Intelligent optimization design method for curtain wall keel structure

Abstract

The invention relates to the technical field of building curtain wall keel structure design, and discloses a curtain wall keel structure intelligent optimization design method, which comprises the following steps: firstly, obtaining geometric parameters, load distribution data and material performance indexes, and verifying the geometric parameters, the load distribution data and the material performance indexes; analyzing the verified data to generate multi-dimensional structure characteristic parameters, and constructing a structure optimization reference model; keel structure data are verified in real time according to the model, and abnormal matching of node displacement and stress distribution is quantitatively evaluated; according to a verification result, generating a structural anomaly index to judge a design defect; and when defects exist, optimization design schemes including keel section adjustment, node reinforcement and the like are automatically generated and sequenced. According to the method, multiple factors are comprehensively considered, multi-objective optimization is achieved, the curtain wall keel structure design quality and efficiency can be improved, the structure safety is guaranteed, and the cost is reduced.

Description

Technical Field

[0001] The invention relates to the technical field of building curtain wall keel structure design, in particular to an intelligent optimization design method for a curtain wall keel structure. Background Art

[0002] With the rapid development of the construction industry, curtain walls are widely used in modern architecture due to their aesthetics and transparency. As the key support system for curtain walls, the rationality and reliability of the curtain wall frame structure are directly related to the curtain wall's safety, stability, and service life. The traditional design of curtain wall keel structures mainly relies on the experience of engineers and conventional mechanical calculation methods. During the design process, it is often difficult to accurately consider the impact of actual factors such as panel size deviation and connector positioning error on structural performance when processing geometric parameters, resulting in a certain deviation between the design and actual construction. In terms of load distribution data processing, complex load factors such as wind pressure gradient changes and temperature stress distribution are usually simplified, which cannot fully and accurately reflect the load conditions that the curtain wall is subjected to during actual use. For material performance indicators, only the average performance parameters of the material are considered, ignoring the actual changes in material properties such as the discrete value of the elastic modulus and the yield strength fluctuation range. As a result, the designed keel structure may have problems of insufficient performance or material waste when facing actual working conditions. In the design verification phase, traditional methods mostly use static and local analysis methods, which makes it difficult to monitor the changes in the mechanical properties of the keel structure under different working conditions in real time. The correlation analysis between node displacement and stress distribution is not in-depth enough, and it is impossible to detect abnormal matching of node displacement and stress distribution in a timely manner, resulting in potential design defects not being discovered and corrected in a timely manner. Moreover, traditional design methods lack systematicity and scientificity when optimizing design solutions. They can often only optimize for a single goal, such as simply pursuing the minimization of structural weight or the minimization of cost. It is difficult to achieve a balance between multiple goals such as structural weight, stiffness and cost, and cannot meet the comprehensive requirements of modern buildings for high performance and low cost of curtain wall keel structures. With the continuous increase in building height, the increasingly complex shape and the diversification of usage environments, the traditional curtain wall keel structure design method has gradually become difficult to meet the actual needs of the project. There is an urgent need for a more intelligent, efficient and scientific design method to improve the design quality and efficiency of curtain wall keel structures. Summary of the Invention

[0003] The purpose of the present invention is to provide an intelligent optimization design method for curtain wall keel structure to solve the problems raised in the above background technology.

[0004] To achieve the above-mentioned object, the present invention provides the following technical solution: a curtain wall keel structure intelligent optimization design method, the method comprising: S1: Obtaining geometric parameters, load distribution data, and material performance indicators of the building curtain wall, and performing data verification on the geometric parameters, load distribution data, and material performance indicators; S2: Analyze the verified data to generate multi-dimensional structural feature parameters, perform feature selection based on the multi-dimensional structural feature parameters to form a feature vector set, and construct a structural optimization benchmark model based on the feature vector set; S3: Real-time verification of keel structural data based on the structural optimization benchmark model, using multi-level data association technology to identify the topological correlation between node displacement and stress distribution, and quantitatively evaluate abnormal matching between node displacement and stress distribution; S4: Generate a structural abnormality index based on the verification results of the structural optimization benchmark model, and determine whether there are design defects based on the structural abnormality index.

[0005] Preferably, the geometric parameters include panel size deviation and connector positioning error; the load distribution data includes wind pressure gradient change and temperature stress distribution; the material performance indicators include discrete values ​​of elastic modulus and yield strength fluctuation range; data verification is performed on the geometric parameters, load distribution data and material performance indicators, and the data verification includes data normalization, outlier correction, dimensional unification and integrity verification; a geometric feature sequence is generated after the geometric parameter verification, and a load feature sequence is generated after the load distribution data verification; the geometric feature sequences of different dimensions are spatially associated with the load feature sequences to form a multi-source structural feature set.

[0006] Preferably, in S2, the following steps are included: S201: Extracting spatial characteristic parameters from the verified geometric characteristic sequence, load characteristic sequence, and material performance index, respectively. The spatial characteristic parameters include stiffness variation coefficient, connection point offset, and stress concentration. Each spatial characteristic parameter is combined according to a preset weight to generate a multi-dimensional structural characteristic parameter. S202: Setting a dynamic threshold for the multi-dimensional structural feature parameters in the multi-source structural feature set, screening feature parameters that meet the threshold range to form an initial feature vector set, and excluding feature parameters that exceed the threshold range; S203: performing cross-condition correlation analysis on each characteristic parameter in the initial characteristic vector set, extracting the deviation of the characteristic parameters in different working conditions under the same load condition, calculating the coefficient of variation of the deviation and marking it as the difference parameter between working conditions; S204: Compare the difference parameters between the working conditions with the preset tolerance threshold, select the characteristic parameters that exceed the tolerance threshold and add them to the structural optimization benchmark model, and add the yield strength mutation characteristics to the structural optimization benchmark model according to the material performance indicators.

