An optimization method and system for reinforced concrete support design

Abstract

The present application provides an optimization method and system for reinforced concrete support design, wherein a multi-dimensional analysis of the load expected to be borne by a building is performed to determine the key stress points in the building structure; based on the key stress points, the behavioral characteristics of reinforced concrete materials at different building structural levels are simulated to generate a multi-level mechanical model; based on the multi-level mechanical model and data obtained from monitoring the building structure, an intelligent decision-making process is established using genetic algorithms, particle swarm optimization, and reinforcement learning frameworks to generate a reinforced concrete support design scheme; a comprehensive evaluation system is established using Bayesian networks and gray system theory to quantify the reinforced concrete support design scheme, obtain feedback information, and use the feedback information to optimize the reinforced concrete support design scheme to generate an optimized reinforced concrete support design scheme. The present application improves the scientificity and rationality of the reinforced concrete support design scheme.

Description

Technical Field

[0001] The embodiments of the present application relate to the technical fields of building structure engineering and intelligent optimization algorithms, and in particular to an optimization method and system for reinforced concrete support design. Background Art

[0002] In modern building structural engineering, as the height and complexity of buildings continue to increase, more stringent requirements are placed on the design of supporting structures. Especially in projects such as high-rise buildings, bridges and large public facilities, it is necessary to accurately analyze the various loads that the building is expected to bear, including static loads, dynamic loads and seismic loads, in order to determine the key stress points in the structure. This is not only to ensure the safety and stability of the building, but also to optimize the use of materials, reduce construction costs and improve economic benefits. Accurately identifying key stress points is crucial for designing reinforced concrete supporting structures that are both safe and economical. In addition, faced with increasingly complex engineering environments and technical requirements, traditional methods have gradually revealed their limitations. There is an urgent need for a new method that can comprehensively consider the impact of multi-dimensional loads and has efficient optimization capabilities.

[0003] Traditional reinforced concrete support design relies primarily on empirical formulas and finite element analysis. Designers conduct preliminary designs based on specifications and standards, combined with historical data and experimental results, and optimize the design through trial and error. In recent years, advanced numerical simulation techniques and intelligent optimization algorithms have begun to be applied to this field. These include hybrid modeling techniques that combine deep learning with physical rules to simulate material behavior, and methods such as genetic algorithms and particle swarm optimization to find optimal solutions. These methods have improved design accuracy and efficiency to a certain extent, but there is still room for improvement. Despite this, existing technologies have significantly advanced the design of building structures and provided strong support for complex projects.

[0004] However, existing solutions still face many challenges in practical applications. First, traditional methods find it difficult to fully consider the impact of multi-dimensional loads, resulting in inaccurate identification of key stress points, which in turn affects the reliability of the overall design. Second, the traditional optimization process based on trial and error is time-consuming and prone to falling into local optimal solutions, and cannot guarantee global optimality. More importantly, existing solutions often lack an effective quantitative evaluation system and feedback mechanism, making it difficult to comprehensively evaluate and continuously optimize design solutions, especially when faced with complex and changeable actual working conditions, and the design flexibility is insufficient. These problems limit the ability of existing methods to provide optimal solutions in high-performance and complex environments. Summary of the Invention

[0005] The embodiments of the present application provide a method and system for optimizing reinforced concrete support design, so as to solve the problems of low scientificity and low rationality of reinforced concrete support design schemes in the prior art.

[0006] In a first aspect, an embodiment of the present application provides a method for optimizing reinforced concrete support design, comprising:

[0007] Analyze the loads expected to be borne by the building in multiple dimensions to determine the critical stress points in the building structure;

[0008] Based on the key stress points, a hybrid modeling technology integrating deep learning and physical rules is used to simulate the behavioral characteristics of reinforced concrete materials at different building structural levels and generate a multi-level mechanical model;

[0009] Based on the multi-level mechanical model and data obtained from monitoring the building structure, an intelligent decision-making process is established using a genetic algorithm, particle swarm optimization, and reinforcement learning framework to generate a reinforced concrete support design solution;

[0010] A comprehensive evaluation system is established using Bayesian network and grey system theory to quantitatively analyze the reinforced concrete support design scheme to obtain feedback information, which is then used to optimize the reinforced concrete support design scheme to generate an optimized reinforced concrete support design scheme.

[0011] Optionally, the method of establishing an intelligent decision-making process based on the multi-level mechanical model and the data obtained from monitoring the building structure using a genetic algorithm, a particle swarm optimization, and a reinforcement learning framework to generate a reinforced concrete support design solution includes:

[0012] Using finite element analysis, the multi-level mechanical model is simulated to generate simulation data. Based on the simulation data, the behavior patterns and potential risk points of the key stress points under different loading conditions are verified. The simulation data are classified and regressed using a random forest algorithm to identify key parameters and change trends that affect the structural performance of the building, thereby obtaining the behavioral characteristics of the key stress points.

[0013] Based on the behavioral characteristics, the data obtained by monitoring the building structure, and the simulation data, principal component analysis dimensionality reduction technology is applied to extract key feature vectors, and a time series modeling method is used to analyze the changing trend of the building structure response under preset working conditions to identify dynamic stress conditions, and a structural response prediction model is generated based on the dynamic stress conditions and the expected load changes of the building structure;

[0014] Based on the structural response prediction model, initial support members are selected according to the behavioral characteristics of the key stress points and the expected load changes, and design parameters of the initial support members are encoded using a genetic algorithm to generate an initial population. The initial population is analyzed using a differential evolution algorithm to generate an optimized design parameter set;

[0015] Based on the changing trend and dynamic stress conditions of the structural response prediction model, a fitness function is defined, and an optimal design parameter combination is iteratively searched using a particle swarm optimization algorithm. Based on the optimal design parameter combination and the optimized design parameter set, an optimal design solution set corresponding to the fitness function is analyzed. Based on the optimal design solution set and combined with an ant colony optimization algorithm, an optimal combination of optimized design parameters is generated;

[0016] According to the reinforcement learning framework, the optimal combination is used as the action space of the intelligent agent in the building structure, and the safety, stability and economy of the building structure are used as reward signals to adjust the strategy parameters of the intelligent agent to generate a reinforced concrete support design scheme.

[0017] Optionally, defining a fitness function based on the change trend and dynamic stress conditions of the structural response prediction model, iteratively searching for an optimal design parameter combination using a particle swarm optimization algorithm, analyzing an optimal design solution set of the fitness function based on the optimal design parameter combination combined with the optimized design parameter set, and generating an optimal combination of optimized design parameters based on the optimal design solution set combined with an ant colony optimization algorithm, including:

[0018] Defining a fitness function based on the change trend and dynamic stress conditions of the structural response prediction model and in combination with different factors of the building structure, wherein the different factors of the building structure include safety, stability, and economy;

[0019] Based on the fitness function, a particle swarm optimization algorithm and a differential evolution algorithm are used to iteratively search for an optimal design parameter combination to calculate the fitness function value corresponding to each particle in the particle swarm optimization algorithm, and an adaptive inertia weight is applied to obtain an optimal design parameter set. Design solutions with fitness function values ​​higher than a preset threshold are screened from the optimal design parameter set to generate an optimal design solution set;

[0020] Based on the optimal design solution set, combined with the ant colony optimization algorithm, the best combination of optimized design parameters is generated.

[0021] Optionally, based on the fitness function, a particle swarm optimization algorithm combined with a differential evolution algorithm is used to iteratively search for an optimal design parameter combination to calculate a fitness function value corresponding to each particle in the particle swarm optimization algorithm, and an adaptive inertia weight is applied to obtain an optimal design parameter set, and design solutions with fitness function values ​​higher than a preset threshold are screened from the optimal design parameter set to generate an optimal design solution set, including:

[0022] Calculating the fitness function value corresponding to each particle in the particle swarm optimization algorithm;

[0023] Applying an adaptive inertia weight adjustment strategy to dynamically adjust the particle velocity update rule to generate an optimized particle swarm, and obtaining an optimal design parameter set based on the optimized particle swarm;

[0024] The design schemes whose fitness function values ​​are higher than a preset threshold are screened out from the optimal design parameter set, and an optimal design scheme set is generated based on the design schemes.

[0025] Optionally, the selecting of initial supports based on the structural response prediction model according to the behavioral characteristics of the key stress points and the expected load changes, encoding the design parameters of the initial supports using a genetic algorithm to generate an initial population, and analyzing the initial population using a differential evolution algorithm to generate an optimized design parameter set includes:

[0026] Selecting initial supports suitable for each of the key stress points according to the structural response prediction model, the behavioral characteristics of the key stress points, and the expected load changes;

[0027] Encoding the design parameters of the initial support parts using a genetic algorithm to obtain an initial support part set, constructing an initial population based on the initial support part set, and applying a differential evolution algorithm to construct and mutate difference vectors of individuals in the initial population to obtain an optimized population;

[0028] Based on the optimized population, an optimized design parameter set is generated.

