An interactive generative bridge parameter design method and system

Abstract

The application relates to the technical field of bridge parameterized design, in particular to an interactive generation type bridge parameter design method and system, which comprises the following steps: based on span, beam height and support reaction force parameters, the influence of the span on the beam height and the influence of the beam height on the support reaction force are analyzed, a node and an edge weight set are constructed through parameter relations, a network structure is formed, and a parameter correlation network weight matrix is generated. In the application, the parameter correlation network is constructed, the parameter relation visualization and analysis efficiency among the span, the beam height and the support reaction force are effectively improved, the optimization and integration of the parameter constraint interval improve the design flexibility, the generation of the dynamic path is more logical and has real-time response capability, the load distribution optimization ensures the accurate adjustment of the design parameters under complex load working conditions, the robustness and global consistency of the design are enhanced, the adaptability and reliability of the parameter design are improved, and the efficiency and precision of the bridge design are significantly improved.

Description

Technical Field

[0001] This invention relates to the field of parametric design technology for bridges, and in particular to an interactive generative bridge parametric design method and system. Background Technology

[0002] The field of parametric bridge design technology encompasses parametric modeling and design methods for bridge structures. The core of this technology lies in modeling and optimizing the structural morphology, construction methods, and engineering constraints of bridges using parametric techniques. The overall technological scope covers the definition of bridge geometric parameters, structural design based on parametric models, and parameter generation methods combined with computer-aided design techniques. Furthermore, related research also includes the integrated processing of bridge structural design, model optimization, and construction schemes based on preset rules, aiming to achieve efficiency and accuracy in the bridge design process.

[0003] Interactive generative bridge parametric design method refers to the use of interactive design technology and parametric modeling methods to automatically generate and dynamically adjust key parameters in the bridge design process to complete the bridge model design. This method encompasses the following technical aspects: parameter acquisition based on user input and interaction, dynamic generation of the bridge structure through parametric modeling tools, visual adjustment of the bridge structure's geometry, and parameter combination generation based on optimization algorithms. Specifically, the entire design process is completed through a combination of parameter input rules, geometric modeling methods, and optimization algorithms, ensuring the real-time nature and accuracy of parameter generation and adjustment.

[0004] Existing technologies struggle to dynamically represent the complex coupling relationships between parameters in parametric modeling. They typically rely on static rules for parameter adjustment, lacking flexibility and dynamic adaptability, making it difficult to identify the global impact of related parameters during design adjustments. Regarding multi-parameter constraint handling, existing technologies often employ single constraints, ignoring the multiple restrictive effects of complex parameter interactions on the design, thus affecting the global consistency of the design scheme. Path generation methods are mostly based on fixed rules or single-step optimization, failing to meet the dynamic adjustment needs of complex bridge design scenarios and limiting the real-time performance and accuracy of path generation. In load analysis, single load models are prone to insufficient response to comprehensive load conditions, failing to effectively support bridge parameter adjustment and optimization under complex load conditions. These shortcomings result in significant deficiencies in design accuracy and adaptability, making it difficult to address the diverse needs of parametric design in complex design scenarios. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies by proposing an interactive generative bridge parameter design method and system.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: an interactive generative bridge parameter design method, comprising the following steps:

[0007] S1: Based on the span, beam height and support reaction parameters, analyze the effect of span on beam height and the influence of beam height on support reaction. Construct a set of node and edge weights through parameter relationships to form a network structure and generate a parameter-related network weight matrix.

[0008] S2: Based on the parameter association network weight matrix, extract the influence range of span and beam height on support reaction force, determine the constraint boundary values ​​between parameters, integrate the constraint results into the network matrix, and generate a set of parameter constraint intervals;

[0009] S3: Based on the set of parameter constraint intervals, combine the span and beam height parameters, calculate the weight of each parameter path, select the path combination that meets the constraint conditions, form a path scheme, and generate a set of dynamic parameter path values.

[0010] S4: Based on the set of dynamic path values ​​of the parameters, the live load, dead load and wind load are allocated to each path, the load response to the span and beam height is calculated, the adjustment range of each parameter under the load is analyzed, and the load response parameter matrix is ​​generated.

[0011] S5: Based on the load response parameter matrix, perform a global check on the parameter adjustment values ​​of the span and beam height combination, select combinations that meet the dynamic path and load response, check all combination results, and generate optimized integrated values ​​for bridge parameters.

[0012] As a further embodiment of the present invention, the parameter association network weight matrix includes a span node set, a beam height node set, a support reaction force node set, and a weight distribution between nodes; the parameter constraint interval set includes span range constraints, beam height range constraints, support reaction force range constraints, and a coupling boundary between span and beam height; the parameter dynamic path value set includes path weight values, support reaction force dynamic range values, and combined paths of span and beam height; the load response parameter matrix includes live load response coefficient, dead load response coefficient, wind load response coefficient, and path response adjustment value; and the bridge parameter optimization integration value includes optimized span value, optimized beam height value, and optimized support reaction force value.

[0013] As a further aspect of the present invention, based on the parameters of span, beam height, and support reaction force, the effects of span on beam height and beam height on support reaction force are analyzed. A set of node and edge weights is constructed through parameter relationships to form a network structure. The specific steps for generating the parameter-related network weight matrix are as follows:

[0014] S101: Based on the parameters of span, beam height and support reaction force, extract the numerical range and distribution characteristics of span and beam height, group the spans according to the parameter distribution law, calculate the variation range and fluctuation range of beam height corresponding to the span, set the span variation threshold, filter the data with large span variation range, and establish the influence characteristics of span variation on beam height.

