Optimization method and system for municipal continuous beam bridges based on deep learning
Through a deep learning-based method, the cross-section variables and stress curves of municipal continuous beam bridges are extracted, and the cross-section design is optimized, which solves the problem of low optimization accuracy in the existing technology, and achieves more efficient material use and design accuracy.
Patent Information
- Application Number
- CN202410373490.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-03-29
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2044-03-29
AI Technical Summary
The existing municipal continuous beam bridge optimization algorithm fails to effectively consider actual engineering requirements, resulting in low optimization accuracy and overly one-sided design, which fails to fully utilize the differences between different types of bridges.
Using a deep learning-based method, the box beam cross-section variables corresponding to the mid-span section and fulcrum section are extracted, the cross-section area and number of reinforcement are calculated, and the stress curve is analyzed using the GA-BP neural network model to optimize the cross-section size to improve optimization accuracy. .
Through deep learning methods, the cross-sectional design of municipal continuous beam bridges is optimized, material usage is reduced, design accuracy and efficiency are improved, potential stress concentration areas and weak areas are discovered, thereby reducing construction costs and material waste.
Smart Images

Figure CN118211303B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of bridge optimization, and in particular to a method and system for optimizing a municipal continuous beam bridge based on deep learning. Background Art
[0002] The municipal continuous beam bridge is a special type of bridge structure, mainly used in urban roads and transportation systems. It uses a continuous beam as its structure and can span longer distances without the need for supporting columns or piers. This bridge design can reduce interference with waterway traffic while providing a larger passage space and can span longer distances. In order to reduce the energy consumption of continuous beam bridges, it is necessary to optimize the materials used in the bridge.
[0003] The existing municipal bridge engineering field usually adopts a more conservative design strategy in design methods to ensure the performance of municipal bridges, which will lead to excessively large cross-sectional dimensions of municipal bridges. Therefore, a larger reinforcement ratio is required to meet the reinforcement requirements of the cross-sectional dimensions. Although there are relevant optimization algorithms that can be used to optimize the cross-sectional dimensions and steel bar configuration of municipal bridges, and take cost, carbon emissions or energy consumption as optimization objective functions, these algorithms have the following problems: 1. They do not take into account the actual requirements of municipal bridge engineering, 2. They are relatively one-sided optimization design calculations, and 3. The differences between different types of municipal bridges lead to low optimization accuracy of municipal continuous beam bridges. Summary of the invention
[0004] The present invention provides a method and system for optimizing a municipal continuous beam bridge based on deep learning, the main purpose of which is to improve the optimization accuracy of the municipal continuous beam bridge.
[0005] To achieve the above purpose, the present invention provides a method for optimizing a municipal continuous beam bridge based on deep learning, comprising:
[0006] Selecting a mid-span section and a support section of the continuous beam bridge to be optimized, respectively extracting box girder section variables corresponding to the mid-span section and the support section, obtaining a mid-span box girder section variable and a support box girder section variable, calculating a mid-span section area corresponding to the mid-span section according to the mid-span box girder section variable, and calculating a support section area corresponding to the support section according to the support box girder section variable;
[0007] According to the mid-span cross-sectional area, the number of mid-span reinforcements corresponding to the mid-span cross-section is calculated, and according to the support cross-sectional area, the number of support reinforcements corresponding to the support cross-section is calculated;
[0008] The optimal mid-span cross-sectional area of the continuous beam bridge is calculated by combining the number of mid-span reinforcement and the mid-span cross-sectional area, and the optimal support cross-sectional area of the continuous beam bridge is calculated by combining the number of support reinforcement and the support cross-sectional area;
[0009] According to the optimal mid-span cross-sectional area and the optimal support cross-sectional area, the cross-sectional dimension information corresponding to the mid-span cross-sectional area and the support cross-sectional area is determined, and the optimized dimension information is input into a pre-trained GA-BP neural network model. The GA-BP neural network model is used to analyze the stress curve corresponding to the continuous beam bridge, and the cross-sectional dimension information is optimized in combination with the stress curve to obtain an optimization result.
[0010] Optionally, calculating the mid-span cross-sectional area corresponding to the mid-span cross-section according to the mid-span box girder cross-sectional variable includes:
[0011] The mid-span cross-sectional area corresponding to the mid-span cross-section is calculated by the following formula:
[0012]
[0013] Among them, A c represents the mid-span cross-sectional area corresponding to the mid-span section, e a represents the mid-span top plate thickness in the mid-span box girder section variable, b represents the mid-span web plate width in the mid-span box girder section variable, e d represents the mid-span bottom plate thickness in the mid-span box girder section variable, e b represents the mid-span web thickness in the mid-span box girder section variable, h represents the mid-span box girder height in the mid-span box girder section variable, d represents the mid-span flange plate width in the mid-span box girder section variable, e c Represents the mid-span root thickness in the mid-span box girder section variable.
[0014] Optionally, calculating the amount of mid-span reinforcement corresponding to the mid-span section according to the mid-span section area includes:
[0015] According to the mid-span cross-sectional area, the mid-span upper core distance and the mid-span lower core distance corresponding to the mid-span section are calculated respectively, and the mid-span steel bar eccentricity corresponding to the mid-span section is queried. According to the mid-span steel bar eccentricity, the mid-span upper edge eccentricity and the mid-span lower edge eccentricity corresponding to the mid-span section are calculated respectively. In combination with the mid-span upper edge eccentricity, the mid-span lower edge eccentricity, the mid-span upper core distance and the mid-span lower core distance, the mid-span reinforcement quantity corresponding to the mid-span section is calculated.