[0029] Optionally, based on the key stress points, a hybrid modeling technology integrating deep learning and physical rules is used to simulate the behavioral characteristics of reinforced concrete materials at different building structural levels to generate a multi-level mechanical model, including:

[0030] Based on the key stress points, a hybrid modeling technology integrating deep learning and physical rules is used to collect key data of each key stress point, generate a basic data set, and use the basic data set to build a hybrid modeling framework. The key data includes relevant experimental data, historical monitoring data, and theoretical calculation results;

[0031] Based on the hybrid modeling framework, a deep learning model is used to simulate the behavior of reinforced concrete materials at the detail level, predict the performance changes of the reinforced concrete materials under different stress states, and obtain a detail-level material behavior model;

[0032] Based on the detailed material behavior model, combined with the structural characteristics of reinforced concrete materials at the component level, the behavior of reinforced concrete components at the structural detail level is simulated, and physical rules are introduced to guide model training to generate a component-level behavior model;

[0033] Extending the behavior in the component-level behavior model to the overall structural level, simulating the response of the entire building structure under different loading conditions, and using physical rules to constrain the output of the deep learning model to generate an overall structural response model;

[0034] A multi-level mechanical model is established based on the detail-level material behavior model, the component-level behavior model, and the overall structure-level response model.

[0035] Optionally, the method of establishing a comprehensive evaluation system using Bayesian network and grey system theory to quantitatively analyze the reinforced concrete support design scheme to obtain feedback information, and optimizing the reinforced concrete support design scheme using the feedback information to generate an optimized reinforced concrete support design scheme includes:

[0036] Defining evaluation indicators based on key factors of the reinforced concrete support design scheme to construct a comprehensive evaluation indicator system, wherein the key factors include safety, stability, and economy;

[0037] Using a Bayesian network, the comprehensive evaluation index system is risk evaluated, and the probability relationship between different risk factors in the reinforced concrete support design scheme is dynamically adjusted to obtain a risk assessment result;

[0038] Applying grey system theory to analyze the uncertainty of the reinforced concrete support design scheme, and using grey correlation analysis method to generate an uncertainty analysis report;

[0039] Arranging the risk assessment results and the uncertainty analysis report to form a comprehensive evaluation system, and quantitatively scoring the reinforced concrete support design scheme based on the comprehensive evaluation system to obtain a quantitative scoring result;

[0040] Analyze data obtained from monitoring the building structure, expert review opinions, and feedback from stakeholders to obtain a summary of feedback information;

[0041] Based on the feedback information summary and the quantitative scoring results, the reinforced concrete support design scheme is adjusted using a genetic algorithm, a particle swarm optimization, and a reinforcement learning framework to generate an optimized reinforced concrete support design scheme.

[0042] In a second aspect, an embodiment of the present application provides an optimization system for reinforced concrete support design, comprising:

[0043] an analysis module for performing multi-dimensional analysis of various loads expected to be borne by the building to determine key stress points in the building structure;

[0044] A simulation module is used to simulate the behavior characteristics of reinforced concrete materials at different building structural levels based on the key stress points and generate a multi-level mechanical model using a hybrid modeling technology that integrates deep learning and physical rules;

[0045] a generation module for generating a reinforced concrete support design based on the multi-level mechanical model in combination with data obtained from monitoring the building structure and establishing an intelligent decision-making process using genetic algorithms, particle swarm optimization, and reinforcement learning frameworks;

[0046] The quantification module is used to establish a comprehensive evaluation system using Bayesian network and grey system theory, quantitatively analyze the reinforced concrete support design scheme, obtain feedback information, optimize the reinforced concrete support design scheme using the feedback information, and generate an optimized reinforced concrete support design scheme.

[0047] In a third aspect, an embodiment of the present application provides a computing device comprising a processor and a memory, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute an optimization method for reinforced concrete support design as described in any one of the first aspects.

[0048] In a fourth aspect, an embodiment of the present application provides a computer storage medium having computer program instructions stored thereon, wherein the computer program instructions, when executed by a processor, implement an optimization method for reinforced concrete support design as described in any one of the first aspects.

[0049] In an embodiment of the present application, a multi-dimensional analysis is performed on the load that a building is expected to bear to determine the key stress points in the building structure; based on the key stress points, a hybrid modeling technology integrating deep learning and physical rules is used to simulate the behavioral characteristics of reinforced concrete materials at different building structural levels to generate a multi-level mechanical model; based on the multi-level mechanical model and the data obtained by monitoring the building structure, an intelligent decision-making process is established using genetic algorithms, particle swarm optimization, and a reinforcement learning framework to generate a reinforced concrete support design scheme; a comprehensive evaluation system is established using Bayesian networks and grey system theory to quantitatively analyze the reinforced concrete support design scheme and obtain feedback information, which is then used to optimize the reinforced concrete support design scheme to generate an optimized reinforced concrete support design scheme.

[0050] The technical solution of this application has the following beneficial effects:

[0051] This application conducts a multi-dimensional analysis of the various loads expected to be borne by a building, accurately identifying key stress points within the structure and providing a scientific basis for subsequent design optimization. Based on these key stress points, a hybrid modeling technique is employed to simulate the behavior of reinforced concrete materials at different structural levels, generating a multi-level mechanical model. This ensures that the design considers the material's behavior at all levels, from microscopic to macroscopic, improving both accuracy and reliability. An intelligent decision-making process is established using genetic algorithms, particle swarm optimization, and a reinforcement learning framework, accelerating optimization and improving the quality of the proposed solution. This approach can quickly identify the optimal solution among a large number of candidate solutions, significantly improving design efficiency and accuracy. A comprehensive evaluation system is established by incorporating Bayesian networks and grey system theory to conduct a comprehensive quantitative assessment of reinforced concrete support design solutions and generate feedback. This feedback is used to further optimize the design, ensuring the feasibility and superior performance of the final solution in practical applications. By seamlessly integrating these steps, this method can generate and optimize reinforced concrete support design solutions, ensuring they are both safe and stable while also being economical. This not only improves the overall performance of the building, but also reduces long-term maintenance costs and enhances the reliability and durability of the structure. This method is applicable to the design requirements of various complex building structures and can flexibly respond to different working conditions and environmental conditions, ensuring that the design scheme not only meets the requirements of the specifications but also effectively copes with the uncertainties in actual projects.

[0052] Furthermore, the embodiment of the present application also uses finite element analysis to simulate the multi-level mechanical model, generates simulation data to verify the behavior patterns and potential risk points of key stress points under different loading conditions, and identifies the key parameters and changing trends that affect the structural performance of the building through the random forest algorithm. Furthermore, the principal component analysis dimensionality reduction technology is applied to extract key feature vectors, and the changing trend of the building structure response is analyzed in combination with the time series modeling method to generate a structural response prediction model. Based on this model, the initial support parts are selected and the design parameters are encoded through the genetic algorithm, and the differential evolution algorithm is used to generate the optimized design parameter set. Subsequently, the fitness function is defined and the particle swarm optimization algorithm is used to iteratively search for the best design parameter combination, and the ant colony optimization algorithm is combined to generate the best combination of optimized design parameters. Finally, the reinforcement learning framework is introduced, the best combination is used as the action space of the intelligent agent, and the safety, stability and economy of the building are used as reward signals to adjust the strategy parameters, and finally a reinforced concrete support design scheme is generated.

[0053] The above-mentioned methods improved the scientific and rationality of the reinforced concrete bracing design. First, through finite element analysis and random forest algorithms, the behavioral patterns and potential risk points of key load-bearing points were accurately identified, ensuring the reliability of the design. Second, the application of principal component analysis dimensionality reduction techniques and time series modeling methods enabled accurate capture of the changing trends in the building's structural response, thereby improving the accuracy of dynamic stress prediction. Furthermore, the combined use of genetic algorithms, differential evolution algorithms, particle swarm optimization algorithms, and ant colony optimization algorithms not only accelerated optimization but also improved the quality of the solution, ensuring the global optimal solution. Finally, a reinforcement learning framework was introduced, using the optimal combination as the action space of the intelligent agent and adjusting the strategy parameters based on safety, stability, and economy as reward signals, achieving adaptive optimization. This approach not only improved design efficiency and quality but also ensured the overall performance and long-term reliability of the building, reduced maintenance costs, and enhanced the durability and flexibility of the structure.

[0054] These and other aspects of the present application will become more readily apparent from the description of the following embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0056] Figure 1 A flowchart of a reinforced concrete support design optimization method provided in an embodiment of the present application;

[0057] Figure 2 A schematic diagram of the structure of an optimization system for reinforced concrete support design provided in an embodiment of the present application;

[0058] Figure 3 A schematic diagram of the structure of a computing device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0059] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0060] In some of the processes described in the specification and claims of this application and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this document or may be executed in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish between different operations, and the serial numbers themselves do not represent any order of execution. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this document are used to distinguish different messages, devices, modules, etc., and do not represent a sequential order, nor do they limit "first" and "second" to being different types.

[0061] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without making creative efforts are within the scope of protection of this application.

[0062] Figure 1 A flowchart of a reinforced concrete support design optimization method is provided for an embodiment of the present application. Figure 1 As shown, the method includes:

[0063] Step 101: Analyze the load expected to be borne by the building in multiple dimensions to determine the key stress points in the building structure;

[0064] In this step, multidimensional analysis involves comprehensive consideration of the various load conditions a building may face during the design phase, including static loads (such as deadweight), dynamic loads (such as wind and seismic forces), and long-term forces (such as temperature fluctuations). Rather than focusing solely on a single type of load, this analysis incorporates multiple factors under different operating conditions, using numerical simulations and theoretical calculations to identify the most vulnerable or critical stress points in the structure. These critical stress points are locations that experience the greatest stress or deformation under specific conditions and are directly related to the safety and stability of the entire building.