[0015] S102: Based on the influence characteristics of the span change on the beam height, analyze the changing trend of the beam height and the support reaction force, fit the relationship between the support reaction force data and the beam height parameters, determine the key node parameters of the beam height change on the support reaction force, adjust the parameter range, calculate the correlation value between the beam height and the support reaction force, and generate a beam height and support reaction force correlation dataset.

[0016] S103: Based on the aforementioned beam height and support reaction force correlation dataset, construct a set of node and edge weights for the parameter correlation network, group the network according to the weight between nodes, filter out regions with large weight fluctuations and adjust edge weights, calculate the value of the overall network weight matrix, and generate the parameter correlation network weight matrix.

[0017] As a further aspect of the present invention, based on the parameter association network weight matrix, the specific steps for extracting the influence range of span and beam height on support reaction force, determining the constraint boundary values ​​between parameters, integrating the constraint results into the network matrix, and generating a set of parameter constraint intervals are as follows:

[0018] S201: Based on the parameter association network weight matrix, extract the interaction range of span, beam height and support reaction force parameters, analyze the interaction weight data between parameters, screen the main range of span and beam height affecting support reaction force according to the weight change trend, calculate the weight distribution of span and beam height on support reaction force and determine its range, and generate a set of the range of span and beam height affecting support reaction force.

[0019] S202: Based on the set of influence ranges of span and beam height on support reaction force, analyze the distribution law of weight within the influence range, combine the numerical boundary characteristics between span, beam height and support reaction force, calculate the boundary numerical range of span and beam height on support reaction force parameters, screen constraint conditions according to numerical changes and judge their effectiveness, adjust the upper and lower bounds of parameters, and generate a set of constraint boundaries between parameters.

[0020] S203: Based on the set of constraint boundaries between parameters, integrate the adjusted constraint boundary data into the parameter association network weight matrix, redistribute the weight values ​​in the network matrix according to the constraint boundaries, calculate the weight data distribution value after matrix adjustment, verify the consistency of the adjusted matrix, and generate a set of parameter constraint intervals.

[0021] As a further aspect of the present invention, based on the set of parameter constraint intervals, the span and beam height parameters are combined, the weight of each parameter path is calculated, and the path combination that satisfies the constraint conditions is selected to form a path scheme. The specific steps for generating a set of dynamic parameter path values ​​are as follows:

[0022] S301: Based on the set of parameter constraint intervals, combine the numerical pairs of span and beam height parameters, gradually assign weight values ​​to each parameter path according to its numerical characteristics, calculate the path weights and analyze their distribution trend values, select the parameter combination range with stable weight distribution, and generate a set of path weights for span and beam height parameters.

[0023] S302: Based on the path weight set of the span and beam height parameters, analyze the distribution characteristics of the path weights, screen the combinations of path weight values ​​that satisfy the parameter constraints, calculate the cumulative weight value of the screened paths, and determine whether they meet the dynamic change conditions. Adjust the priority order of the path combinations and generate a set of path combinations that satisfy the constraints.

[0024] S303: Based on the set of path combinations that satisfy the constraints, optimize the part of the path combination with uneven weights, redistribute the parameter weight values ​​within the path according to the optimization results, integrate the optimized path combinations to form the final path scheme, calculate the dynamic weight cumulative value of the path scheme, and generate a set of dynamic path value parameters.

[0025] As a further aspect of the present invention, the formula for calculating the cumulative weight value is as follows:

[0026]

[0027] Among them, Q tot Represents the cumulative weight value, q p,j z represents the weight of the p-th parameter point in the j-th group of path combinations. p,j This represents the path parameter value of the p-th parameter point in the j-th group of path combinations. t represents the average value of the path parameter values ​​in the j-th group of path combinations, and t represents the total number of parameter points in the j-th group of path combinations.

[0028] As a further aspect of the present invention, based on the aforementioned set of dynamic path values ​​for parameters, the live load, dead load, and wind load are allocated to each path, the load response to span and beam height is calculated, the adjustment range of each parameter under load is analyzed, and the specific steps for generating the load response parameter matrix are as follows:

[0029] S401: Based on the set of dynamic path values ​​of the parameters, allocate live load, dead load and wind load to each parameter point in the path, analyze the numerical distribution characteristics and range of the load, adjust the load values ​​one by one to meet the bearing capacity of the path, calculate the force data of the load on the path, and select loads with stable distribution to generate a set of path load allocation results.

[0030] S402: Based on the set of path load distribution results, calculate the response values ​​of live load, dead load and wind load to span and beam height point by point, analyze the response trend under load according to the dynamic change range of parameters, extract the numerical range of load that has a significant impact on span and beam height, and generate a load response numerical dataset.

[0031] S403: Based on the load response numerical dataset, analyze the adjustment range of each load on the span and beam height parameters, integrate the parameter response range according to the numerical changes between path nodes, construct the adjustment interval matrix of span and beam height under load, and generate the load response parameter matrix.

[0032] As a further aspect of the present invention, the formula for calculating the corrected force data is specifically as follows:

[0033]

[0034] Among them, F k,i The value of the corrected force at the k-th parameter point in the path under the load distribution is represented by l. k w represents the actual span of the k-th point in the path. j,k P represents the weight of the j-th type of load at the k-th parameter point in the path. j,k A represents the distribution value of the j-th type of load at point k in the path. j,k R represents the area of ​​action of the j-th type of load at point k in the path. k represents the bearing capacity threshold of the k-th point in the path, and m represents the total number of load types.

[0035] As a further aspect of the present invention, based on the load response parameter matrix, the parameter adjustment values ​​of the span and beam height combination are globally checked, combinations that conform to the dynamic path and load response are selected, all combination results are checked, and the specific steps for generating optimized and integrated bridge parameter values ​​are as follows:

[0036] S501: Based on the load response parameter matrix, analyze the adjustment range of the combination of span and beam height parameters one by one, check the validity of the combination values ​​point by point according to the constraints of the dynamic path, remove the parameters that do not meet the requirements, and select the parameter combinations that meet the conditions to generate a set of parameter combinations that meet the path constraints.

