evaluation function is established based on the positive incentive relationship and the negative restriction relationship, the target evaluation function including individual evaluation functions and a global evaluation function; the associated concrete components are evaluated by using the individual evaluation functions and the global evaluation function to obtain individual evaluation values and global evaluation values; it is judged whether the individual evaluation values and the global evaluation values meet the target parameter requirement, and when the target parameter requirement is not met, component optimization is performed through an optimization space to obtain an optimized component group, wherein the optimization space is constructed based on the target evaluation function and a preset optimization rule; the current concrete components are adjusted and optimized by using the optimized component group. Thus, the positive incentive relationship and the negative restriction relationship of the concrete components and the evaluation parameters are constructed, the individual evaluation functions and the global evaluation function are established to optimize the concrete components, the global optimization is improved, the stability of the concrete performance is improved, and the adaptability of the concrete performance to the actual requirement is improved.

[0010] The above description is only a summary of the technical solutions of the present application. In order to enable one skilled in the art to better understand the technical means of the present application, the content of the specification can be implemented, and in order to enable the above and other purposes, features and advantages of the present application to be more obvious and easy to understand, the specific embodiments of the present application are described below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are only exemplary, and for those skilled in the art, other drawings can be obtained without creative labor on the basis of the provided drawings.

[0012] Figure 1 Flowchart of the concrete component adjustment method based on multi-dimensional parameter analysis of the present application;

[0013] Figure 2 Structure diagram of the concrete component adjustment system based on multi-dimensional parameter analysis of the present application.

[0014] Reference signs: mapping relationship list construction module 11, multi-dimensional decomposition module 12, associated concrete component acquisition module 13, target evaluation function establishment module 14, evaluation module 15, component optimization module 16, adjustment optimization module 17. DETAILED DESCRIPTION

[0015] The present application provides a concrete component adjustment method and system based on multi-dimensional parameter analysis, which solves the technical problem in the prior art that the lack of systematic optimization method easily leads to unstable concrete performance and difficulty in meeting actual requirements. By constructing positive incentive relationship and negative limiting relationship between concrete components and evaluation parameters, individual evaluation function and global evaluation function are established to optimize concrete components, so as to improve global optimization, improve concrete performance stability, and further improve the adaptability of concrete performance to actual requirements.

[0016] Hereinafter, the technical solutions in the present application will be described clearly and completely with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. It should be understood that the present application is not limited by the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of protection of the present application. In addition, it should be noted that, for convenience of description, only parts related to the present application are shown in the drawings, not all.

[0017] Embodiment one

[0018] Please refer to the accompanying drawings Figure 1 The present application provides a concrete component adjustment method based on multi-dimensional parameter analysis, wherein the method is applied to a concrete component adjustment system based on multi-dimensional parameter analysis, and the method specifically includes the following steps:

[0019] Step one: construct a mapping relationship list of concrete components and evaluation parameters, the mapping relationship list includes the influence relationship of concrete components and evaluation parameters, and the influence relationship includes positive incentive relationship and negative limiting relationship;

[0020] Specifically, the concrete components usually include water, cement, aggregate (sand, stone, etc.), air conditioner, antifreeze, etc. The evaluation parameters are parameters for evaluating the performance of concrete, including compressive strength, durability, etc.

[0021] The positive incentive relationship and the negative limiting relationship are both ways of describing the interaction between two variables, but the directions and effects are different. The positive incentive relationship means that when one variable increases, the other variable also increases. In this application, the positive incentive relationship refers to the increase of a certain concrete component leading to the positive improvement of an evaluation parameter, for example, the increase of cement content improves the compressive strength of concrete. The negative limiting relationship means that when one variable increases, the other variable decreases or is limited. In this application, the negative limiting relationship may refer to the increase of a certain concrete component leading to the decrease or limitation of an evaluation parameter, for example, too high water-cement ratio leads to the decrease of compressive strength. Based on this, by analyzing the influence relationship between the evaluation parameters and the components of the concrete, the concrete components and the evaluation parameters with the positive incentive relationship and the negative limiting relationship are obtained, which constitute the mapping relationship list.

[0022] Step two: obtaining a target parameter requirement, multi-dimensionally decomposing the target parameter requirement to determine multi-dimensional target evaluation parameters;

[0023] Specifically, the target parameter requirement refers to the performance requirement of concrete, which needs to be determined by a person skilled in the art in combination with the actual application scene, including compressive strength, durability, etc. The target parameter requirement may include various types of performance parameters, which are decomposed to obtain multi-dimensional target evaluation parameters.

[0024] Step three: comparing the multi-dimensional target evaluation parameters with the evaluation parameters of the mapping relationship list to obtain associated concrete components;

[0025] Specifically, the multi-dimensional target evaluation parameters are compared with the evaluation parameters of the mapping relationship list to obtain the concrete components with the positive incentive relationship and the negative limiting relationship between the mapping relationship list and the multi-dimensional target evaluation parameters as the associated concrete components. Subsequently, the content of the associated concrete components can be analyzed to optimize the multi-dimensional target evaluation parameters.