[0065] In practice, the process begins by collecting and integrating information from multiple sources, including historical data, geological survey reports, and meteorological data, to construct a comprehensive load database. Next, the building structure is modeled using finite element analysis software. Various possible load combinations are input, and a simulation program is run to generate detailed stress distribution maps. Finally, based on the simulation results, critical locations that experience significant stress under extreme conditions are identified, providing a basis for subsequent design optimization.

[0066] For example, in one high-rise residential project, engineers first collected meteorological data from the area over the past fifty years, including information on wind speed and precipitation, and conducted geological surveys to understand foundation conditions. They then used finite element analysis software to create a 3D model of the entire building structure, inputting all the aforementioned parameters for simulation. The results revealed that the frame columns in the lower floors of the building experienced significant shear forces from strong winds, while the top floors faced primarily vertical compression. Based on this information, the designers decided to employ stronger materials and technical measures in these critical locations to ensure structural safety.

[0067] Step 102: Based on the key stress points, a hybrid modeling technology integrating deep learning and physical rules is used to simulate the behavior characteristics of reinforced concrete materials at different building structural levels to generate a multi-level mechanical model;

[0068] In this step, a hybrid modeling technique integrating deep learning and physical rules is used. This approach combines the powerful pattern recognition capabilities of machine learning with the precise descriptive power of traditional engineering mechanics. Deep learning can process large and complex data sets and automatically extract features from them; physical rules, based on known scientific principles, provide accurate predictions of system behavior. This combination simulates the behavior of reinforced concrete materials at three different scales: microscopic, mesoscopic, and macroscopic, from cement hydration reactions to overall structural response, generating a flexible and reliable multi-level mechanical model.

[0069] In practice, experimental data and theoretical formulas are first prepared as training samples, and a deep neural network is trained to capture the implicit relationship between the material's internal structure and external properties. Simultaneously, a physical model is established based on classical mechanics theory to define the behavioral laws at each level. Next, the two are combined, and appropriate constraints are introduced at each level to ensure that the model can both capture complex behavior and conform to actual physical laws. Finally, through repeated verification and adjustment, the resulting multi-level mechanical model is ensured to accurately reproduce the true properties of reinforced concrete materials.

[0070] For example, continuing with the aforementioned high-rise residential project example, the designers conducted further detailed material property studies based on the key stress points identified in the first step. They collected samples of steel and concrete used at the construction site and conducted comprehensive mechanical tests, acquiring data at all levels from microscopic to macroscopic. The team then developed a hybrid model consisting of a deep learning module and a physical rule module. The former is responsible for learning the implicit relationship between the internal structure and external performance of the material, while the latter ensures that the prediction results conform to basic physical laws. After multiple iterative optimizations, the model successfully simulated the nonlinear response of reinforced concrete under different loading conditions, providing a reliable basis for subsequent design.

[0071] Step 103: Based on the multi-level mechanical model and the data obtained from monitoring the building structure, an intelligent decision-making process is established using a genetic algorithm, a particle swarm optimization, and a reinforcement learning framework to generate a reinforced concrete support design solution;

[0072] In this step, the intelligent decision-making process involves integrating multiple intelligent optimization algorithms, such as genetic algorithms, particle swarm optimization, and reinforcement learning frameworks, to build a system that can automatically explore the optimal solution space and make the best choice. Genetic algorithms are used for global search, particle swarm optimization is used for local refinement, and the reinforcement learning framework is responsible for adjusting strategy parameters based on real-time feedback. This combination not only enables rapid identification of the optimal design solution that meets multiple objectives, but also adapts to changing operating conditions, achieving adaptive optimization.

[0073] In practice, the multi-level mechanical model generated in step 102 is first used as a foundation, combined with actual monitoring data (such as strain gauge readings and displacement sensor outputs) to define initial design variables and constraints. Next, a genetic algorithm is used to generate a random initial population, and through crossover and mutation operations, a progressively optimized solution is evolved. These solutions are then further refined using a particle swarm optimization algorithm to find the local optimal solution. Building on this foundation, a reinforcement learning framework is introduced, treating the current design solution as the action space of an "agent." Using the building's safety, stability, and economic efficiency as reward signals, the strategy parameters are dynamically adjusted until convergence to the global optimal solution. Ultimately, a fully optimized reinforced concrete brace design solution is generated.

[0074] For example, continuing with a previous high-rise residential project, the designers initiated an intelligent decision-making process using a previously established multi-level mechanical model and real-time monitoring data provided by a sensor network installed on-site. They first generated a set of preliminary designs using a genetic algorithm, covering different types of supports and their layout. As the number of iterations increased, the particle swarm optimization algorithm gradually narrowed the search scope and found several potential optimal solutions. At the same time, the reinforcement learning framework continuously adjusted the strategy parameters based on actual monitoring data to ensure that each solution achieved a balance between safety, stability, and economy. Ultimately, the team selected a design solution that met strength requirements while controlling costs, and successfully applied it to actual construction.

[0075] Step 104: Establish a comprehensive evaluation system using Bayesian network and grey system theory, quantitatively analyze the reinforced concrete support design scheme, obtain feedback information, and optimize the reinforced concrete support design scheme using the feedback information to generate an optimized reinforced concrete support design scheme.

[0076] In this step, a Bayesian network, a probabilistic graphical model, is used to represent causal relationships between variables and to update the probability distribution of these relationships based on new observations. Grey system theory is suitable for handling uncertainty and fuzzy information, helping to quantify factors that are difficult to measure directly. Combining these two methods can effectively evaluate the technical feasibility, reliability, and economic efficiency of design solutions, providing objective, quantitative feedback to support further optimization.

[0077] In practice, a series of evaluation indicators covering safety, stability, and economic efficiency are first defined based on existing technology and experience. A Bayesian network is then used to establish logical relationships between these indicators, forming a dynamically updated probabilistic model. Factors that are uncertain or difficult to measure directly are quantified using grey system theory. Next, the performance of different design solutions within this system is compared to identify their strengths and weaknesses. Finally, based on the feedback obtained, design parameters are adjusted specifically until the optimal state is achieved. The entire process emphasizes continuous improvement and closed-loop management to ensure that the design solution is always in optimal condition.

[0078] For example, in a high-rise residential project, designers used the multiple candidate design solutions generated in step 103 to establish a comprehensive evaluation system based on Bayesian networks and grey system theory. They set several evaluation indicators, including compressive strength, seismic performance, and material costs, and obtained relevant data through expert review and field testing. The Bayesian network helped the team clearly demonstrate the interplay between various indicators, while grey system theory effectively addressed the issue of incomplete data for some indicators. Through quantitative analysis, the team discovered that one design solution, while initially costing more, demonstrated greater stability and durability in the long term. Therefore, they decided to adopt this solution and, based on the feedback, further optimized details, such as adjusting the size and layout of supports, ultimately achieving their goals.

[0079] By integrating these four steps, this method not only improves the scientific and rationality of reinforced concrete bracing design solutions but also significantly enhances the intelligence of the design process. Multidimensional analysis ensures the accurate identification of critical stress points, hybrid modeling technology enables comprehensive simulation of material behavior, intelligent decision-making accelerates optimization and improves solution quality, and a comprehensive evaluation system provides strong support for continuous improvement. Ultimately, this approach not only ensures the safety and stability of buildings, but also takes into account economic benefits, reduces maintenance costs, enhances the durability and flexibility of the structure, and provides a new solution for modern construction projects.

[0080] To address the limitations of traditional design methods when faced with complex structures and variable loading conditions, in some embodiments, the step 103 utilizes a genetic algorithm, particle swarm optimization, and reinforcement learning framework to establish an intelligent decision-making process based on the multi-level mechanical model and the data obtained from monitoring the building structure to generate a reinforced concrete support design solution, including:

[0081] The multi-level mechanical model is simulated by finite element analysis to generate simulation data. Based on the simulation data, the behavior patterns and potential risk points of the key stress points under different loading conditions are verified, and the simulation data are classified and regressed using a random forest algorithm to identify the key parameters and change trends that affect the structural performance of the building, and obtain the behavior characteristics of the key stress points; based on the behavior characteristics, the data obtained by monitoring the building structure and the simulation data, the principal component analysis dimensionality reduction technology is applied to extract the key feature vectors, and the time series modeling method is used to analyze the change trend of the building structure response under preset working conditions to identify the dynamic stress conditions, and based on the dynamic stress conditions and the expected load changes of the building structure, a structural response prediction model is generated; based on the structural response prediction model, according to the behavior characteristics and expected load changes of the key stress points Initial supports are selected according to load changes, and the design parameters of the initial supports are encoded using a genetic algorithm to generate an initial population. The initial population is analyzed using a differential evolution algorithm to generate an optimized design parameter set. A fitness function is defined based on the changing trend and dynamic stress conditions of the structural response prediction model, and a particle swarm optimization algorithm is used to iteratively search for the best design parameter combination. Based on the best design parameter combination combined with the optimized design parameter set, an optimal design solution set corresponding to the fitness function is analyzed. Based on the optimal design solution set combined with an ant colony optimization algorithm, an optimal combination of optimized design parameters is generated. According to a reinforcement learning framework, the optimal combination is used as the action space of the intelligent agent in the building structure, and the safety, stability and economy of the building structure are used as reward signals to adjust the strategy parameters of the intelligent agent to generate a reinforced concrete support design solution.

