Engineering structural design method, medium and equipment of ecc-concrete composite beam

Abstract

The application relates to the technical field of architectural design, in particular to an ECC-concrete composite beam engineering structure design method, medium and equipment, which comprises the following steps: construction and integration of a special database; intelligent prediction model building: a distributed gradient boosting library and a classification feature enhancement agent model optimized through system evaluation are arranged on a model layer, and the model layer is composed of three modules, namely a generator, an evaluator and an optimizer, which work cooperatively; automatic generation of a composite beam structure scheme; autonomous optimization and evaluation of material proportioning; mechanical response prediction and damage interval division; and output of achievements in an application layer: determination of an optimal structure scheme and a damage monitoring interval based on a bending moment-deflection curve. The application solves the performance and cost imbalance problem caused by single target optimization in traditional design, and realizes the collaborative optimization of ECC material proportioning and composite beam structure design.

Description

Technical Field

[0001] This invention relates to the field of architectural design technology, and in particular to an engineering structural design method, medium, and equipment for ECC-concrete composite beams. Background Technology

[0002] Key structures in thermal power plants, such as main plant buildings, boiler foundations, and cooling towers, are subjected to harsh environments involving high temperatures, vibrations, corrosion, and heavy loads. New energy power plants also face demands for large deformations, high durability, and long service life. Both types of power structures place higher standards on the load-bearing capacity, ductility, crack resistance, and durability of concrete components. Traditional concrete has low tensile strength, high brittleness, and is prone to cracking. Under alternating loads and extreme environments, it easily develops through-cracks, leading to steel corrosion, stiffness degradation, and durability decline, making it difficult to meet the long-term safe operation requirements of thermal power structures and the full life-cycle service requirements of new energy structures. Existing mix design often relies on experience or single-objective optimization, making it difficult to balance strength, deformation, and cost. Structural design and material selection are disconnected, failing to achieve optimal synergy between performance and economy.

[0003] High-ductility concrete (ECC), with its strain hardening, multi-crack characteristics, high ductility, and high corrosion resistance, offers a new material pathway for upgrading power engineering structures. However, existing ECC formulations and applications have significant shortcomings: mix proportions rely on trial mixing, have a single objective, and do not consider the actual stress and service scenarios of power structures. There is a lack of integrated intelligent design and performance-cost evaluation methods for thermal power, new energy, and other scenarios, and insufficient deep integration of structure and materials limits its large-scale application in power engineering. Summary of the Invention

[0004] The main objective of this invention is to provide an engineering structural design method, medium, and equipment for ECC-concrete composite beams, aiming to solve the technical problems of existing ECC formulations, such as reliance on trial mixing, single objective, lack of consideration for the actual stress and service scenarios of power structures, and insufficient deep integration of structure and materials.

[0005] To achieve the above objectives, this invention proposes an engineering structural design method for ECC-concrete composite beams, comprising the following steps:

[0006] S1. Construction and integration of dedicated databases;

[0007] S2. Intelligent prediction model construction: The model layer deploys a distributed gradient boosting library and a classification feature enhancement proxy model that have been evaluated and optimized by the system. It consists of three modules that work together: generator, evaluator and optimizer.

[0008] S3. Automatic generation of composite beam structure schemes: Based on the user-input load-bearing capacity improvement target and structural constraints, the generator automatically explores the design space and generates several sets of candidate ECC structure schemes.

[0009] S4. Autonomous optimization and evaluation of material proportions: Taking the material performance parameters output by the ECC structural scheme obtained in S3 as the target, the optimizer automatically starts a multi-objective optimization process and generates a Pareto standard optimal proportion solution set under the constraints of a dedicated database. Subsequently, based on preset performance indicators, it automatically traverses and quantifies all solutions on the Pareto standard frontier, autonomously identifies and locks the globally optimal material design scheme.

[0010] S5. Mechanical Response Prediction and Damage Range Division: Based on the structural parameters determined by the ECC structural scheme obtained in S3, the system automatically predicts the yield capacity, peak capacity and corresponding deflection of the composite beam; according to the prediction results, the system generates a complete moment-deflection curve and automatically divides the beam into three quantitative damage ranges based on the deflection threshold.