[0007] Preferably, in S3, quantitatively evaluating the abnormal matching between node displacement and stress distribution includes the following steps: S301: Real-time monitoring of the displacement change of the keel node and the stress change at the corresponding position, and calculating the difference in the rate of change between the two. If the difference in the rate of change exceeds a preset reasonable range, it is determined to be an abnormal mechanical event; S302: Count the number of abnormal mechanical events within a preset analysis period, record it as M, and simultaneously obtain the elastic modulus fluctuation amplitude of the material performance index, record it as E; S303: Based on the correlation between the number of abnormal mechanical events M and the elastic modulus fluctuation amplitude E, a mechanical anomaly index is generated using nonlinear function mapping, wherein the product component and the ratio component of M and E are superimposed to generate a comprehensive evaluation value; If the mechanical anomaly index exceeds the preset safety threshold, a design defect flag is triggered.

[0008] Preferably, in S4, the verification results of the structural optimization benchmark model in the historical design data of the curtain wall of the same building are extracted, and the deviation values ​​of each benchmark model are weighted and accumulated to generate a structural abnormality index. If the structural abnormality index exceeds the preset trigger number continuously, it is determined that there is a design defect.

[0009] Preferably, in S202, the method for setting the dynamic threshold includes: statistically analyzing the distribution law of each characteristic parameter based on the historical engineering database, using the probability density function to calculate the confidence interval of the characteristic parameter, using the upper and lower limits of the confidence interval as the initial value of the dynamic threshold, and dynamically adjusting the threshold range according to the characteristic parameter distribution of the current project.

[0010] Preferably, in S303, the specific implementation method of the nonlinear function mapping is: establishing a coupling function with the number of abnormal mechanical events M as the independent variable and the elastic modulus fluctuation amplitude E as the dependent variable, and obtaining a prediction model of the mechanical anomaly index through machine learning algorithm training. The prediction model adopts a neural network structure, the input layer includes the normalized values ​​of M and E, and the output layer is the mechanical anomaly index.

[0011] Preferably, the method further includes S5: when it is determined that there is a design defect, an optimized design scheme is automatically generated, the optimized design scheme includes suggestions for adjusting the keel cross-section size, connection node reinforcement measures and material replacement schemes, and the priority of the optimized design schemes is sorted according to the size of the structural abnormality index.

[0012] Preferably, in S5, the method for generating the optimized design scheme includes: constructing a multi-objective optimization function, taking minimizing structural weight, maximizing stiffness and minimizing cost as optimization objectives, using a genetic algorithm to perform multi-scheme iterative solution, and outputting a Pareto optimal solution set as a candidate optimized design scheme.

[0013] Compared with the prior art, the present invention has the following beneficial effects: The intelligent optimization design method for curtain wall keel structures provided by this invention achieves improved comprehensiveness and accuracy in data processing. By acquiring the geometric parameters, load distribution data, and material performance indicators of the building curtain wall, and performing rigorous data verification such as data normalization, outlier correction, dimensional unification, and integrity testing, the method fully considers practical factors such as panel size deviations, wind pressure gradient changes, and discrete values ​​of the elastic modulus. Data from different dimensions are spatially correlated to form a multi-source structural feature set, enabling subsequent design to be based on data that better reflects actual working conditions, avoiding design deviations caused by inaccurate or incomplete data. In terms of model construction, by extracting spatial characteristic parameters and scientifically combining them to form multi-dimensional structural characteristic parameters, a structural optimization benchmark model was constructed using dynamic threshold screening and cross-condition correlation analysis. This model comprehensively considers the impact of multiple factors on structural performance and more accurately reflects the mechanical properties of curtain wall keel structures in actual use than traditional models, providing a reliable basis for subsequent design verification and optimization. During the design verification process, multi-level data correlation technology is used to verify the keel structure data in real time, quantitatively evaluating any abnormal mismatches between node displacements and stress distribution. Through real-time monitoring and precise calculations, potential structural issues under various operating conditions can be promptly identified. Compared to traditional static, localized analysis methods, design flaws can be identified earlier, buying time to ensure curtain wall structural safety. In terms of design defect judgment and optimization, design defects are judged by generating structural abnormality indicators, and an optimized design scheme is automatically generated when defects are determined to exist. This scheme is optimized with multiple objectives such as minimizing structural weight, maximizing stiffness, and minimizing cost. A genetic algorithm is used to solve the Pareto optimal solution set, which is then sorted according to the structural abnormality index. This method changes the limitations of traditional single-objective optimization and can find a balance between multiple important objectives, ensuring the safety and performance of the structure while achieving effective cost control, improving material utilization, and reducing resource waste. At the same time, the mechanism of automatically generating and sorting schemes greatly improves the efficiency of design optimization, reduces the time and energy cost of manual design, and provides an efficient, scientific, and intelligent solution for the design of building curtain wall keel structures, promoting the development of building curtain wall design technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 This is a working principle diagram of the intelligent optimization design method for curtain wall keel structure according to the present invention; Figure 2 Generate graphs for data validation and feature sets; Figure 3 Flowchart for building a benchmark model for structural optimization. DETAILED DESCRIPTION

[0015] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0016] See also Figure 1-Figure 3 The present invention relates to an intelligent optimization design method for curtain wall keel structure, and the specific implementation steps are as follows: Step S1: Obtain the geometric parameters, load distribution data, and material performance indicators of the building curtain wall, and perform data verification on the geometric parameters, load distribution data, and material performance indicators. Geometric parameters include panel size deviations and connector positioning errors; load distribution data covers wind pressure gradient changes and temperature stress distribution; and material performance indicators involve discrete values ​​of elastic modulus and yield strength fluctuation ranges. The data verification process includes data normalization, outlier correction, dimensional unification, and integrity verification. After geometric parameter verification, a geometric feature sequence is generated; after load distribution data verification, a load feature sequence is generated; and geometric feature sequences of different dimensions are spatially associated with load feature sequences to form a multi-source structural feature set.