[0016] Optionally, the calculating the amount of mid-span reinforcement corresponding to the mid-span section by combining the mid-span upper edge eccentricity, the mid-span lower edge eccentricity, the mid-span upper core distance and the mid-span lower core distance comprises:
[0017] The amount of mid-span reinforcement corresponding to the mid-span section can be calculated using the following formula:
[0018]
[0019] Among them, N s Indicates the amount of mid-span reinforcement corresponding to the mid-span section, e s represents the eccentricity of the upper edge of the mid-span, Mmax represents the maximum bending moment of the mid-span section, M min represents the minimum bending moment at the mid-span section, e x represents the eccentricity of the lower edge of the mid-span, k s represents the mid-span upper core distance, k x Indicates the mid-span lower core distance, A p Represents the cross-sectional area of a single steel bar, σ pe Represents the cross-sectional area of a single steel bar.
[0020] Optionally, calculating the number of support reinforcements corresponding to the support cross section according to the support cross section area includes:
[0021] According to the cross-sectional area of the fulcrum, the upper core distance and the lower core distance of the fulcrum corresponding to the fulcrum cross-section are calculated respectively, and the eccentricity of the fulcrum steel bar corresponding to the fulcrum cross-section is queried. According to the eccentricity of the fulcrum steel bar, the upper edge eccentricity and the lower edge eccentricity of the fulcrum corresponding to the fulcrum cross-section are calculated. In combination with the upper edge eccentricity of the fulcrum, the lower edge eccentricity of the fulcrum, the upper core distance of the fulcrum and the lower core distance of the fulcrum, the amount of fulcrum reinforcement corresponding to the fulcrum cross-section is calculated.
[0022] Optionally, the calculating the number of support reinforcements corresponding to the support cross section by combining the upper edge eccentricity of the support, the lower edge eccentricity of the support, the upper core distance of the support and the lower core distance of the support comprises:
[0023] The number of support reinforcements corresponding to the support section is calculated using the following formula:
[0024]
[0025] Among them, N y Indicates the number of support reinforcements corresponding to the support section, k y Indicates the core distance on the fulcrum, e y Indicates the eccentricity of the upper edge of the fulcrum, G max Indicates the maximum bending moment of the support section, G min represents the minimum bending moment of the support section, e z Indicates the eccentricity of the lower edge of the fulcrum, k z Indicates the core distance below the fulcrum, A p Represents the cross-sectional area of a single steel bar, σpe Represents the cross-sectional area of a single steel bar.
[0026] Optionally, the calculating the optimal mid-span cross-sectional area of the continuous beam bridge by combining the mid-span reinforcement quantity and the mid-span cross-sectional area includes:
[0027] According to the number of mid-span reinforcements, the steel bar area corresponding to the mid-span section is calculated to obtain the mid-span steel bar area; according to the mid-span steel bar area, the mid-span concrete area corresponding to the mid-span section is calculated; according to the mid-span concrete area, the concrete width corresponding to the mid-span section is determined to obtain the mid-span concrete width; according to the mid-span concrete width, the mid-span compression height corresponding to the mid-span section is calculated; according to the mid-span compression height and the mid-span concrete width, the section bending bearing capacity corresponding to the mid-span section is calculated; according to the section bending bearing capacity and a preset bearing capacity threshold, the optimal mid-span section area of the continuous beam bridge is screened out from the mid-span section area.
[0028] Optionally, analyzing the stress curve corresponding to the continuous beam bridge using the GA-BP neural network model includes:
[0029] The standard size information corresponding to the GA-BP neural network model is obtained, and the input layer in the GA-BP neural network model is used to normalize the optimized size information to obtain the target size information, and the convolution layer in the GA-BP neural network model is used to extract the features of the target size information to obtain the size information features, and the fitness function in the GA-BP neural network model is used to calculate the fitness value between the size information features and the standard size information, and the activation function in the GA-BP neural network model is used to calculate the bridge stress value corresponding to the continuous beam bridge according to the fitness value and the target size information, and the stress curve corresponding to the continuous beam bridge is drawn according to the bridge stress value.
[0030] Optionally, the calculating the fitness value between the size information feature and the standard size information by using the fitness function in the GA-BP neural network model includes:
[0031] The fitness function includes:
[0032]
[0033] Among them, Q represents the adaptation value between the size information feature and the standard size information, T represents the number of features of the size information feature, i and i+1 represent the serial numbers of the size information feature and the standard size information respectively, and R i,i+1 Indicates the feature distance value corresponding to the i-th size information feature and the i+1-th standard size information, U irepresents the feature mapping value corresponding to the i-th size information feature, U i+1 Indicates the feature map value corresponding to the i+1th standard size information.