[0037] S502: Based on the set of parameter combinations that meet the path constraints, calculate the numerical fluctuation of each parameter combination in the load response, analyze the differences in the numerical distribution of the load on the parameter combination, screen the parameter combination with balanced load action and the smallest fluctuation range, integrate the screened parameter combinations, and generate a set of balanced load parameter combinations.

[0038] S503: Based on the set of balanced load parameter combinations, globally check the adjustment values ​​of the span and beam height parameter combinations, correct the parameter ranges of span and beam height according to the load influence value of each combination, integrate the corrected parameter combinations, calculate the overall optimized values, and generate the optimized integrated values ​​of bridge parameters.

[0039] An interactive generative bridge parameter design system, comprising:

[0040] The parameter association module analyzes the effect of span on beam height and the influence of beam height on support reaction force based on span, beam height and support reaction force parameters. It constructs a set of node and edge weights through parameter relationships and generates a parameter association network weight matrix.

[0041] Based on the parameter association network weight matrix, the constraint interval module extracts the influence range of span and beam height on support reaction force, determines the constraint boundary values ​​between parameters, and generates a set of parameter constraint intervals.

[0042] The path weighting module combines the span and beam height parameters based on the set of parameter constraint intervals, calculates the weight of each parameter path, selects the path combination that meets the constraint conditions, and generates a set of dynamic parameter path values.

[0043] Based on the set of dynamic path values ​​of the parameters, the load response module allocates live load, dead load and wind load to each path, calculates the load response to the span and beam height, and generates a load response parameter matrix.

[0044] Based on the load response parameter matrix, the parameter optimization module performs a global check on the parameter adjustment values ​​of the span and beam height combination, selects combinations that conform to the dynamic path and load response, and generates optimized integrated values ​​for bridge parameters.

[0045] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0046] In this invention, by constructing a parameter association network, the visualization and analytical efficiency of the parameter relationship between span, beam height and support reaction force are effectively improved. The optimization and integration of parameter constraint intervals enhances design flexibility, makes the generation of dynamic paths more logical and responsive in real time, and the load distribution optimization ensures the accurate adjustment of design parameters under complex load conditions, enhances the robustness and global consistency of the design, improves the adaptability and reliability of parameter design, and significantly improves the efficiency and accuracy of bridge design. Attached Figure Description

[0047] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1 This is a schematic diagram of the steps of the present invention;

[0049] Figure 2 This is a flowchart of steps S1 of the present invention;

[0050] Figure 3 This is a flowchart of steps S2 of the present invention;

[0051] Figure 4 This is a flowchart of steps S3 of the present invention;

[0052] Figure 5 This is a flowchart of step S4 of the present invention;

[0053] Figure 6 This is a flowchart of steps S5 of the present invention;

[0054] Figure 7 This is a system module diagram of the present invention. Detailed Implementation

[0055] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0056] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0057] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, their intended meanings are consistent. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, their intended meanings are consistent.

[0058] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0059] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0060] Please see Figure 1 An interactive generative bridge parameter design method includes the following steps:

[0061] S1: Based on the span, beam height and support reaction parameters, analyze the effect of span on beam height and the influence of beam height on support reaction. Construct a set of node and edge weights through parameter relationships to form a network structure and generate a parameter-related network weight matrix.

[0062] S2: Based on the parameter association network weight matrix, extract the influence range of span and beam height on support reaction force, determine the constraint boundary values ​​between parameters, integrate the constraint results into the network matrix, and generate a set of parameter constraint intervals;

[0063] S3: Based on the set of parameter constraint intervals, the span and beam height parameters are combined, the weight of each parameter path is calculated, the path combination that meets the constraint conditions is selected, a path scheme is formed, and a set of dynamic parameter path values ​​is generated.

[0064] S4: Based on the set of dynamic path values, live load, dead load and wind load are allocated to each path, the load response to span and beam height is calculated, the adjustment range of each parameter under load is analyzed, and a load response parameter matrix is ​​generated.

[0065] S5: Based on the load response parameter matrix, the parameter adjustment values ​​of the span and beam height combination are globally checked, combinations that meet the dynamic path and load response are selected, all combination results are checked, and optimized integrated values ​​of bridge parameters are generated.

[0066] The parameter association network weight matrix includes the span node set, beam height node set, support reaction force node set, and node weight distribution; the parameter constraint interval set includes span range constraints, beam height range constraints, support reaction force range constraints, and the coupling boundary between span and beam height; the parameter dynamic path value set includes path weight values, support reaction force dynamic range values, and combined span and beam height paths; the load response parameter matrix includes live load response coefficient, dead load response coefficient, wind load response coefficient, and path response adjustment values; and the bridge parameter optimization integration values ​​include optimized span values, optimized beam height values, and optimized support reaction force values.

[0067] Please see Figure 2 The specific steps of S1 are as follows:

[0068] S101: Based on the parameters of span, beam height and support reaction force, extract the numerical range and distribution characteristics of span and beam height, group the spans according to the parameter distribution law, calculate the variation range and fluctuation range of beam height corresponding to the span, set the span variation threshold, filter the data with large span variation range, and establish the influence characteristics of span variation on beam height.