[0011] S6. Output results at the application layer: Based on the globally optimal material design scheme and damage range, output the optimal ECC material mix ratio, structural design parameters of composite beam, determine the optimal structural scheme, and damage monitoring range based on the moment-deflection curve.

[0012] The engineering structural design method for ECC-concrete composite beams of the present invention is further improved in that the dedicated database includes an ECC material mix proportion database and an ECC-concrete composite beam database. The ECC-concrete composite beam database includes twelve sets of input variables and five structural response indicators.

[0013] The twelve sets of input variables are: beam section height, beam section width, beam span, distance from the point of application of concentrated force to the beam support, longitudinal reinforcement ratio of the upper part of the beam, reinforcement ratio of the lower part of the beam, stirrup diameter, stirrup spacing, ECC layer thickness at the bottom of the beam, ECC tensile strength, ECC tensile strain and ECC compressive strength.

[0014] The five structural response indicators are: initial stiffness, yield strength, deflection corresponding to yield strength, peak strength, and deflection corresponding to peak strength of the ECC-concrete composite beam.

[0015] The engineering structural design method for ECC-concrete composite beams of the present invention is further improved in that S2 specifically includes:

[0016] S201, Generator: Receives user-input constraints, automatically samples within the design space, and generates an initial set of candidate solutions;

[0017] S202, Evaluator: Real-time call to the proxy model of the model layer to perform batch predictions on each candidate solution produced by the generator, and feedback the fitness value to provide a quantitative basis for optimization iteration;

[0018] S203, Optimizer: Built-in multi-objective optimization algorithm, with the dual objectives of maximizing structural performance and minimizing material cost, drives the evolution of the solution; in each iteration, the optimizer automatically filters non-dominated solutions based on the fitness value fed back by the evaluator, guiding the search direction to converge toward the Pareto standard front.

[0019] The engineering structural design method of the ECC-concrete composite beam of the present invention is further improved in that the optimized distributed gradient boosting library and the classification feature enhancement surrogate model in S2 are both subjected to hyperparameter optimization through random search and Fourier optimization.

[0020] The engineering structural design method for ECC-concrete composite beams of the present invention is further improved in that the hyperparameters in the optimized distributed gradient boosting library in S2 are: n_estimators=600, max_depth=3, learning_rate=0.16, booster='gbtree', gamma=0.5, reg_alpha=0.1, reg_lambda=0, min_child_weight=3, subsample=0.9, colsample_bytree=0.8, random_state=200; and the hyperparameters for classification feature enhancement are: iterations=400, depth=3, learning_rate=0.1, loss_function='Poisson', od_type='IncToDec', od_wait=50.

[0021] The engineering structural design method of ECC-concrete composite beam of the present invention is further improved in that the multi-objective optimization algorithm in S203 is a multi-objective optimization method based on multi-objective particle swarm optimization algorithm; the multi-objective optimization algorithm aims to maximize the compressive strength of ECC, maximize the tensile strength of ECC and minimize the material cost, and the constraints limit the total mass of ECC raw materials and aggregate content parameters. Based on the target tensile strength and tensile strain requirements, a set of Pareto standard optimal solutions is generated.

[0022] The engineering structural design method of the ECC-concrete composite beam of the present invention is further improved in that the ECC structural scheme includes different ECC layer thicknesses, ECC tensile strengths, ECC tensile strains and ECC compressive strengths.

[0023] The engineering structural design method for ECC-concrete composite beams of the present invention is further improved in that the optimal structural scheme utilizes the obtained ECC mix proportion cost. With ECC layer thickness The performance index of the ECC concrete composite beam is calculated using the following expression:

[0024] ;

[0025] in, and These are the width and length of the composite beam, respectively. For performance index; The smaller the value, the better the corresponding design scheme.

[0026] In addition, the present invention provides a readable storage medium storing a computer program adapted to be loaded by a processor and executed as described above for the engineering structural design method of ECC-concrete composite beams.