[0017] Step S2: Analyze the verified data to generate multi-dimensional structural characteristic parameters, perform feature selection based on the multi-dimensional structural characteristic parameters to form a characteristic vector set, and construct a structural optimization benchmark model based on the characteristic vector set. In specific implementation, first extract spatial characteristic parameters from the verified geometric characteristic sequence, load characteristic sequence, and material performance index, respectively. The spatial characteristic parameters include stiffness variation coefficient, connection point offset, and stress concentration. Combine each spatial characteristic parameter according to a preset weight to generate a multi-dimensional structural characteristic parameter. Then, set a dynamic threshold for the multi-dimensional structural characteristic parameters in the multi-source structural characteristic set, filter the characteristic parameters that meet the threshold range to form an initial characteristic vector set, and exclude the characteristic parameters that exceed the threshold range. Then, perform cross-condition correlation analysis on each characteristic parameter in the initial characteristic vector set, extract the deviation of the characteristic parameters in different working conditions under the same load condition, calculate the coefficient of variation of the deviation, and mark it as the difference parameter between working conditions. Finally, compare the difference parameter between working conditions with the preset tolerance threshold, filter the characteristic parameters that exceed the tolerance threshold to add to the structural optimization benchmark model, and supplement the yield strength mutation feature to the structural optimization benchmark model based on the material performance index. Step S3: Verify the keel structure data in real time based on the structural optimization benchmark model, use multi-level data association technology to identify the topological correlation between node displacement and stress distribution, and quantitatively evaluate the abnormal matching of node displacement and stress distribution. The specific operation is to monitor the change in keel node displacement and the change in stress at the corresponding position in real time, calculate the difference in the rate of change between the two, and determine it as an abnormal mechanical event if the difference in the rate of change exceeds the preset reasonable range; the number of abnormal mechanical events within the preset analysis period is recorded as M, and the elastic modulus fluctuation amplitude in the material performance index is obtained simultaneously and recorded as E; based on the correlation between the number of abnormal mechanical events M and the elastic modulus fluctuation amplitude E, a nonlinear function mapping is used to generate a mechanical anomaly index, where the product component and ratio component of M and E are superimposed to generate a comprehensive evaluation value; if the mechanical anomaly index exceeds the preset safety threshold, a design defect flag is triggered. Step S4: A structural anomaly index is generated based on the verification results of the structural optimization benchmark model. The structural anomaly index is then used to determine whether a design flaw exists. Specifically, the verification results of the structural optimization benchmark model from the historical design data for the same building curtain wall are extracted. The deviation values ​​of each benchmark model are weighted and accumulated to generate the structural anomaly index. If the structural anomaly index exceeds a preset triggering number of times consecutively, a design flaw is determined.

[0018] The present invention will be further described below in conjunction with Examples 1 to 5: Example

[0019] During data verification, panel size deviations within geometric parameters are considered. During actual production and installation, panel sizes may vary due to factors such as manufacturing processes and measurement errors. These deviations may also occur in different measurement units or representations. During normalization, the original units and representations of the data must be clearly defined. For example, some panel size deviation data may be in millimeters, while others may be in inches, or some data may be presented as a percentage deviation from a standard size. Based on different scenarios, appropriate conversion rules are employed to convert all panel size deviation data into a unified unit and standard format, ensuring comparability and consistency in subsequent processing.

[0020] Connector positioning errors present similar challenges, with data coming from diverse sources and representations. During normalization, precise calculations and conversions are performed based on the connector's designed and measured positions within the building's curtain wall structure, unifying the positioning error data into a common metric. This eliminates interference caused by differences in data representation and lays the foundation for accurate subsequent analysis and processing of geometric parameters.

[0021] In terms of outlier correction, statistical methods are used to conduct in-depth analysis of panel dimensional deviations and connector positioning error data. First, statistical quantities such as the mean, median, and standard deviation are calculated to describe the overall distribution characteristics of the data. By setting a reasonable threshold range, usually based on the mean and standard deviation, for example, using the mean plus or minus a certain number of standard deviations as the threshold range, data points that significantly deviate from the normal range are identified. When abnormal data points are discovered, they are not simply deleted. Instead, they are appropriately corrected based on the overall distribution characteristics of the data. For example, if an abnormal data point is caused by measurement error and the surrounding data exhibits a certain distribution pattern, the values ​​and distribution trends of adjacent data points can be referenced to adjust the abnormal data to make it conform to the overall data distribution, thereby ensuring data reliability and accuracy.

[0022] During the dimensional unification process, geometric parameters are converted based on physical principles and engineering specifications. For example, length-related parameters, such as panel dimensional deviations and connector positioning errors, are uniformly converted to meters in the International System of Units (SI). During dimensional conversion, accuracy must be ensured to avoid conversion errors that could affect subsequent calculations and analysis. Furthermore, the interrelationships between different parameters must be considered to ensure that the unified data accurately reflects the geometric characteristics of the building's curtain wall structure.

[0023] Completeness verification is a crucial step in data validation. For geometric parameters, carefully check that all data fields are complete to ensure that no critical information is missing. For example, when recording panel dimensional deviations, it's important to record not only the deviations in length and width but also the deviation in thickness. For connector positioning errors, it's important to record positioning deviations in all directions, as well as relevant connector properties. This comprehensive inspection allows for the timely identification and supplementation of missing data, ensuring data integrity and providing a complete data foundation for subsequent analysis and modeling.

