Underground assembly type structure optimization design method and system considering stratum effect

Abstract

The invention relates to the technical field of structure optimization modeling, in particular to an underground assembly type structure optimization design method and system considering the stratigraphic effect. In the method, node numbers in a path are constructed into a topological structure, path recursion is executed based on a rigidity difference value and section parameter combination, and the optimal design of the underground assembly type structure is achieved. A multi-target screening mechanism of a non-dominated sorting genetic algorithm is introduced in the component screening process, evolution convergence processing is carried out on component path combination in combination with three control targets of structural rigidity difference, section layout coordination and path coherence, the coordination of component arrangement and the overall performance stability are improved, and the component screening efficiency is improved. The assembly sequence with the minimum rotation direction is screened through the node coordinate difference value and the component included angle, optimization of angle change control in an assembly path is achieved, grids are divided in combination with component density distribution and a node sudden change area, a joint distribution area is judged based on serial number continuity, joint arrangement is deduced through the overall offset trend, and the joint distribution is optimized. And constructing a corresponding relation between the stratum response and the component combination.

Description

Technical Field

[0001] The present invention relates to the technical field of structural optimization modeling, and in particular to an optimization design method and system for underground prefabricated structures taking stratum effects into consideration. Background Art

[0002] The field of structural optimization modeling technology aims to optimize the parameter configuration, construction form and layout of the structure to achieve the best performance design result by establishing physical simulation models and mathematical optimization models of the structural system while meeting the strength, stiffness, stability, safety and construction constraints.

[0003] An optimization design method for underground prefabricated structures that considers stratum effects is centered around explicitly introducing the influencing factors of stratum effects during the optimization design process. Relying on a coupling model of geological parameters and structural responses, the method identifies the impact of geological factors such as stratum stiffness, foundation nonlinearity, and stratified settlement on the mechanical properties of the structure. This method aims to achieve overall optimization among structural layout, component parameters, and foundation reaction forces while ensuring bearing capacity and stability, thereby reducing the amount of structural materials used and improving the collaborative bearing efficiency of the foundation.

[0004] Existing technologies use separate modeling of structure and foundation. Structural parameter optimization focuses on controlling component internal forces, failing to clearly define the interaction between ground parameters and structural response during the generation phase. Connection structure identification is based on component type classification, ignoring the coupling relationship between path scale and component physical parameters. This makes it impossible to effectively screen for areas with weak displacement release capacity. Furthermore, the lack of dynamic feedback on path stiffness gradient changes leads to sudden local stiffness changes within the path, impacting the coordinated deformation capacity of components. Existing layouts are often executed in a numbered sequence, without screening component angle changes or node connection directions, increasing the risk of interference during on-site assembly. During joint layout, the division method is based on the number of components, failing to consider spatial distribution density and offset mutation trends. This can cause a mismatch between joint locations and deformation areas, limiting overall performance improvements. Summary of the Invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose an optimization design method and system for underground prefabricated structures taking into account stratum effects.

[0006] In order to achieve the above-mentioned object, the present invention adopts the following technical solution: a method for optimizing the design of underground prefabricated structures taking into account the stratum effect, comprising the following steps: S1: Based on the 3D layout of underground prefabricated structural components, an ant colony algorithm is used to retrieve the node component number set, coordinate set and boundary stiffness value, generate the principal axis direction and screen the angle, perform projection analysis of the principal axis direction and the principal value direction of the shear tensor, extract the stress continuous path segment, and establish a coupled shear path network map; S2: Based on the coupled shear stress path network map, extract the path node connection component numbers and connection types, calculate and sort the thickness difference and path length ratios, determine the connection type and mark the hinged components, and output the component connection stability risk component set; S3: Based on the component set with connection stability risk, a non-dominated sorting genetic algorithm is used to extract node numbers and construct a topological graph, recursively deduce component stiffness differences, combine section parameters and arrange paths, determine the stiffness extreme difference and select the optimal one, and establish a coordinated component stiffness layout sequence; S4: Based on the coordinated component stiffness layout sequence, extract the assembly component node numbers, calculate the coordinate difference and angle, screen the out-of-tolerance combinations and reconstruct the sequence, determine the minimum rotation angle sequence, and output a small-angle assembly sequence combination table; S5: Based on the small-angle assembly sequence combination table, extract node and component information, divide the grid area and identify the offset mutation points, select the area with continuous components and dense nodes, and establish the prefabricated structure joint zoning layout results.

[0007] As a further solution of the present invention, the specific steps of generating the coupled shear response path network map are: Based on the 3D layout of underground prefabricated structural components, the node component numbers are matched with the coordinate points one by one, and the coordinate differences of the start and end nodes in the three-axis directions are extracted. The node connection direction vector is calculated using a vector construction method and normalized to generate a node principal axis vector sequence. Based on the node principal axis vector sequence, the angle between the direction vector and the formation gravity gradient vector is calculated, and a critical angle range is set for judgment and screening. The component set structure is established using the information of the node components in the screen, and the output is the main response node component set; Based on the main response node component set, the ant colony algorithm is used to extract the shear stress tensor field value of each node and map it to the component direction vector. Based on the mapping, the continuous shear stress minimum path node sequence is extracted, connected to form a direction segment set, and a coupled shear stress path network map is established.