[0027] In addition, the present invention also provides a computer device, the computer device including a memory and a processor, the memory storing a computer program, which, when executed by the processor, runs the engineering structure design method for ECC-concrete composite beams as described above.

[0028] The technical solution of the present invention has the following beneficial effects:

[0029] The engineering structural design method for ECC-concrete composite beams of this invention, by deploying an optimized distributed gradient boosting library (XGBoost) and a classification feature enhancement (CatBoost) proxy model that has been systematically evaluated and selected, along with three collaborative modules—generator, evaluator, and optimizer—automates the entire process of candidate scheme generation, batch prediction of mechanical performance and material cost, iterative optimization of schemes, and optimization of material proportions. This eliminates the need for extensive manual trial mixing and tedious calculations, significantly reducing the workload of designers, shortening the design cycle, and effectively avoiding design errors caused by human experience bias, thus greatly improving the accuracy and reliability of the design scheme. Employing a multi-objective optimization algorithm based on multi-objective particle swarm optimization (MOPSO), with the core objectives of maximizing structural bearing capacity and minimizing material cost, and combining the mechanical performance requirements of ECC materials to generate a Pareto optimal solution set, then quantifying and scoring it using a preset performance index (PI), ultimately locking in the globally optimal design scheme. This effectively solves the performance-cost imbalance problem caused by single-objective optimization in traditional design, achieving synergistic optimization of ECC material proportions and composite beam structural design. This method achieves a closed-loop intelligent design process, encompassing ECC material mix optimization, composite beam structure design, mechanical response prediction, and damage zone classification. The output of optimal ECC mix proportions and composite beam structural parameters directly guides engineering construction without requiring additional scheme optimization. Based on the predicted moment-deflection curves, and according to the deflection corresponding to yield strength and peak strength, it automatically classifies damage zones into three categories: no repair required, repairable, and unrepairable. This clearly defines the service status of the composite beam, providing clear quantitative standards for later monitoring, maintenance, and repair, thus helping to extend the structural service life and reduce operation and maintenance costs. The method is highly adaptable, flexibly generating suitable design schemes based on different constraints such as user-input cross-sectional dimensions, reinforcement limitations, and durability requirements. It is particularly suitable for power engineering scenarios such as new energy power generation and thermal power, which have high requirements for structural bearing capacity, durability, and economy. It provides reliable technical support for structural upgrades in the construction of new power systems and has broad engineering application value and promotion prospects. Attached Figure Description

[0030] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0031] Figure 1 This is a flowchart of the engineering structural design method for the ECC-concrete composite beam of the present invention. Detailed Implementation

[0032] The technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0033] Furthermore, in this invention, descriptions involving "first," "second," etc., are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0034] In this invention, unless otherwise explicitly specified and limited, the terms "connection," "fixed," etc., should be interpreted broadly. For example, "fixed" can mean a fixed connection, a detachable connection, or an integral part; it can mean a mechanical connection or an electrical connection; it can mean a direct connection or an indirect connection through an intermediate medium; it can mean the internal communication of two components or the interaction between two components, unless otherwise explicitly limited. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0035] Furthermore, the technical solutions of the various embodiments of the present invention can be combined with each other, but only if they are feasible for those skilled in the art. If the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such combination of technical solutions does not exist and is not within the scope of protection claimed by the present invention.

[0036] like Figure 1 As shown, this invention proposes an engineering structural design method for ECC-concrete composite beams, comprising the following steps:

[0037] S1. Construction and Integration of Dedicated Databases: The data layer is the foundation of the intelligent design system. Two high-quality dedicated databases were specifically constructed and integrated: the ECC material mix proportion database (covering 18 input variables and 3 mechanical performance indicators) and the ECC-concrete composite beam database (covering 12 input variables and 5 structural response indicators). These two databases provide complete data support for subsequent model training and optimization.