[0024] For load distribution data, wind pressure gradient change data can be affected by various factors during the collection process, such as meteorological conditions and measurement equipment, leading to discrepancies and non-standardization. During normalization, wind pressure gradient change data from different sources and under different measurement conditions are converted to a unified standard format and units based on wind pressure data collection principles and actual application requirements. For example, wind pressure data can be uniformly converted to Pascals (Pa) and organized according to specified time intervals and spatial locations.

[0025] Temperature stress distribution data also requires normalization. Temperature changes cause material expansion and contraction, generating stress. Various factors can lead to data inconsistencies during temperature measurement and stress calculation. Through appropriate conversion and processing, temperature stress distribution data can be converted into a unified representation for easier analysis and comparison.

[0026] Regarding outlier correction, statistical methods are also used to identify abnormal data points in load distribution data. By analyzing the data distribution patterns and setting appropriate thresholds, we can identify data that deviates from the normal range. For these abnormal data, we combine actual project conditions with relevant theories to determine the cause. If the cause is measurement error or abnormal weather conditions, we apply appropriate correction methods to ensure that the load distribution data accurately reflects the actual load conditions.

[0027] Dimensional unification is also crucial for load distribution data. Based on mechanical principles and engineering specifications, the dimensions of data such as wind pressure gradients and temperature stress distributions are unified to meet the requirements of structural analysis. For example, harmonizing the dimensions of wind pressure data with those of stress data allows for accurate calculation and comparison during structural force analysis.

[0028] During the integrity inspection, all aspects of the load distribution data are thoroughly checked, including wind pressure gradient changes at different locations and times, as well as temperature stress distribution data at different locations. This ensures that no critical load information is missed and that the data can fully describe the stress conditions of the building curtain wall under different load conditions.

[0029] When acquiring material performance indicators, such as discrete values ​​of elastic modulus and yield strength fluctuation range data, data may be inaccurate and inconsistent due to factors such as the material's inherent characteristics and testing methods. During normalization, the discrete values ​​of elastic modulus and yield strength fluctuation range data obtained under different testing conditions are uniformly processed according to the standards and specifications for material performance testing. For example, elastic modulus data is uniformly converted to Pascals (Pa) in the International System of Units and organized according to specified accuracy requirements.

[0030] Regarding outlier correction, statistical methods are used to analyze material performance data and identify abnormal data points. The rationality of these abnormal data points is determined based on factors such as the material's production process and testing procedures. If the abnormal data is caused by testing errors, appropriate corrections are made using appropriate methods to ensure that the material performance data accurately reflects the material's actual performance.

[0031] During the dimensional unification process, based on the principles of material mechanics and engineering specifications, the dimensions of data such as the discrete values ​​of elastic modulus and yield strength fluctuation range were unified to a form that meets structural design requirements. This ensures that material performance index data can be accurately calculated and evaluated during structural design and analysis.

[0032] During integrity testing, carefully examine all aspects of material performance data, including performance data from different batches of materials, data obtained using different testing methods, etc. This ensures that no critical material performance information is missed, providing accurate and complete material performance data for subsequent structural design and optimization.

[0033] After verifying the geometric parameters, load distribution data, and material performance indicators, the geometric feature sequences of different dimensions are spatially correlated with the load feature sequences. First, a three-dimensional spatial coordinate system is established, with the actual spatial position of the building curtain wall as a reference, and the coordinate system's origin and coordinate axis directions are determined. Then, parameters in the geometric feature sequence, such as panel dimensional deviation and connector positioning error, are mapped to corresponding positions in the spatial coordinate system based on their actual positions in the curtain wall structure. Similarly, for the load feature sequence, data such as wind pressure gradient changes and temperature stress distribution are mapped to the spatial coordinate system based on the position and direction of the load. In this way, the geometric features and load features are spatially correlated and integrated, so that each spatial position corresponds to corresponding geometric and load feature information, thus forming a multi-source structural feature set. This multi-source structural feature set comprehensively describes the geometric and load characteristics of the building curtain wall structure, providing a rich and accurate data foundation for the subsequent construction of a structural optimization benchmark model. This model more realistically reflects the actual conditions of the building curtain wall and provides strong support for subsequent structural optimization design. Example

[0034] When constructing a structural optimization benchmark model, spatial characteristic parameters are first extracted from the verified geometric feature sequences, load feature sequences, and material performance indicators. For the geometric feature sequences, stiffness varies across different structural components due to differences in shape, size, and connection methods. To calculate the stiffness variation coefficient, the structural divisions must be determined. Based on project realities, the curtain wall keel structure is rationally partitioned into panels, connection areas, and other areas. Then, for each zone, the stiffness of each zone is analyzed using mechanical principles, based on data such as panel dimensional deviation and connector positioning error from the geometric feature sequences, combined with information such as the discrete values ​​of the elastic modulus from material performance indicators. By comparing the stiffness values ​​of different zones, the degree of stiffness variation across the entire structure is calculated, resulting in the stiffness variation coefficient. Determining the connection point offset relies on the precise measurement and comparison of the actual and designed connector positions. During curtain wall construction, high-precision measurement equipment such as total stations and laser rangefinders is used to obtain the actual coordinates of the connectors in three dimensions. These coordinates are then compared with the coordinates in the design drawings to determine the offset values ​​for each connector in each direction, which are then used to determine the connection point offset. Stress concentration analysis primarily focuses on structural nodes, as nodes are critical locations for force transmission and are subject to complex forces. Finite element analysis and other methods are used to simulate the stresses on the structure under different loading conditions, analyze the stress distribution at the nodes, determine the degree of stress concentration, and ultimately obtain stress concentration parameters.