[0008] As a further solution of the present invention, the ant colony algorithm is based on the formula:

[0009] in: Indicates the starting node number in the current search path. Indicates the target node number in the current search path, Represents the ant colony slave node Move to Node The path cost value, Represents the nodes in the shear stress tensor field To Node Shear stress mapping value along the member direction, Representation node To Node The reciprocal of the cosine of the angle between the direction vector of and the direction vector of the main response component, Representation node With node The normalized difference in thickness of the component, Indicates the principal components of the shear stress tensor at the node With node The ratio of local gradient changes between .

[0010] As a further solution of the present invention, the specific steps of generating the component connection stability risk component set are: Based on the coupled shear path network graph, the path node numbers are extracted and the component numbers are connected according to the topological order, the connection type between the nodes and the components is queried and numbered, the connection properties are classified and sorted, and a path component connection classification table is generated; Based on the path component connection classification table, component cross-sectional parameters are extracted and component path segment lengths are measured. Calculation and sorting are performed based on the difference ratio between the path component cross-sectional thickness value and the path segment length, and the node component combination with the largest range is screened out to generate a high-variance path node combination. Based on the high-variability path node combination, the component connection attribute information is queried to determine whether it is an articulated connection, and the corresponding node component number is marked. The component set structure is combined and the identification field is integrated to output the component connection stability risk component set.

[0011] As a further solution of the present invention, the specific steps of generating the coordinated component stiffness layout sequence are: Based on the component connection stability risk component set, the start and end node coordinates are extracted and numbered, and the adjacent components are indexed in node order by establishing a node connection table, and forward index merging is performed to establish a node topology path diagram; Based on the node topology path graph, the path component number traversal is performed, the cross-section width and height and moment of inertia parameters are extracted, the component stiffness difference calculation is performed through the node connection sequence, and the cumulative summation processing is performed to generate a path component stiffness difference sequence; Based on the stiffness difference sequence of the path components, a non-dominated sorting genetic algorithm is used to screen the path groups with stiffness differences in the lower limit interval, and the serial number comparison and spatial relative position screening of the involved node components are performed to reconstruct the component layout number table and establish a coordinated component stiffness layout sequence.

[0012] As a further solution of the present invention, the non-dominated sorting genetic algorithm is according to the formula:

[0013] in: represents the fitness value of the path component, represents the weight coefficient of the stiffness difference factor, represents the weight coefficient of the layout number offset factor, Represents the weight coefficient of the spatial distance factor, Represents the weight coefficient of the connection angle deviation factor, Represents the weight coefficient of the node normal reaction ratio factor, Indicates the stiffness difference of a component relative to the minimum stiffness component in the path. Indicates the serial number offset value of the component and the connecting component in the layout number table. represents the Euclidean distance between components and connected components in three-dimensional space, Indicates the connection angle deviation factor between the component and the connected component, It represents the ratio of the normal reaction force of the component node to the average reaction force of the path.

[0014] As a further solution of the present invention, the execution sequence number comparison and spatial relative position screening are performed, and the component arrangement number table is reconstructed. The components included in the screened path are compared according to the node number order, and the component pairs with discontinuous numbers and component connections that do not meet the requirements of the unidirectional topological path are screened out. On the basis of the comparison, the unit vector of the main axis direction in the three-dimensional rectangular coordinate system is calculated according to the node space coordinates of each component, and the components are sorted and screened according to the angle between the unit vector and the specified direction vector. The screening condition is that the angle is less than 45°, and the screened components are renumbered in ascending order according to the starting node in the Z-axis coordinate direction.

[0015] As a further solution of the present invention, the specific steps of generating the small-angle assembly sequence combination table are: Based on the coordinated component stiffness layout sequence, the node numbers at both ends of all components are extracted and the three-dimensional coordinates of the nodes are matched in the order of the numbers. The numerical differences between the coordinate dimensions are calculated and merged into a node difference set. The vector modulus values are extracted to generate a component spatial displacement data table. Based on the component spatial displacement data table, a node connection vector is constructed through three-dimensional coordinate differences and a vector angle value is calculated. Component number combinations are screened based on angles greater than a threshold value and the number sequence is rearranged. Component direction change marks are matched to establish an assembly deviation reconstruction sequence set. Based on the assembly deviation reconstruction sequence set, the three-dimensional coordinates of each pair of nodes are extracted to calculate the rotation axis and the reference angle value, the rotation angle sequence between the components is determined and the component numbering sequence is rearranged, the minimum angle combination component set is extracted, and a small angle assembly sequence combination table is output.

[0016] As a further solution of the present invention, the specific steps of generating the partition layout result of the prefabricated structure joint are: Based on the small-angle assembly sequence combination table, node numbers and three-dimensional coordinate sets are extracted, coordinate axis direction range segmentation processing is performed and a three-dimensional coordinate grid index table is generated, a mapping relationship between node numbers and index positions is constructed, and an assembly node grid partition information table is generated; Based on the assembly node grid partition information table, extracting each region node set according to the grid index, calculating the difference in each coordinate direction between adjacent nodes and extracting the difference mutation pairs, screening the mutation point number list and merging the boundary node set to generate the node offset mutation region set; Based on the node offset mutation area set, the component information corresponding to the mutation node number is extracted and the node continuous numbering is judged, the component number density value is calculated and the spatial position group is matched, the joint insertion node set is constructed, and the prefabricated structure joint partition layout result is established.