[0026] Step four: establishing a target evaluation function based on the positive incentive relationship and the negative limiting relationship, the target evaluation function including individual evaluation functions and a global evaluation function;

[0027] Specifically, the individual evaluation function is for the evaluation parameter corresponding to each concrete component of the associated concrete components. For each concrete component, an evaluation system can be established as an individual evaluation function according to the positive incentive relationship or the negative limiting relationship, such as a relationship formula for the compressive strength and the cement component content. Each concrete component in the associated concrete components corresponds to an individual evaluation function.

[0028] The global evaluation function is a weighted sum of the individual evaluation functions, and considers the influence of all components in the associated concrete components on different evaluation parameters. The weights of the individual evaluation functions can be set by the person skilled in the art based on the importance of the evaluation parameters having positive incentive relationships or negative limiting relationships corresponding to the concrete components, and engineering requirements.

[0029] Step five: evaluating the associated concrete components using the individual evaluation function and the global evaluation function to obtain a single evaluation value and a global evaluation value;

[0030] Specifically, the associated concrete components include the content ratio information of each associated concrete component in the currently used concrete formula. Thus, each concrete component is substituted into the individual evaluation function to calculate the individual evaluation function value of each concrete component as the single evaluation value.

[0031] The single evaluation values corresponding to all concrete components in the associated concrete components are weighted and summed according to the global evaluation function to obtain the global evaluation value.

[0032] Step six: determining whether the single evaluation value and the global evaluation value meet the target parameter requirement. When the requirement is not met, component optimization is performed through an optimization space to obtain an optimized component composition, wherein the optimization space is constructed based on the target evaluation function and a preset optimization rule.

[0033] Specifically, the single evaluation value reflects the advantages and disadvantages of each concrete component in the performance having the influence relationship, and the global evaluation reflects the comprehensive performance of the entire associated concrete components. Thus, whether the single evaluation value and the global evaluation value meet the target parameter requirement is determined. If the requirement is not met, component optimization is performed through the optimization space. The component optimization is performed based on the existing particle swarm optimization algorithm. The optimization space includes all possible concrete component compositions. The target evaluation function is used for optimization, and the component combination with the optimal single evaluation value and global evaluation value is selected as the optimized component composition. The preset optimization rule refers to the rule for constructing a new component combination based on the particle swarm optimization algorithm.

[0034] Step seven: adjusting and optimizing the current concrete components using the optimized component composition.

[0035] Specifically, the optimized component composition is the adjustment target of the concrete components. The current concrete components are adjusted and optimized according to the optimized component composition to improve the performance of the concrete.

[0036] Further, step one of the present application further includes:

[0037] The concrete components are classified, the basic components and additional components are determined; individual parameter adjustment experiments are performed according to the basic components, and a basic component experimental data set is obtained; individual parameter adjustment experiments are sequentially performed on the additional components according to the basic component experimental data set, and an additional component experimental data set is obtained; the basic component experimental data set and the additional component experimental data set are fitted for the component-evaluation parameter relationship respectively, and an evaluation parameter curve of each component is obtained; the positive excitation relationship or the negative inhibition relationship is determined according to the trend of the evaluation parameter curve of each component, and the influence relationship parameter value is determined according to the curvature of the evaluation parameter curve of each component; a mapping relationship between the components and the evaluation parameters is established, and a mapping relationship list of the concrete components and the evaluation parameters is constructed according to the positive excitation relationship or the negative inhibition relationship and the corresponding influence relationship parameter value.

[0038] Specifically, the basic components are essential parts of concrete and play a decisive role in the basic performance of concrete. Generally, the basic components include cement, aggregate (sand, stone) and water. The additional components are added to improve or enhance the performance of concrete, and are usually not essential but can be selected and adjusted according to specific needs. Common additional components include additives (such as water reducing agent, reinforcing agent, etc.) and admixtures (such as fly ash, silica fume, etc.). The component table of the basic components and the additional components of the concrete can be constructed by the skilled person in the art, so that the basic components and the additional components are matched and classified based on the concrete components, and the basic components and the additional components are obtained.

[0039] For the determined basic components, individual parameter adjustment experiments are performed, including changing the amount of cement, the amount of aggregate, the amount of water, etc., and the performance of the concrete after changing the basic components is tested based on the existing technology, and a basic component experimental data set is obtained. The basic component experimental data set contains the performance of the concrete under different basic components. On the basis of the basic component experiment, individual parameter adjustment experiments are sequentially performed on the additional components, i.e. only the parameter of one additional component (such as the amount of additive) is changed each time, while the other components remain unchanged, and then the performance of the concrete is tested based on the existing technology to obtain an additional component experimental data set.