[0035] For load characteristic sequences, relevant spatial characteristic parameters are extracted based on mechanical principles and actual engineering practices. Taking into account changes in wind pressure gradients and temperature stress distribution, the impact of loads on various parts of the structure at different locations and times is analyzed to determine characteristic parameters related to structural performance. For material performance indicators, parameters related to the spatial performance of the structure are extracted in combination with the material's mechanical properties, such as the effect of changes in the material's elastic modulus on structural stiffness.

[0036] Each extracted spatial characteristic parameter is combined according to preset weights to generate multi-dimensional structural characteristic parameters. The determination of preset weights is a rigorous process that requires comprehensive consideration of the importance of each parameter in affecting structural performance. Experts in the fields of structural design, materials science, engineering mechanics, etc. are organized to evaluate each spatial characteristic parameter based on previous engineering experience, theoretical analysis, and in-depth research on the performance of building curtain wall structures. For example, in certain curtain wall structures with high stiffness requirements, the weight of the stiffness variation coefficient may be relatively high; and in structures where problems are prone to occur at connection nodes, the weights of the connection point offset and stress concentration will be increased accordingly. Through scientific methods such as expert scoring and hierarchical analysis, the specific weight value of each spatial characteristic parameter is determined, and the parameters are combined according to the weights to form multi-dimensional structural characteristic parameters that comprehensively reflect the performance characteristics of the structure in different aspects.

[0037] Dynamic thresholds are set for the multi-dimensional structural feature parameters in a multi-source structural feature set. The dynamic thresholds are set based on a historical project database that collects a large amount of data on building curtain wall projects of various types and under various environmental conditions. Through statistical analysis of the historical data, a probability density function is used to calculate the confidence intervals for each feature parameter. The probability density function describes the probability distribution of a feature parameter in the dataset, and the calculated confidence interval reflects the range of values ​​that the feature parameter normally takes. The upper and lower limits of the confidence interval serve as the initial values ​​for the dynamic thresholds. In practice, the initial threshold range is dynamically adjusted based on the specific distribution of feature parameters in the current project. For example, if the material properties of the current project differ significantly from those of the historical project, or if the building's environmental conditions are unique, resulting in changes in the distribution of certain feature parameters, the threshold range is adjusted accordingly to ensure that the threshold accurately reflects the actual conditions of the current project. Based on the adjusted threshold range, feature parameters that meet the requirements are selected to form an initial feature vector set. Feature parameters that fall outside the threshold range are temporarily excluded, ensuring that the initial feature vector set contains parameters that best represent the structural characteristics of the current project.

[0038] Perform cross-condition correlation analysis on each characteristic parameter in the initial characteristic vector set. In the actual use of building curtain walls, there will be many different working conditions, such as temperature changes in different seasons, wind pressure effects of different intensities, etc. Under the same load conditions, analyze the deviation of characteristic parameters in different working conditions. For example, under the same wind pressure load conditions, analyze the changes in characteristic parameters such as the stiffness variation coefficient and connection point offset of the structure under high temperature conditions in summer and low temperature conditions in winter. By calculating the coefficient of variation of the deviation of the characteristic parameters under different working conditions, the difference parameters between working conditions are obtained. The coefficient of variation can reflect the degree of discreteness of the data. By calculating the coefficient of variation of the deviation, the degree of difference in the change of the characteristic parameters under different working conditions can be quantified, thereby evaluating the stability and reliability of the characteristic parameters under different working conditions.

[0039] Finally, the difference parameters between working conditions are compared with the preset tolerance threshold. The preset tolerance threshold is determined according to engineering design standards and actual use requirements, and is a limit value used to measure whether the changes in characteristic parameters under different working conditions are reasonable. The characteristic parameters that exceed the tolerance threshold are screened out. These characteristic parameters change more significantly under different working conditions and have a greater impact on structural performance, and are added to the structural optimization benchmark model. At the same time, based on the relevant data in the material performance indicators, the yield strength mutation characteristics are supplemented to the structural optimization benchmark model. Yield strength mutation is an important change characteristic of material properties that may occur under certain special circumstances, and has an important impact on the safety and stability of the structure. By comprehensively considering these factors, the construction of the structural optimization benchmark model is completed, so that it can accurately reflect the performance characteristics of the building curtain wall keel structure under different working conditions and conditions, and provide a reliable basic model for subsequent structural optimization design. Example

[0040] When quantitatively evaluating the abnormal matching of node displacement and stress distribution, real-time monitoring must be performed first. High-precision displacement sensors and stress sensors are installed at key node locations of the building curtain wall keel structure. These sensors have the characteristics of high precision, high reliability and long-term stable operation. They can accurately collect the displacement change of the keel node and the stress change data at the corresponding position in real time. During the installation process, it is necessary to accurately determine the installation position of the sensor to ensure that it can truly reflect the mechanical state of the node. For displacement sensors, they should be installed in a position that can accurately measure the displacement changes of the node in all directions; for stress sensors, they should be installed in key locations that can effectively measure the stress changes of the node. After installation, the sensor is calibrated and debugged to ensure the accuracy and reliability of its measurement data.