[0038] The 18 input variables in the ECC material mix design database are: cement, water, silica sand, other sand, coarse aggregate, fiber volume fraction, fiber weight, water-binder ratio, total amount of mineral admixtures, silica fume, slag, fly ash, other active mineral admixtures, water-reducing agent, fiber diameter, fiber length, fiber elongation at break, and fiber tensile strength. The three mechanical properties are ECC tensile strength, ECC tensile strain, and ECC compressive strength.

[0039] The 12 sets of input variables in the ECC-concrete composite beam database are: beam section height, beam section width, beam span, distance from the point of application of concentrated force to the beam support, longitudinal reinforcement ratio of the upper part of the beam, reinforcement ratio of the lower part of the beam, stirrup diameter, stirrup spacing, ECC layer thickness at the bottom of the beam, ECC tensile strength, ECC tensile strain, and ECC compressive strength.

[0040] The five structural response indices of the ECC-concrete composite beam database are: initial stiffness, yield strength, deflection at yield strength, peak strength, and deflection at peak strength.

[0041] S2. Intelligent Prediction Model Construction: The model layer deploys a distributed gradient boosting library (XGBoost) and a classification feature enhancement (CatBoost) surrogate model that have undergone systematic evaluation and optimization. Both XGBoost and CatBoost surrogate models have undergone hyperparameter optimization through random search and Fourier optimization, enabling them to achieve an accuracy of over 95% in predicting ECC material properties and ECC-concrete composite beams. The intelligent prediction model consists of three collaborative modules: a generator, an evaluator, and an optimizer.

[0042] S201, Generator: Receives user-input constraints (including cross-sectional size limits, reinforcement limits, durability requirements, and load-bearing capacity improvement targets), automatically samples within the design space, and generates an initial candidate scheme set (covering ECC layer thickness and its corresponding material performance parameters).

[0043] S202, Evaluator: Real-time calls the proxy model of the model layer to perform batch predictions on each candidate solution produced by the generator and feed back fitness values. Fitness values ​​include mechanical properties (bearing capacity, deflection, etc.) and material costs, providing a quantitative basis for optimization iteration;

[0044] S203, Optimizer: Built-in multi-objective optimization algorithm with the dual objectives of maximizing structural performance and minimizing material cost to drive the evolution of the solution; in each iteration, the optimizer automatically filters non-dominated solutions based on the fitness value fed back by the evaluator, guiding the search direction to converge toward the Pareto front.

[0045] The hyperparameters in the optimized distributed gradient boosting library (XGBoost) in S2 are:

[0046] n_estimators=600, max_depth=3, learning_rate=0.16, booster='gbtree', gamma=0.5, reg_alpha=0.1, reg_lambda=0, min_child_weight=3, subsample=0.9, colsample_bytree=0.8, random_state=200; The hyperparameters for CatBoost are: iterations=400, depth=3, learning_rate=0.1, loss_function='Poisson', od_type='IncToDec', od_wait=50.

[0047] Specifically, the multi-objective optimization algorithm in S203 is a multi-objective optimization method based on the multi-objective particle swarm optimization algorithm. The multi-objective optimization algorithm aims to maximize the compressive strength of ECC, maximize the tensile strength of ECC, and minimize the material cost. The constraints limit the total mass of ECC raw materials and the aggregate content parameters. Based on the target tensile strength and tensile strain requirements, a set of Pareto optimal solutions is generated.

[0048] S3. Automatic generation of composite beam structure schemes: Based on the user-input load-bearing capacity improvement target and structural constraints, the generator automatically explores the design space and generates several sets of candidate ECC structure schemes; the ECC structure schemes include different ECC layer thicknesses, ECC tensile strength, ECC tensile strain and ECC compressive strength.

[0049] S4. Autonomous optimization and evaluation of material proportions: Taking the material performance parameters output by the ECC structural scheme obtained in S3 as the target, the optimizer automatically starts a multi-objective optimization process and generates a Pareto standard optimal proportion solution set under the constraints of a dedicated database. Subsequently, based on preset performance indicators, it automatically traverses and quantifies all solutions on the Pareto standard frontier, autonomously identifies and locks the globally optimal material design scheme.