[0017] A system for optimizing the design of underground prefabricated structures taking into account stratum effects is provided. The system is used to execute the above-mentioned method for optimizing the design of underground prefabricated structures taking into account stratum effects. The system comprises: Shear stress path identification module: Calls component number, node coordinates and boundary stiffness value, calculates the angle between the main axis direction and the shear stress tensor, selects component paths with angles less than 30 degrees, calls the shear stress path construction solution based on ant colony algorithm to execute path generation, and establishes the shear stress path network map; Connection risk extraction module: Based on the shear stress path network map, extract the ratio of component thickness difference to path length, determine whether the connection type is hinged, filter component pair numbers and output the connection stability risk component set; Component stiffness coordination optimization module: Based on the connection stability risk component set, the inertia moment is extracted and the difference sequence is calculated. According to the component combination rules set by the component stiffness coordination optimization scheme based on the non-dominated sorting genetic algorithm, the component path screening operation is performed to generate the component stiffness coordination path sequence; Assembly sequence generation module: Based on the component stiffness coordination path sequence, calculate the component node angles and sort them, establish a rotation direction chain and remove duplicate combinations, and output the component rotation assembly sequence table; Joint partition construction module: Based on the component rotation assembly sequence table, the coordinate and number information is extracted, the grid is divided and the node mutation points are located, the inversion optimization scheme based on the coupling mapping of equivalent stiffness and foundation reaction force is called to perform reaction force matching screening, and the prefabricated structure joint partition pattern is established.

[0018] Compared with the prior art, the advantages and positive effects of the present invention are: In the present invention, by retrieving the node component number, coordinate set and boundary stiffness value, performing angle analysis between the principal axis direction and the principal value direction of the shear stress tensor and screening the continuous component path segments, the shear stress path network is established in combination with the ant colony mechanism to realize active identification of stress continuity. In this invention, the node numbers within the path are constructed into a topological structure, and path recursion is performed based on the combination of stiffness difference and section parameters. In the component screening process, a multi-objective screening mechanism of the non-dominated sorting genetic algorithm is introduced. Combining the three control objectives of structural stiffness difference, section layout coordination, and path coherence, the component path combination is subjected to evolutionary convergence processing to improve the coordination of component layout and overall performance stability. In this method, the assembly sequence with the smallest rotation direction is selected by the node coordinate difference and the component angle, which optimizes the control of angle changes in the assembly path. The grid is divided based on the component density distribution and the node mutation area. The joint distribution area is determined based on the number continuity. The joint layout is deduced from the overall offset trend, and the corresponding relationship between the formation response and the component combination is established. In the present invention, continuous principal direction shear stress analysis and path iterative control based on the heuristic mechanism are introduced into the path generation method, which enhances the coverage ability of the path for the main control deformation direction. A joint judgment criterion of thickness difference and path length is established in connection identification to improve the exposure range of local unstable sections. Multi-objective solution set screening and cross-section combination strategy are introduced in component sorting to optimize the overall layout of components. The assembly process improves the directional coordination of node docking through dual screening of angle difference and rotation sequence. Joint generation is based on density distribution within the grid and identification of spatial mutation points, which enhances the consistency between the joint area and the deformation control area. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a schematic diagram of the workflow of the present invention; Figure 2 It is a system flow chart of the present invention. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0021] In the description of the present invention, it should be understood that the terms "length," "width," "up," "down," "front," "back," "left," "right," "vertical," "horizontal," "top," "bottom," "inside," "outside," and the like, indicating positions or relationships, are based on the positions or relationships shown in the accompanying drawings and are intended only to facilitate the description of the present invention and simplify the description. They do not indicate or imply that the devices or elements referred to must have a specific orientation, be constructed, or operate in a specific orientation. Therefore, they should not be construed as limiting the present invention. Furthermore, in the description of the present invention, "plurality" means two or more, unless otherwise expressly and specifically defined.

[0022] Example 1 See also Figure 1 The present invention provides a technical solution: an optimization design method for underground prefabricated structures considering stratum effects, comprising the following steps: S1: Based on the 3D layout of underground prefabricated structural components, an ant colony algorithm is used to retrieve the node component number set, coordinate set and boundary stiffness value, generate the principal axis direction and screen the angle, perform projection analysis of the principal axis direction and the principal value direction of the shear tensor, extract the stress continuous path segment, and establish a coupled shear path network map; S2: Based on the coupled shear stress path network map, the path node connection component numbers and connection types are extracted, the thickness difference and path length ratio are calculated and sorted, the connection type is determined and the hinged components are marked, and the component connection stability risk set is output; S3: Based on the component set with connection stability risk, a non-dominated sorting genetic algorithm is used to extract node numbers and construct a topological map. The component stiffness difference is recursively calculated, the cross-sectional parameters are combined and the paths are arranged. The extreme stiffness difference is determined and the optimal selection is made to establish a coordinated component stiffness layout sequence. S4: Based on the coordinated component stiffness layout sequence, extract the assembly component node numbers, calculate the coordinate difference and angle, screen the out-of-tolerance combinations and reconstruct the sequence, determine the minimum rotation angle sequence, and output the small-angle assembly sequence combination table; S5: Based on the small-angle assembly sequence combination table, extract node and component information, divide the grid area and identify the offset mutation points, select areas with continuous components and dense nodes, and establish the prefabricated structure joint zoning layout results.