[0082] In this embodiment, finite element analysis is a numerical simulation method that discretizes a continuum into a finite number of units or elements to facilitate the solution of complex physical problems by computer. The simulation data includes information on the behavior patterns of key stress points and potential risk points under different loading conditions. This information is used to verify the theoretical model and provide behavioral characteristics under actual working conditions. The random forest algorithm is an integrated learning method that performs classification and regression prediction by voting on a large number of decision trees. It can effectively process high-dimensional data sets and identify key parameters and changing trends that affect the structural performance of buildings. Principal component analysis is a statistical method used to reduce data dimensions while retaining as much original data information as possible, thereby extracting key feature vectors. Time series modeling helps understand the changes in structural response over time, which is crucial for capturing dynamic stress conditions. The fitness function defines the criteria for evaluating the quality of individuals, and is used here to evaluate the quality of different support design parameters. The ant colony optimization algorithm simulates the foraging behavior of ants and finds the optimal path or solution through a pheromone update mechanism.

[0083] In an embodiment of the present application, a multi-level mechanical model is first simulated by finite element analysis to generate detailed simulation data, and then the random forest algorithm is used to classify and regress these data to determine which factors most affect the structural performance of the building. Next, the principal component analysis dimensionality reduction technique is applied to extract the most important eigenvectors, and combined with the time series modeling method to predict the changing trend of the structural response under preset working conditions. Based on these prediction results, the initial support members are selected and the design parameters are encoded by the genetic algorithm to form an initial population. Subsequently, the differential evolution algorithm is used to analyze the initial population to obtain an optimized design parameter set. The particle swarm optimization algorithm is then used to iteratively search for the best design parameter combination, and the corresponding optimal design solution set is selected from the optimized design parameter set. Finally, the ant colony optimization algorithm is introduced to further refine the best design parameter combination, and the strategy parameters of the intelligent agent are adjusted through the reinforcement learning framework to ensure that the final solution is optimal in terms of safety, stability and economy.

[0084] The following is a specific embodiment:

[0085] In the design project for a large commercial complex, engineers faced a highly complex structure encompassing multiple functional areas. They first constructed a 3D model of the entire building using finite element analysis software and conducted detailed simulations to determine how key load-bearing points would perform under various possible operating conditions. The team then used a random forest algorithm to analyze the simulation data and discovered that the placement of supports in certain locations significantly impacted overall structural performance. To better understand and predict long-term performance, they applied principal component analysis and time series modeling, successfully streamlining the data analysis process and accurately predicting the evolving structural response over the next few years. Based on this, the engineers selected several candidate supports and, through a series of intelligent optimization algorithms, found the optimal design that met strength requirements while controlling costs. Ultimately, using a reinforcement learning framework to continuously adjust strategy parameters, they ensured the safety, stability, and cost-effectiveness of the design, laying a solid foundation for the project's success.

[0086] In order to solve the problem that traditional optimization methods are difficult to take into account multiple objectives under complex working conditions, in some embodiments, step 103 defines a fitness function based on the change trend and dynamic stress conditions of the structural response prediction model, and uses a particle swarm optimization algorithm to iteratively search for an optimal design parameter combination. Based on the optimal design parameter combination and the optimized design parameter set, an optimal design solution set of the fitness function is analyzed. Based on the optimal design solution set and in combination with an ant colony optimization algorithm, an optimal combination of optimized design parameters is generated, which further includes:

[0087] A fitness function is defined based on the changing trends and dynamic stress conditions of the structural response prediction model, combined with various structural factors of the building, including safety, stability, and economy. Based on this fitness function, a particle swarm optimization algorithm and a differential evolution algorithm are employed to iteratively search for the optimal design parameter combination. This algorithm calculates the fitness function value corresponding to each particle in the particle swarm optimization algorithm, and applies adaptive inertia weights to obtain an optimal design parameter set. From this optimal design parameter set, design solutions with fitness function values ​​above a preset threshold are screened to generate an optimal design solution set. Based on this optimal design solution set, the ant colony optimization algorithm is employed to generate the optimal combination of optimized design parameters. Optionally, based on the fitness function, a particle swarm optimization algorithm is used in combination with a differential evolution algorithm to iteratively search for the optimal design parameter combination to calculate the fitness function value corresponding to each particle in the particle swarm optimization algorithm, and an adaptive inertia weight is applied to obtain an optimal design parameter set, and design schemes with fitness function values ​​higher than a preset threshold are screened out from the optimal design parameter set to generate an optimal design scheme set, including: calculating the fitness function value corresponding to each particle in the particle swarm optimization algorithm; applying an adaptive inertia weight adjustment strategy to dynamically adjust the speed update rule of the particles to generate an optimized particle swarm, and obtaining an optimal design parameter set based on the optimized particle swarm; screening out design schemes with fitness function values ​​higher than a preset threshold from the optimal design parameter set, and generating an optimal design scheme set based on the design schemes.

[0088] In this embodiment, the fitness function is a quantitative indicator used to evaluate the quality of an individual. In this embodiment, it comprehensively considers multiple factors such as the safety, stability, and economy of the building structure. Safety includes compressive strength and seismic performance; stability involves structural deformation control and stiffness maintenance; and economy covers factors such as material cost and construction difficulty. These different factors are assigned different weights to form a comprehensive scoring system to guide the optimization process. By combining the changing trends and dynamic stress conditions of the structural response prediction model, the fitness function can more accurately reflect the performance of the design scheme in actual application. In addition, the particle swarm optimization algorithm simulates the foraging behavior of birds and finds the optimal solution through group collaboration, while the differential evolution algorithm introduces mutation operations to increase population diversity. The adaptive inertia weight adjustment strategy dynamically changes the particle speed update rules to help the algorithm strike a balance between exploration and exploitation.

[0089] In the embodiment of the present application, first, according to the changing trend of the structural response prediction model and the dynamic stress conditions, combined with different factors of the building structure, a fitness function is defined. Then, the particle swarm optimization algorithm is combined with the differential evolution algorithm to iteratively search for the best design parameter combination. Specifically, the fitness function value corresponding to each particle is calculated, and the adaptive inertia weight adjustment strategy is applied to dynamically adjust the particle speed update rule to generate an optimized particle swarm. Based on the optimized particle swarm, the best design parameter set is obtained. From this set, design schemes with fitness function values ​​higher than the preset threshold are screened out to generate an optimal design scheme set. Finally, based on the optimal design scheme set, the ant colony optimization algorithm is combined to further optimize the optimal combination of design parameters. The ant colony optimization algorithm enhances the global search capability by simulating the foraging path selection mechanism of ants, ensuring that the final solution not only meets the technical requirements but also maximizes economic benefits.

[0090] The following is a specific embodiment:

[0091] During a large-scale bridge construction project, engineers needed to ensure the bridge remained safe and stable under extreme weather conditions while also managing construction costs. They first developed a structural response prediction model to analyze the bridge's behavior under varying loads. They then defined a fitness function that encompassed safety, stability, and economy, with safety taking the highest priority, followed by stability, and finally economy. The team then used a particle swarm optimization algorithm and differential evolution for joint optimization. In each iteration, the fitness function value for each particle was calculated, and an adaptive inertia weight adjustment strategy was applied to dynamically adjust the particle's velocity update rule to generate an optimized particle swarm. After multiple iterations, a set of optimal design parameters was obtained. From this set, designs with fitness function values ​​exceeding a preset threshold were selected to form the optimal set of design solutions. To further improve the quality of the solutions, the team also employed an ant colony optimization algorithm to fine-tune the optimal design, ultimately generating a safe and economical bridge support design. This solution not only met all technical specifications but also achieved optimal performance within budget, providing strong support for the successful implementation of the project.

[0092] In order to solve the problem that traditional design methods are difficult to effectively handle complex structures and variable load conditions, in some embodiments, the step 103 includes selecting initial supports based on the structural response prediction model, the behavioral characteristics of the key stress points, and the expected load changes, encoding the design parameters of the initial supports using a genetic algorithm to generate an initial population, and analyzing the initial population using a differential evolution algorithm to generate an optimized design parameter set, further comprising:

[0093] According to the structural response prediction model, according to the behavioral characteristic description of the key stress points and the expected load changes, the initial support members suitable for each of the key stress points are selected; the design parameters of the initial support members are encoded using a genetic algorithm to obtain an initial support member set; based on the initial support member set, an initial population is constructed; the differential evolution algorithm is applied to construct and mutate the difference vectors of the individuals in the initial population to obtain an optimized population; based on the optimized population, an optimized design parameter set is generated.

[0094] In this embodiment, the structural response prediction model is a mathematical model based on a multi-level mechanical model and actual monitoring data, used to predict the dynamic behavior of a building under different operating conditions. The behavioral characteristics of key stress points include information such as the stress distribution, deformation patterns, and potential risk points that these locations may exhibit under various loading conditions. The expected load variation covers the various types of loads (such as static loads, dynamic loads, and seismic loads) that a building may encounter during its lifecycle, as well as their changing trends over time. Initial supports are selected based on the above-mentioned prediction model and behavioral characteristics, combined with the expected load variations, to determine the type and size of support components that best suit each key stress point. A genetic algorithm is a global search algorithm that simulates the process of natural selection. It encodes design parameters to form "chromosomes" and generates new solutions through operations such as crossover and mutation. Differential evolution is another evolutionary algorithm that enhances population diversity by constructing difference vectors and introducing mutation operations, thereby improving search efficiency and the ability to find better solutions.