[0041] After collecting real-time data on displacement changes at the keel node and stress changes at the corresponding locations, the difference in their rates of change is calculated. This rate difference calculation is crucial for assessing whether the node's mechanical condition is abnormal. By analyzing the displacement and stress changes over a period of time, their respective rates of change are calculated. The displacement rate of change reflects the rate of change of the node position over time, while the stress rate of change reflects the rate of change of the node stress over time. Subtracting these two rates of change yields the rate difference. If this rate difference exceeds a preset acceptable range, an abnormal mechanical event is identified. Determining this acceptable range is a complex process that requires a comprehensive consideration of structural mechanics theory and engineering experience. Structural mechanics theory provides the theoretical basis for determining this acceptable range. Mechanical analysis of the curtain wall keel structure determines the reasonable relationship between the displacement rate of change and the stress rate of change under normal operating conditions. Engineering experience, based on the actual operational performance of similar projects, refines and improves the theoretical calculation results to determine a preset acceptable range that meets actual project requirements.

[0042] Next, the number of abnormal mechanical events within the preset analysis period is counted, denoted as M. The selection of the preset analysis period should be determined based on the characteristics of the curtain wall keel structure, the operating environment, and the actual project requirements. Within this period, all cases determined to be abnormal mechanical events are counted to obtain the number of abnormal mechanical events M. Simultaneously, the fluctuation amplitude of the elastic modulus in the material performance index is obtained, denoted as E. The elastic modulus is a key mechanical performance indicator of the material, and its fluctuation amplitude reflects the stability of the material's properties. Through real-time monitoring and analysis of the material's properties, the fluctuation amplitude E of the elastic modulus within the preset analysis period is obtained.

[0043] Based on the correlation between the number of abnormal mechanical events, M, and the elastic modulus fluctuation amplitude, E, a nonlinear function mapping is used to generate the mechanical anomaly index. Specifically, a coupling function is established with the number of abnormal mechanical events, M, as the independent variable and the elastic modulus fluctuation amplitude, E, as the dependent variable. This coupling function describes the intrinsic relationship between M and E. The specific form of the function is determined through analysis and mining of large amounts of historical data. To obtain an accurate prediction model for the mechanical anomaly index, a machine learning algorithm is used to train the coupling function. Machine learning algorithms have powerful data analysis and pattern recognition capabilities and can learn the complex relationship between M and E from large amounts of data.

[0044] During the training process, a predictive model is constructed using a neural network structure. The input layer of this neural network includes the normalized values ​​of M and E. Normalization is performed to eliminate the influence of dimensions and orders of magnitude between different variables, allowing the model to treat each input variable more fairly. By mapping the values ​​of M and E to a specific interval, their normalized values ​​are obtained. The output layer of the neural network is the mechanical anomaly index. During the neural network training process, the network weights and biases are continuously adjusted to ensure that the model output reflects the correlation between M and E as accurately as possible.

[0045] In the process of generating the mechanical anomaly index, the product and ratio of M and E are superimposed to produce a comprehensive evaluation value. The product reflects the combined effect of M and E, while the ratio reflects their relative relationship. These two components are superimposed to produce the final mechanical anomaly index. This superposition method comprehensively and comprehensively considers the impact of M and E on the mechanical properties of the structure.

[0046] Finally, the generated mechanical anomaly index is compared with a preset safety threshold. This threshold, determined based on engineering design standards and actual operational requirements, serves as a threshold for determining the safety of the structural mechanical properties. If the mechanical anomaly index exceeds the threshold, a design defect is flagged. This indicates that under the current operating conditions, there is an abnormal mismatch between the node displacement and stress distribution in the curtain wall keel structure, potentially indicating a design flaw requiring further analysis and resolution. This quantitative assessment method enables timely and accurate identification of potential problems in the curtain wall keel structure, providing strong support for structural optimization and safety assessment. Example

[0047] In the process of generating structural abnormality indicators and judging design defects, it is necessary to first establish a fully functional historical data storage database. Taking the building curtain wall project of a large commercial complex as an example, the project includes curtain wall designs in multiple different areas, and the curtain walls in each area are different in geometric parameters, load conditions and material selection. In the process of project advancement, each time the design of the curtain wall keel structure of a region is completed, a structural optimization benchmark model will be constructed according to the method described in the present invention, and the model will be fully verified. At this time, all kinds of data generated during the verification process, such as node displacement monitoring data, stress distribution calculation results, model parameter settings, etc., will be stored completely and accurately in the historical data storage database in accordance with a unified data format and standard. The storage architecture of the database adopts a hierarchical classification method, and is classified and stored according to information such as project name, design stage, curtain wall area, etc., to facilitate subsequent retrieval and calling of data.

[0048] When identifying design flaws in a newly designed curtain wall keel structure, verification results for a structural optimization benchmark model from the same building's historical curtain wall design data are extracted from the historical data storage database. For example, if the current design involves the curtain wall of a new floor in the same commercial complex, verification results for structural optimization benchmark models for previous curtain wall designs for each floor and area of ​​the building are retrieved from the database. These results encompass analysis and verification data on curtain wall structures from different periods and by different design teams, providing a rich reference for subsequent analysis.

[0049] Next, the deviation values ​​for each benchmark model are weighted and accumulated to generate a structural anomaly index. The deviation value is determined by comparing the newly designed structural optimization benchmark model with the historical benchmark model. For example, for the design of curtain wall stud connections, the stress distribution of a certain type of connection node in the new design differs from the stress distribution of the same node in the historical design under the same load conditions. This difference is part of the deviation value. For different structural parameters, such as the stud cross-sectional dimensions and the elastic modulus of the material, the deviation value is calculated using the corresponding comparison method.

[0050] During the weighted accumulation process, the weight of each deviation is determined based on its significance to structural performance. For example, the cross-sectional dimensions of the keel directly affect the structure's load-bearing capacity and stiffness, significantly impacting structural performance. Therefore, when calculating the structural anomaly index, the deviation of the keel's cross-sectional dimensions is given a higher weight. On the other hand, the dimensional deviation of some minor decorative components, which has a relatively small impact on structural performance, is given a lower weight. The determination of these weights requires a professional team composed of structural engineers, materials experts, and others to conduct in-depth analysis and evaluation of each structural parameter based on engineering experience, structural mechanics principles, and relevant design specifications.