[0040] The existing statistical method or machine learning algorithm is used to fit the component-evaluation parameter relationship of the basic component experimental data set and the additional component experimental data set respectively, and the evaluation parameter is the concrete performance that changes simultaneously after the component changes, and thus the evaluation parameter curve of the evaluation parameter and each component is drawn, which can intuitively show the influence of the component change on the performance parameter. Based on the trend of the evaluation parameter curve, the positive excitation relationship or negative inhibition relationship between the component and the performance parameter can be determined. The positive excitation relationship means that the increase of the component will improve the performance, and the negative inhibition relationship means that the increase of the component will reduce the performance. The curvature of the evaluation parameter curve is identified, and the curvature is used as the influence relationship parameter value. The greater the curvature, the more significant the influence of the component on the performance parameter, and therefore the influence relationship parameter value should also be greater.

[0041] The mapping relationship between the concrete components and the evaluation parameters is established based on the positive excitation relationship or negative inhibition relationship determined in the above steps and the corresponding influence relationship parameter value, and the established mapping relationship is arranged in the form of a list to obtain a mapping relationship list, which contains the basic components and additional components, and the positive excitation or negative inhibition relationship and the influence relationship parameter value between them and the evaluation parameters, providing support for the optimization of the concrete formula.

[0042] Further, step five of the present application further comprises:

[0043] According to the associated concrete components, the positive excitation relationship and the negative inhibition relationship are obtained; the evaluation parameters are used as clustering centers to cluster the positive excitation relationship and the negative inhibition relationship, and the relationship network of each evaluation parameter is constructed; the individual evaluation function is used to evaluate the relationship network of each evaluation parameter respectively to obtain the individual evaluation value; and the global evaluation function is used to comprehensively evaluate all the positive excitation relationships and negative inhibition relationships of the associated concrete components to obtain the global evaluation value.

[0044] Specifically, the process of evaluating the associated concrete components by using the individual evaluation function and the global evaluation function to obtain the individual evaluation value and the global evaluation value is as follows:

[0045] Based on the mapping relationship list, the influence relationship between the associated concrete components and the evaluation parameters is determined, including the positive excitation relationship and the negative inhibition relationship. The evaluation parameters (such as compressive strength, impermeability, etc.) are used as clustering centers to perform clustering analysis on the positive excitation relationship and the negative inhibition relationship. For example, the existing clustering algorithms such as K-means, hierarchical clustering, etc. are used to realize the classification of the positive excitation relationship or the negative inhibition relationship that affects the same evaluation parameter into a class, and the relationship network of each evaluation parameter is constructed based on the clustering results, which shows the influence of different components on the same evaluation parameter.

[0046] For each evaluation parameter network, an individual evaluation function is used for evaluation, and the influence of different components on the same evaluation parameter is calculated to obtain a single evaluation value. The global evaluation function is used to weight the single evaluation values of different components to obtain the global evaluation value. The global evaluation value comprehensively considers all performance parameters and component relationships, and reflects the overall performance of the concrete formulation. It is convenient for comprehensive evaluation and optimization analysis of concrete performance.

[0047] Further, step six of the present application further comprises:

[0048] A plurality of optimization particles corresponding to the associated concrete components are configured; based on the individual evaluation function, the plurality of optimization particles are subjected to target maximum screening to obtain individual particle optimization values; based on the individual particle optimization values of the plurality of optimization particles, the global evaluation function is subjected to target maximum screening to obtain a global particle group optimization value; the individual particle optimization values and the global particle group optimization value are used for individual particle data expansion to obtain expanded individual particle values; the expanded individual particle values are iteratively expanded until a preset iteration number or a convergence condition is reached to obtain individual particle value solution spaces; all individual particle value solution spaces and the global evaluation function are used for global particle comprehensive optimization, and a global evaluation value maximum solution is screened as the optimization component group.

[0049] Further, the present application further comprises the following steps:

[0050] The historical composition record information is obtained, and the particle evolution relationship is analyzed, and the particle position parameters and the particle initial step length are matched, the particle position parameters are component characteristic values, and the particle initial step length is used for adjusting the step length during expansion; the individual expansion learning factor and the global expansion learning factor are configured, the individual expansion learning factor is smaller than the global expansion learning factor; the individual particle record values are obtained according to the historical composition record information, and the deviation values are calculated respectively with the individual particle optimization values and the global particle group optimization values to obtain individual deviation values and global deviation values; based on the individual deviation values, the global deviation values, the particle position parameters and the particle initial step length, an expansion update function is constructed; according to the individual particle optimization values and the global particle group optimization values, the individual particle optimization values are expanded by using the expansion update function to generate the expanded individual particle values.

[0051] Further, the present application further comprises the following steps:

[0052] The expansion update function expression is:

[0053] ;

[0054] Wherein, characterizes the modification step length of the f+1th kth individual particle optimization value, a modification step of the kth individual particle optimization value in the fth time, when f = 0, is a particle initial step, is an individual expansion learning factor, is a global expansion learning factor, ∈ [0, 2], is an individual bias value, is a global bias value, and is a random number between 0 and 1.