[0095] In an embodiment of the present application, first, based on the structural response prediction model, engineers select initial supports suitable for each key stress point based on the behavioral characteristics description of the key stress points and the expected load changes. Then, a genetic algorithm is used to encode the design parameters of these initial supports to obtain an initial support set. Based on this set, an initial population is constructed, that is, a set of candidate solutions. Next, the differential evolution algorithm is applied to construct and mutate the difference vectors of the individuals in the initial population to generate an optimized population. In this process, the differential evolution algorithm increases the diversity of the population by introducing a mutation mechanism, avoiding premature convergence to a local optimal solution. Finally, based on the optimized population, an optimized design parameter set is generated. These parameters can better adapt to the actual working conditions and ensure that the design scheme is optimal in terms of safety, stability and economy.

[0096] The following is a specific embodiment:

[0097] In a high-rise office building construction project, engineers needed to ensure the building's safety and stability under diverse loads. They first used a structural response prediction model to analyze the building's dynamic behavior under various loading conditions, specifically the performance of critical load points under extreme conditions. Based on these analysis results, the team selected initial support components appropriate for each critical load point, such as using high-strength steel in some locations and prestressed concrete in others. They then used a genetic algorithm to encode the design parameters of these initial support components, forming an initial support set. Based on this set, they constructed an initial population containing multiple candidate solutions. The team then applied a differential evolution algorithm to construct and mutate difference vectors for each individual in the initial population, generating an optimized population. This approach not only increased the diversity of the population but also identified a series of optimal design parameter combinations. Ultimately, based on the optimized population, they generated a set of optimized design parameters that not only met all technical specifications but also achieved optimal performance within budget, strongly supporting the successful implementation of the project. Furthermore, the optimized support design took into account long-term maintenance costs and ease of construction, further enhancing the overall project's economic benefits.

[0098] To address the problem that traditional modeling methods are unable to fully capture the behavioral characteristics of reinforced concrete materials at different levels, in some embodiments, step 102 uses a hybrid modeling technology that integrates deep learning and physical rules based on the key stress points to simulate the behavioral characteristics of reinforced concrete materials at different building structural levels and generate a multi-level mechanical model, including:

[0099] Based on the key stress points, a hybrid modeling technology integrating deep learning and physical rules is used to collect key data of each key stress point to generate a basic data set. The basic data set is used to construct a hybrid modeling framework, and the key data include relevant experimental data, historical monitoring data and theoretical calculation results; based on the hybrid modeling framework, a deep learning model is used to simulate the behavior of reinforced concrete materials at the detail level, predict the performance changes of the reinforced concrete materials under different stress states, and obtain a detail-level material behavior model; according to the detail-level material behavior model, combined with the structural characteristics of reinforced concrete materials at the component level, the behavior of reinforced concrete components at the structural detail level is simulated, and the component-level behavior model is generated by introducing physical rules to guide model training; the behavior in the component-level behavior model is extended to the overall structural level, and the response of the entire building structure under different loading conditions is simulated. The output of the deep learning model is constrained by physical rules to generate an overall structural level response model; based on the detail-level material behavior model, the component-level behavior model and the overall structural level response model, a multi-level mechanical model is established.

[0100] In this embodiment, key data includes relevant experimental data, historical monitoring data, and theoretical calculation results, which are used to describe and predict the performance changes of reinforced concrete materials under various stress states. Experimental data typically comes from laboratory tests, such as compressive strength and elastic modulus. Historical monitoring data comes from long-term monitoring systems of completed buildings, providing information on material performance under actual working conditions. Theoretical calculation results are mathematical formulas and analytical solutions derived from the principles of classical mechanics. The basic dataset integrates these three types of data to form a comprehensive database, which serves as the foundation for subsequent modeling. The hybrid modeling framework combines the powerful pattern recognition capabilities of deep learning with the precise description of physical rules, enabling it to handle complex nonlinear problems while ensuring that model outputs conform to physical laws. The detail-level material behavior model focuses on microscopic simulations, such as the cement hydration reaction process and the interface effects between aggregate and matrix, to predict material performance changes under different stress states. The component-level behavior model considers the overall structural characteristics of reinforced concrete components, such as size, shape, and reinforcement. By incorporating physical rules to guide model training, the accuracy of simulation results is ensured. The overall structural hierarchical response model extends the component-level behavior to the entire building structure, evaluates its response under different loading conditions, and uses physical rules to constrain the output of the deep learning model to ensure the reliability of the model.

[0101] In the embodiment of the present application, first, based on the key stress points, a hybrid modeling technology integrating deep learning and physical rules is adopted to collect key data of each key stress point and generate a basic data set. These data cover information at all levels from micro to macro, forming a comprehensive data resource library. Next, a hybrid modeling framework is constructed using the basic data set, in which the deep learning model is used to capture the implicit relationship between the internal structure and external performance of the material, while the physical rules provide an accurate description of the behavior of the system. Then, based on the hybrid modeling framework, the deep learning model is used to simulate the behavior of reinforced concrete materials at the detail level, predict its performance changes under different stress states, and obtain a detail level material behavior model. According to this model, combined with the structural characteristics of reinforced concrete materials at the component level, the behavior of reinforced concrete components at the structural detail level is simulated, and the model training is guided by introducing physical rules to generate a component level behavior model. Finally, the behavior in the component level behavior model is extended to the overall structural level to simulate the response of the entire building structure under different loading conditions, and the output of the deep learning model is constrained by physical rules to generate an overall structural level response model. Finally, based on the detail-level material behavior model, component-level behavior model, and overall structure-level response model, a complete multi-level mechanical model was established, which can fully reflect the true performance of reinforced concrete materials.

[0102] The following is a specific embodiment:

[0103] In a high-rise residential project, engineers needed to accurately predict the behavior of reinforced concrete at various levels to ensure the building's safety and durability. They first identified several key stress points, such as ground-floor columns and beam-slab joints, and collected relevant experimental data (e.g., compressive strength and elastic modulus), historical monitoring data (e.g., strain gauge readings and displacement sensor outputs), and theoretical calculations (e.g., finite element analysis). This data was integrated into a base dataset to construct a hybrid modeling framework. The team then developed a hybrid model consisting of a deep learning module that learned the implicit relationship between the material's internal structure and external properties, while the physics-based model ensured that the predictions adhered to underlying physical laws. Through multiple iterations of optimization, the model successfully simulated the nonlinear response of reinforced concrete under various loading conditions. Specifically, at the microscopic level, it accurately captured the cement hydration reaction and the aggregate-matrix interface effects. Furthermore, they simulated the behavior of reinforced concrete components at the structural level of detail, generating a component-level behavioral model. This model was then extended to the overall structural level to simulate the response of the entire building under extreme conditions such as earthquakes and wind loads. Ultimately, a multi-level mechanical model was developed based on detail-level material behavior models, component-level behavior models, and overall structural-level response models, providing reliable decision support for the design team. This approach not only improved design accuracy but also provided valuable reference for future maintenance and renovations.

[0104] In order to solve the problem that traditional evaluation methods are difficult to fully quantify the risks and uncertainties of reinforced concrete support design solutions, as well as the lack of an effective feedback mechanism during the optimization process, in some embodiments, the comprehensive evaluation system is established by using Bayesian networks and grey system theory in step 104 to quantitatively analyze the reinforced concrete support design solution, obtain feedback information, and optimize the reinforced concrete support design solution using the feedback information to generate an optimized reinforced concrete support design solution, including:

[0105] Based on the key factors of the reinforced concrete support design scheme, evaluation indicators are defined to construct a comprehensive evaluation index system, wherein the key factors include safety, stability and economy; a risk assessment is performed on the comprehensive evaluation index system using a Bayesian network, and the probability relationship between different risk factors in the reinforced concrete support design scheme is dynamically adjusted to obtain a risk assessment result; the grey system theory is applied to analyze the uncertainty of the reinforced concrete support design scheme, and a grey correlation analysis method is used to generate an uncertainty analysis report; the risk assessment results and the uncertainty analysis report are sorted out to form a comprehensive evaluation system, and the reinforced concrete support design scheme is quantitatively scored according to the comprehensive evaluation system to obtain a quantitative scoring result; the data obtained from monitoring the building structure, expert review opinions and feedback information from relevant parties are analyzed to obtain a feedback information summary; based on the feedback information summary and the quantitative scoring result, the reinforced concrete support design scheme is adjusted using a genetic algorithm, a particle swarm optimization and a reinforcement learning framework to generate an optimized reinforced concrete support design scheme.