[0069] First, the span and beam height values ​​for each measuring point were extracted from the engineering database. These datasets reflect the dimensional distribution of different structures. Subsequently, statistical analysis was performed on the data to determine its distribution characteristics, such as mean and standard deviation. Based on these statistical indicators, the spans were divided into several groups, each with similar dimensional characteristics. To accurately control the design and construction standards of the beams, a change threshold was introduced to identify which span changes were significant. That is, by calculating whether the span change exceeded the set threshold, data with large changes that might affect structural safety were selected. This process underwent rigorous quantitative analysis through mathematical models. For example, the threshold was set based on the percentile of span changes in historical data, ensuring that only the most representative change data were selected for further analysis. Finally, based on the selected data, a detailed characteristic model describing the impact of span changes on beam height was established.

[0070] S102: Based on the influence characteristics of span variation on beam height, analyze the variation trend of beam height and support reaction force, fit the variation relationship between support reaction force data and beam height parameters, determine the key node parameters of beam height variation on support reaction force, adjust the parameter range, calculate the correlation degree between beam height and support reaction force, and generate a beam height and support reaction force correlation dataset.

[0071] By performing regression analysis on the relationship between support reaction force data and beam height, key beam height parameters affecting the variation of support reaction force are identified. This analysis process involves complex mathematical modeling and statistical verification. First, the relationship between beam height and support reaction force is fitted using collected structural monitoring data. Linear or nonlinear regression models are used to describe the dependence between the two. Then, the model parameters are adjusted to improve the accuracy and reliability of predictions, and the model's fitness is evaluated. The parameter range is adjusted to cover more variations in beam height. At the same time, the correlation values ​​of these parameters are calculated to provide a scientific basis for support design, thereby generating a comprehensive dataset on the correlation between beam height and support reaction force. This dataset provides important input parameters for subsequent structural analysis and optimization.

[0072] S103: Based on the dataset of beam height and support reaction force association, construct a set of node and edge weights for the parameter association network, group the network according to the weight between nodes, filter out areas with large weight fluctuations and adjust the edge weights, calculate the value of the overall network weight matrix, and generate the parameter association network weight matrix.

[0073] When constructing the parameter association network, the associated dataset of beam height and support reaction force is processed. Data points are treated as nodes using network analysis tools, and the interaction strength between nodes is used as the edge weight. The edge weight is determined by the correlation degree between nodes. Nodes in the network are grouped according to their weights, and areas with large weight fluctuations are analyzed and adjusted in detail. To calculate the overall network weight matrix, the formula W = A·D is used. -1 The calculation is performed, where A represents the adjacency matrix and D represents the inverse of the degree matrix. This process allows for a clearer identification of which nodes play a core role in the structure, thereby optimizing the configuration of the entire network. Ultimately, a parametric network weight matrix is ​​generated, providing an optimized structural connection scheme for bridge design.

[0074] Consider a simple network with three nodes. The connections between the nodes are as follows: node 1 is connected to nodes 2 and 3, and node 2 is connected to node 3. The adjacency matrix A is:

[0075]

[0076] The degree matrix D represents the number of connections for each node, i.e.:

[0077]

[0078] Inverse D of the degree matrix -1 for:

[0079]

[0080] Therefore, the weight matrix W is calculated as follows:

[0081]

[0082] The result shows that the connection weight between each node and other nodes is 0.5, which reflects the equal importance of each node in the undirected graph.

[0083] Please see Figure 3 The specific steps of S2 are as follows:

[0084] S201: Based on the parameter association network weight matrix, extract the interaction range of span, beam height and support reaction force parameters, analyze the interaction weight data between parameters, screen the main range of span and beam height affecting support reaction force according to the weight change trend, calculate the weight distribution of span and beam height on support reaction force and determine its range, and generate a set of the influence range of span and beam height on support reaction force.

[0085] The network weight matrix is ​​decomposed into submatrices to analyze the weight distribution characteristics of each pair of parameters. The highly correlated regions of span and beam height in the influence of support reaction force are screened out item by item. The influence range is determined by combining the weight change trend. By constructing a weight distribution model, the weight gradient of span and beam height is analyzed by matrix operation, and the high and low boundary value range of the parameters is calculated. The screening results are further validated and the consistency is evaluated. Finally, the set of influence ranges of span and beam height on support reaction force is generated.

[0086] S202: Based on the set of influence ranges of span and beam height on support reaction force, analyze the distribution law of weight within the influence range, combine the numerical boundary characteristics between span, beam height and support reaction force, calculate the boundary numerical range of span and beam height on support reaction force parameters, screen constraint conditions based on numerical changes and judge their effectiveness, adjust the upper and lower bounds of parameters, and generate a set of constraint boundaries between parameters.

[0087] Based on the set of influence ranges of span and beam height on support reactions, according to the formula

[0088]

[0089] Calculate the boundary numerical range of the span and beam height on the support reaction parameters.

[0090] In the formula, M(x,y) represents the integral result, f(x,y) represents the distribution of the influence factors of the span parameter, g(x,y) represents the distribution of the influence factors of the beam height parameter, and x1,x2 and y1,y2 are the boundary ranges of the span and beam height, respectively.

[0091] The span range is determined by monitoring the physical dimensions of the span and taking its place in actual bridge design, for example, 10 meters to 50 meters, defined as x1 = 10 and x2 = 50. The beam height range is calculated based on the bridge's load capacity; assuming a beam height of 1 meter to 5 meters, defined as y1 = 1 and y2 = 5. The influence factor distribution functions f(x,y) and g(x,y) are fitted using experimental data, assuming f(x,y) = 2x and g(x,y) = 3y. 2 Substitute it into the formula:

[0092]

[0093] First, integrate with respect to x:

[0094]

[0095] Integrate over y again:

[0096]

[0097] Multiply the two:

[0098] M(x,y)=2400·124=297600;

[0099] The result shows that the integral of the span and beam height with respect to the support reaction force is 297600, which provides a basis for determining the parameter boundary interval. The upper and lower limits of the parameters are adjusted in combination with this value to generate a set of constraint boundaries.