[0050] S5. Mechanical Response Prediction and Damage Range Division: Based on the structural parameters determined by the ECC structural scheme obtained in S3, the system automatically predicts the yield strength, peak load capacity, and corresponding deflection of the composite beam. According to the prediction results, the system generates a complete moment-deflection curve and automatically divides the beam into three quantitative damage ranges based on the deflection threshold. The moment-deflection curve is used to describe the stress process of the beam, and is divided into elastic and plastic stages according to the yield strength and corresponding deflection, and peak strength and corresponding deflection of the composite beam.

[0051] Based on the defined moment-deflection curve, the damage level of the beam can be classified:

[0052] Before the yield point, the beam only develops minor cracks that do not affect its normal performance. Therefore, the damage state before yielding is classified as the area that does not require repair.

[0053] After reaching the yield point, the beam begins to undergo irreversible plastic deformation, and the cracks gradually expand, and the reinforcing steel may enter the strengthening stage; the damage in this stage is still controllable and is classified as a repairable range.

[0054] After reaching peak load capacity, damage and deformation continue to increase, resulting in irreversible damage that is difficult to control. Therefore, this stage is classified as the irreparable zone.

[0055] S6. Output results at the application layer: Based on the globally optimal material design scheme and damage range, output the optimal ECC material mix ratio, structural design parameters of composite beam, determine the optimal structural scheme, and damage monitoring range based on the moment-deflection curve, realizing intelligent closed-loop design of the entire process from "material-structure-performance-monitoring".

[0056] Specifically, the optimal structural scheme utilizes the obtained ECC mix proportion cost. With ECC layer thickness The performance index of the ECC concrete composite beam is calculated using the following expression:

[0057] ;

[0058] in, and These are the width and length of the composite beam, respectively. For performance index; The smaller the value, the better the corresponding design scheme.

[0059] In addition, the present invention provides a readable storage medium storing a computer program adapted to be loaded by a processor and executed as described above for the engineering structural design method of ECC-concrete composite beams.

[0060] In addition, the present invention also provides a computer device, the computer device including a memory and a processor, the memory storing a computer program, which, when executed by the processor, runs the engineering structure design method for ECC-concrete composite beams as described above.

[0061] This invention provides a specific application example of a beam specimen, which includes the following:

[0062] This invention uses a beam specimen with a length of 3m and a cross-sectional dimension of 440mm × 200mm. The stirrups have a diameter of 8mm and a spacing of 150mm; two longitudinal reinforcing bars with a diameter of 16mm are placed at the top and bottom of the beam specimen. Furthermore, two concentrated load points are applied to the beam specimen, each 1000mm from its corresponding support. The bearing capacity of this beam specimen is approximately 72kN. Due to the need to increase the bearing capacity and the limitations imposed by the beam's cross-sectional dimensions, its bearing capacity needs to be increased by 50%.

[0063] The generator module at the system application layer, driven by a target bearing capacity of 108 kN, automatically samples within the design space. Based on the XGBoost surrogate model at the model layer, the generator rapidly iterates to explore feasible combinations of ECC layer thickness and ECC tensile strength, generating a set of candidate solutions. To facilitate subsequent experimental verification and discussion, five representative sampling points are selected from the continuous Pareto front generated by the system for presentation:

[0064] ;

[0065] ;

[0066] ;

[0067] ;

[0068] ;

[0069] in: ~ Each represents one of the five design schemes. Indicates the tensile strength of ECC. ~ These represent five groups of ECC layers with different thicknesses. These five schemes are merely representative points in the design space generated by the system; the complete Pareto front will be explored autonomously by the optimizer in subsequent steps.

[0070] The system then enters the material optimization module.

[0071] Objective function: A three-objective optimization framework is established, including: 1) maximizing ECC compressive strength; 2) maximizing ECC tensile strength; 3) minimizing material cost.