[0023] The specific steps to generate the coupled shear stress path network map are: Based on the 3D layout of underground prefabricated structural components, the node component numbers are matched with the coordinate points one by one, and the coordinate differences of the start and end nodes in the three-axis directions are extracted. The node connection direction vector is calculated using a vector construction method and normalized to generate a node principal axis vector sequence. Based on the node principal axis vector sequence, the angle between the direction vector and the formation gravity gradient vector is calculated, and a critical angle range is set for judgment and screening. The component set structure is established using the node component information in the screen, and the output is the main response node component set; Based on the main response node component set, the ant colony algorithm is used to extract the shear stress tensor field value of each node and map it to the component direction vector. Based on the mapping, the continuous shear stress minimum path node sequence is extracted and connected to form a direction segment set to establish a coupled shear stress path network map. Based on the three-dimensional layout diagram of underground prefabricated structural components, the start and end node numbers of all components are extracted and matched one by one with the node coordinates. The three-axis coordinate differences of each pair of nodes in the horizontal and vertical directions are extracted. Using the vector calculation method, the coordinate differences are sequentially merged into a set of direction quantities. According to the sum of the coordinate differences of the direction quantities of each component, the length of the opposite vector is calculated and scaled according to the length. The length of each vector is proportionally adjusted to unity and unified into a unit length vector to generate a node principal axis vector sequence. Based on the node principal axis vector sequence, the angle between each direction vector and the formation gravity gradient direction is compared. The direction angle screening method is used to measure the angle between the direction vector and the gradient direction. The angle judgment interval threshold is set between 30 and 60 degrees. The component number and node number corresponding to the direction vector within the range are recorded, and the component number index set is sorted and output as the main response node component set. Based on the set of main response node components, an ant colony algorithm is used to construct an initial information table containing node numbers, node coordinates, and principal values of the shear stress tensor. The information update parameters are set to the pheromone initial value of 1 and the pheromone attenuation value of 0.3. The total number of ant individuals in the path search process is 50, and the number of path search cycles is set to 100. When each ant individual selects the next hop node for the current position node, an ant colony search method based on pheromone intensity and direction matching factor is adopted. The direction matching factor is calculated by the angle between the current path direction vector and the direction vector of the candidate connection component. The direction matching factor is then combined with the principal value of the shear stress tensor corresponding to the current node as the stress response indicator. The two values are combined in a weighted proportion to calculate the path attraction score. The score is used as the input factor for the path selection probability to perform the selection operation. The next node direction is determined according to the set rules. Continuous path registration is performed for all node connection components in the path. After the path registration is completed, the information of all component connection segments is recorded in the order of the direction vector. A component path segment set is established, and finally a coupled shear response path network map is formed.

[0024] Ant colony algorithm, according to the formula:

[0025] in: Indicates the starting node number in the current search path. Indicates the target node number in the current search path, Represents the ant colony slave node Move to Node The path cost value, Represents the nodes in the shear stress tensor field To Node Shear stress mapping value along the member direction, Representation node To Node The reciprocal of the cosine of the angle between the direction vector of and the direction vector of the main response component, Representation node With node The normalized difference in thickness of the component, Indicates the principal components of the shear stress tensor at the node With node The ratio of local gradient changes between Execution process: First, extract the direction vector of the main response component and With node Obtain the principal components of the shear stress tensor at the location, and then map the principal components to the component direction as the shear stress mapping value , according to the node To Node The angle between the connecting direction and the main component direction, calculate the cosine value and take the reciprocal as the direction consistency factor , then, read the node from the component design parameters With node The thickness value of the component to which it belongs, calculate its normalized difference as the thickness adjustment factor , which is used to reflect the influence of the variability of the stratum thickness at the component interface on the path stability. Then, by extracting the equidistant differential points around the node, the stress gradient change rate is calculated using the differential results of the shear stress principal value in the node neighborhood as the shear stress change factor , which is used to measure the degree of interference of formation disturbance on path continuity. Finally, the above four parameters are integrated and applied to the path evaluation system to guide the ant colony algorithm to perform path construction and global optimization process under complex formations and component structures.

[0026] The specific steps for generating component connection stability risk component set are: Based on the coupled shear path network graph, the path node numbers are extracted and the component numbers are connected according to the topological order. The connection types between nodes and components are queried and numbered. The connection properties are classified and sorted to generate a path component connection classification table. Based on the path component connection classification table, the component cross-section parameters are extracted and the component path segment length is measured. The difference ratio between the path component cross-section thickness value and the path segment length is calculated and sorted to screen out the node component combination with the largest range and generate the high-variance path node combination. Based on the high-variability path node combination, the component connection attribute information is queried to determine whether it is a hinged connection, and the corresponding node component number is marked. The component set structure is combined and the identification field is integrated to output the component connection stability risk component set; Based on the coupled shear path network graph, the node number and the connected component number of each path are extracted. The path topology sorting method is used, and the starting point of the path connection is set as the node with the lowest number. All nodes are arranged in ascending order of number. The connected component numbers are retrieved one by one and the component connection sequence is reconstructed. The component attribute information table is called to perform the connection type identification operation on each component number in turn. The connection form is queried according to the connection structure field of the node at both ends of the component. The connection attribute label is recorded for the components identified as rigid or hinged. The component number, start and end node numbers and connection type identifier are integrated to generate a path component connection classification table. Based on the path component connection classification table, a cross-section and length deviation sorting method is used to extract the cross-section thickness value and path segment length value of each component in the path. All components are paired and arranged in the order of path numbers. The ratio of the thickness difference between adjacent components to the path segment length difference is calculated. The ratio sequence is sorted and the top 5% of the component number combinations in the ratio sorting results are selected. The node index of the component combination is recorded by number and output in ascending order of number to generate high-variance path node combinations. Based on the high-variability path node combination, the connection attribute matching and identification method is adopted to retrieve the connection attribute label field registered for the corresponding component number in the path component connection classification table. The component number with the attribute value equal to the hinge is marked, and the node numbers corresponding to all marked components are combined into a set structure. The component number, node number and connection attribute fields are written into the set, and the components are integrated into a structural form in ascending order of component numbers to output the component connection stability risk component set.