[0034] A system for optimizing municipal continuous beam bridges based on deep learning, characterized in that the system comprises:
[0035] A section calculation module is used to select the mid-span section and the support section of the continuous beam bridge to be optimized, respectively extract the box girder section variables corresponding to the mid-span section and the support section, obtain the mid-span box girder section variable and the support box girder section variable, calculate the mid-span section area corresponding to the mid-span section according to the mid-span box girder section variable, and calculate the support section area corresponding to the support section according to the support box girder section variable;
[0036] A reinforcement quantity calculation module, used to calculate the number of mid-span reinforcements corresponding to the mid-span section according to the mid-span section area, and calculate the number of support reinforcements corresponding to the support section according to the support section area;
[0037] An optimal cross-section calculation module, used to calculate the optimal mid-span cross-section area of the continuous beam bridge by combining the number of mid-span reinforcement and the mid-span cross-section area, and to calculate the optimal support cross-section area of the continuous beam bridge by combining the number of support reinforcement and the support cross-section area;
[0038] The optimization analysis module is used to determine the cross-sectional size information corresponding to the mid-span section and the support section according to the optimal mid-span cross-sectional area and the optimal support cross-sectional area, input the optimized size information into a pre-trained GA-BP neural network model, use the GA-BP neural network model to analyze the stress curve corresponding to the continuous beam bridge, and optimize the cross-sectional size information in combination with the stress curve to obtain an optimization result.
[0039] The present invention calculates the mid-span cross-sectional area corresponding to the mid-span cross-section according to the mid-span box girder cross-sectional variable, and can obtain the cross-sectional area corresponding to the mid-span cross-section, which is convenient for the subsequent calculation of the number of mid-span reinforcements. The present invention calculates the number of mid-span reinforcements corresponding to the mid-span cross-sectional area according to the mid-span cross-sectional area, and can obtain the number of steel bars required for the mid-span cross-sectional area, which is convenient for the subsequent calculation of the cross-sectional area of the continuous beam bridge. The present invention calculates the optimal mid-span cross-sectional area of the continuous beam bridge by combining the number of mid-span reinforcements and the mid-span cross-sectional area, so as to minimize the material usage of the continuous beam bridge, thereby reducing construction costs and material waste. The present invention analyzes the stress curve corresponding to the continuous beam bridge by using the GA-BP neural network model, and can find potential stress concentration areas and weak parts, and then optimize the design of the continuous beam bridge, thereby improving the optimization accuracy of the continuous beam bridge. Therefore, the optimization method and system for realizing municipal continuous beam bridges based on deep learning provided by the embodiment of the present invention can improve the optimization accuracy of continuous beam bridges. BRIEF DESCRIPTION OF THE DRAWINGS
[0040] Figure 1 A schematic flow chart of a method for optimizing a municipal continuous beam bridge based on deep learning provided by one embodiment of the present invention;
[0041] Figure 2 A functional module diagram of a system for optimizing municipal continuous beam bridges based on deep learning provided in one embodiment of the present invention.
[0042] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION
[0043] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.
[0044] The embodiment of the present application provides an optimization method for municipal continuous beam bridges based on deep learning. In the embodiment of the present application, the execution subject of the optimization method for municipal continuous beam bridges based on deep learning includes but is not limited to at least one of the electronic devices such as a server and a terminal that can be configured to execute the method provided by the embodiment of the present application. In other words, the optimization method for municipal continuous beam bridges based on deep learning can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc. The server can be an independent server, or it can be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), and big data and artificial intelligence platforms.
[0045] Reference Figure 1 FIG. 1 is a flow chart of a method for optimizing a municipal continuous beam bridge based on deep learning according to an embodiment of the present invention. In this embodiment, the method for optimizing a municipal continuous beam bridge based on deep learning includes steps S1 to S4.
[0046] S1. Select the mid-span section and support section of the continuous beam bridge to be optimized, extract the box girder section variables corresponding to the mid-span section and the support section respectively, obtain the mid-span box girder section variable and the support box girder section variable, calculate the mid-span cross-sectional area corresponding to the mid-span section according to the mid-span box girder section variable, and calculate the support cross-sectional area corresponding to the support section according to the support box girder section variable.
[0047] The present invention calculates the mid-span cross-sectional area corresponding to the mid-span cross-section based on the mid-span box girder cross-sectional variable, and can obtain the cross-sectional area corresponding to the mid-span cross-section, which is convenient for the subsequent calculation of the number of mid-span reinforcements, wherein the mid-span cross-section is the cross-sectional area at the middle position corresponding to the continuous beam bridge, the support cross-sectional area is the beam cross-sectional area closest to the support of the continuous beam bridge, and is also the position that bears the maximum shear force and reverse bending moment, the box girder cross-sectional variable is a parameter or variable that describes the shape and size of the box girder cross-sectional area corresponding to the mid-span cross-sectional area and the support cross-sectional area, the mid-span cross-sectional area is the specific area corresponding to the mid-span cross-sectional area, and the support cross-sectional area is the area corresponding to the support cross-sectional area. Optionally, the selection of the mid-span cross-sectional area and the support cross-sectional area of the continuous beam bridge to be optimized can be achieved by a simulated annealing algorithm, and the calculation principle of the support cross-sectional area is the same as that of the mid-span cross-sectional area, which will not be elaborated herein.
[0048] As an embodiment of the present invention, the mid-span cross-sectional area corresponding to the mid-span cross-section is calculated according to the mid-span box girder cross-sectional variable, including:
[0049] The mid-span cross-sectional area corresponding to the mid-span cross-section is calculated by the following formula:
[0050]
[0051] Among them, A c represents the mid-span cross-sectional area corresponding to the mid-span section, e a represents the mid-span top plate thickness in the mid-span box girder section variable, b represents the mid-span web plate width in the mid-span box girder section variable, e d represents the mid-span bottom plate thickness in the mid-span box girder section variable, e b represents the mid-span web thickness in the mid-span box girder section variable, h represents the mid-span box girder height in the mid-span box girder section variable, d represents the mid-span flange plate width in the mid-span box girder section variable, e c Represents the mid-span root thickness in the mid-span box girder section variable.