[0055] Specifically, a plurality of optimization particles are configured, each of which represents a possible associated concrete component, which can be extracted based on historical concrete component parameters. Based on an individual evaluation function, the performance of the plurality of optimization particles is evaluated, and the optimization particle with the maximum individual evaluation function value is selected as the individual particle optimization value. According to the individual particle optimization values of the plurality of optimization particles, a weighted calculation is performed based on a global evaluation function, and the optimization particle combination with the maximum global evaluation function value is obtained as the global particle swarm optimization value.

[0056] Further, according to the individual particle optimization value and the global particle swarm optimization value, an individual particle data expansion is performed to obtain an expanded individual particle value, and the specific process is as follows: obtaining historical composition record information, which includes the proportions of different components. By analyzing the historical composition record information, the trend of the particle, i.e., the concrete component, in the evolution process is identified, and how the concrete component changes. The component feature value in the historical composition record information is taken as the particle position parameter, and the content of the component is taken as the component feature value. The change interval of the component feature value is taken as the particle initial step. An individual expansion learning factor and a global expansion learning factor are configured, wherein the individual expansion learning factor and the global expansion learning factor ∈ [0, 2], and the individual expansion learning factor is less than the global expansion learning factor, indicating that the particle expansion pays more attention to global experience.

[0057] According to the historical composition record information, a record value corresponding to any component is obtained as an individual particle record value, and the deviation between the individual particle record value and the individual particle optimization value is calculated as an individual bias value. The deviation between the individual bias value and the optimization value corresponding to the component in the global particle swarm optimization value is calculated as a global bias value.

[0058] Based on the individual bias value, the global bias value, the particle position parameter, and the particle initial step, an expansion update function is constructed. The expression of the expansion update function is:

[0059] ;

[0060] wherein, a modification step of the kth individual particle optimization value in the fth time, a modification step of the kth individual particle optimization value in the fth iteration, when f = 0, is an initial step of the particle, is an individual expansion learning factor, is a global expansion learning factor, ∈ [0, 2], is an individual bias value, is a global bias value, and is a random number between 0 and 1, and the random number is set to make the expansion result more comprehensive. The individual particle is expanded by using the expansion update function to generate an expanded individual particle value, improve the sample data amount of the concrete component optimization, and thus improve the global optimization.

[0061] Further according to the expanded individual particle value, iterative expansion is performed, that is, the individual particle data expansion is repeatedly performed until a preset iteration number or a convergence condition is reached, wherein the preset iteration number or the convergence condition is set by a person skilled in the art, such as 30 times. All the expanded individual particle values constitute an individual particle value solution space. Then, all the individual particle value solution spaces and the global evaluation function are used for comprehensive optimization of all particles, that is, the individual particle value solution space contains the optimal solution of a single concrete component, but the combination of all single optimal solutions is not necessarily optimal. Therefore, the expanded individual particle value corresponding to each concrete component is randomly selected from the individual particle value solution space for combination, and then each combination is globally evaluated by the global evaluation function. The combination corresponding to the maximum global evaluation value is used as the optimization component, thereby improving the accuracy of global optimization and further improving the performance of the concrete.

[0062] Further, the present application further includes the following steps:

[0063] obtaining a target parameter tolerance of the target parameter requirement; setting an evaluation parameter forced coefficient according to the target parameter tolerance; setting a constraint condition based on a positive incentive relationship and a negative limit relationship of the evaluation parameter according to the evaluation parameter forced coefficient; and adding the constraint condition to the optimization space.

[0064] Specifically, the target parameter requirement target parameter requirement refers to the performance requirement of the concrete, including compressive strength, durability, etc. In order to consider the fault tolerance in actual production, the tolerance of these target parameters needs to be determined, i.e. the allowed parameter variation range, which is determined by the professional technical personnel in the field combined with actual experience, and the allowed parameter variation range is taken as the target parameter tolerance. According to the tolerance of the target parameter, the evaluation parameter mandatory coefficient is set, which is the proportion coefficient of different components based on the tolerance of the target parameter, for example, according to the concrete components corresponding to a certain target parameter, the components of the concrete within the range of the target parameter tolerance are obtained, and the proportion coefficient of the components of the concrete in the entire concrete is calculated as the evaluation parameter mandatory coefficient.

[0065] The constraint condition is set according to the evaluation parameter mandatory coefficient and the positive incentive relationship and the negative limit relationship of the evaluation parameter, and the constraint condition is used to limit the search range of the optimization particle, and ensure that the search is performed under the premise of meeting the performance requirement. Specifically, for the evaluation parameter with a positive incentive relationship, we can set its lower limit value to ensure that the optimization result is not lower than a certain performance level; for the evaluation parameter with a negative limit relationship, we can set its upper limit value to avoid the optimization result exceeding the allowed range. Add the constraint condition to the optimization space, and in the subsequent particle expansion process, data expansion can be performed under the premise of meeting the constraint condition, so as to find a concrete component combination that meets the performance requirement and conforms to the actual production condition, and improve the pertinence and practicality of the optimization result.