[0027] The specific steps to generate a coordinated component stiffness layout sequence are: Based on the component set with component connection stability risk, the coordinates of the start and end nodes are extracted and numbered. By establishing a node connection table, adjacent components are indexed in node order, and forward indexes are merged to establish a node topology path diagram. Based on the node topology path graph, the path component number traversal is performed, the cross-section width and height and moment of inertia parameters are extracted, the component stiffness difference calculation is performed through the node connection sequence, and the cumulative summation is performed to generate the path component stiffness difference sequence; Based on the stiffness difference sequence of path components, a non-dominated sorting genetic algorithm is used to screen the path groups with stiffness differences in the lower limit interval. The sequence numbers of the involved node components are compared and the spatial relative positions are screened. The component layout number table is reconstructed to establish a coordinated component stiffness layout sequence. Based on the component connection stability risk component set, the start and end node numbers associated with the components are extracted, and the corresponding node three-dimensional coordinate values are extracted. Using the node connection table construction method, all node numbers are sorted in ascending order. The corresponding connection component number of each node is searched in turn and a connection index relationship is established. The component number connection matrix is filled in order, and pairing operations are performed on adjacent components in order of node numbers. A forward merge is performed on the component index in the node list, and the component path records are superimposed in reverse order according to the direction of the smallest node number. The component number sequence group with direct connection relationships between all nodes is output, and a node topology path diagram is established. Based on the node topology path diagram, a component path parameter recursion method is used to traverse all component numbers in the path. The cross-sectional width, cross-sectional height, and moment of inertia parameter values of each component are extracted, and a path arrangement list is constructed in numerical order. The moment of inertia difference calculation operation is performed on each pair of adjacent components. The moment of inertia value of the previous component is subtracted from the moment of inertia value of the next component and the absolute value of the difference is recorded. All differences are sequentially accumulated to form a difference accumulation vector sequence structure, generating a path component stiffness difference sequence. Based on the stiffness difference sequence of path components, a non-dominated sorting genetic algorithm is adopted. The objective function is set as the minimum stiffness difference of path components. The constraints are the consistency of node numbering sequence and the compactness constraint of path space. The population size is set to 100, the number of iterations is 150, the crossover probability is 0.8, and the mutation probability is 0.15. A chromosome structure containing four parameters: component number, node number, section width, and section height is constructed. After population initialization, selection and crossover operations are performed. In each round of iteration, sorting is used to determine whether the current solution is in the non-inferior frontier of the target space, and elite individuals are retained to form a new generation of population. Component number groups that ultimately meet the stiffness difference in the lower limit interval are screened, and then the component sequence is judged for consistency, and the spatial coordinate difference of the corresponding node is calculated. Component combinations with spatial differences less than the set threshold are clustered and renumbered and sorted. The numbering matrix is reconstructed and the component arrangement relationship is listed in order to establish a coordinated component stiffness layout sequence.

[0028] Non-dominated sorting genetic algorithm, according to the formula:

[0029] in: represents the fitness value of the path component, represents the weight coefficient of the stiffness difference factor, represents the weight coefficient of the layout number offset factor, Represents the weight coefficient of the spatial distance factor, Represents the weight coefficient of the connection angle deviation factor, Represents the weight coefficient of the node normal reaction ratio factor, Indicates the stiffness difference of a component relative to the minimum stiffness component in the path. Indicates the serial number offset value of the component and the connecting component in the layout number table. represents the Euclidean distance between components and connected components in three-dimensional space, Indicates the connection angle deviation factor between the component and the connected component, It represents the ratio of the normal reaction force at the component node to the average reaction force along the path; Execution process: First, extract the equivalent stiffness value of each component and calculate its difference with the minimum stiffness component in the path to obtain the stiffness difference term Then, obtain the number information of the target component and its connected components from the component arrangement number table, and calculate the number difference to obtain , then based on the three-dimensional space coordinate model, extract the coordinates of the target component and the starting and ending nodes of the connected component, and calculate the spatial Euclidean distance as Then, the angle between the component axis and the connection component direction is obtained and the angle is normalized to the dimensionless deviation factor , which is used to measure the degree of layout broken line. Subsequently, the normal reaction force value of the component node is extracted under the simulation of stratum action, and the ratio operation is performed with the average normal reaction force of the component node in the path to obtain the mechanical factor reflecting the stability of the component under stratum constraint. , and finally multiply by the weight coefficient to , weighted sum to generate component fitness value , as the core basis for the selection operation in the non-dominated sorting genetic algorithm, is used to screen out the component path combination with the best comprehensive performance in terms of stiffness coordination, layout order, spatial continuity, structural turning degree and formation stability.

[0030] Perform serial number comparison and spatial relative position screening, reconstruct the component layout number table, and compare the components included in the screened path according to the node number sequence. Filter out component pairs with discontinuous numbers and component connections that do not meet the requirements of a unidirectional topological path. Based on the comparison, calculate the unit vector of the main axis direction in the three-dimensional rectangular coordinate system according to the node space coordinates of each component. Sorting and screening are performed based on the angle between the unit vector and the specified direction vector. The screening condition is that the angle is less than 45°, and the screened components are renumbered in ascending order according to the starting node in the Z-axis coordinate direction.