[0100] S203: Based on the set of constraint boundaries between parameters, integrate the adjusted constraint boundary data into the parameter association network weight matrix, redistribute the weight values ​​in the network matrix according to the constraint boundaries, calculate the weight data distribution value after matrix adjustment, verify the consistency of the adjusted matrix, and generate a set of parameter constraint intervals.

[0101] The adjusted constraint boundary data is redistributed to the parameter association network weight matrix, the boundary set data is mapped to each weight node in the matrix, a piecewise linear relationship model is established to calculate the distribution value of the adjusted weight data in the matrix, consistency verification is performed on the weight node data, the error range after weight adjustment is analyzed and error compensation is performed, and finally the parameter constraint interval set is output.

[0102] Please see Figure 4 The specific steps of S3 are as follows:

[0103] S301: Based on the set of parameter constraint intervals, combine the numerical pairs of span and beam height parameters, gradually assign weight values ​​to each parameter path according to its numerical characteristics, calculate the path weights and analyze their distribution trend values, screen out the parameter combination range with stable weight distribution, and generate a set of path weights for span and beam height parameters.

[0104] By gradually assigning weight values ​​to analyze parameter paths, a set of parameter paths for span and beam height is generated based on numerical characteristics. When calculating path weights, the interaction characteristics of parameters are extracted through the correlation matrix, and a path weight distribution map is constructed. The local and overall trends of path weights are analyzed, and path combinations with large fluctuations are eliminated based on trend changes. Finally, parameter combinations with stable weight distribution are sorted out, and a set of parameter path weights for span and beam height is output.

[0105] S302: Based on the path weight set of span and beam height parameters, analyze the distribution characteristics of path weights, screen the combinations of path weight values ​​that satisfy the parameter constraints, calculate the cumulative weight value of the screened paths, and determine whether they meet the dynamic change conditions. Adjust the priority order of path combinations and generate a set of path combinations that satisfy the constraints.

[0106] The formula for calculating the cumulative weight value is as follows:

[0107]

[0108] Among them, Q tot Represents the cumulative weight value, q p,jz represents the weight of the p-th parameter point in the j-th group of path combinations. p,j This represents the path parameter value of the p-th parameter point in the j-th group of path combinations. t represents the average value of the path parameter values ​​in the j-th group of path combinations, and t represents the total number of parameter points in the j-th group of path combinations.

[0109] In the above formula, each parameter is analyzed and assigned a specific value. The j-th group of parameters in the path combination contains 5 points, with parameter values ​​of 38, 42, 36, 40, and 44, and weight values ​​of 0.2, 0.25, 0.15, 0.3, and 0.1, respectively. The weights are set based on the importance of the parameter points, and these values ​​are determined by the proportion of the load capacity of path points in historical data. The numerical range is selected through surveys as a uniform distribution of 0.1 to 0.3. The average value is calculated based on the path parameter values, obtained from monitoring, and then calculated using the arithmetic mean formula.

[0110] Parameter description:

[0111] Total number of parameter points: t = 5

[0112] Path parameter value: z 1,j =38, z 2,j =42, z 3,j =36, z 4,j =40, z 5,j =44

[0113] Weight value: q 1,j =0.2, q 2,j =0.25, q 3,j =0.15, q 4,j =0.3, q 5,j =0.1

[0114] Average path parameter value:

[0115] Calculation process:

[0116] Calculate the numerator:

[0117]

[0118] Calculate the denominator:

[0119]

[0120]

[0121] Calculation results:

[0122]

[0123] Numerical results analysis:

[0124] The results show that the cumulative weight value of the selected path in the j-th group is 0.3. This value reflects the combined influence of the deviation of the path parameter values ​​and the weight distribution. The lower cumulative weight value indicates that the distribution of path parameter points is more uniform and the weight allocation is reasonable. It is related to the possibility that the path combination meets the constraint conditions and provides a basis for adjusting the path priority order.

[0125] S303: Based on the set of path combinations that satisfy the constraints, optimize the part of the path combination with uneven weights, redistribute the parameter weight values ​​within the path according to the optimization results, integrate the optimized path combinations to form the final path scheme, calculate the dynamic weight cumulative value of the path scheme, and generate a set of dynamic path value parameters.

[0126] By analyzing the balance of path weight distribution, the weight values ​​of path parameters are adjusted for areas with uneven weight distribution. During the adjustment process, the weight distribution is optimized by combining the dynamic characteristics of the path and the constraints. The weight values ​​within the path are redistributed using a segment-by-segment adjustment method. The adjusted path weight distribution scheme is determined through repeated verification. The changing trend of the optimized path weight is further analyzed. Finally, the optimized path is integrated to form a dynamic weight cumulative value, and a set of dynamic path value parameters is generated.

[0127] Please see Figure 5 The specific steps of S4 are as follows:

[0128] S401: Based on the set of dynamic path values, assign live load, dead load and wind load to each parameter point in the path, analyze the numerical distribution characteristics and range of the load, adjust the load values ​​one by one to meet the bearing capacity of the path, calculate the force data of the load on the path, and select loads with stable distribution to generate a set of path load allocation results.

[0129] The specific formula for calculating the corrected force data is as follows:

[0130]

[0131] Among them, F k,i The value of the corrected force at the k-th parameter point in the path under the load distribution is represented by l. k w represents the actual span of the k-th point in the path. j,k P represents the weight of the j-th type of load at the k-th parameter point in the path. j,k A represents the distribution value of the j-th type of load at point k in the path. j,k R represents the area of ​​action of the j-th type of load at point k in the path. k represents the bearing capacity threshold of the k-th point in the path, and m represents the total number of load types.