[0051] Taking the curtain wall design of this commercial complex as an example, when analyzing the curtain wall design of a newly added floor, the deviation values ​​of multiple structural parameters are calculated by comparing historical data and accumulated according to weights. Assuming that the calculated structural anomaly index is a specific value, this value is compared with the preset trigger count. The preset trigger count is set based on the actual project conditions and design standards. For projects like this commercial complex that require high building safety, the preset trigger count may be set to three consecutive times. If the structural anomaly index exceeds the preset trigger count for consecutive times, the curtain wall keel structure design is determined to be flawed.

[0052] In practice, situations may arise where the structural anomaly index obtained during the first calculation exceeds the preset value, but remains within the normal range during the second calculation and exceeds it again during the third calculation. In this case, since the preset trigger count has not been exceeded continuously, a design flaw is not determined and design monitoring and analysis continues. Only when the structural anomaly index reaches or exceeds the preset trigger count continuously is the current curtain wall keel structural design clearly identified as flawed, allowing timely optimization and improvement measures to ensure the safety and reliability of the building curtain wall. Example

[0053] When the system determines that there are defects in the curtain wall keel structure design based on the structural abnormality index, it enters the process of automatically generating an optimized design plan. First, a multi-objective optimization function is constructed, which focuses on the three core goals of minimizing structural weight, maximizing rigidity, and minimizing cost. In actual operation, minimizing structural weight aims to reduce the amount of material used and reduce the overall load burden of the building. Taking the curtain wall of a large commercial building as an example, the keel structure is usually composed of a large amount of metal materials. By rationally optimizing the cross-sectional shape, size and layout of the keel, the material usage can be reduced as much as possible while meeting the structural safety requirements, thereby achieving a reduction in structural weight.

[0054] Maximizing rigidity is key to ensuring curtain wall stability under various loads. Curtain walls must possess sufficient rigidity to resist deformation under varying environmental conditions, such as strong winds and temperature fluctuations. For example, in the design of high-rise building curtain walls, strong winds generate wind pressure loads that exert significant forces on the curtain wall keel structure. Optimizing the keel connection method and adding support structures can effectively improve the overall rigidity of the structure, ensuring the curtain wall remains stable even in extreme weather conditions.

[0055] Cost minimization comprehensively considers multiple factors, including material costs, processing costs, and installation costs. In material selection, in addition to considering material performance indicators, prices from different suppliers are compared, selecting cost-effective materials while ensuring quality. Regarding processing costs, processing costs are reduced by optimizing the process, eliminating unnecessary steps, and improving production efficiency. Installation costs are closely related to the design of the keel structure. Reasonable simplification of the keel installation process, reducing installation difficulty and time, can effectively reduce installation costs. The multi-objective optimization function comprehensively considers these three objectives, balancing the relationships between them to achieve the optimal solution for the curtain wall keel structure in terms of performance and economy.

[0056] A genetic algorithm is used to iteratively solve multiple solutions. Simulating the biological evolution process, the genetic algorithm first generates a set of initial curtain wall keel structural optimization design solutions, which are like individuals in a biological population. Each solution includes a series of design parameters such as keel cross-sectional dimensions, connection node form, and material type. Next, a selection operation is performed. Based on the performance of each solution in the multi-objective optimization function, the solution with higher fitness is selected, just as in nature, organisms that are more adaptable to the environment are more likely to survive and reproduce. Fitness is evaluated primarily based on the comprehensive performance of each solution in terms of the three objectives of structural weight, stiffness, and cost. The better the performance, the higher the probability of selection.

[0057] The crossover operation simulates the biological reproduction process, swapping and combining some design parameters of two selected solutions to generate a new solution. For example, if one solution has a more reasonable keel cross-sectional design, while another has an advantage in the connection node form, the crossover operation can combine the advantages of these two solutions to form a new solution with different characteristics. The mutation operation makes random, small changes to some of the design parameters in the solution, increasing the diversity of solutions and preventing the algorithm from falling into a local optimal solution. This is similar to the genetic mutation that occurs during biological evolution, which may produce new characteristics that are more adaptable to the environment.

[0058] By repeatedly performing selection, crossover, and mutation operations, multiple rounds of iteration are performed, with each round generating a new generation of design solutions. As the iterations progress, the performance of the solutions under the multi-objective optimization function gradually improves, meaning that the structural weight continues to decrease, the stiffness continues to increase, and the cost continues to decrease. The final Pareto optimal solution set output includes multiple solutions that strike a balance between different objectives. These solutions are not absolutely superior or inferior; solutions that perform well on some objectives may be slightly inferior on others. For example, one solution may have the lightest structural weight but a relatively high cost; another solution may have the highest stiffness but use more material.

[0059] After obtaining the Pareto optimal solution set, the optimization design schemes are prioritized according to the size of the structural abnormality index. The structural abnormality index reflects the severity of the design defects. The larger the index, the more serious the impact of the design defects on the structural performance, and the corresponding optimization design schemes need to be given priority for implementation. Taking a curtain wall project as an example, if a certain scheme proposes effective optimization measures for design defects that seriously affect structural safety, and the structural abnormality index corresponding to the scheme is large, then it will be ranked high in the priority ranking. In this way, a scientific and reasonable scheme selection can be provided for the optimization of the curtain wall keel structure design, ensuring that within limited resources and time, the scheme that can best solve the design defect problem is implemented first, thereby improving the safety and economy of the curtain wall keel structure.