[0031] The specific steps to generate the small corner assembly sequence combination table are: Based on the coordinated component stiffness layout sequence, the node numbers at both ends of all components are extracted and the three-dimensional coordinates of the nodes are matched in the order of the numbers. The numerical differences between the coordinate dimensions are calculated and merged into a node difference set. The vector modulus values are extracted to generate a component spatial displacement data table. Based on the component spatial displacement data table, the node connection vector is constructed through the three-dimensional coordinate difference and the vector angle value is calculated. The component number combinations are filtered according to the angle greater than the threshold and the number sequence is rearranged. The component direction change marks are matched to establish the assembly deviation reconstruction sequence set. Based on the assembly deviation, the sequence set is reconstructed, the three-dimensional coordinates of each pair of nodes are extracted to calculate the rotation axis and reference angle value, the rotation angle sequence between components is determined and the component numbering sequence is rearranged, the minimum angle combination component set is extracted, and a small angle assembly sequence combination table is output; Based on the coordinated component stiffness layout sequence, the start and end node numbers of all components are extracted, and the three-dimensional coordinates of the nodes are matched in ascending order of the numbers. The three-axis direction difference of each pair of start and end node coordinates is calculated and merged into a three-element numerical group. The modulus calculation method is used to calculate the modulus by squaring the three-axis difference and then adding the square root to perform the modulus calculation. The calculated vector modulus of each component is written into the table field to generate the component spatial displacement data table; Based on the component spatial displacement data table, a three-dimensional vector angle screening method is used to construct the coordinates of the start and end nodes of each component into a spatial vector structure. The angle between two vector segments is calculated continuously according to the component number. The angle threshold is set to 45 degrees. The component number groups with angle values greater than the threshold are identified based on the angle value. The component groups that meet the conditions are reordered, and the original component direction changes are set as angle change mark fields and written into the new sequence to establish the assembly deviation reconstruction sequence set. Based on the assembly deviation reconstruction sequence set, the component rotation angle sorting method is adopted to extract the three-dimensional coordinates of the start and end nodes of each pair of components. The rotation axis fitting operation is performed on the component direction vector, and the direction of the three-axis coordinate change is constructed as a scalar vector in the rotation axis direction. The reference vector is set as the vertical unit axis. The angle between the rotation axis and the reference vector is calculated according to the component direction vector. The angle values are sorted according to the numbering sequence and the component number with the minimum angle is marked. The component numbers of the smallest angle combination are aggregated and reordered, and a small-angle assembly sequence combination table is output.

[0032] The specific steps for generating the partition layout results of prefabricated structure joints are as follows: Based on the small-angle assembly sequence combination table, the node numbers and three-dimensional coordinate sets are extracted, the coordinate axis direction range is segmented and a three-dimensional coordinate grid index table is generated. The mapping relationship between the node numbers and index positions is constructed, and the assembly node grid partition information table is generated. Based on the assembly node grid partition information table, extract the node set of each region according to the grid index, calculate the difference in each coordinate direction between adjacent nodes and extract the difference mutation pairs, filter out the mutation point number list and merge the boundary node set to generate the node offset mutation region set; Based on the node offset mutation area set, the component information corresponding to the mutation node number is extracted and the node continuous number is judged. The component number density value is calculated and matched with the spatial position group. The joint insertion node set is constructed to establish the prefabricated structure joint partition layout result. Based on the small-angle assembly sequence combination table, all component node numbers and their three-dimensional coordinate sets are extracted. Using the coordinate axis segmented grid index method, all nodes are set to a segment interval length of one in the X, Y, and Z directions. Each coordinate value is divided by the interval and rounded down to obtain the grid coordinate index value. The X, Y, and Z index values of each node are spliced into a triple key-value structure. The node number and the triple index value are constructed as a mapping key-value pair structure and written into the table. The node numbers are sorted in ascending order and written into the record field to generate the assembly node grid partition information table. Based on the assembly node grid partition information table, a node coordinate difference mutation detection method is adopted to extract the node sets corresponding to each triple grid index one by one. In each set, the difference calculation operation between any two node coordinates in the X, Y, and Z directions is performed. The difference sequence is sorted by absolute value and the point pair numbers exceeding the threshold are recorded. The difference mutation threshold is set to three. In each set, the node numbers that meet the conditions are extracted and added to the mutation point list. The node number sets of each grid boundary are extracted according to the index boundary value and merged into the list. The total number set structure is merged to generate the node offset mutation area set. Based on the node offset mutation area set, the component density mapping and number connectivity judgment method are adopted to extract the component number set connected by the mutation node number. For each component number, it is checked whether the corresponding node is a continuous structure with the previous and next numbers. A combination identifier is formed for the continuously numbered components. The start and end node coordinate ranges of each component in three-dimensional space are extracted and the component space bounding box structure is constructed. The number of overlapping areas at the overlapping positions of all component spatial ranges is counted, and the component density value is assigned according to the counting result. The component numbers with density values exceeding the set number threshold are marked as high-density groups. The spatial overlapping components and node combinations are recorded and matched in ascending order of number. All node numbers in the combination are extracted as joints and inserted into the candidate node set to establish the zoning layout result of the prefabricated structure joints.