[0132] Parameter assignment:

[0133] There are 5 parameter points along the path. The k=3rd parameter point has a span of 10 meters. There are 3 types of loads: live load, dead load, and wind load, with distribution weights of 0.4, 0.35, and 0.25 respectively. These weights are set based on the proportion of each load's contribution to the stress at the path points as recorded in the monitoring data. Distribution values ​​(unit: kN / m) 2 The effective areas are 20, 15, and 10 respectively (unit: m²). 2 The values ​​are 25, 30, and 40 respectively, with a bearing capacity threshold of 800kN.

[0134] Specific parameter value: l k =10w 1,3 =0.4,w 2,3 =0.35,w 3,3 =0.25P 1,3 =20,P 2,3 =15,P 3,3 =10A 1,3 =25,A 2,3 =30,A 3,3 =40R k =800

[0135] Calculation process:

[0136] Calculate the correction value for each load item by item:

[0137] w 1,3 ·|P 1,3 ·A 1,3 -R k |=0.4·|20·25-800|=0.4·|500-800|=0.4·

[0138] 300 = 120;

[0139] w 2,3 ·|P 2,3 ·A 2,3 -R k |=0.35·|15·30-800|=0.35·|450-800|=

[0140] 0.35·350=122.5;

[0141] w 3,3 ·|P 3,3 ·A 3,3 -R k |=0.25·|10·40-800|=0.25·|400-800|=

[0142] 0.25·400=100;

[0143] Summation:

[0144]

[0145] Calculate the final force value:

[0146]

[0147] Numerical results analysis:

[0148] The results indicate that the corrected force value at the third parameter point in the path is 34.25, reflecting the combined impact of different load types and parameter points. This value is directly related to the path bearing capacity verification results, providing a basis for screening loads with stable distribution, and further generating a set of path load distribution results.

[0149] S402: Based on the path load distribution result set, calculate the response values ​​of live load, dead load and wind load on span and beam height point by point, analyze the response trend under load according to the dynamic change range of parameters, extract the numerical range of load that has a significant impact on span and beam height, and generate load response numerical dataset.

[0150] Based on the path load allocation result set, according to the formula

[0151]

[0152] Calculate the numerical response of live load, dead load, and wind load to span and beam height.

[0153] In the formula, F ij L represents the load response value. i and H j t1 and t2 are the distribution values ​​of live load and dead load, respectively, V(t) is the time function of wind load, and t1 and t2 are the time range of wind load action.

[0154] Assume the span of the path is 40 meters, the beam height is 2 meters, the live load distribution value is 2000, the dead load distribution value is 3000, the wind load time function is V(t) = 50·sin(t), and the wind load action time range is t1 = 0 seconds to t2 = π seconds. Substituting into the formula:

[0155]

[0156] F ij =300000000·2=600000000;

[0157] The results show that the response values ​​of live load, dead load and wind load to span and beam height are 600000000, which can be used to analyze the parameter response trend under load and extract the numerical range that has a significant impact on span and beam height.

[0158] S403: Based on the load response numerical dataset, analyze the adjustment range of each load on the span and beam height parameters, integrate the parameter response range according to the numerical changes between path nodes, construct the adjustment interval matrix of span and beam height under load, and generate the load response parameter matrix.

[0159] The adjustment range of load on span and beam height is refined to each node of the path. The parameter response range is integrated based on the numerical change pattern between nodes. The adjustment range of each path node under load is determined by constructing an adjustment interval matrix. The adjusted matrix data is compared item by item with the dynamic change range of parameters to finally form the adjustment interval matrix of span and beam height under load and generate the load response parameter matrix.

[0160] Please see Figure 6 The specific steps of S5 are as follows:

[0161] S501: Based on the load response parameter matrix, analyze the adjustment range of the combination of span and beam height parameters one by one, check the validity of the combination values ​​point by point according to the constraints of the dynamic path, remove the parameters that do not meet the requirements, and select the parameter combinations that meet the conditions to generate a set of parameter combinations that meet the path constraints.

[0162] The adjustment range of the span and beam height parameter combinations is analyzed group by group, and the constraints of each parameter combination in the dynamic path are extracted. The parameter combination values ​​are checked against the constraints point by point. Parameter combinations that do not meet the requirements are eliminated. The screening results are further verified item by item by combining the changing trends of the constraints and parameter combinations after adjustment. Finally, the set of parameter combinations that meet the path constraints is output.

[0163] S502: Based on the set of parameter combinations that meet path constraints, calculate the numerical fluctuation of each parameter combination in the load response, analyze the differences in the numerical distribution of the load on the parameter combination, screen the parameter combination with balanced load action and the smallest fluctuation range, integrate the screened parameter combinations, and generate a set of balanced load parameter combinations.

[0164] Based on the set of parameter combinations that meet path constraints, according to the formula

[0165]

[0166] Calculate the numerical fluctuation of each parameter combination in the load response.

[0167] In the formula, S i R represents the fluctuation value of the parameter combination. ij Let j be the response value. Let be the average response value of the i-th parameter combination, and n be the number of responses.

[0168] Given a parameter combination with response values ​​of 50, 55, 52, 48, and 53, calculate the fluctuation value S. i Number of responses n=5, average response value for:

[0169]

[0170] Substitute into the formula:

[0171]

[0172] Calculate the square of the difference for each term:

[0173] (50-51.6) 2 =2.56, (55-51.6) 2 =11.56, (52-51.6) 2 =0.16, (48-51.6) 2 =12.96, (53-51.6) 2 =1.96;

[0174] Summing and substituting:

[0175]

[0176] The results show that the parameter combination has a numerical fluctuation of 2.42 in the load response. The parameter combination with the smallest fluctuation range and balanced load action is used to screen out the parameter combination and integrate the screened parameter combination.