[0052] S2. Calculate the number of mid-span reinforcements corresponding to the mid-span section according to the mid-span cross-sectional area, and calculate the number of support reinforcements corresponding to the support section according to the support cross-sectional area.
[0053] The present invention calculates the number of mid-span reinforcements corresponding to the mid-span section according to the mid-span cross-sectional area, and can obtain the number of steel bars required for the mid-span section, thereby facilitating the subsequent calculation of the cross-sectional area of the continuous beam bridge, wherein the number of mid-span reinforcements is the number of steel bars required for the mid-span section.
[0054] As an embodiment of the present invention, the method of calculating the amount of mid-span reinforcement corresponding to the mid-span section according to the mid-span cross-sectional area includes: calculating the upper mid-span core distance and the lower mid-span core distance corresponding to the mid-span section according to the mid-span cross-sectional area, querying the mid-span steel bar eccentricity corresponding to the mid-span section, calculating the upper mid-span eccentricity and the lower mid-span eccentricity corresponding to the mid-span section according to the mid-span steel bar eccentricity, and calculating the amount of mid-span reinforcement corresponding to the mid-span section in combination with the upper mid-span eccentricity, the lower mid-span eccentricity, the upper mid-span core distance and the lower mid-span core distance.
[0055] Among them, the upper core distance at mid-span is the distance between the center of the beam section of the mid-span section and the top surface, the lower core distance at mid-span is the distance between the center of the beam section of the mid-span section and the bottom surface, the mid-span steel bar eccentricity is the distance from the center of gravity of the steel bar section in the mid-span section to the edge of the mid-span section, the upper edge eccentricity at mid-span is the offset distance of the center of the mid-span section relative to the upper edge, and the lower edge eccentricity at mid-span is the offset distance of the center of the mid-span section relative to the lower edge.
[0056] Optionally, the calculation formula for the upper core distance at mid-span is: upper core distance at mid-span = moment of inertia of section / (mid-span cross-sectional area × distance from upper edge of mid-span to longitudinal neutral axis of mid-span section), the calculation formula for the lower core distance at mid-span is: lower core distance at mid-span = moment of inertia of section / (mid-span cross-sectional area × distance from lower edge of mid-span to longitudinal neutral axis of mid-span section), the moment of inertia of section describes the influence of the mid-span section on the rotational inertia of the object, in structural mechanics, the moment of inertia is also called the second-order moment of section or the cross-sectional area moment, the mid-span steel bar eccentricity corresponding to the mid-span section can be obtained by querying the bridge database, the calculation formula for the upper edge eccentricity at mid-span is: upper edge eccentricity at mid-span = distance from upper edge of mid-span of the mid-span section to longitudinal neutral axis - mid-span steel bar eccentricity, the calculation formula for the lower edge eccentricity at mid-span is: lower edge eccentricity at mid-span = distance from lower edge of mid-span to longitudinal neutral axis of mid-span section - mid-span steel bar eccentricity.
[0057] Optionally, as an optional embodiment of the present invention, the calculation of the amount of mid-span reinforcement corresponding to the mid-span section by combining the mid-span upper edge eccentricity, the mid-span lower edge eccentricity, the mid-span upper core distance and the mid-span lower core distance includes:
[0058] The amount of mid-span reinforcement corresponding to the mid-span section can be calculated using the following formula:
[0059]
[0060] Among them, N s Indicates the amount of mid-span reinforcement corresponding to the mid-span section, e s represents the eccentricity of the upper edge of the mid-span, Mmax represents the maximum bending moment of the mid-span section, M min represents the minimum bending moment at the mid-span section, e x represents the eccentricity of the lower edge of the mid-span, k s represents the mid-span upper core distance, k x Indicates the mid-span lower core distance, A p Represents the cross-sectional area of a single steel bar, σ pe Represents the cross-sectional area of a single steel bar.
[0061] The present invention calculates the number of support reinforcements corresponding to the support cross-section based on the support cross-sectional area, and can obtain the number of steel bars required for the support cross-section, thereby improving the calculation accuracy of the cross-sectional area of the continuous beam bridge, wherein the number of support reinforcements is the required number of steel bars corresponding to the support cross-section.
[0062] As an embodiment of the present invention, the method of calculating the number of fulcrum reinforcement corresponding to the fulcrum cross-section according to the fulcrum cross-sectional area includes: calculating the fulcrum upper core distance and the fulcrum lower core distance corresponding to the fulcrum cross-section according to the fulcrum cross-sectional area, querying the fulcrum steel bar eccentricity corresponding to the fulcrum cross-section, calculating the fulcrum upper edge eccentricity and the fulcrum lower edge eccentricity corresponding to the fulcrum cross-section according to the fulcrum steel bar eccentricity, and calculating the number of fulcrum reinforcement corresponding to the fulcrum cross-section in combination with the fulcrum upper edge eccentricity, the fulcrum lower edge eccentricity, the fulcrum upper core distance and the fulcrum lower core distance.
[0063] Among them, the upper core distance of the fulcrum and the lower core distance of the fulcrum are respectively the distances between the center of the beam section of the fulcrum section and the top surface and the bottom surface, the fulcrum steel bar eccentricity is the distance from the center of gravity of the steel bar section in the fulcrum section to the edge of the fulcrum section, the upper edge eccentricity of the fulcrum and the lower edge eccentricity of the fulcrum are respectively the offset distances of the center of the fulcrum section relative to the upper edge and the lower edge, optionally, the upper core distance of the fulcrum and the lower core distance of the fulcrum corresponding to the fulcrum section are calculated in accordance with the same principles as the above-mentioned mid-span upper core distance and mid-span lower core distance, and no further details are given here. The upper edge eccentricity of the fulcrum and the lower edge eccentricity of the fulcrum corresponding to the fulcrum section are calculated in accordance with the same principles as the above-mentioned mid-span upper edge eccentricity and mid-span lower edge eccentricity, and no further details are given here.