[0106] In this embodiment, the comprehensive evaluation index system refers to a series of evaluation indicators defined based on key factors of reinforced concrete bracing design solutions (such as safety, stability, and economy). These indicators are used to measure the performance of design solutions in different aspects, forming a comprehensive evaluation framework. A Bayesian network is a probabilistic graphical model that represents causal relationships between variables and dynamically updates the probability distribution of these relationships based on new observations. It can help identify and assess the mutual influence of different risk factors, providing an evidence-based risk assessment tool. Gray system theory is suitable for handling uncertainty and fuzzy information, especially when data is incomplete or fluctuates significantly. Gray correlation analysis can be used to quantify uncertainty. The risk assessment results include the changes in the probabilistic relationships between risk factors, reflecting the potential risk level of the design solution. The uncertainty analysis report provides a detailed description of the uncertainty of the solution, helping decision makers understand potential biases and unknown factors. The quantitative scoring result is a comprehensive score calculated by scoring the performance of the design solution on each evaluation indicator, which is used to compare the advantages and disadvantages of different solutions.

[0107] In the embodiments of the present application, first, evaluation indicators are defined based on the key factors of the reinforced concrete support design scheme (such as safety, stability, and economy), and a comprehensive evaluation index system is constructed. Next, a Bayesian network is used to conduct a risk assessment on this comprehensive evaluation index system, dynamically adjusting the probabilistic relationship between different risk factors in the design scheme to obtain a risk assessment result. Then, the gray system theory is applied to analyze the uncertainty of the design scheme, and a gray correlation analysis method is used to generate an uncertainty analysis report. The above risk assessment results and uncertainty analysis report are organized to form a comprehensive evaluation system, and the design scheme is quantitatively scored based on this system to obtain a quantitative scoring result. In addition, the team also analyzes data obtained from monitoring the building structure, expert review opinions, and feedback information from relevant parties to obtain a feedback information summary. Finally, based on the feedback information summary and quantitative scoring results, the reinforced concrete support design scheme is adjusted using genetic algorithms, particle swarm optimization, and reinforcement learning frameworks to generate an optimized design scheme.

[0108] The following is a specific embodiment:

[0109] In a large commercial complex design project, engineers needed to ensure the support structure was both safe and economical. They first defined a series of evaluation indicators based on key factors (such as compressive strength, seismic performance, and material cost) and constructed a comprehensive evaluation index system. Next, they used a Bayesian network to conduct a risk assessment of this system, finding that the placement of support components in certain locations significantly affected the safety of the overall structure. They also applied grey system theory to analyze the uncertainties of the design solutions, particularly the fuzzy factors in long-term performance predictions, and generated a detailed uncertainty analysis report. After compiling this information, the team developed a comprehensive evaluation system and quantitatively scored multiple candidate solutions. Furthermore, they collected on-site monitoring data, expert review opinions, and feedback from the owner and other stakeholders, generating a detailed feedback summary. Based on this information, the team used genetic algorithms, particle swarm optimization, and reinforcement learning frameworks to adjust the original design. For example, the genetic algorithm encoded the design parameters and generated an initial population. The particle swarm optimization algorithm iteratively searched for the optimal design parameter combination, while the reinforcement learning framework continuously adjusted the strategy parameters based on feedback to ensure the final solution achieved the optimal combination of safety, stability, and economic efficiency. This approach not only improves the quality of the design solution, but also enhances the feasibility and reliability of the project, laying a solid foundation for the success of the project.

[0110] This application considers that the traditional particle swarm optimization algorithm is prone to falling into local optimal solutions when dealing with complex multi-objective optimization problems, and its parameter settings have a significant impact on the algorithm performance. Therefore, a new optional solution is proposed, which includes:

[0111] Based on the fitness function, an improved particle swarm optimization algorithm combined with a differential evolution algorithm is used to iteratively search for the optimal design parameter combination to calculate the fitness function value corresponding to each particle in the particle swarm optimization algorithm, and an adaptive inertia weight is applied to obtain an optimal design parameter set. Design solutions with fitness function values ​​higher than a preset threshold are screened from the optimal design parameter set to generate an optimal design solution set, including:

[0112] Using the fitness function, according to the improved particle swarm optimization algorithm, the fitness function value corresponding to each particle in the particle swarm optimization algorithm is calculated to evaluate the performance of each particle, and an adaptive inertia weight adjustment strategy is applied to each particle. , calculate the fitness value of the particle according to the fitness function, and introduce the adaptive inertia weight , according to the current number of iterations The inertia weight formula is defined as:

[0113] ;

[0114] in, Indicates in The inertia weight at the iteration, is the base weight, and are the maximum and minimum values ​​of the inertia weight, is the number of intermediate iterations, is the coefficient that controls the rate of change of weights, is the amplitude, is the angular frequency, is the phase difference;

[0115] The following is a detailed explanation of each parameter:

[0116] Base weights represent the minimum value of the underlying weights without additional adjustments. This value is typically set through experimentation or experience. A conservative base value can be chosen based on early experimental results to ensure sufficient exploration capability in the early stages of the algorithm.

[0117] and Indicates the maximum and minimum values ​​of the inertia weight, respectively limiting the range of variation of the inertia weight. Set based on experience with the specific problem. The maximum value should be large enough to ensure strong global search capabilities in the early stages, while the minimum value ensures fine local search in the later stages.

[0118] Represents the number of intermediate iterations, indicating the critical point for transitioning from global search to local search. It can be set to half the expected number of iterations, or the optimal transition point can be determined through preliminary experiments.

[0119] The coefficient that controls the rate of change of weights determines the speed at which the inertia weights change over time. A larger value means a slower rate of change, The value changes faster.

[0120] Amplitude represents the magnitude of the sine wave. It is set based on the random perturbation intensity as needed. The appropriate amplitude value can be determined through experimentation to maintain a certain level of exploration capability.

[0121] This represents the angular frequency, which determines the period length of the sine wave. This is set based on the desired periodic frequency variation. A smaller value is generally chosen to avoid excessive frequency disturbances.

[0122] This represents the phase difference, which determines the initial position of the sine wave. It can be set randomly or adjusted according to the specific situation to ensure a certain degree of variability between different runs.

[0123] The following is an introduction to the design reasons of each sub-item:

[0124] The purpose is to provide a stable basic value to ensure that the inertia weight is not too low, thereby maintaining a certain exploration capability.

[0125] By introducing the Sigmoid function, the inertia weight gradually decreases from a large value to a small value as the number of iterations increases. This change helps to conduct a broad global search in the early stage and turn to a more refined local search in the later stage.

[0126] By introducing sine waves, periodic random disturbances are provided to prevent the algorithm from falling into local optimal solutions and enhance the global search capability.

[0127] By adding up the sub-items, each sub-item represents a different regulatory mechanism, which works together to achieve better optimization results.

[0128] According to the result of the inertia weight, the speed update rule of the particle is dynamically adjusted according to the particle's own historical best position. The particle speed and position are updated with the historical best position gbest of the group to generate the optimized particle swarm, where the speed update formula is:

[0129] ;

[0130] in, Indicates in The particle speed, Indicates in The particle speed, and is the acceleration constant, and is a random number in the interval [0, 1], Indicates in The particle Current location, and is an exponential factor, gbest represents the best historical position in the group, Indicates in The particle new location;

[0131] Among them, the acceleration constant adjustment and The calculation formula is:

[0132] ;

[0133] in, and They are The minimum and maximum values ​​of is the coefficient that controls the rate of change, and They are The minimum and maximum values ​​of is the coefficient that controls the rate of change; where the exponential factor and The calculation formula is:

[0134] ;

[0135] in, and They are The minimum and maximum values ​​of and They are The minimum and maximum values ​​of and is the coefficient that controls the rate of change of the acceleration constant and the exponential factor, and is the amplitude, and is the angular frequency, and is the phase difference.

[0136] The following is a detailed explanation of each parameter:

[0137] Indicates the The particle The speed is calculated by the above formula and depends on current and historical information.

[0138] Indicates the The particle The speed is obtained by the result of the previous iteration and can be set randomly or set to zero during initialization.

[0139] Inertia weight, which is dynamically adjusted in each iteration. It is calculated using the adaptive inertia weight formula.

[0140] and Represents the acceleration constant, which is used to adjust the speed at which a particle moves toward its own historical best position and the group's historical best position. Dynamic calculations are performed using the acceleration constant adjustment formula to ensure appropriate search intensity at different stages.

[0141] and Represents a random number in the interval [0, 1]. This introduces randomness to prevent the algorithm from falling into a local optimum. It is obtained by randomly generating it at each iteration.

[0142] Represents particles The best historical position of the particle itself. This is obtained by updating the particle's fitness value in each iteration.

[0143] Indicates the The particle The current position of . The acquisition method is based on the result of the previous iteration and can be randomly set during initialization.

[0144] and Represents the exponential factor, which is used to adjust the particle The acceleration of gbest movement is obtained by dynamically calculating the exponential factor adjustment formula to enhance the control of the search process.

[0145] gbest represents the best historical position in the population. It is obtained by updating the fitness values ​​of all particles in the entire population.

[0146] Indicates the The particle The new position is obtained by calculating the speed update formula.

[0147] represents the acceleration constant The minimum and maximum values ​​of . The acquisition method is set based on the experience of the specific problem.

[0148] and This is the coefficient that controls the rate of change. It is obtained through experimentation or experience. Smaller values ​​mean slower change, while larger values ​​speed up the change.

[0149] and and Represents the exponential factor and The minimum and maximum values ​​of . The acquisition method is set based on the experience of the specific problem.

[0150] and This is the coefficient that controls the rate of change. It is obtained through experimentation or experience. Smaller values ​​mean slower change, while larger values ​​speed up the change.