[0066] Further, the present application further comprises the following steps:

[0067] According to the evaluation parameter mandatory coefficient, the mandatory parameter particle is determined from the positive incentive relationship, and the taboo condition of the mandatory parameter particle is configured based on the evaluation parameter mandatory coefficient, wherein the matching degree between the evaluation value of the mandatory parameter particle and the evaluation parameter mandatory coefficient reaches a preset condition; when the expansion individual particle value reaches the taboo condition, the expansion individual particle value is added to the taboo table, and the corresponding optimization area is stopped; the neighborhood is determined based on the optimization area, and the individual particle expansion is performed in the neighborhood; when the taboo condition is reached, the expansion individual particle value is added to the taboo, if the taboo condition is not reached, the optimal expansion particle value in the neighborhood is obtained and added to the taboo table, and the iteration optimization is performed until the convergence condition is reached; the particle value in the taboo table is used to construct the individual particle value solution space.

[0068] Specifically, the expansion individual particle value is iteratively expanded until a preset iteration number or a convergence condition is reached, and the process of obtaining the individual particle value solution space further comprises:

[0069] According to the evaluation parameter forced coefficient, the component with the evaluation parameter forced coefficient greater than the preset threshold is screened out from the positive incentive relationship, and the particle of the component with the evaluation parameter forced coefficient greater than the preset threshold is taken as a forced parameter particle. The forced parameter particle is configured with a taboo condition, and the taboo condition is that the value of the extended individual particle corresponding to the forced parameter particle reaches the evaluation parameter forced coefficient, that is, when the value of the extended individual particle corresponding to the forced parameter particle reaches the evaluation parameter forced coefficient, that is, the set component proportion coefficient is reached, the value of the extended individual particle is added to the taboo table, and the optimization of the corresponding optimization area is stopped. The taboo table is used to record the extended individual particle values that have been evaluated, so as to avoid repeated utilization.

[0070] Based on the optimization area, the neighborhood is determined, and the individual particle is expanded in the neighborhood. The extended individual particle value is generated by adjusting the position parameter and the step length of the particle. During the neighborhood expansion process, if the newly generated extended individual particle value reaches the taboo condition, it is also added to the taboo table, and the search in this direction is stopped. If the taboo condition is not reached, the particle value is continuously evaluated, and the optimal extended particle value in the neighborhood is obtained and added to the taboo table. The above-mentioned way of continuously iterating the expansion of the particle, the evaluation of the particle value and the processing of the taboo condition is continued until the convergence condition is reached. The convergence condition can be a preset iteration number, which is set by the professional technical personnel. The particle values in the taboo table are used to construct the individual particle value solution space. The individual particle value solution space contains the particle values found in the optimization process that meet the performance requirements and the taboo condition, and provides a basis for subsequent global comprehensive optimization, which facilitates the improvement of the globality of the optimization and more accurately finds the concrete component combination that meets the target parameter requirements.

[0071] In summary, the concrete component adjustment method based on multi-dimensional parameter analysis provided in the present application has the following technical effects:

[0072] mapping relationship list of the concrete components and the evaluation parameters, the mapping relationship list comprising an influence relationship between the concrete components and the evaluation parameters, the influence relationship comprising a positive incentive relationship and a negative restriction relationship; obtaining a target parameter requirement, performing multi-dimensional decomposition on the target parameter requirement to determine a multi-dimensional target evaluation parameter; comparing the multi-dimensional target evaluation parameter with an evaluation parameter of the mapping relationship list to obtain associated concrete components; establishing a target evaluation function based on the positive incentive relationship and the negative restriction relationship, the target evaluation function comprising an individual evaluation function and a global evaluation function; evaluating the associated concrete components by using the individual evaluation function and the global evaluation function to obtain a single evaluation value and a global evaluation value; determining whether the single evaluation value and the global evaluation value meet the target parameter requirement, and when the single evaluation value and the global evaluation value do not meet the target parameter requirement, performing component optimization by using an optimization space to obtain an optimized component group, wherein the optimization space is constructed based on the target evaluation function and a preset optimization rule; and adjusting and optimizing the current concrete components by using the optimized component group. Thus, by constructing the positive incentive relationship and the negative restriction relationship between the concrete components and the evaluation parameters, the individual evaluation function and the global evaluation function are established to optimize the concrete components, so that the global optimization is improved, the stability of the concrete performance is improved, and the adaptability of the concrete performance to the actual demand is improved.

[0073] Embodiment Two

[0074] Based on the same inventive concept as the concrete component adjustment method based on multi-dimensional parameter analysis in the foregoing embodiments, the present application also provides a concrete component adjustment system based on multi-dimensional parameter analysis. Please refer to the accompanying drawings Figure 2 , which comprises:

[0075] A mapping relationship list construction module 11 is configured to construct a mapping relationship list of the concrete components and the evaluation parameters, the mapping relationship list comprising an influence relationship between the concrete components and the evaluation parameters, the influence relationship comprising a positive incentive relationship and a negative restriction relationship.