[0072] Constraints: All constraints are handled using the penalty function method to ensure the feasibility of the optimization scheme: 1) Economical polyethylene fiber is preferred; 2) The cement usage per cubic meter is constrained to be <650kg; 3) The total mass of ECC raw materials per cubic meter is constrained to be 1300 kg~2500 kg, which is determined based on the workability and density requirements of ECC materials in actual engineering; 4) The usage of other sand and coarse aggregates per cubic meter is constrained to be 0, which is determined based on the material composition characteristics of high ductility ECC and the consistency of the collected dataset to ensure the multi-crack and strain hardening performance of ECC; 5) The remaining mix proportion variables are constrained within the minimum and maximum values ​​of the collected dataset.

[0073] The optimizer employs the Multi-Objective Particle Swarm Optimization (MOPSO) algorithm, with the following core parameters set: number of particles n_particles=100, number of iterations n_iters=200, and inertia weight. =0.65, acceleration coefficients c1=1.6 and c2=1.6, velocity limit v_clamp=0.25, archive_size=300. After multiple iterations, the system automatically converges to the Pareto front, generating the optimal mix design corresponding to the five target intensity values, as shown in Table 1:

[0074] Table 1. Five design schemes (composition of ECC per cubic meter)

[0075]

[0076] The accuracy of the five sets of generated results was determined by experiments. The specific experimental results are shown on the right side of Table 1. The error between the measured tensile strength and the target strength predicted by the system is within 10%, which verifies the accuracy and robustness of the proposed system in multi-objective optimization.

[0077] The system enters the final decision-making stage. Based on preset performance indicators ( The evaluator within the application layer automatically traverses all feasible solutions on the Pareto front and performs quantification scoring. Performance metrics ( This comprehensively reflects the combined benefits of material cost and structural design; the smaller the value, the better the solution. The evaluation results show that Solution 1 has the lowest performance index value. The system autonomously identifies and locks Solution 1 as the globally optimal solution, and then outputs it to the user interface. Thus, the system completes a fully automated closed-loop process from "target input" to "optimal solution output."

[0078] Based on the load-bearing capacity prediction results of Scheme 1, the yield strength and corresponding deflection, peak strength and corresponding deflection of the ECC-concrete composite beam were derived, and the damage range was defined based on these values. In summary, for subsequent deformation monitoring of the composite beam: if the deflection is less than 12.5 mm, the beam is basically in the elastic stage and requires no repair; if the deflection is between 12.5 mm and 54.6 mm, the beam undergoes plastic deformation and is in a repairable state; if the deflection exceeds 54.6 mm, due to excessive damage and deformation of both the reinforcement and concrete, the beam is determined to be unrepairable. Specific results are shown in Table 2.

[0079] Table 2. Damage Zone Classification of ECC-Concrete

[0080]

[0081] The above description is only a preferred embodiment of the present invention and does not limit the scope of the present invention. All equivalent structural transformations made using the contents of the present invention under the inventive concept of the present invention, or direct / indirect applications in other related technical fields, are included within the protection scope of the present invention.

Claims

1. An engineering structural design method for ECC-concrete composite beams, characterized in that, Includes the following steps: S1. Construction and integration of dedicated databases; S2. Intelligent prediction model construction: The model layer deploys a distributed gradient boosting library and a classification feature enhancement proxy model that have been evaluated and optimized by the system. It consists of three modules that work together: generator, evaluator and optimizer. S3. Automatic generation of composite beam structure schemes: Based on the user-input load-bearing capacity improvement target and structural constraints, the generator automatically explores the design space and generates several sets of candidate ECC structure schemes. S4. Autonomous optimization and evaluation of material proportions: Taking the material performance parameters output by the ECC structural scheme obtained in S3 as the objective, the optimizer automatically starts a multi-objective optimization process and generates a Pareto standard optimal proportion solution set under the constraints of a dedicated database. Subsequently, based on preset performance indicators, all solutions on the Pareto standard front are automatically traversed and quantified, and the globally optimal material design scheme is autonomously identified and locked. S5. Mechanical Response Prediction and Damage Range Division: Based on the structural parameters determined by the ECC structural scheme obtained in S3, the system automatically predicts the yield capacity, peak capacity and corresponding deflection of the composite beam; according to the prediction results, the system generates a complete moment-deflection curve and automatically divides the beam into three quantitative damage ranges based on the deflection threshold. S6. Output results at the application layer: Based on the globally optimal material design scheme and damage range, output the optimal ECC material mix ratio, structural design parameters of composite beam, determine the optimal structural scheme, and damage monitoring range based on the moment-deflection curve.