[0177] S503: Based on the set of balanced load parameter combinations, the adjustment values ​​of the span and beam height parameter combinations are globally checked. The parameter ranges of span and beam height are corrected according to the load influence value of each combination. The corrected parameter combinations are integrated, and the overall optimized values ​​are calculated to generate the optimized integrated values ​​of bridge parameters.

[0178] The load influence values ​​of each combination are correlated with the parameter range data. The parameter ranges of span and beam height are gradually corrected for the load influence values ​​of each parameter combination. The piecewise linear interpolation method is used to optimize the adjustment process. The corrected parameter combinations are verified by dynamic load action. The response results of the corrected parameter combinations are further integrated and the overall optimized values ​​are calculated. Finally, the optimized integrated values ​​of bridge parameters are generated.

[0179] Please see Figure 7 An interactive generative bridge parameter design system, comprising:

[0180] The parameter association module analyzes the effect of span on beam height and the influence of beam height on support reaction force based on span, beam height and support reaction force parameters. It constructs a set of node and edge weights through parameter relationships and generates a parameter association network weight matrix.

[0181] The constraint interval module extracts the influence range of span and beam height on support reaction force based on the parameter association network weight matrix, determines the constraint boundary values ​​between parameters, and generates a set of parameter constraint intervals.

[0182] The path weighting module combines the span and beam height parameters based on the set of parameter constraint intervals, calculates the weight of each parameter path, selects the path combination that meets the constraint conditions, and generates a set of dynamic parameter path values.

[0183] The load response module allocates live load, dead load and wind load to each path based on the set of parameter dynamic path values, calculates the load response to span and beam height, and generates a load response parameter matrix.

[0184] The parameter optimization module performs a global check on the parameter adjustment values ​​of the span and beam height combination based on the load response parameter matrix, selects combinations that meet the dynamic path and load response, and generates optimized integrated values ​​for bridge parameters.

[0185] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. An interactive generative bridge parameter design method, characterized in that, Includes the following steps: S1: Based on the span, beam height and support reaction parameters, analyze the effect of span on beam height and the influence of beam height on support reaction. Construct a set of node and edge weights through parameter relationships to form a network structure and generate a parameter-related network weight matrix. S2: Based on the parameter association network weight matrix, extract the influence range of span and beam height on support reaction force, determine the constraint boundary values ​​between parameters, integrate the constraint results into the network matrix, and generate a set of parameter constraint intervals; S3: Based on the set of parameter constraint intervals, combine the span and beam height parameters, calculate the weight of each parameter path, select the path combination that meets the constraint conditions, form a path scheme, and generate a set of dynamic parameter path values. The specific steps are as follows: S301: Based on the set of parameter constraint intervals, combine the numerical pairs of span and beam height parameters, gradually assign weight values ​​to each parameter path according to its numerical characteristics, calculate the path weights and analyze their distribution trend values, select the parameter combination range with stable weight distribution, and generate a set of path weights for span and beam height parameters. S302: Based on the path weight set of the span and beam height parameters, analyze the distribution characteristics of the path weights, screen the combinations of path weight values ​​that satisfy the parameter constraints, calculate the cumulative weight value of the screened paths, and determine whether they meet the dynamic change conditions. Adjust the priority order of the path combinations and generate a set of path combinations that satisfy the constraints. S303: Based on the set of path combinations that satisfy the constraints, optimize the part of the path combination with uneven weights, redistribute the parameter weight values ​​within the path according to the optimization results, integrate the optimized path combinations to form the final path scheme, calculate the dynamic weight cumulative value of the path scheme, and generate a set of dynamic path value parameters. The formula for calculating the cumulative weight value is as follows: ； in, Represents the cumulative weight value. Represents the path combination number The first in the group The weights of each parameter point Represents the path combination number The first in the group The path parameter values ​​for each parameter point. Represents the path combination number The average value of the path parameter values ​​in the group. Represents the path combination number The total number of parameter points in the group; S4: Based on the set of dynamic path values ​​of the parameters, the live load, dead load and wind load are allocated to each path, the load response to the span and beam height is calculated, the adjustment range of each parameter under the load is analyzed, and the load response parameter matrix is ​​generated. S5: Based on the load response parameter matrix, perform a global check on the parameter adjustment values ​​of the span and beam height combination, select combinations that meet the dynamic path and load response, check all combination results, and generate optimized integrated values ​​for bridge parameters.

2. The interactive generative bridge parameter design method according to claim 1, characterized in that, The parameter association network weight matrix includes a span node set, a beam height node set, a support reaction force node set, and a weight distribution between nodes. The parameter constraint interval set includes span range constraints, beam height range constraints, support reaction force range constraints, and the coupling boundary between span and beam height. The parameter dynamic path value set includes path weight values, support reaction force dynamic range values, and combined paths of span and beam height. The load response parameter matrix includes live load response coefficient, dead load response coefficient, wind load response coefficient, and path response adjustment value. The bridge parameter optimization and integration value includes optimized span value, optimized beam height value, and optimized support reaction force value.