[0076] A multi-dimensional decomposition module 12 is configured to obtain a target parameter requirement, perform multi-dimensional decomposition on the target parameter requirement to determine a multi-dimensional target evaluation parameter.

[0077] An associated concrete component acquisition module 13 is configured to compare the multi-dimensional target evaluation parameter with an evaluation parameter of the mapping relationship list to obtain associated concrete components.

[0078] A target evaluation function establishment module 14 is configured to establish a target evaluation function based on the positive incentive relationship and the negative restriction relationship, the target evaluation function comprising an individual evaluation function and a global evaluation function.

[0079] An evaluation module 15 is configured to evaluate the associated concrete components by using the individual evaluation function and the global evaluation function, to obtain individual evaluation values and global evaluation values;

[0080] A component optimization module 16 is configured to determine whether the individual evaluation values and the global evaluation values meet the target parameter requirement, and when the target parameter requirement is not met, to perform component optimization by using an optimization space, to obtain an optimized component composition, wherein the optimization space is constructed based on the target evaluation function and a preset optimization rule;

[0081] An adjustment optimization module 17 is configured to adjust and optimize the current concrete components by using the optimized component composition.

[0082] Further, the mapping relationship list construction module 11 in the system is further configured to:

[0083] divide the concrete components into types, to determine basic components and additional components;

[0084] perform individual parameter adjustment experiments on the basic components, to obtain a basic component experiment data set;

[0085] perform individual parameter adjustment experiments on the additional components in sequence based on the basic component experiment data set, to obtain an additional component experiment data set;

[0086] fit the component-evaluation parameter relationship of the basic component experiment data set and the additional component experiment data set respectively, to obtain evaluation parameter curves of the components;

[0087] determine positive excitation relationships or negative inhibition relationships based on the evaluation parameter curves of the components, and determine impact relationship parameter values based on the curvatures of the evaluation parameter curves of the components;

[0088] establish a mapping relationship between the components and the evaluation parameters, and construct a mapping relationship list of the concrete components and the evaluation parameters based on the positive excitation relationships or the negative inhibition relationships and the corresponding impact relationship parameter values.

[0089] Further, the evaluation module 15 in the system is further configured to:

[0090] obtain positive excitation relationships and negative limitation relationships based on the associated concrete components;

[0091] cluster the positive excitation relationships and the negative limitation relationships by taking the evaluation parameters as clustering centers, to construct a relationship network of the evaluation parameters;

[0092] evaluate the relationship network of the evaluation parameters by using the individual evaluation function respectively, to obtain the individual evaluation values;

[0093] The global evaluation function is used to comprehensively evaluate all positive incentive relations and negative restriction relations of the associated concrete components, and a global evaluation value is obtained.

[0094] Further, the component optimization module 16 in the system is also used for:

[0095] A plurality of optimization particles corresponding to the associated concrete components are configured.

[0096] Based on the individual evaluation function, target maximum screening is performed on the plurality of optimization particles to obtain individual particle optimization values.

[0097] Based on the global evaluation function, target maximum screening is performed on the individual particle optimization values of the plurality of optimization particles to obtain a global particle swarm optimization value.

[0098] Based on the individual particle optimization values and the global particle swarm optimization value, individual particle data expansion is performed to obtain expanded individual particle values.

[0099] Iterative expansion is performed according to the expanded individual particle values until a preset iteration number or a convergence condition is reached, and individual particle value solution spaces are obtained.

[0100] All individual particle value solution spaces and the global evaluation function are used for comprehensive optimization of all particles, and a global evaluation value maximum solution is screened as the optimization component composition.

[0101] Further, the component optimization module 16 in the system is also used for:

[0102] Historical composition record information is obtained, particle evolution relation analysis is performed, and particle position parameters and particle initial step lengths are matched, the particle position parameters being component characteristic values, and the particle initial step lengths being used for adjusting step lengths during expansion.

[0103] Individual expansion learning factors and global expansion learning factors are configured, the individual expansion learning factors being smaller than the global expansion learning factors.

[0104] Individual particle record values are obtained according to the historical composition record information, and deviation values are calculated respectively from the individual particle optimization values and the global particle swarm optimization values, to obtain individual deviation values and global deviation values.

[0105] Based on the individual deviation values, the global deviation values, particle position parameters, and particle initial step lengths, an expansion update function is constructed.

[0106] According to the individual particle optimization values and the global particle swarm optimization values, the expansion update function is used to expand the individual particle optimization values to generate the expanded individual particle values.

[0107] Further, the component optimization module 16 in the system is further configured to:

[0108] The extended update function expression is:

[0109] ;

[0110] wherein, a modification step length of the f+1th kth individual particle optimization value, a modification step length of the fth kth individual particle optimization value, when f=0, is a particle initial step length, is an individual extended learning factor, is a global extended learning factor, ∈[0,2], is an individual bias value, is a global bias value, and represents a random number between 0 and 1.