2. The engineering structural design method for ECC-concrete composite beams according to claim 1, characterized in that, The dedicated database includes an ECC material mix proportion database and an ECC-concrete composite beam database. The ECC-concrete composite beam database includes twelve sets of input variables and five structural response indicators. The twelve sets of input variables are: beam section height, beam section width, beam span, distance from the point of application of concentrated force to the beam support, longitudinal reinforcement ratio of the upper part of the beam, reinforcement ratio of the lower part of the beam, stirrup diameter, stirrup spacing, ECC layer thickness at the bottom of the beam, ECC tensile strength, ECC tensile strain and ECC compressive strength. The five structural response indicators are: initial stiffness, yield strength, deflection corresponding to yield strength, peak strength, and deflection corresponding to peak strength of the ECC-concrete composite beam.

3. The engineering structural design method for ECC-concrete composite beams according to claim 1, characterized in that, S2 specifically includes: S201, Generator: Receives user-input constraints, automatically samples within the design space, and generates an initial set of candidate solutions; S202, Evaluator: Real-time call to the proxy model of the model layer to perform batch predictions on each candidate solution produced by the generator, and feedback the fitness value to provide a quantitative basis for optimization iteration; S203, Optimizer: Built-in multi-objective optimization algorithm, with the dual objectives of maximizing structural performance and minimizing material cost, drives the evolution of the solution; in each iteration, the optimizer automatically filters non-dominated solutions based on the fitness value fed back by the evaluator, guiding the search direction to converge toward the Pareto standard front.

4. The engineering structural design method for ECC-concrete composite beams according to claim 3, characterized in that, In S2, both the optimized distributed gradient boosting library and the classification feature enhancement surrogate model undergo hyperparameter optimization through random search and Fourier optimization.

5. The engineering structural design method for ECC-concrete composite beams according to claim 4, characterized in that, The hyperparameters in the optimized distributed gradient boosting library in S2 are: n_estimators=600, max_depth=3, learning_rate=0.16, booster='gbtree', gamma=0.5, reg_alpha=0.1, reg_lambda=0, min_child_weight=3, subsample=0.9, colsample_bytree=0.8, random_state=200; The hyperparameters for the classification feature enhancement are: iterations=400, depth=3, learning_rate=0.1, loss_function='Poisson', od_type='IncToDec', od_wait=50.

6. The engineering structural design method for ECC-concrete composite beams according to claim 3, characterized in that, The multi-objective optimization algorithm in S203 is a multi-objective optimization method based on the multi-objective particle swarm optimization algorithm. The multi-objective optimization algorithm aims to maximize the compressive strength of ECC, maximize the tensile strength of ECC, and minimize the material cost. The constraints limit the total mass of ECC raw materials and the aggregate content parameters. Based on the target tensile strength and tensile strain requirements, a set of Pareto standard optimal solutions is generated.

7. The engineering structural design method for ECC-concrete composite beams according to claim 1, characterized in that, ECC structural options include different ECC layer thicknesses, ECC tensile strength, ECC tensile strain, and ECC compressive strength.

8. The engineering structural design method for ECC-concrete composite beams according to claim 7, characterized in that, The optimal structural scheme utilizes the obtained ECC mix proportion cost With ECC layer thickness The performance index of the ECC concrete composite beam is calculated using the following expression: ； in, and These are the width and length of the composite beam, respectively. For performance index; The smaller the value, the better the corresponding design scheme.

9. A readable storage medium, characterized in that, The readable storage medium stores a computer program that is adapted to be loaded by a processor and executed as an engineering structural design method for ECC-concrete composite beams according to any one of claims 1-8.

10. A computer device, characterized in that, The computer device includes a memory and a processor. The memory stores a computer program, which, when executed by the processor, runs the engineering structural design method for ECC-concrete composite beams according to any one of claims 1-8.

Images