[0064] Optionally, as an optional embodiment of the present invention, the calculation of the number of support reinforcements corresponding to the support cross section by combining the upper edge eccentricity of the support, the lower edge eccentricity of the support, the upper core distance of the support, and the lower core distance of the support includes:
[0065] The number of support reinforcements corresponding to the support section is calculated using the following formula:
[0066]
[0067] Among them, N y Indicates the number of support reinforcements corresponding to the support section, k y Indicates the core distance on the fulcrum, e y Indicates the eccentricity of the upper edge of the fulcrum, G max Indicates the maximum bending moment of the support section, G min represents the minimum bending moment of the support section, e z Indicates the eccentricity of the lower edge of the fulcrum, kz Indicates the core distance below the fulcrum, A p Represents the cross-sectional area of a single steel bar, σ pe Represents the cross-sectional area of a single steel bar.
[0068] S3. Calculate the optimal mid-span cross-sectional area of the continuous beam bridge by combining the amount of mid-span reinforcement and the mid-span cross-sectional area, and calculate the optimal support cross-sectional area of the continuous beam bridge by combining the amount of support reinforcement and the support cross-sectional area.
[0069] The present invention calculates the optimal mid-span cross-sectional area of the continuous beam bridge by combining the mid-span reinforcement quantity and the mid-span cross-sectional area, so as to minimize the material usage of the continuous beam bridge, thereby reducing construction costs and material waste. Optionally, the optimal support cross-sectional area is calculated on the same principle as the optimal mid-span cross-sectional area, which will not be elaborated herein.
[0070] As an embodiment of the present invention, the optimal mid-span cross-sectional area of the continuous beam bridge is calculated by combining the mid-span reinforcement quantity and the mid-span cross-sectional area, including: calculating the steel bar area corresponding to the mid-span cross-section according to the mid-span reinforcement quantity to obtain the mid-span steel bar area; calculating the mid-span concrete area corresponding to the mid-span cross-section according to the mid-span steel bar area; determining the concrete width corresponding to the mid-span cross-section according to the mid-span concrete area to obtain the mid-span concrete width; calculating the mid-span compression height corresponding to the mid-span cross-section according to the mid-span concrete width; calculating the section bending bearing capacity corresponding to the mid-span section according to the mid-span compression height and the mid-span concrete width; and screening the optimal mid-span cross-sectional area of the continuous beam bridge from the mid-span cross-sectional area according to the section bending bearing capacity and a preset bearing capacity threshold.
[0071] Among them, the mid-span steel bar area is the total area of the steel bar section corresponding to the mid-span section, the mid-span concrete area is the area of concrete required for the mid-span section, the concrete width is the width corresponding to the concrete in the mid-span section, the mid-span compressive height is the height of the area where the mid-span section is deformed due to pressure, the section bending bearing capacity is the ability of the mid-span section to resist bending deformation and damage when subjected to bending moment, and the preset bearing capacity threshold is the design bearing capacity corresponding to the mid-span section.
[0072] Optionally, the formula for calculating the steel bar area corresponding to the mid-span section is: mid-span steel bar area = the number of mid-span reinforcements * the area of a single steel bar. The mid-span concrete area can be obtained by subtracting the mid-span steel bar area from the mid-span section. The concrete thickness can be calculated by combining the cross-sectional dimension information in the mid-span box girder section variable and the mid-span steel bar dimension information. The concrete width corresponding to the mid-span section is obtained by dividing the mid-span concrete area by the concrete thickness. The mid-span compressive height corresponding to the mid-span section is calculated as follows: mid-span compressive height = f pd A p_all -f sd A s / f cd B, fpd is the design value of prestressed steel bar compressive strength; fsd is the design value of ordinary steel bar tensile strength; fcd is the design value of concrete compressive strength, and B represents the mid-span concrete width; the above three strength design values can be found in the bridge design specifications according to the needs of actual projects, and the section bending bearing capacity can be realized by a bearing capacity calculator, which is compiled by a script language.
[0073] S4. Determine the cross-sectional size information corresponding to the mid-span cross-sectional area and the support cross-sectional area according to the optimal mid-span cross-sectional area, input the optimized size information into a pre-trained GA-BP neural network model, use the GA-BP neural network model to analyze the stress curve corresponding to the continuous beam bridge, and optimize the cross-sectional size information in combination with the stress curve to obtain an optimized result.
[0074] The present invention analyzes the stress curve corresponding to the continuous beam bridge by using the GA-BP neural network model, and can find potential stress concentration areas and weak points, and then optimize the design of the continuous beam bridge, thereby improving the optimization accuracy of the continuous beam bridge, wherein the cross-sectional dimension information is the dimension information corresponding to the optimal mid-span cross-sectional area and the optimal support cross-sectional area of the continuous beam bridge, such as cross-sectional width, length and height, etc. The GA-BP neural network model is a model for analyzing the corresponding force curve of the continuous beam bridge when it is subjected to external loads, and the stress curve is the force change curve of the continuous beam bridge. Optionally, the design requirements corresponding to the continuous beam bridge are queried, and the design shape corresponding to the continuous beam bridge is determined according to the design requirements, and the cross-sectional dimension information corresponding to the mid-span section and the support section is determined according to the design shape and the optimal mid-span cross-sectional area and the optimal support cross-sectional area.