[0111] Further, the system further comprises a constraint condition setting module, which is configured to:

[0112] obtain a target parameter tolerance of the target parameter requirement;

[0113] set an evaluation parameter forced coefficient according to the target parameter tolerance;

[0114] set a constraint condition based on a positive incentive relationship and a negative limit relationship of the evaluation parameter according to the evaluation parameter forced coefficient;

[0115] add the constraint condition to the optimization space.

[0116] Further, the component optimization module 16 in the system is further configured to:

[0117] determine a forced parameter particle from the positive incentive relationship according to the evaluation parameter forced coefficient, and configure a taboo condition of the forced parameter particle based on the evaluation parameter forced coefficient, wherein the taboo condition is that a matching degree of an evaluation value of the forced parameter particle and the evaluation parameter forced coefficient reaches a preset condition;

[0118] when the extended individual particle value reaches the taboo condition, add the extended individual particle value to a taboo table, and stop the corresponding optimization area;

[0119] Determine a neighborhood based on the optimization region, expand individual particles in the neighborhood, when the taboo condition is reached, add the expanded individual particle value to the taboo, if the taboo condition is not reached, obtain the optimal expanded particle value in the neighborhood and add it to the taboo table, and iterate the optimization until the convergence condition is reached.

[0120] Construct the individual particle value solution space using the particle values in the taboo table.

[0121] Each embodiment in the specification is described in a progressive manner, and each embodiment focuses on the difference from other embodiments. The foregoing Figure 1 The concrete component adjustment method based on multi-dimensional parameter analysis in embodiment one and the specific examples are also applicable to the concrete component adjustment system based on multi-dimensional parameter analysis in the present embodiment. Based on the foregoing detailed description of the concrete component adjustment method based on multi-dimensional parameter analysis, those skilled in the art can clearly understand the concrete component adjustment system based on multi-dimensional parameter analysis in the present embodiment. Therefore, for the sake of brevity of the specification, it will not be described in detail here. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant part can be referred to the method part description.

[0122] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0123] Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application also intends to include these modifications and variations.

Claims

1. A method for adjusting concrete composition based on multidimensional parameter analysis, characterized in that, The method includes: Construct a mapping relationship list between concrete composition and evaluation parameters. The mapping relationship list includes the influence relationship between concrete composition and evaluation parameters, including positive incentive relationship and negative constraint relationship. Obtain the target parameter requirements, perform multidimensional decomposition on the target parameter requirements, and determine multidimensional target evaluation parameters; By comparing the multidimensional target evaluation parameters with the evaluation parameters in the mapping relationship list, the associated concrete components are obtained; Based on the positive incentive relationship and the negative constraint relationship, a target evaluation function is established, which includes an individual evaluation function and a global evaluation function. The associated concrete components are evaluated using the individual evaluation function and the global evaluation function to obtain individual evaluation values ​​and global evaluation values; Determine whether the individual evaluation value and the global evaluation value meet the target parameter requirements. If they do not meet the requirements, perform component optimization through the optimization space to obtain the optimized component composition. The optimization space is constructed based on the target evaluation function and preset optimization rules. The current concrete composition is adjusted and optimized using the aforementioned optimal component composition; The step of optimizing components through the optimization space to obtain the optimal component composition includes: Multiple optimized particles are configured, and the multiple optimized particles correspond to the associated concrete components; Based on the individual evaluation function, the target maximum screening is performed on the multiple optimization particles to obtain the individual particle optimization value; Based on the individual particle optimization values ​​of multiple optimized particles, the target maximum screening is performed based on the global evaluation function to obtain the overall particle swarm optimization value; Based on the individual particle optimization value and the overall particle swarm optimization value, the individual particle data is expanded to obtain the expanded individual particle value; The individual particle value is iteratively expanded according to the expanded individual particle value until a preset number of iterations or a convergence condition is reached, thereby obtaining the individual particle value solution space. The solution space of all individual particle values ​​and the global evaluation function are used to perform a comprehensive optimization of all particles, and the solution with the largest global evaluation value is selected as the optimization component. The step of augmenting individual particle data based on the individual particle optimization value and the overall particle swarm optimization value to obtain augmented individual particle values ​​includes: Historical composition record information is obtained, particle evolution relationship analysis is performed, and particle position parameters and initial particle step size are matched. The particle position parameters are component feature values, and the initial particle step size is used to adjust the step size during expansion. Configure individual expanded learning factors and global expanded learning factors, wherein the individual expanded learning factors are smaller than the global expanded learning factors; Based on the historical composition record information, the individual particle record value is obtained, and the deviation value is calculated by comparing it with the individual particle optimization value and the whole particle swarm optimization value to obtain the individual deviation value and the global deviation value. Based on the individual deviation value, global deviation value, particle position parameters, and initial particle step size, an expanded update function is constructed. Based on the individual particle optimization value and the overall particle swarm optimization value, the individual particle optimization value is expanded using the expansion update function to generate the expanded individual particle value. The expression for the expanded update function is: ； in, The step size representing the modification of the optimization value of the k-th individual particle in the (f+1)-th iteration. The step size for modifying the optimization value of the k-th individual particle in the f-th iteration, when f=0. The initial step size of the particle. To expand learning factors for individuals, To expand the learning factors globally, ∈[0,2], Individual deviation value, This is the global deviation value. and Represents a random number between 0 and 1; Obtain the target parameter tolerance required for the target parameters; The evaluation parameter forced coefficients are set according to the tolerance of the target parameters; Based on the forced coefficients of the evaluation parameters, constraints are set according to the positive incentive relationship and negative constraint relationship of the evaluation parameters; Add the constraints to the optimization space; The iterative expansion based on the expanded individual particle values, until a preset number of iterations or a convergence condition is reached, to obtain the individual particle value solution space, includes: Based on the evaluation parameter mandatory coefficient, a mandatory parameter particle is determined from the positive excitation relationship, and a taboo condition for the mandatory parameter particle is configured based on the evaluation parameter mandatory coefficient, wherein the taboo condition is that the matching degree between the evaluation value of the mandatory parameter particle and the evaluation parameter mandatory coefficient reaches a preset condition. When the expanded individual particle value reaches the taboo condition, the expanded individual particle value is added to the taboo table, and the corresponding optimization region is stopped; Based on the optimization region, a neighborhood is determined, and individual particles are expanded in the neighborhood. When the taboo condition is met, the expanded individual particle value is added to the taboo. If the taboo condition is not met, the optimal expanded particle value in the neighborhood is obtained and added to the taboo table. This process is repeated iteratively until the convergence condition is met. The solution space of the individual particle values ​​is constructed using the particle values ​​in the taboo table.