[0151] and Indicates the amplitude, which indicates the magnitude of the sine and cosine waves. The acquisition method is to set the random perturbation intensity as needed.

[0152] and Represents the angular frequency, which determines the period length of the sine and cosine waves. The acquisition method is set according to the desired periodic frequency.

[0153] and The phase difference determines the initial position of the sine and cosine waves. The acquisition method can be set randomly or adjusted according to the specific situation.

[0154] The following is an introduction to the design reasons of each sub-item:

[0155] The purpose of the design is to maintain the original motion trend of particles, prevent particles from changing direction too quickly, and help maintain global search capabilities.

[0156] The purpose of the design is to guide particles to move to their own historical best position and enhance individual learning ability. and exponential factors Provides flexibility and diversity.

[0157] The purpose of the design is to guide particles to move to the best historical position of the group and enhance the group's cooperation ability. and exponential factors Provides flexibility and diversity.

[0158] By summing up the various sub-items, each representing a different control mechanism, they work together to achieve better optimization results: The inertia component preserves the particle's original motion trend, ensuring that particles do not suddenly change direction. The individual learning component guides particles toward their own optimal position, enhancing individual learning capabilities. The group collaboration component guides particles toward the group's optimal position, strengthening group collaboration. By summing up, particle velocity updates not only consider the original motion trend but also incorporate historical best information from both individuals and the group, thereby improving the algorithm's exploration and development capabilities.

[0159] Here's a specific example:

[0160] Imagine you are designing a high-rise residential project. You need to choose the most appropriate support arrangement to ensure the building's safety and cost-effectiveness. You have identified several critical load points and constructed a multi-layered mechanical model to simulate the structural response under different loading conditions.

[0161] Parameter settings:

[0162] Initial population size , speed range , location range , inertia weight base value , the maximum value of inertia weight , minimum inertia weight , the number of intermediate iterations , controlling the weight change rate coefficient ,amplitude Angular frequency , phase difference , the minimum value of the acceleration constant , the maximum value of the acceleration constant , control the rate of change coefficient , the minimum exponential factor , the maximum value of the exponential S factor , control the rate of change coefficient ,amplitude , angular frequency , phase difference Calculation process:

[0163] 50 particles are randomly generated, each particle represents a possible support arrangement scheme. The fitness value of each particle is calculated according to the fitness function . Assume that at the first iteration:

[0164] ;

[0165] Assume that the particle Current location and speed are 0.3 and 0.2 respectively, the historical best position of the group gbest is 0.7, and the historical best position of the particle itself is is 0.5. At the second iteration:

[0166] ;

[0167] Assumptions

[0168]

[0169] (If the position range is exceeded, it will be clamped within the range)

[0170] After multiple iterations, the design schemes with fitness function values ​​higher than the preset threshold are screened out from the obtained optimal design parameter set to generate the optimal design scheme set.

[0171] The calculation results show that the improved particle swarm optimization algorithm combined with the differential evolution algorithm not only accelerates the search but also increases the likelihood of finding the global optimal solution. In particular, the application of adaptive inertia weights, dynamically adjusted acceleration constants, and exponential factors enables the algorithm to achieve a good balance between exploration and exploitation, ultimately generating an optimal design solution that meets safety and stability requirements while achieving cost control. This approach provides strong support for the successful implementation of high-rise residential projects and provides valuable reference for future maintenance and renovation.

[0172] Figure 2 A structural diagram of an optimization system for reinforced concrete support design is provided for the embodiment of the present application. Figure 2 As shown, the system includes:

[0173] An analysis module 21 is used to perform multi-dimensional analysis on various loads expected to be borne by the building, so as to determine the key stress points in the building structure;

[0174] A simulation module 22 is configured to simulate the behavior of reinforced concrete materials at different building structural levels based on the key stress points and generate a multi-level mechanical model by using a hybrid modeling technology that integrates deep learning and physical rules;

[0175] a generation module 23 for generating a reinforced concrete support design solution based on the multi-level mechanical model in combination with data obtained from monitoring the building structure and establishing an intelligent decision-making process using genetic algorithms, particle swarm optimization, and reinforcement learning frameworks;

[0176] The quantification module 24 is used to establish a comprehensive evaluation system using Bayesian network and grey system theory, quantitatively analyze the reinforced concrete support design scheme, obtain feedback information, optimize the reinforced concrete support design scheme using the feedback information, and generate an optimized reinforced concrete support design scheme.

[0177] Figure 2 The optimization system for reinforced concrete support design can be executed Figure 1 The implementation principles and technical effects of the reinforced concrete support design optimization method described in the illustrated embodiment are not further elaborated. The specific manner in which the various modules and units perform operations in the reinforced concrete support design optimization system in the above embodiment have been described in detail in the relevant embodiments of the method and will not be further elaborated here.

[0178] In one possible design, Figure 2 The optimization system for reinforced concrete support design of the embodiment shown can be implemented as a computing device, such as Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32;

[0179] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 32 .

[0180] The processing component 32 is used to: perform a multi-dimensional analysis of the load that the building is expected to bear to determine the key stress points in the building structure; based on the key stress points, use a hybrid modeling technology that integrates deep learning and physical rules to simulate the behavioral characteristics of reinforced concrete materials at different building structural levels to generate a multi-level mechanical model; based on the multi-level mechanical model and the data obtained by monitoring the building structure, use genetic algorithms, particle swarm optimization and reinforcement learning frameworks to establish an intelligent decision-making process to generate a reinforced concrete support design scheme; use Bayesian networks and grey system theory to establish a comprehensive evaluation system, quantitatively analyze the reinforced concrete support design scheme, obtain feedback information, and use the feedback information to optimize the reinforced concrete support design scheme to generate an optimized reinforced concrete support design scheme.

[0181] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component may also be implemented as one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above method.

[0182] The storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile memory device, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk, or optical disk.

[0183] Of course, a computing device may also include other components, such as input / output interfaces, display components, communication components, etc.

[0184] The input / output interface provides an interface between the processing component and the peripheral interface module, which can be an output device, an input device, etc.

[0185] The communication component is configured to facilitate, among other things, wired or wireless communications between the computing device and other devices.

[0186] Among them, the computing device can be a physical device or an elastic computing host provided by a cloud computing platform, etc. In this case, the computing device can refer to a cloud server, and the above-mentioned processing components, storage components, etc. can be basic server resources rented or purchased from the cloud computing platform.

[0187] The present application also provides a computer storage medium storing a computer program, wherein the computer program can achieve the above-mentioned Figure 1 The embodiment shown is an optimization method for reinforced concrete support design.

[0188] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0189] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0190] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods described in each embodiment or certain portions of the embodiments.

[0191] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A reinforced concrete support design optimization method, characterized in that: include: Analyze the loads expected to be borne by the building in multiple dimensions to determine the critical stress points in the building structure; Based on the key stress points, a hybrid modeling technology integrating deep learning and physical rules is used to simulate the behavioral characteristics of reinforced concrete materials at different building structural levels and generate a multi-level mechanical model; Based on the multi-level mechanical model and data obtained from monitoring the building structure, an intelligent decision-making process is established using a genetic algorithm, particle swarm optimization, and reinforcement learning framework to generate a reinforced concrete support design solution; establishing a comprehensive evaluation system using Bayesian networks and grey system theory, quantitatively analyzing the reinforced concrete support design scheme, obtaining feedback information, optimizing the reinforced concrete support design scheme using the feedback information, and generating an optimized reinforced concrete support design scheme; The method includes establishing an intelligent decision-making process based on the multi-level mechanical model and data obtained from monitoring the building structure using a genetic algorithm, particle swarm optimization, and reinforcement learning framework to generate a reinforced concrete support design solution, including: Using finite element analysis, the multi-level mechanical model is simulated to generate simulation data. Based on the simulation data, the behavior patterns and potential risk points of the key stress points under different loading conditions are verified. The simulation data are classified and regressed using a random forest algorithm to identify key parameters and change trends that affect the structural performance of the building, thereby obtaining the behavioral characteristics of the key stress points. The data obtained by monitoring the building structure and the simulation data according to the behavioral characteristics are used to extract key feature vectors using principal component analysis dimensionality reduction technology, and a time series modeling method is used to analyze the changing trend of the building structure response under preset working conditions to identify dynamic stress conditions, and a structural response prediction model is generated based on the dynamic stress conditions and the expected load changes of the building structure; Based on the structural response prediction model, initial support members are selected according to the behavioral characteristics of the key stress points and the expected load changes, and design parameters of the initial support members are encoded using a genetic algorithm to generate an initial population. The initial population is analyzed using a differential evolution algorithm to generate an optimized design parameter set; Based on the changing trend and dynamic stress conditions of the structural response prediction model, a fitness function is defined, and an optimal design parameter combination is iteratively searched using a particle swarm optimization algorithm. Based on the optimal design parameter combination and the optimized design parameter set, an optimal design solution set corresponding to the fitness function is analyzed. Based on the optimal design solution set and combined with an ant colony optimization algorithm, an optimal combination of optimized design parameters is generated; According to a reinforcement learning framework, the optimal combination is used as an action space of an agent in the building structure, and the safety, stability, and economy of the building structure are used as reward signals to adjust the strategy parameters of the agent to generate a reinforced concrete support design solution; The method further comprises: defining a fitness function according to a change trend and dynamic stress condition of the structural response prediction model, iteratively searching for an optimal design parameter combination using a particle swarm optimization algorithm, analyzing an optimal design solution set for the fitness function based on the optimal design parameter combination in combination with the optimized design parameter set, and generating an optimal combination of optimized design parameters based on the optimal design solution set in combination with an ant colony optimization algorithm, including: Defining a fitness function based on the change trend and dynamic stress conditions of the structural response prediction model and in combination with different factors of the building structure, wherein the different factors of the building structure include safety, stability, and economy; Based on the fitness function, a particle swarm optimization algorithm and a differential evolution algorithm are used to iteratively search for an optimal design parameter combination to calculate the fitness function value corresponding to each particle in the particle swarm optimization algorithm, and an adaptive inertia weight is applied to obtain an optimal design parameter set. Design solutions with fitness function values ​​higher than a preset threshold are screened from the optimal design parameter set to generate an optimal design solution set; Based on the optimal design solution set, combined with the ant colony optimization algorithm, the best combination of optimized design parameters is generated.