[0075] As an embodiment of the present invention, the method of analyzing the stress curve corresponding to the continuous beam bridge using the GA-BP neural network model includes: obtaining standard size information corresponding to the GA-BP neural network model, and using the input layer in the GA-BP neural network model to normalize the optimized size information to obtain target size information, using the convolution layer in the GA-BP neural network model to extract features of the target size information to obtain size information features, using the fitness function in the GA-BP neural network model to calculate the fitness value between the size information features and the standard size information, and using the activation function in the GA-BP neural network model to calculate the bridge stress value corresponding to the continuous beam bridge according to the fitness value and the target size information, and drawing the stress curve corresponding to the continuous beam bridge according to the bridge stress value.
[0076] Among them, the standard size feature is the size feature stored during training in the GA-BP neural network model, the target size information is the information adapted to the GA-BP neural network model obtained after the optimized size information is adjusted by the input layer, the size information feature is the feature information in the target size information, such as size parameters, shape and geometric features, etc., the adaptation value represents the degree of adaptation between the size information feature and the standard size information, and the activation function is a function used to calculate the stress value corresponding to the target size information.
[0077] Optionally, the optimized size information can be normalized by the normalization function in the input layer, and the target size information can be feature extracted by the convolution kernel in the convolution layer. According to the fitness value, the function parameter value in the activation function in the GA-BP neural network model is adjusted, such as the linear value, and the target size information is input into the adjusted activation function to calculate the bridge stress value corresponding to the continuous beam bridge. Drawing the stress curve corresponding to the continuous beam bridge can be achieved by drawing tools, such as Visio drawing tools.
[0078] Optionally, as an optional embodiment of the present invention, the calculating the fitness value between the size information feature and the standard size information by using the fitness function in the GA-BP neural network model includes:
[0079] The fitness function includes:
[0080]
[0081] Among them, Q represents the adaptation value between the size information feature and the standard size information, T represents the number of features of the size information feature, i and i+1 represent the serial numbers of the size information feature and the standard size information respectively, and R i,i+1Indicates the feature distance value corresponding to the i-th size information feature and the i+1-th standard size information, U i represents the feature mapping value corresponding to the i-th size information feature, U i+1 Indicates the feature map value corresponding to the i+1th standard size information.
[0082] The present invention calculates the mid-span cross-sectional area corresponding to the mid-span cross-section according to the mid-span box girder cross-sectional variable, and can obtain the cross-sectional area corresponding to the mid-span cross-section, which is convenient for the subsequent calculation of the number of mid-span reinforcements. The present invention calculates the number of mid-span reinforcements corresponding to the mid-span cross-sectional area according to the mid-span cross-sectional area, and can obtain the number of steel bars required for the mid-span cross-sectional area, which is convenient for the subsequent calculation of the cross-sectional area of the continuous beam bridge. The present invention calculates the optimal mid-span cross-sectional area of the continuous beam bridge by combining the number of mid-span reinforcements and the mid-span cross-sectional area, so as to minimize the material usage of the continuous beam bridge, thereby reducing construction costs and material waste. The present invention analyzes the stress curve corresponding to the continuous beam bridge by using the GA-BP neural network model, and can find potential stress concentration areas and weak parts, and then optimize the design of the continuous beam bridge, thereby improving the optimization accuracy of the continuous beam bridge. Therefore, the optimization method for realizing municipal continuous beam bridges based on deep learning provided by the embodiment of the present invention can improve the optimization accuracy of continuous beam bridges.
[0083] like Figure 2 As shown, it is a functional module diagram of a system for optimizing municipal continuous beam bridges based on deep learning provided by one embodiment of the present invention.
[0084] The optimization system 100 for realizing municipal continuous beam bridge based on deep learning of the present invention can be installed in an electronic device. According to the functions implemented, the optimization system 100 for realizing municipal continuous beam bridge based on deep learning can include a section calculation module 101, a reinforcement quantity calculation module 102, an optimal section calculation module 103 and an optimization analysis module 104. The module of the present invention can also be referred to as a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.
[0085] In this embodiment, the functions of each module / unit are as follows:
[0086] The section calculation module 101 is used to select the mid-span section and the support section of the continuous beam bridge to be optimized, respectively extract the box girder section variables corresponding to the mid-span section and the support section, obtain the mid-span box girder section variable and the support box girder section variable, calculate the mid-span section area corresponding to the mid-span section according to the mid-span box girder section variable, and calculate the support section area corresponding to the support section according to the support box girder section variable;
[0087] The reinforcement quantity calculation module 102 is used to calculate the number of mid-span reinforcements corresponding to the mid-span section according to the mid-span section area, and calculate the number of support reinforcements corresponding to the support section according to the support section area;
[0088] The optimal cross-section calculation module 103 is used to calculate the optimal mid-span cross-section area of the continuous beam bridge by combining the mid-span reinforcement quantity and the mid-span cross-section area, and to calculate the optimal support cross-section area of the continuous beam bridge by combining the support reinforcement quantity and the support cross-section area;
[0089] The optimization analysis module 104 is used to determine the cross-sectional size information corresponding to the mid-span section and the support section according to the optimal mid-span cross-sectional area and the optimal support cross-sectional area, input the optimized size information into a pre-trained GA-BP neural network model, use the GA-BP neural network model to analyze the stress curve corresponding to the continuous beam bridge, and optimize the cross-sectional size information in combination with the stress curve to obtain an optimization result.