3. The interactive generative bridge parameter design method according to claim 1, characterized in that, Based on the parameters of span, beam height, and support reaction force, the effects of span on beam height and beam height on support reaction force are analyzed. A set of node and edge weights is constructed through parameter relationships to form a network structure. The specific steps for generating the parameter-related network weight matrix are as follows: S101: Based on the parameters of span, beam height and support reaction force, extract the numerical range and distribution characteristics of span and beam height, group the spans according to the parameter distribution law, calculate the variation range and fluctuation range of beam height corresponding to the span, set the span variation threshold, filter the data with large span variation range, and establish the influence characteristics of span variation on beam height. S102: Based on the influence characteristics of the span change on the beam height, analyze the changing trend of the beam height and the support reaction force, fit the relationship between the support reaction force data and the beam height parameters, determine the key node parameters of the beam height change on the support reaction force, adjust the parameter range, calculate the correlation value between the beam height and the support reaction force, and generate a beam height and support reaction force correlation dataset. S103: Based on the aforementioned beam height and support reaction force correlation dataset, construct a set of node and edge weights for the parameter correlation network, group the network according to the weight between nodes, filter out regions with large weight fluctuations and adjust edge weights, calculate the value of the overall network weight matrix, and generate the parameter correlation network weight matrix.

4. The interactive generative bridge parameter design method according to claim 1, characterized in that, Based on the parameter association network weight matrix, the specific steps for extracting the influence range of span and beam height on support reaction force, determining the constraint boundary values ​​between parameters, integrating the constraint results into the network matrix, and generating a set of parameter constraint intervals are as follows: S201: Based on the parameter association network weight matrix, extract the interaction range of span, beam height and support reaction force parameters, analyze the interaction weight data between parameters, screen the main range of span and beam height affecting support reaction force according to the weight change trend, calculate the weight distribution of span and beam height on support reaction force and determine its range, and generate a set of the range of span and beam height affecting support reaction force. S202: Based on the set of influence ranges of span and beam height on support reaction force, analyze the distribution law of weight within the influence range, combine the numerical boundary characteristics between span, beam height and support reaction force, calculate the boundary numerical range of span and beam height on support reaction force parameters, screen constraint conditions according to numerical changes and judge their effectiveness, adjust the upper and lower bounds of parameters, and generate a set of constraint boundaries between parameters. S203: Based on the set of constraint boundaries between parameters, integrate the adjusted constraint boundary data into the parameter association network weight matrix, redistribute the weight values ​​in the network matrix according to the constraint boundaries, calculate the weight data distribution value after matrix adjustment, verify the consistency of the adjusted matrix, and generate a set of parameter constraint intervals.

5. The interactive generative bridge parameter design method according to claim 1, characterized in that, Based on the aforementioned set of dynamic path values, the live load, dead load, and wind load are allocated to each path. The load response to span and beam height is calculated, the adjustment range of each parameter under load is analyzed, and the specific steps for generating the load response parameter matrix are as follows: S401: Based on the set of dynamic path values ​​of the parameters, allocate live load, dead load and wind load to each parameter point in the path, analyze the numerical distribution characteristics and range of the load, adjust the load values ​​one by one to meet the bearing capacity of the path, calculate the force data of the load on the path, and select loads with stable distribution to generate a set of path load allocation results. S402: Based on the set of path load distribution results, calculate the response values ​​of live load, dead load and wind load to span and beam height point by point, analyze the response trend under load according to the dynamic change range of parameters, extract the numerical range of load that has a significant impact on span and beam height, and generate a load response numerical dataset. S403: Based on the load response numerical dataset, analyze the adjustment range of each load on the span and beam height parameters, integrate the parameter response range according to the numerical changes between path nodes, construct the adjustment interval matrix of span and beam height under load, and generate the load response parameter matrix.

6. The interactive generative bridge parameter design method according to claim 1, characterized in that, Based on the load response parameter matrix, the parameter adjustment values ​​for the combination of span and beam height are globally checked, combinations that conform to the dynamic path and load response are selected, all combination results are checked, and the specific steps for generating optimized and integrated bridge parameter values ​​are as follows: S501: Based on the load response parameter matrix, analyze the adjustment range of the combination of span and beam height parameters one by one, check the validity of the combination values ​​point by point according to the constraints of the dynamic path, remove the parameters that do not meet the requirements, and select the parameter combinations that meet the conditions to generate a set of parameter combinations that meet the path constraints. S502: Based on the set of parameter combinations that meet the path constraints, calculate the numerical fluctuation of each parameter combination in the load response, analyze the differences in the numerical distribution of the load on the parameter combination, screen the parameter combination with balanced load action and the smallest fluctuation range, integrate the screened parameter combinations, and generate a set of balanced load parameter combinations. S503: Based on the set of balanced load parameter combinations, globally check the adjustment values ​​of the span and beam height parameter combinations, correct the parameter ranges of span and beam height according to the load influence value of each combination, integrate the corrected parameter combinations, calculate the overall optimized values, and generate the optimized integrated values ​​of bridge parameters.

7. An interactive generative bridge parameter design system, characterized in that, According to any one of claims 1-6, the interactive generative bridge parameter design method, the system comprises: The parameter association module analyzes the effect of span on beam height and the influence of beam height on support reaction force based on span, beam height and support reaction force parameters. It constructs a set of node and edge weights through parameter relationships and generates a parameter association network weight matrix. Based on the parameter association network weight matrix, the constraint interval module extracts the influence range of span and beam height on support reaction force, determines the constraint boundary values ​​between parameters, and generates a set of parameter constraint intervals. The path weighting module combines the span and beam height parameters based on the set of parameter constraint intervals, calculates the weight of each parameter path, selects the path combination that meets the constraint conditions, and generates a set of dynamic parameter path values. Based on the set of dynamic path values ​​of the parameters, the load response module allocates live load, dead load and wind load to each path, calculates the load response to the span and beam height, and generates a load response parameter matrix. Based on the load response parameter matrix, the parameter optimization module performs a global check on the parameter adjustment values ​​of the span and beam height combination, selects combinations that conform to the dynamic path and load response, and generates optimized integrated values ​​for bridge parameters.

Images