[0060] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "includes," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0061] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A curtain wall keel structure intelligent optimization design method, characterized in that: The method comprises the following steps: S1: Obtaining geometric parameters, load distribution data, and material performance indicators of the building curtain wall, and performing data verification on the geometric parameters, load distribution data, and material performance indicators; S2: Analyze the verified data to generate multi-dimensional structural feature parameters, perform feature selection based on the multi-dimensional structural feature parameters to form a feature vector set, and construct a structural optimization benchmark model based on the feature vector set; S3: Real-time verification of keel structural data based on the structural optimization benchmark model, using multi-level data association technology to identify the topological correlation between node displacement and stress distribution, and quantitatively evaluate abnormal matching between node displacement and stress distribution; S4: Generate a structural abnormality index based on the verification results of the structural optimization benchmark model, and determine whether there are design defects based on the structural abnormality index.

2. The intelligent optimization design method for curtain wall keel structure according to claim 1, characterized in that: The geometric parameters include panel size deviation and connector positioning error; the load distribution data includes wind pressure gradient change and temperature stress distribution; the material performance indicators include discrete values ​​of elastic modulus and yield strength fluctuation range; data verification is performed on the geometric parameters, load distribution data and material performance indicators, and the data verification includes data normalization, outlier correction, dimensional unification and integrity verification; a geometric feature sequence is generated after the geometric parameter verification, and a load feature sequence is generated after the load distribution data verification; the geometric feature sequences of different dimensions are spatially associated with the load feature sequences to form a multi-source structural feature set.

3. The intelligent optimization design method for curtain wall keel structure according to claim 2, characterized in that: In S2, the following steps are included: S201: Extracting spatial characteristic parameters from the verified geometric characteristic sequence, load characteristic sequence, and material performance index, respectively. The spatial characteristic parameters include stiffness variation coefficient, connection point offset, and stress concentration. Each spatial characteristic parameter is combined according to a preset weight to generate a multi-dimensional structural characteristic parameter. S202: Setting a dynamic threshold for the multi-dimensional structural feature parameters in the multi-source structural feature set, screening feature parameters that meet the threshold range to form an initial feature vector set, and excluding feature parameters that exceed the threshold range; S203: performing cross-condition correlation analysis on each characteristic parameter in the initial characteristic vector set, extracting the deviation of the characteristic parameters in different working conditions under the same load condition, calculating the coefficient of variation of the deviation and marking it as the difference parameter between working conditions; S204: Compare the difference parameters between the working conditions with the preset tolerance threshold, select the characteristic parameters that exceed the tolerance threshold and add them to the structural optimization benchmark model, and add the yield strength mutation characteristics to the structural optimization benchmark model according to the material performance indicators.

4. The intelligent optimization design method for curtain wall keel structure according to claim 3, characterized in that: In S3, quantitative evaluation of abnormal matching between node displacement and stress distribution includes the following steps: S301: Real-time monitoring of the displacement change of the keel node and the stress change at the corresponding position, and calculating the difference in the rate of change between the two. If the difference in the rate of change exceeds a preset reasonable range, it is determined to be an abnormal mechanical event; S302: Count the number of abnormal mechanical events within a preset analysis period, record it as M, and simultaneously obtain the elastic modulus fluctuation amplitude of the material performance index, record it as E; S303: Based on the correlation between the number of abnormal mechanical events M and the elastic modulus fluctuation amplitude E, a mechanical anomaly index is generated using nonlinear function mapping, wherein the product component and the ratio component of M and E are superimposed to generate a comprehensive evaluation value; If the mechanical anomaly index exceeds the preset safety threshold, a design defect flag is triggered.

5. The intelligent optimization design method for curtain wall keel structure according to claim 4, characterized in that: In S4, the verification results of the structural optimization benchmark model in the historical design data of the same building curtain wall are extracted, and the deviation values ​​of each benchmark model are weighted and accumulated to generate a structural abnormality index. If the structural abnormality index exceeds the preset trigger number continuously, it is determined that there is a design defect.

6. The intelligent optimization design method for curtain wall keel structure according to claim 3, characterized in that: In S202, the method for setting the dynamic threshold includes: statistically analyzing the distribution law of each characteristic parameter based on the historical engineering database, using the probability density function to calculate the confidence interval of the characteristic parameter, using the upper and lower limits of the confidence interval as the initial value of the dynamic threshold, and dynamically adjusting the threshold range according to the characteristic parameter distribution of the current project.

7. The intelligent optimization design method for curtain wall keel structure according to claim 4, characterized in that: In S303, the specific implementation method of the nonlinear function mapping is: establish a coupling function with the number of abnormal mechanical events M as the independent variable and the elastic modulus fluctuation amplitude E as the dependent variable, and obtain a prediction model of the mechanical anomaly index through machine learning algorithm training. The prediction model adopts a neural network structure, the input layer includes the normalized values ​​of M and E, and the output layer is the mechanical anomaly index.

8. The intelligent optimization design method for curtain wall keel structure according to claim 1, characterized in that: It also includes S5: when it is determined that there are design defects, an optimized design plan is automatically generated, and the optimized design plan includes a suggestion for adjusting the keel section size, connection node reinforcement measures and material replacement plan, and the priority of the optimized design plan is sorted according to the size of the structural abnormality index.

9. The intelligent optimization design method for curtain wall keel structure according to claim 8, characterized in that: In S5, the method for generating the optimization design scheme includes: constructing a multi-objective optimization function, taking minimizing structural weight, maximizing stiffness and minimizing cost as optimization goals, using a genetic algorithm to perform multi-scheme iterative solution, and outputting a Pareto optimal solution set as a candidate optimization design scheme.

Images