[0090] In detail, each module described in the system 100 for optimizing a municipal continuous beam bridge based on deep learning described in the embodiment of the present application is used in the same manner as described above. Figure 1 The technical means described in the article are the same as the optimization method for municipal continuous beam bridges based on deep learning, and can produce the same technical effects, so I will not go into details here.
[0091] In the several embodiments provided by the present invention, it should be understood that the provided methods and systems can be implemented in other ways. For example, the method embodiments described above are only illustrative, for example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation.
[0092] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.
Claims
1. An optimization method for municipal continuous beam bridges based on deep learning. It is characterized in that The method comprises: Selecting a mid-span section and a support section of the continuous beam bridge to be optimized, respectively extracting box girder section variables corresponding to the mid-span section and the support section, obtaining a mid-span box girder section variable and a support box girder section variable, calculating a mid-span section area corresponding to the mid-span section according to the mid-span box girder section variable, and calculating a support section area corresponding to the support section according to the support box girder section variable; The step of calculating the mid-span cross-sectional area corresponding to the mid-span cross-section according to the mid-span box girder cross-sectional variable comprises: The mid-span cross-sectional area corresponding to the mid-span cross-section is calculated by the following formula: Among them, A c represents the mid-span cross-sectional area corresponding to the mid-span section, e a represents the mid-span top plate thickness in the mid-span box girder section variable, b represents the mid-span web plate width in the mid-span box girder section variable, e d represents the mid-span bottom plate thickness in the mid-span box girder section variable, e b represents the mid-span web thickness in the mid-span box girder section variable, h represents the mid-span box girder height in the mid-span box girder section variable, d represents the mid-span flange plate width in the mid-span box girder section variable, e c represents the mid-span root thickness in the mid-span box girder section variable; According to the mid-span cross-sectional area, the number of mid-span reinforcements corresponding to the mid-span cross-section is calculated, and according to the support cross-sectional area, the number of support reinforcements corresponding to the support cross-section is calculated; The optimal mid-span cross-sectional area of the continuous beam bridge is calculated by combining the number of mid-span reinforcement and the mid-span cross-sectional area, and the optimal support cross-sectional area of the continuous beam bridge is calculated by combining the number of support reinforcement and the support cross-sectional area; According to the optimal mid-span cross-sectional area and the optimal support cross-sectional area, the cross-sectional dimension information corresponding to the mid-span cross-section and the support cross-sectional area is determined, and the cross-sectional dimension information is input into a pre-trained GA-BP neural network model. The GA-BP neural network model is used to analyze the stress curve corresponding to the continuous beam bridge, and the cross-sectional dimension information is optimized in combination with the stress curve to obtain an optimization result.
2. The method for optimizing a municipal continuous beam bridge based on deep learning as claimed in claim 1, It is characterized in that Calculating the amount of mid-span reinforcement corresponding to the mid-span section according to the mid-span section area includes: According to the mid-span cross-sectional area, the mid-span upper core distance and the mid-span lower core distance corresponding to the mid-span section are calculated respectively, and the mid-span steel bar eccentricity corresponding to the mid-span section is queried. According to the mid-span steel bar eccentricity, the mid-span upper edge eccentricity and the mid-span lower edge eccentricity corresponding to the mid-span section are calculated respectively. In combination with the mid-span upper edge eccentricity, the mid-span lower edge eccentricity, the mid-span upper core distance and the mid-span lower core distance, the mid-span reinforcement quantity corresponding to the mid-span section is calculated.
3. The method for optimizing a municipal continuous beam bridge based on deep learning as claimed in claim 2, It is characterized in that The calculation of the number of mid-span reinforcement corresponding to the mid-span section by combining the mid-span upper edge eccentricity, the mid-span lower edge eccentricity, the mid-span upper core distance and the mid-span lower core distance comprises: The amount of mid-span reinforcement corresponding to the mid-span section can be calculated using the following formula: Among them, N s Indicates the amount of mid-span reinforcement corresponding to the mid-span section, e s represents the eccentricity of the upper edge of the mid-span, Mmax represents the maximum bending moment of the mid-span section, M min represents the minimum bending moment at the mid-span section, e x represents the eccentricity of the lower edge of the mid-span, k s represents the mid-span upper core distance, k x Indicates the mid-span lower core distance, A p Represents the cross-sectional area of a single steel bar, σ pe Represents the cross-sectional area of a single steel bar.
4. The method for optimizing a municipal continuous beam bridge based on deep learning as claimed in claim 1, It is characterized in that The step of calculating the number of support reinforcements corresponding to the support cross section according to the support cross section area includes: According to the cross-sectional area of the fulcrum, the upper core distance and the lower core distance of the fulcrum corresponding to the fulcrum cross-section are calculated respectively, and the eccentricity of the fulcrum steel bar corresponding to the fulcrum cross-section is queried. According to the eccentricity of the fulcrum steel bar, the upper edge eccentricity and the lower edge eccentricity of the fulcrum corresponding to the fulcrum cross-section are calculated. In combination with the upper edge eccentricity of the fulcrum, the lower edge eccentricity of the fulcrum, the upper core distance of the fulcrum and the lower core distance of the fulcrum, the amount of fulcrum reinforcement corresponding to the fulcrum cross-section is calculated.