2. The method as described in claim 1, characterized in that, The mapping relationship list between the constructed concrete components and evaluation parameters includes: The concrete composition is classified into different types, and the basic components and additional components are determined. Individual parameter adjustment experiments were conducted based on the aforementioned basic components to obtain the basic component experimental dataset. Based on the basic component experimental dataset, individual parameter adjustment experiments were sequentially performed on the additional components to obtain the additional component experimental dataset. The component-evaluation parameter relationship was fitted to the basic component experimental dataset and the additional component experimental dataset respectively to obtain the evaluation parameter curve of each component; Based on the trend of the evaluation parameter curves of each component, determine the positive incentive relationship or the negative inhibition relationship; based on the curvature of the evaluation parameter curves of each component, determine the value of the influence relationship parameter. Establish a mapping relationship between components and evaluation parameters, and construct a mapping relationship list between concrete components and evaluation parameters based on the positive excitation relationship or negative inhibition relationship and the corresponding influence relationship parameter values.

3. The method as described in claim 1, characterized in that, The associated concrete components are evaluated using the individual evaluation function and the global evaluation function to obtain individual evaluation values ​​and global evaluation values, including: Based on the associated concrete composition, obtain the positive incentive relationship and the negative constraint relationship; Using the evaluation parameters as cluster centers, the positive incentive relationships and negative constraint relationships are clustered to construct a relationship network for each evaluation parameter; The individual evaluation function is used to evaluate the relationship network of each evaluation parameter to obtain the individual evaluation value; The global evaluation function is used to comprehensively evaluate all positive incentive relationships and negative constraint relationships of the associated concrete components to obtain the global evaluation value.

4. A concrete composition adjustment system based on multidimensional parameter analysis, characterized in that, The system comprises: steps for implementing the method according to any one of claims 1 to 3, wherein the system includes: The mapping relationship list construction module is used to construct a mapping relationship list between concrete components and evaluation parameters. The mapping relationship list includes the influence relationship between concrete components and evaluation parameters, and the influence relationship includes positive incentive relationship and negative constraint relationship. The multidimensional decomposition module is used to obtain the target parameter requirements, perform multidimensional decomposition on the target parameter requirements, and determine the multidimensional target evaluation parameters. The associated concrete composition acquisition module is used to compare the multidimensional target evaluation parameters with the evaluation parameters of the mapping relationship list to obtain the associated concrete composition. The target evaluation function establishment module is used to establish a target evaluation function based on the positive incentive relationship and the negative constraint relationship. The target evaluation function includes an individual evaluation function and a global evaluation function. The evaluation module is used to evaluate the associated concrete components using the individual evaluation function and the global evaluation function, and to obtain individual evaluation values ​​and global evaluation values. The component optimization module is used to determine whether the individual evaluation value and the global evaluation value meet the target parameter requirements. When they do not meet the requirements, the module performs component optimization through the optimization space to obtain the optimized component composition. The optimization space is constructed based on the target evaluation function and preset optimization rules. The adjustment and optimization module is used to adjust and optimize the current concrete composition using the optimized component composition.

Images