[0033] See also Figure 2 , an optimization design system for underground prefabricated structures considering stratum effects, the system includes: Shear stress path identification module: Calls component number, node coordinates and boundary stiffness value, calculates the angle between the main axis direction and the shear stress tensor, selects component paths with angles less than 30 degrees, calls the shear stress path construction solution based on ant colony algorithm to execute path generation, and establishes the shear stress path network map; Connection risk extraction module: Based on the shear stress path network map, it extracts the ratio of component thickness difference to path length, determines whether the connection type is hinged, filters the component pair numbers, and outputs the connection stability risk component set; Component stiffness coordination optimization module: Based on the set of connection stability risk components, the module extracts the moment of inertia and calculates the difference sequence. Based on the component combination rules set by the component stiffness coordination optimization scheme based on the non-dominated sorting genetic algorithm, the module performs component path screening operations and generates a component stiffness coordination path sequence. Assembly sequence generation module: Based on the component stiffness coordination path sequence, it calculates and sorts the component node angles, establishes a rotation direction chain and removes duplicate combinations, and outputs a component rotation assembly sequence table; Joint partition construction module: Based on the component rotation and assembly sequence table, the coordinate and number information is extracted, the grid is divided and the node mutation points are located. The inverse optimization scheme based on the coupling mapping of equivalent stiffness and foundation reaction force is called to perform reaction matching screening and establish the prefabricated structure joint partition pattern.

[0034] The above are merely preferred embodiments of the present invention and do not limit the present invention in any other form. Any technician familiar with the profession may use the technical content disclosed above to change or modify it into an equivalent embodiment with equivalent changes and apply it to other fields. However, any simple modification, equivalent change and modification made to the above embodiment based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall still fall within the scope of protection of the technical solution of the present invention.

Claims

1. A method for optimizing the design of underground prefabricated structures taking into account stratum effects, characterized in that: The following steps are involved: S1: Based on the 3D layout of underground prefabricated structural components, an ant colony algorithm is used to retrieve the node component number set, coordinate set and boundary stiffness value, generate the principal axis direction and screen the angle, perform projection analysis of the principal axis direction and the principal value direction of the shear tensor, extract the stress continuous path segment, and establish a coupled shear path network map; S2: Based on the coupled shear stress path network map, extract the path node connection component numbers and connection types, calculate and sort the thickness difference and path length ratios, determine the connection type and mark the hinged components, and output the component connection stability risk component set; S3: Based on the component set with connection stability risk, a non-dominated sorting genetic algorithm is used to extract node numbers and construct a topological graph, recursively deduce component stiffness differences, combine section parameters and arrange paths, determine the stiffness extreme difference and select the optimal one, and establish a coordinated component stiffness layout sequence; S4: Based on the coordinated component stiffness layout sequence, extract the assembly component node numbers, calculate the coordinate difference and angle, screen the out-of-tolerance combinations and reconstruct the sequence, determine the minimum rotation angle sequence, and output a small-angle assembly sequence combination table; S5: Based on the small-angle assembly sequence combination table, extract node and component information, divide the grid area and identify the offset mutation points, select the area with continuous components and dense nodes, and establish the prefabricated structure joint zoning layout results.

2. The underground prefabricated structure optimization design method considering stratum effect according to claim 1 is characterized in that: The specific steps of generating the coupled shear stress path network map are as follows: Based on the 3D layout of underground prefabricated structural components, the node component numbers are matched with the coordinate points one by one, and the coordinate differences of the start and end nodes in the three-axis directions are extracted. The node connection direction vector is calculated using a vector construction method and normalized to generate a node principal axis vector sequence. Based on the node principal axis vector sequence, the angle between the direction vector and the formation gravity gradient vector is calculated, and a critical angle range is set for judgment and screening. The component set structure is established using the information of the node components in the screen, and the output is the main response node component set; Based on the main response node component set, the ant colony algorithm is used to extract the shear stress tensor field value of each node and map it to the component direction vector. Based on the mapping, the continuous shear stress minimum path node sequence is extracted, connected to form a direction segment set, and a coupled shear stress path network map is established.

3. The underground prefabricated structure optimization design method considering stratum effect according to claim 1 is characterized in that: The ant colony algorithm is based on the formula: ， in: Indicates the starting node number in the current search path. Indicates the target node number in the current search path, Represents the ant colony slave node Move to Node The path cost value, Represents the nodes in the shear stress tensor field To Node Shear stress mapping value along the member direction, Representation node To Node The reciprocal of the cosine of the angle between the direction vector of and the direction vector of the main response component, Representation node With node The normalized difference in thickness of the component, Indicates the principal components of the shear stress tensor at the node With node The ratio of local gradient changes between .

4. The underground prefabricated structure optimization design method considering stratum effect according to claim 1 is characterized in that: The specific steps of generating the component connection stability risk component set are: Based on the coupled shear path network graph, the path node numbers are extracted and the component numbers are connected according to the topological order, the connection type between the nodes and the components is queried and numbered, the connection properties are classified and sorted, and a path component connection classification table is generated; Based on the path component connection classification table, component cross-sectional parameters are extracted and component path segment lengths are measured. Calculation and sorting are performed based on the difference ratio between the path component cross-sectional thickness value and the path segment length, and the node component combination with the largest range is screened out to generate a high-variance path node combination. Based on the high-variability path node combination, the component connection attribute information is queried to determine whether it is an articulated connection, and the corresponding node component number is marked. The component set structure is combined and the identification field is integrated to output the component connection stability risk component set.