2. The method according to claim 1, characterized in that Based on the fitness function, a particle swarm optimization algorithm combined with a differential evolution algorithm is used to iteratively search for an optimal design parameter combination to calculate the fitness function value corresponding to each particle in the particle swarm optimization algorithm, and an adaptive inertia weight is applied to obtain an optimal design parameter set. Design solutions with fitness function values ​​higher than a preset threshold are screened from the optimal design parameter set to generate an optimal design solution set, including: Calculating the fitness function value corresponding to each particle in the particle swarm optimization algorithm; Applying an adaptive inertia weight adjustment strategy to dynamically adjust the particle velocity update rule to generate an optimized particle swarm, and obtaining an optimal design parameter set based on the optimized particle swarm; The design schemes whose fitness function values ​​are higher than a preset threshold are screened out from the optimal design parameter set, and an optimal design scheme set is generated based on the design schemes.

3. The method according to claim 1, characterized in that The method comprises: selecting initial support members based on the structural response prediction model and the behavioral characteristics of the key stress points and the expected load changes; encoding the design parameters of the initial support members using a genetic algorithm to generate an initial population; and analyzing the initial population using a differential evolution algorithm to generate an optimized design parameter set, including: Selecting initial supports suitable for each of the key stress points according to the structural response prediction model, the behavioral characteristics of the key stress points, and the expected load changes; Encoding the design parameters of the initial support parts using a genetic algorithm to obtain an initial support part set, constructing an initial population based on the initial support part set, and applying a differential evolution algorithm to construct and mutate difference vectors of individuals in the initial population to obtain an optimized population; Based on the optimized population, an optimized design parameter set is generated.

4. The method according to claim 1, wherein Based on the key stress points, a hybrid modeling technology integrating deep learning and physical rules is used to simulate the behavioral characteristics of reinforced concrete materials at different building structural levels to generate a multi-level mechanical model, including: Based on the key stress points, a hybrid modeling technology integrating deep learning and physical rules is used to collect key data of each key stress point, generate a basic data set, and use the basic data set to build a hybrid modeling framework. The key data includes relevant experimental data, historical monitoring data, and theoretical calculation results; Based on the hybrid modeling framework, a deep learning model is used to simulate the behavior of reinforced concrete materials at the detail level, predict the performance changes of the reinforced concrete materials under different stress states, and obtain a detail-level material behavior model; Based on the detailed material behavior model, combined with the structural characteristics of reinforced concrete materials at the component level, the behavior of reinforced concrete components at the structural detail level is simulated, and physical rules are introduced to guide model training to generate a component-level behavior model; Extending the behavior in the component-level behavior model to the overall structural level, simulating the response of the entire building structure under different loading conditions, and using physical rules to constrain the output of the deep learning model to generate an overall structural response model; A multi-level mechanical model is established based on the detail-level material behavior model, the component-level behavior model, and the overall structure-level response model.

5. The method according to claim 1, wherein The method of establishing a comprehensive evaluation system using Bayesian network and grey system theory, quantitatively analyzing the reinforced concrete support design scheme, obtaining feedback information, optimizing the reinforced concrete support design scheme using the feedback information, and generating an optimized reinforced concrete support design scheme includes: Defining evaluation indicators based on key factors of the reinforced concrete support design scheme to construct a comprehensive evaluation indicator system, wherein the key factors include safety, stability, and economy; Using a Bayesian network, the comprehensive evaluation index system is risk evaluated, and the probability relationship between different risk factors in the reinforced concrete support design scheme is dynamically adjusted to obtain a risk assessment result; Applying grey system theory to analyze the uncertainty of the reinforced concrete support design scheme, and using grey correlation analysis method to generate an uncertainty analysis report; Arranging the risk assessment results and the uncertainty analysis report to form a comprehensive evaluation system, and quantitatively scoring the reinforced concrete support design scheme based on the comprehensive evaluation system to obtain a quantitative scoring result; Analyze data obtained from monitoring the building structure, expert review opinions, and feedback from stakeholders to obtain a summary of feedback information; Based on the feedback information summary and the quantitative scoring results, the reinforced concrete support design scheme is adjusted using a genetic algorithm, a particle swarm optimization, and a reinforcement learning framework to generate an optimized reinforced concrete support design scheme.

6. An optimization system for reinforced concrete support design, characterized in that: include: An analysis module for performing multi-dimensional analysis of the various loads expected to be borne by the building to determine the critical stress points in the building structure; A simulation module is used to simulate the behavior characteristics of reinforced concrete materials at different building structural levels based on the key stress points and generate a multi-level mechanical model using a hybrid modeling technology that integrates deep learning and physical rules; a generation module for generating a reinforced concrete support design based on the multi-level mechanical model in combination with data obtained from monitoring the building structure and establishing an intelligent decision-making process using genetic algorithms, particle swarm optimization, and reinforcement learning frameworks; a quantification module, configured to establish a comprehensive evaluation system using Bayesian networks and grey system theory, quantitatively analyze the reinforced concrete support design scheme, obtain feedback information, optimize the reinforced concrete support design scheme using the feedback information, and generate an optimized reinforced concrete support design scheme; The method includes establishing an intelligent decision-making process based on the multi-level mechanical model and data obtained from monitoring the building structure using a genetic algorithm, particle swarm optimization, and reinforcement learning framework to generate a reinforced concrete support design solution, including: Using finite element analysis, the multi-level mechanical model is simulated to generate simulation data. Based on the simulation data, the behavior patterns and potential risk points of the key stress points under different loading conditions are verified. The simulation data are classified and regressed using a random forest algorithm to identify key parameters and change trends that affect the structural performance of the building, thereby obtaining the behavioral characteristics of the key stress points. The data obtained by monitoring the building structure and the simulation data according to the behavioral characteristics are used to extract key feature vectors using principal component analysis dimensionality reduction technology, and a time series modeling method is used to analyze the changing trend of the building structure response under preset working conditions to identify dynamic stress conditions, and a structural response prediction model is generated based on the dynamic stress conditions and the expected load changes of the building structure; Based on the structural response prediction model, initial support members are selected according to the behavioral characteristics of the key stress points and the expected load changes, and design parameters of the initial support members are encoded using a genetic algorithm to generate an initial population. The initial population is analyzed using a differential evolution algorithm to generate an optimized design parameter set; Based on the changing trend and dynamic stress conditions of the structural response prediction model, a fitness function is defined, and an optimal design parameter combination is iteratively searched using a particle swarm optimization algorithm. Based on the optimal design parameter combination and the optimized design parameter set, an optimal design solution set corresponding to the fitness function is analyzed. Based on the optimal design solution set and combined with an ant colony optimization algorithm, an optimal combination of optimized design parameters is generated; According to a reinforcement learning framework, the optimal combination is used as an action space of an agent in the building structure, and the safety, stability, and economy of the building structure are used as reward signals to adjust the strategy parameters of the agent to generate a reinforced concrete support design solution; The method further comprises: defining a fitness function according to a change trend and dynamic stress condition of the structural response prediction model, iteratively searching for an optimal design parameter combination using a particle swarm optimization algorithm, analyzing an optimal design solution set for the fitness function based on the optimal design parameter combination in combination with the optimized design parameter set, and generating an optimal combination of optimized design parameters based on the optimal design solution set in combination with an ant colony optimization algorithm, including: Defining a fitness function based on the change trend and dynamic stress conditions of the structural response prediction model and in combination with different factors of the building structure, wherein the different factors of the building structure include safety, stability, and economy; Based on the fitness function, a particle swarm optimization algorithm and a differential evolution algorithm are used to iteratively search for an optimal design parameter combination to calculate the fitness function value corresponding to each particle in the particle swarm optimization algorithm, and an adaptive inertia weight is applied to obtain an optimal design parameter set. Design solutions with fitness function values ​​higher than a preset threshold are screened from the optimal design parameter set to generate an optimal design solution set; Based on the optimal design solution set, combined with the ant colony optimization algorithm, the best combination of optimized design parameters is generated.

7. A computing device, characterized in that It comprises a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement an optimization method for reinforced concrete support design as described in any one of claims 1 to 5.

8. A computer storage medium, characterized in that A computer program is stored, and when the computer program is executed by a computer, the optimization method for reinforced concrete support design according to any one of claims 1 to 5 is implemented.

Images