5. The method for optimizing a municipal continuous beam bridge based on deep learning as claimed in claim 4, It is characterized in that The calculation of the number of support reinforcements corresponding to the support cross section by combining the upper edge eccentricity of the support, the lower edge eccentricity of the support, the upper core distance of the support and the lower core distance of the support includes: The number of support reinforcements corresponding to the support section is calculated using the following formula: Among them, N y Indicates the number of support reinforcements corresponding to the support section, k y Indicates the core distance on the fulcrum, e y Indicates the eccentricity of the upper edge of the fulcrum, G max Indicates the maximum bending moment of the support section, G min represents the minimum bending moment of the support section, e z Indicates the eccentricity of the lower edge of the fulcrum, k z Indicates the core distance below the fulcrum, A p Represents the cross-sectional area of a single steel bar, σ pe Represents the cross-sectional area of a single steel bar.
6. The method for optimizing a municipal continuous beam bridge based on deep learning as claimed in claim 1, It is characterized in that The step of calculating the optimal mid-span cross-sectional area of the continuous beam bridge by combining the mid-span reinforcement quantity and the mid-span cross-sectional area comprises: According to the number of mid-span reinforcements, the steel bar area corresponding to the mid-span section is calculated to obtain the mid-span steel bar area; according to the mid-span steel bar area, the mid-span concrete area corresponding to the mid-span section is calculated; according to the mid-span concrete area, the concrete width corresponding to the mid-span section is determined to obtain the mid-span concrete width; according to the mid-span concrete width, the mid-span compression height corresponding to the mid-span section is calculated; according to the mid-span compression height and the mid-span concrete width, the section bending bearing capacity corresponding to the mid-span section is calculated; according to the section bending bearing capacity and a preset bearing capacity threshold, the optimal mid-span section area of the continuous beam bridge is screened out from the mid-span section area.
7. The method for optimizing a municipal continuous beam bridge based on deep learning as claimed in claim 1, It is characterized in that The method of analyzing the stress curve corresponding to the continuous beam bridge by using the GA-BP neural network model includes: The standard size information corresponding to the GA-BP neural network model is obtained, and the input layer in the GA-BP neural network model is used to normalize the cross-sectional size information to obtain the target size information, and the convolution layer in the GA-BP neural network model is used to extract the features of the target size information to obtain the size information features, and the fitness function in the GA-BP neural network model is used to calculate the fitness value between the size information features and the standard size information, and the activation function in the GA-BP neural network model is used to calculate the bridge stress value corresponding to the continuous beam bridge according to the fitness value and the target size information, and the stress curve corresponding to the continuous beam bridge is drawn according to the bridge stress value.
8. The method for optimizing a municipal continuous beam bridge based on deep learning as claimed in claim 7, It is characterized in that The calculating the fitness value between the size information feature and the standard size information by using the fitness function in the GA-BP neural network model includes: The fitness function includes: Among them, Q represents the adaptation value between the size information feature and the standard size information, T represents the number of features of the size information feature, i and i+1 represent the serial numbers of the size information feature and the standard size information respectively, and R i,i+1 Indicates the feature distance value corresponding to the i-th size information feature and the i+1-th standard size information, U i Indicates the feature mapping value corresponding to the i-th size information feature, U i+1 Indicates the feature map value corresponding to the i+1th standard size information.
9. An optimization system for municipal continuous beam bridges based on deep learning. It is characterized in that The system comprises: A section calculation module is used to select the mid-span section and the support section of the continuous beam bridge to be optimized, respectively extract the box girder section variables corresponding to the mid-span section and the support section, obtain the mid-span box girder section variable and the support box girder section variable, calculate the mid-span section area corresponding to the mid-span section according to the mid-span box girder section variable, and calculate the support section area corresponding to the support section according to the support box girder section variable; The step of calculating the mid-span cross-sectional area corresponding to the mid-span cross-section according to the mid-span box girder cross-sectional variable comprises: The mid-span cross-sectional area corresponding to the mid-span cross-section is calculated by the following formula: Among them, A c represents the mid-span cross-sectional area corresponding to the mid-span section, e a represents the mid-span top plate thickness in the mid-span box girder section variable, b represents the mid-span web plate width in the mid-span box girder section variable, e d represents the mid-span bottom plate thickness in the mid-span box girder section variable, e b represents the mid-span web thickness in the mid-span box girder section variable, h represents the mid-span box girder height in the mid-span box girder section variable, d represents the mid-span flange plate width in the mid-span box girder section variable, e c represents the mid-span root thickness in the mid-span box girder section variable; A reinforcement quantity calculation module, used to calculate the number of mid-span reinforcements corresponding to the mid-span section according to the mid-span section area, and calculate the number of support reinforcements corresponding to the support section according to the support section area; An optimal cross-section calculation module, used to calculate the optimal mid-span cross-section area of the continuous beam bridge by combining the number of mid-span reinforcement and the mid-span cross-section area, and to calculate the optimal support cross-section area of the continuous beam bridge by combining the number of support reinforcement and the support cross-section area; The optimization analysis module is used to determine the cross-sectional size information corresponding to the mid-span section and the support section according to the optimal mid-span cross-sectional area and the optimal support cross-sectional area, input the cross-sectional size information into a pre-trained GA-BP neural network model, use the GA-BP neural network model to analyze the stress curve corresponding to the continuous beam bridge, and optimize the cross-sectional size information in combination with the stress curve to obtain an optimization result.