5. The underground prefabricated structure optimization design method considering stratum effect according to claim 1 is characterized in that: The specific steps of generating the coordinated component stiffness layout sequence are: Based on the component connection stability risk component set, the start and end node coordinates are extracted and numbered, and the adjacent components are indexed in node order by establishing a node connection table, and forward index merging is performed to establish a node topology path diagram; Based on the node topology path graph, the path component number traversal is performed, the cross-section width and height and moment of inertia parameters are extracted, the component stiffness difference calculation is performed through the node connection sequence, and the cumulative summation processing is performed to generate a path component stiffness difference sequence; Based on the stiffness difference sequence of the path components, a non-dominated sorting genetic algorithm is used to screen the path groups with stiffness differences in the lower limit interval. The serial number comparison and spatial relative position screening of the involved node components are performed, the component layout number table is reconstructed, and a coordinated component stiffness layout sequence is established.

6. The underground prefabricated structure optimization design method considering stratum effect according to claim 1 is characterized in that: The non-dominated sorting genetic algorithm is based on the formula: ， in: represents the fitness value of the path component, represents the weight coefficient of the stiffness difference factor, represents the weight coefficient of the layout number offset factor, Represents the weight coefficient of the spatial distance factor, Represents the weight coefficient of the connection angle deviation factor, Represents the weight coefficient of the node normal reaction ratio factor, Indicates the stiffness difference of a component relative to the minimum stiffness component in the path. Indicates the serial number offset value of the component and the connecting component in the layout number table. represents the Euclidean distance between components and connected components in three-dimensional space, Indicates the connection angle deviation factor between the component and the connected component, It represents the ratio of the normal reaction force of the component node to the average reaction force of the path.

7. The underground prefabricated structure optimization design method considering stratum effect according to claim 1 is characterized in that: The execution sequence number comparison and spatial relative position screening are performed to reconstruct the component arrangement number table. The components included in the screened path are compared according to the node number sequence, and component pairs with discontinuous numbers and component connections that do not meet the requirements of a unidirectional topological path are screened out. Based on the comparison, the unit vector of the main axis direction in the three-dimensional rectangular coordinate system is calculated according to the node space coordinates of each component. The components are sorted and screened according to the angle between the unit vector and the specified direction vector. The screening condition is that the angle is less than 45°, and the screened components are renumbered in ascending order according to the starting node in the Z-axis coordinate direction.

8. The underground prefabricated structure optimization design method considering stratum effect according to claim 1 is characterized in that: The specific steps for generating the small corner assembly sequence combination table are: Based on the coordinated component stiffness layout sequence, the node numbers at both ends of all components are extracted and the three-dimensional coordinates of the nodes are matched in the order of the numbers. The numerical differences between the coordinate dimensions are calculated and merged into a node difference set. The vector modulus values are extracted to generate a component spatial displacement data table. Based on the component spatial displacement data table, node connection vectors are constructed through three-dimensional coordinate differences and vector angle values are calculated. Component number combinations are screened based on angles greater than a threshold and the number sequence is rearranged. Component direction change marks are matched to establish an assembly deviation reconstruction sequence set. Based on the assembly deviation reconstruction sequence set, the three-dimensional coordinates of each pair of nodes are extracted to calculate the rotation axis and the reference angle value, the rotation angle sequence between the components is determined and the component numbering sequence is rearranged, the minimum angle combination component set is extracted, and a small angle assembly sequence combination table is output.

9. The underground prefabricated structure optimization design method considering stratum effect according to claim 1 is characterized in that: The specific steps for generating the partition layout result of the prefabricated structure joint are as follows: Based on the small-angle assembly sequence combination table, node numbers and three-dimensional coordinate sets are extracted, coordinate axis direction range segmentation processing is performed and a three-dimensional coordinate grid index table is generated, a mapping relationship between node numbers and index positions is constructed, and an assembly node grid partition information table is generated; Based on the assembly node grid partition information table, extracting each region node set according to the grid index, calculating the difference in each coordinate direction between adjacent nodes and extracting the difference mutation pairs, screening out the mutation point number list and merging the boundary node set to generate the node offset mutation region set; Based on the node offset mutation area set, the component information corresponding to the mutation node number is extracted and the node continuous numbering is judged, the component number density value is calculated and the spatial position group is matched, the joint insertion node set is constructed, and the prefabricated structure joint partition layout result is established.

10. An optimization design system for underground prefabricated structures taking into account stratum effects, characterized in that: According to any one of claims 1 to 9, the method for optimizing the design of underground prefabricated structures taking into account stratum effects comprises: Shear stress path identification module: Calls component number, node coordinates and boundary stiffness value, calculates the angle between the main axis direction and the shear stress tensor, selects component paths with angles less than 30 degrees, calls the shear stress path construction solution based on ant colony algorithm to execute path generation, and establishes the shear stress path network map; Connection risk extraction module: Based on the shear stress path network map, extract the ratio of component thickness difference to path length, determine whether the connection type is hinged, filter component pair numbers and output the connection stability risk component set; Component stiffness coordination optimization module: Based on the connection stability risk component set, the inertia moment is extracted and the difference sequence is calculated. According to the component combination rules set by the component stiffness coordination optimization scheme based on the non-dominated sorting genetic algorithm, the component path screening operation is performed to generate the component stiffness coordination path sequence; Assembly sequence generation module: Based on the component stiffness coordination path sequence, calculate the component node angles and sort them, establish a rotation direction chain and remove duplicate combinations, and output the component rotation assembly sequence table; Joint partition construction module: Based on the component rotation assembly sequence table, the coordinate and number information is extracted, the grid is divided and the node mutation points are located, the inversion optimization scheme based on the coupling mapping of equivalent stiffness and foundation reaction force is called to perform reaction force matching screening, and the prefabricated structure joint partition pattern is established.

Images